Package Details: google-earth 7.1.4.1529-1

Git Clone URL: https://aur.archlinux.org/google-earth.git (read-only)
Package Base: google-earth
Description: A 3D interface to view satellite images of Earth and other objects
Upstream URL: https://www.google.com/earth/index.html
Licenses: custom:earth
Submitter: None
Maintainer: Det
Last Packager: Det
Votes: 971
Popularity: 5.942508
First Submitted: 2008-02-19 07:36
Last Updated: 2016-04-12 13:31

Pinned Comments

Det commented on 2016-04-12 13:31

Please DO NOT flag out-of-date, unless there's a new version for Linux also:

$ curl -s https://dl.google.com/dl/earth/client/current/google-earth-stable_current_x86_64.rpm | head -c96 | strings | cut -d "-" -f4

Det commented on 2016-02-24 18:38

I got right on that. Evidence: https://aur.archlinux.org/packages/google-earth-pro-wine/

aurelieng commented on 2015-01-31 06:32

Anyone interested in packaging Google Earth Pro, which is now free ?

http://google-latlong.blogspot.fr/2015/01/google-earth-pro-is-now-free.html

Latest Comments

Det commented on 2016-04-12 13:31

Please DO NOT flag out-of-date, unless there's a new version for Linux also:

$ curl -s https://dl.google.com/dl/earth/client/current/google-earth-stable_current_x86_64.rpm | head -c96 | strings | cut -d "-" -f4

Det commented on 2016-04-12 13:28

@manuelschneid3r, 7.1.5 is for Windows only.

wrongmetal commented on 2016-04-01 08:03

Get it to work with Panoramio pictures working and without crashing.
On x86_64 (with nvidia-beta drivers - 364.12)

Pre-install Freeimage (aur)
Pre-install Libpng15 (aur) **built it manually with: makepkg -si --skippgpcheck
(make sure all other dependencies are installed also)

Install Google-Earth with yaourt. (Read the PKGBUILD)
**Modify PKGBUILD (and enable panoramio fix=1 and 64bit=1)
(save it and continue the install)

After succesful install (and before first tome run of google-earth)
make sure to apply: sudo rm -r ~/.googleearth/Cache/

Run Google-Earth and voila, everything works, no lagging, and no crashes :)

...or use the wine-google-earth-pro version if your lazy from the AUR.

Det commented on 2016-03-02 09:10

Read the PKGBUILD.

pmros commented on 2016-03-02 09:09

Could you make a package for 64 bit version? I mean https://dl.google.com/earth/client/current/google-earth-stable_current_amd64.deb instead https://dl.google.com/earth/client/current/google-earth-stable_current_i386.deb.
Thank you.

Det commented on 2016-02-24 18:38

I got right on that. Evidence: https://aur.archlinux.org/packages/google-earth-pro-wine/

Det commented on 2015-10-26 03:55

@eazar001, like just stated, it isn't for Linux.

Det commented on 2015-10-02 03:42

@lmouillart, 7.1.5 isn't for Linux.

Det commented on 2015-08-04 10:50

What's "fully patched arch with kernel 4.0.4-2"? Latest one is 4.1.4-1: https://www.archlinux.org/packages/core/x86_64/linux/

The Qt/Architecture tweaks can be enabled on top of the PKGBUILD. The "Texture Compression" trick is also on the Wiki: https://wiki.archlinux.org/index.php/Google_Earth

jhsnyder666 commented on 2015-05-27 20:57

I see the same error as rigon when I recompiled per hutou's suggestions (64b etc).

I think I saw the same error with i386, but I didn't keep notes.

According to Alchemist251 at this thread:

https://productforums.google.com/forum/#!topic/earth/U8SlQy2nJLM

the problem is buggy qt libraries bundled with GE7. Alchemist251 points to this thread:

https://productforums.google.com/forum/?fromgroups=#!category-topic/earth/linux/_h4t6SpY_II%5B1-25-false%5D

I haven't followed up on it yet, so this may or may not be a solution, but a) the alleged problem is specific and not black magic; and b) I've tried several (recent) suggestions posted here as well as suggestions posted in the first link above ("sudo", deleting ~/.googleearth), but GE7 continues to crash.

Including the first link above for the OP statement that GE7 runs fine if he shuts down the network before invoking GE7.

Running fully patched arch with kernel 4.0.4-2, FX-4100, Nvidia GeForce GTX 750, nvidia proprietary drivers, 16G mainstore.

I doubt that it's relevant, but I have a 4K display (which requires special configuration, so who knows?)

fanisg commented on 2015-04-30 19:29

thank you hutou, it worked now.

attempted fix had worked in the past but since an update was not working. after a new update and your post I gave an other try again and works!!!

so good news for everyone.

dom0 commented on 2015-04-20 12:11

@rigon: The "Not implemented reached in bool net::HaveOnlyLoopbackAddresses()" isn't the issue. I see it, too, with the latest version but it doesn't crash for me.

I'm using Mesa 10.5.2 on an AMD card.

Det commented on 2015-04-17 16:58

Google: http://extra.linuxmint.com/pool/main/g/google-earth-stable/

lesebas commented on 2015-04-17 16:53

Is there any way to download the previous deb package 7.1.2.2041?

rigon commented on 2015-04-16 23:08

I also got the same error after trying to install google-earth-stable_7.1.4.1529_i386

Here is the output when I try to run google-earth:

[0416/235907:ERROR:net_util.cc(2195)] Not implemented reached in bool net::HaveOnlyLoopbackAddresses()
Google Earth has caught signal 11.


And here is the crashlog:

Major Version 7
Minor Version 1
Build Number 0004
Build Date Mar 30 2015
Build Time 22:21:27
OS Type 3
OS Major Version 3
OS Minor Version 19
OS Build Version 3
OS Patch Version 0
Crash Signal 11
Crash Time 1429225147
Up Time 0.555398

Stacktrace from glibc:
./libgoogleearth_free.so(+0x1e8d3b)[0xf7363d3b]
./libgoogleearth_free.so(+0x1e8f71)[0xf7363f71]
linux-gate.so.1(__kernel_sigreturn+0x0)[0xf775ebe0]

lesebas commented on 2015-04-16 11:39

The last version doesn't work on my x86_64 arch, the previous version was working well. I've tried with _64=1 and _attempt_fix=1 but it's always the same problem, crash signal 11 after splash screen.

Please change the optional depends : lib32-nouveau-dri no more exist and is replaced by lib32-mesa

hutou commented on 2015-04-12 20:08

On Arch64, a standard install from AUR succeeded, but GE crashed on startup (error = "unknown graphic card")
I then recompiled the package with _64=1 : GE did launch, but 2 issues remained : no panoramio and search not working.
The first can be solved by installing the packages freeimage and libpng15, and then recompiling the package google-earth with the variables _64 and _attempt_fix set to 1 in the PKGBUILD.
The second issue is related to the decimal separator of some locales (fr for me), and is easily solved by inserting the line
export LC_NUMERIC=en_US.UTF-8
just before the line
LD_LIBRARY_PATH=.:$LD_LIBRARY_PATH ./googleearth-bin "$@"
in the file /opt/google/earth/free/googleearth
This worked for me : GE works fine, no crashes.

hamelg commented on 2015-04-10 19:11

one more version, and still no Panoramio :(

Det commented on 2015-03-21 17:18

From the top of the PKGBUILD (and as hinted by the Sources section):

# Build an unstable 64-bit build: "1" to enable.
_64=0

But you're not gonna want it. Whatever 10-20% rendering improvements it would bring, they are nullified by the 5 first times it crashes on startup.

greyltc commented on 2015-03-21 17:15

Why is "_arch=i386" hardcoded into this PKGBUILD?
Am I reading that right?
In any case, google-earth-stable_7.1.2.2041_i386.deb is being downloaded on my 64 bit machine so something seems to be wrong here.

Det commented on 2015-02-11 02:32

@grugnog, umm, no, "starting with the version 4 beta Google Earth functions under Linux, as a native port using the Qt toolkit." (https://en.wikipedia.org/wiki/Google_Earth#Linux_version), and yes, the "Pro" .bin is old, as I've said twice now.

A Wine version would be a pretty easy thing to do with a helper script like that from utorrent.

grugnog commented on 2015-02-11 02:25

I think to use Google Earth Pro we would need to use the GoogleEarthProWin.exe running under wine, rather than the .bin version (which is the same outdated version as the deb, and just runs under wine internally anyway). Not sure how to systematically install this as an AUR yet though - perhaps we can find another wine package for reference.

Det commented on 2015-02-07 16:38

I'm Det, not grugnog?

qKUqm3wtY4 commented on 2015-02-07 11:32

@grugnog - I'll take your word for it, but it doesn't make much sense to me. If they're sending me a link to a download in reply to my request for Google Earth Pro, why would they send a link to Google Earth? Oh well.

Det commented on 2015-02-06 19:24

@qKUqm3wtY4, like I already said, there is no Google Earth Pro for Linux. It links to an old Google Earth Free. Not the Pro.

qKUqm3wtY4 commented on 2015-02-06 17:28

@grugnog There is a Linux version of Google Earth Pro. When you request access from Google, they'll send you the bin file, along with a username and password to access the license.

I have no clue how to install the bin file successfully under arch. I've tried, but I've had no success. If would really be great if someone would take the time to make an AUR package which asks for the bin file directiory (if that's possible).

grugnog commented on 2015-02-02 22:33

There isn't a Linux version of Earth Pro, but the windows version seems to work just fine (at least, I couldn't find anything not working). I guess it's up to the maintainer of this package if they want to morph it into a Earth Pro package, or stick with the deb install, and perhaps we can make a separate google-earth-pro AUR.

In any case, this worked for me to install, using the default wine package (and a clean prefix):
wget http://dl.google.com/earth/client/advanced/current/GoogleEarthProWin.exe
WINEARCH=win32 wine GoogleEarthProWin.exe

Det commented on 2015-01-31 12:23

That only seems to link to the free version of 6.0 for Linux for now...:

- Linux: http://www.google.com/earth/download/thanks.html#os=linux#version=pro#updater=yes
https://dl.google.com/earth/client/current/GoogleEarthLinux.bin

- Mac: http://www.google.com/earth/download/thanks.html#os=mac#version=pro#updater=yes
https://dl.google.com/earth/client/advanced/current/GoogleEarthProMac-Universal.dmg

- Windows: http://www.google.com/earth/download/thanks.html#os=win#version=pro#updater=yes
https://dl.google.com/tag/s/appguid={65E60E95-0DE9-43FF-9F3F-4F7D2DFF04B5}&iid={E6800BAF-1E9C-D23F-997D-5B8D0ED05690}&lang=undefined&browser=4&usagestats=0&appname=Google%20Earth%20Pro&needsadmin=True&brand=GGGE/earth/client/GoogleEarthProSetup.exe

E: Actually, there apparently is no Linux version of Google Earth Pro at all: https://productforums.google.com/forum/#!topic/earth/qGeaQ3LQYEs

Det commented on 2015-01-31 12:16

That only seems to link to the free version of 6.0 for Linux for now...:

- Linux: http://www.google.com/earth/download/thanks.html#os=linux#version=pro#updater=yes
https://dl.google.com/earth/client/current/GoogleEarthLinux.bin

- Mac: http://www.google.com/earth/download/thanks.html#os=mac#version=pro#updater=yes
https://dl.google.com/earth/client/advanced/current/GoogleEarthProMac-Universal.dmg

- Windows: http://www.google.com/earth/download/thanks.html#os=win#version=pro#updater=yes
https://dl.google.com/tag/s/appguid={65E60E95-0DE9-43FF-9F3F-4F7D2DFF04B5}&iid={E6800BAF-1E9C-D23F-997D-5B8D0ED05690}&lang=undefined&browser=4&usagestats=0&appname=Google%20Earth%20Pro&needsadmin=True&brand=GGGE/earth/client/GoogleEarthProSetup.exe

Det commented on 2015-01-31 12:14

That only seems to link to the free version of 6.0 for Linux for now...:

- Linux: http://www.google.com/earth/download/thanks.html#os=linux#version=pro#updater=yes
https://dl.google.com/earth/client/current/GoogleEarthLinux.bin

- Mac: http://www.google.com/earth/download/thanks.html#os=mac#version=pro#updater=yes
https://dl.google.com/earth/client/advanced/current/GoogleEarthProMac-Universal.dmg

- Windows: http://www.google.com/earth/download/thanks.html#os=windows#version=pro#updater=yes
https://dl.google.com/tag/s/appguid={65E60E95-0DE9-43FF-9F3F-4F7D2DFF04B5}&iid={E6800BAF-1E9C-D23F-997D-5B8D0ED05690}&lang=undefined&browser=4&usagestats=0&appname=Google%20Earth%20Pro&needsadmin=True&brand=GGGE/earth/client/GoogleEarthProSetup.exe

aurelieng commented on 2015-01-31 06:32

Anyone interested in packaging Google Earth Pro, which is now free ?

http://google-latlong.blogspot.fr/2015/01/google-earth-pro-is-now-free.html

Det commented on 2015-01-21 20:13

Is https://aur.archlinux.org/packages/go/google-earth/google-earth.tar.gz blocked in China?

Andy_Crowd commented on 2015-01-21 19:57

Link to baifaao.cpp, google-earth-mimetypes.xml,googleearth
https://github.com/archlinuxcn/repo/tree/master/google-earth

jaro3 commented on 2015-01-17 02:15

Google Earth 6 can use proxy on Linux...
Google Earth 7 can use proxy on Windows, why Google broke it on Linux and cannot fix it is beyond my understanding.

Google Earth 6 got other issues (e.g. needs the old msfonts versus the new ttf-ms-win8 fonts) and doesn't show so detailed pictures and 3D buildings so new version is preferred.

Det commented on 2015-01-16 12:09

Doesn't work with Google Earth 6?: https://aur.archlinux.org/packages/google-earth6/

jaro3 commented on 2015-01-16 06:14

Anyone found some workaround or patch for the proxy issue?

http://code.google.com/p/earth-issues/issues/detail?id=1695

Det commented on 2014-12-09 09:16

You could try: http://pastebin.com/89Qk4Yp7 to see where it's failing at.

Most likely the curl line.

Det commented on 2014-12-09 09:16

You could try: http://pastebin.com/89Qk4Yp7 to see where it's failing at.

mabier commented on 2014-12-09 03:40

Can not install either google-earth 6 or google-earth 7, terminal output:
"-> Moving stuff in place
==> ERROR: A failure occurred in package().
Aborting...
==> ERROR: Makepkg was unable to build google-earth."

luuuciano commented on 2014-10-21 14:16

@grayich, to use panoramio pictures properly you will have to use google-earth6 instead this...

grayich commented on 2014-10-21 14:07

Does the white background open instead of photographs for everyone?
http://i.imgur.com/62ElPGh.jpg

Det commented on 2014-10-15 06:12

Well, people often like to use the latest version just because it sounds nicer to have it, and hence it must be better.

But the thing is, if you don't use Google Earth to get a sense what buildings look like from the sky, or as a tour guide, you will be much happier with the stabilized version, because it's actually going to work.

valentin.brasov commented on 2014-10-15 06:01

Version 7 of google-earth always crashed for me on my arch 64bit, no matter if I enabled any of the 2 flags. I know that version 6 works, but I am not keen to go back to that.

This is a bit of an irony: I mean what keeps me from switching fully from Windows to Linux? Google! Yeah! By not providing for linux the same tools they provide for windows, like google drive (I know of some poor alternatives like grive) or providing low quality versions for linux (like the version 7 of google-earth)!

But thank you for your effort.

Det commented on 2014-10-12 15:08

Do you mean it still crashes after 4 days of nothing changing in the package?

The last upstream update was almost a year ago (Oct 31st '13), which is why I recommended you to try out the legacy Google Earth (v6), which will show you the Panoramas just fine.

valentin.brasov commented on 2014-10-12 15:03

Still crashes with both flags set (64 bit and the attempt fix):
/usr/bin/google-earth: Zeile 17: 6431 Speicherzugriffsfehler (Speicherabzug geschrieben) LD_PRELOAD=/usr/lib/libfreeimage.so.3:./baifaao.so:/usr/lib/libpng15.so LD_LIBRARY_PATH=.:$LD_LIBRARY_PATH ./googleearth-bin "$@"

"Speicherzugriffsfehler" means "memory access error" and "Speicherabzug geschrieben" means "written dump".

Det commented on 2014-10-08 22:18

It pretty much either works with those two options or it doesn't.

There's also the previous Google Earth in AUR: https://aur.archlinux.org/packages/google-earth6/

valentin.brasov commented on 2014-10-08 17:31

Hi,

I re-installed google-earth with the attempt fix flag set to 1 and then when I try displaying a image it crashes with a message "Google Earth has caught signal 11." in the console and a crash log like follows:

Major Version 7
Minor Version 1
Build Number 0002
Build Date Oct 7 2013
Build Time 12:17:02
OS Type 3
OS Major Version 3
OS Minor Version 16
OS Build Version 3
OS Patch Version 0
Crash Signal 11
Crash Time 1412788931
Up Time 25,8312

Stacktrace from glibc:
./libgoogleearth_free.so(+0x1e8d0b)[0xf7067d0b]
./libgoogleearth_free.so(+0x1e8f41)[0xf7067f41]
linux-gate.so.1(__kernel_sigreturn+0x0)[0xf77c4d70]
/usr/lib32/qt/plugins/imageformats/libqjpeg.so(+0x2aed)[0xed7afaed]
/usr/lib32/libQtWebKit.so.4(+0xbe7138)[0xf51c2138]
/usr/lib32/libQtWebKit.so.4(+0xbe71aa)[0xf51c21aa]


Any ideas, please?

Thank you.

valentin.brasov commented on 2014-10-08 17:30

Hi,

I re-installed google-earth with the attempt fix flag set to 1 and then when I try displaying a image it crashes with a message {{ic|code}}Google Earth has caught signal 11.{{ic}} in the console and a crash log like follows:

{{bc|code}}
Major Version 7
Minor Version 1
Build Number 0002
Build Date Oct 7 2013
Build Time 12:17:02
OS Type 3
OS Major Version 3
OS Minor Version 16
OS Build Version 3
OS Patch Version 0
Crash Signal 11
Crash Time 1412788931
Up Time 25,8312

Stacktrace from glibc:
./libgoogleearth_free.so(+0x1e8d0b)[0xf7067d0b]
./libgoogleearth_free.so(+0x1e8f41)[0xf7067f41]
linux-gate.so.1(__kernel_sigreturn+0x0)[0xf77c4d70]
/usr/lib32/qt/plugins/imageformats/libqjpeg.so(+0x2aed)[0xed7afaed]
/usr/lib32/libQtWebKit.so.4(+0xbe7138)[0xf51c2138]
/usr/lib32/libQtWebKit.so.4(+0xbe71aa)[0xf51c21aa]
{{bc}}

Any ideas, please?

Thank you.

valentin.brasov commented on 2014-10-08 17:30

Hi,

I re-installed google-earth with the attempt fix flag set to 1 and then when I try displaying a image it crashes with a message {{ic|code}}Google Earth has caught signal 11.{{ic|code}} in the console and a crash log like follows:

{{bc|code}}
Major Version 7
Minor Version 1
Build Number 0002
Build Date Oct 7 2013
Build Time 12:17:02
OS Type 3
OS Major Version 3
OS Minor Version 16
OS Build Version 3
OS Patch Version 0
Crash Signal 11
Crash Time 1412788931
Up Time 25,8312

Stacktrace from glibc:
./libgoogleearth_free.so(+0x1e8d0b)[0xf7067d0b]
./libgoogleearth_free.so(+0x1e8f41)[0xf7067f41]
linux-gate.so.1(__kernel_sigreturn+0x0)[0xf77c4d70]
/usr/lib32/qt/plugins/imageformats/libqjpeg.so(+0x2aed)[0xed7afaed]
/usr/lib32/libQtWebKit.so.4(+0xbe7138)[0xf51c2138]
/usr/lib32/libQtWebKit.so.4(+0xbe71aa)[0xf51c21aa]
{{bc|code}}

Any ideas, please?

Thank you.

Det commented on 2014-10-05 16:21

You should have a look at the first lines in the PKGBUILD. It explains it all.

valentin.brasov commented on 2014-10-05 15:13

Hello,

The pictures are STILL not displayed (empty ballons). Any WORKING workaround for it? I have a 64bit arch installation. Trying to fix this by adding

export LD_PRELOAD=/usr/lib64/libfreeimage.so.3

before the last line in google-earth (as described in https://productforums.google.com/forum/#!msg/earth/tZfKSs2AaZc/r_rBDl5djIMJ) does not work because this google earth package is 32bit and libfreeimage is 64bit!

In my opinion maintaining a 32bit package for a 64bit system is asking for trouble!

Thanks.

Det commented on 2014-08-05 16:49

You're probably using an AUR tool that can't handle the PKGBUILD or the system really is 64-bit.

bezerker commented on 2014-08-05 14:06

Hrm. Using an old 32bit system here. makepkg is set to i686, but it's trying to source in the dependencies for lib32-dependencygoeshere which obviously fails on a 32bit system.

:(

Det commented on 2014-07-13 22:30

And really, the link always pointed to the amd64 deb on 64-bit systems in the past. It just included 32-bit binaries.

Det commented on 2014-07-09 17:35

Sure. The 64-bit version was so buggy on Linux ever since it was introduced in November 2012 (7.0.1.8244 beta: http://filehippo.com/download_google_earth/changelog/13615/) that I had to re-create the previous one (https://aur.archlinux.org/packages/google-earth6/) so people could actually have a working Google Earth.

I've since recentlish (24th April) moved the 64-bit build behind the _64=1 switch for those who insist, as can be seen from the comments and the PKGBUILD, respectively.

Gabe commented on 2014-07-09 17:24

Is there a reason that this package installs a 32-bit version of Earth on x86_64 when there is a 64-bit version available?

Det commented on 2014-05-07 17:19

I know it sucks. Stick to Google Earth 6: https://aur.archlinux.org/packages/google-earth6/

fanisg commented on 2014-04-30 09:51

unfortunately the attempted fix does not work now.

on my desktop (64bit system) almost works and does convince me that it does something right because it does try to open the picture (normal install does not even try) but then it crashes. To me it does not crashes immediately, only when I try to see a picture.

on my laptop (32bit system) it does not even install.

So I am reduced to not have the attempted fix and not be able to see panoramio pictures.

Det commented on 2014-04-29 22:46

@fanisg, here you go. Still crashes immediately on me, though.

Det commented on 2014-04-29 19:30

lol

persio commented on 2014-04-29 19:30

Det: You are right, I'm terribly sorry: CARCH="x86_64".

Det commented on 2014-04-29 19:19

$ grep CARCH /etc/makepkg.conf? Are you using some non-standard makepkg?

persio commented on 2014-04-29 18:43

I downloaded google-earth 7.1.2.2041-2 from AUR and when I try to install it, I can't because these dependencies are missing:

ld-lsb>=3-5
lib32-fontconfig
lib32-glu
lib32-libgl
lib32-libsm
lib32-libxrender
lib32-mesa

I'm not running 64-bits, I'm on 686. I'm posting here coming from here: https://bbs.archlinux.org/viewtopic.php?pid=1410296#p1410296

Det commented on 2014-04-29 17:13

You shouldn't even talk about it as "reimplementing" it, because it's such a simple change. I'll get to it later today.

fanisg commented on 2014-04-29 16:32

sorry about that click, it was accidental.

are you reimplementing the attempted fix?

Det commented on 2014-04-29 16:18

It's not bloody out-of-date..

Det commented on 2014-04-29 14:25

I'll hit the gym first? It's a matter of adding just a few lines anyway.

fanisg commented on 2014-04-29 14:23

pretty please implement again the attempted fix flag!!!!

the good news is that google-earth out of the box with this update works better than ever. The bad news is that I can not see any panoramio picture (but everything else works fantastic). With the attempted fix the pictures were working. So I would really appreciate if you could reimpliment that attempted fix please (and I will report if it works again)

saultdon commented on 2014-04-28 07:30

Perfect. Thanks!

Det commented on 2014-04-27 21:08

Reinstall ld-lsb.

saultdon commented on 2014-04-27 20:49

GE (7.1.2.2041-2) doesn't launch with the error:
% google-earth
/usr/bin/google-earth: line 43: ./googleearth-bin: No such file or directory

Det commented on 2014-04-25 08:04

No, 'msg2' is what I wanted them to be, but since they don't work in .install, I've reverted them to 'msg_blue'.

kerberizer commented on 2014-04-24 23:54

@Det: sed -i s/msg2/note/ google-earth.install ;)

Det commented on 2014-04-24 22:23

Hmh. My fault really for not doing it sooner.

You're welcome, tho.

vugri commented on 2014-04-24 22:22

Now works Ok with latest fglrx(beta).Finally.
Thanks Det!!!

Loubrix commented on 2014-04-24 21:17

Good idea, everybody can choose.
Thank you ;)

Det commented on 2014-04-24 20:42

The 32-bit version is now the default. 64-bit version can be enabled with the _64 flag on top of the PKGBUILD.

Det commented on 2014-04-24 01:54

The AMD64 package has included 64-bit binaries ever since the first beta of 7.0 (October 2012). We wouldn't have a problem, if it didn't.

Loubrix commented on 2014-04-15 16:34

Hello,
the amd64 deb package don't work even on deb distros (like Ubuntu); so I've found a workaround: build the package for 32 and 64 bits systems only with the i386 deb package (and all the necessary lib32); after all, the amd64 google package is a i386 deb with lib32 deps.
here is a modified PKGBUILD:

http://pastebin.archlinux.fr/499621

hope this help ;)

mrechte commented on 2014-04-10 18:56

Hello,

When starting GE, I get a crash and the following output:

[0410/205332:ERROR:net_util.cc(2195)] Not implemented reached in bool net::HaveOnlyLoopbackAddresses()
libGL error: dlopen /usr/lib32/xorg/modules/dri/i965_dri.so failed (/usr/lib32/xorg/modules/dri/i965_dri.so: mauvaise classe ELF : ELFCLASS32)
libGL error: unable to load driver: i965_dri.so
libGL error: driver pointer missing
libGL error: failed to load driver: i965
libGL error: dlopen /usr/lib32/xorg/modules/dri/swrast_dri.so failed (/usr/lib32/xorg/modules/dri/swrast_dri.so: mauvaise classe ELF : ELFCLASS32)
libGL error: unable to load driver: swrast_dri.so
libGL error: failed to load driver: swrast
[0410/205332:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0410/205332:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0410/205332:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0410/205332:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0410/205332:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0410/205332:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0410/205332:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0410/205332:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0410/205333:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0410/205333:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0410/205333:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0410/205333:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0410/205338:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0410/205338:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0410/205338:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0410/205338:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0410/205338:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
Google Earth has caught signal 11.



We apologize for the inconvenience, but Google Earth has crashed.
This is a bug in the program, and should never happen under normal
circumstances. A bug report and debugging data have been written
to this text file:

/home/mrechte/.googleearth/crashlogs/crashlog-5346e8b2.txt

Please include this file if you submit a bug report to Google.

Is this a 32-bit related problem (I am on a 64-bit Arch)

Thanks

Det commented on 2014-04-08 02:52

Maybe. I used to have that as a flag, but replaced it with the _attempt_fix, which was supposed to let us keep the 64-bit version and even fix Panoramio.

olive commented on 2014-04-05 09:08

This version crash for me on a 64 bits install. However the 32 bits version (as downloaded from Google) works fine with the appropriate multilib libraries installed. Does anyone have success with this version on a 64 bits archlinux? It might be worthwhile to package the 32 bits version even for machine running the 64 bits version of archlinux. I use a AMD graphic card (with the catalyst driver).

Det commented on 2014-02-26 08:46

Well, at least then it doesn't crash.

If you already checked the thread links in the Wiki, then you're probably better off using google-earth6.

Nestor_013 commented on 2014-02-26 08:42

I'v installed the latest kernel and made a full update (main and AUR repo's).
GE is still running fine (except the panoramio, which only works for one picture, then displays blanks).

Any feedback from other user would be appreciated I suppose.

So to help further is there any output you could need.
I'm far from being a specialist, but is there any error messages on console on top of the ones I always got:
[0226/094014:ERROR:net_util.cc(2195)] Not implemented reached in bool net::HaveOnlyLoopbackAddresses()
[0226/094014:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.

keepitsimpleengr commented on 2014-02-25 22:14

BTW - thanks to all who labor to allow others to use google-earth..!

Det commented on 2014-02-25 09:06

I sort of wish that.

Thanks for your contribution so far.

E: By the way, Nvidia in [extra] still isn't updated to .49.

Det commented on 2014-02-25 09:05

I sort of wish that.

Thanks for your contribution so far.

Nestor_013 commented on 2014-02-25 08:51

Of course all is relative :-)

Edit:

Package is running fine out of the box, after toggling the
"_attempt_fix" switch to 1.

Next step for me will be to upgrade the kernel and nvidia driver to be sure it's not comming from there.

Nestor_013 commented on 2014-02-24 16:36

Of course all is relative :-)
I'll try tomorrow at work to see if the package works by itself.

Det commented on 2014-02-24 16:29

From my point of view it's strange that for you it _does_ work.

Anyway, I've added the 'baifaao.so' workaround behind the "_attempt_fix" variable. I keep saying 'attempted fix' because for me it really does fix nothing.

Maybe you'll see something wrong in my procedure?

Nestor_013 commented on 2014-02-24 10:43

Strange...
I've put
http://cpm-ws4.ulb.ac.be/test/baifaao.so (md5:8d85aeec96d5c342241825cd920f8c8c)

OK didn't noticed lib64 was linked to lib
I swear I'll "pastebin" next time :-)

PS: I've recompiled baifaao.so to get the same md5sum as yours and it still runs.

Nestor_013 commented on 2014-02-24 10:07

Strange...
I've put
http://cpm-ws4.ulb.ac.be/test/baifaao.so (md5:8d85aeec96d5c342241825cd920f8c8c)

OK didn't noticed lib64 was linked to lib
I swear I'll "pastebin" next time :-)

Det commented on 2014-02-24 09:38

For what it's worth, even with the stock kernel (3.13.5-1-ARCH) and nvidia (331.38), preloading baifaao.so just makes Google Earth either crash, freeze or not finish loading the Earth surface (even with texture compression disabled).

Maybe you could upload your baifaao.so somewhere (provided its md5sum isn't 'd590f35ed578083fd21b9267e9865c56')?

ps. You don't need 'lib64' for libfreeimage.so.3. We provide it as a symlink to /usr/lib.
pps. Pastebin.

Det commented on 2014-02-24 08:55

I so wish you woulda used Pastebin..

Nestor_013 commented on 2014-02-24 08:49

Heelo Det, wrong assumptions: I've a full x86_64 system (3.12.9-2-ARCH) and an nvidia GTX-560 (331.38). BTW I applied it also on a SuSe at home without nvidia but with the same successful result.

Compilation without restrictions (except the include path):
gcc -I /usr/include/qt4/ -O3 -fPIC --shared baifaao.cpp -o baifaao.so

Just in case it could shade some light, here is an ldd of my running version made in the same conditions as in the launch script (ie with the PRELOAD):

LD_PRELOAD=/usr/lib64/libfreeimage.so.3:./baifaao.so:/usr/lib/libpng15.so LD_LIBRARY_PATH=.:$LD_LIBRARY_PATH ldd ./googleearth-bin

linux-vdso.so.1 (0x00007fffca922000)
/usr/lib64/libfreeimage.so.3 (0x00007f56f949c000)
./baifaao.so (0x00007f56f929b000)
/usr/lib/libpng15.so (0x00007f56f906e000)
libpthread.so.0 => /usr/lib/libpthread.so.0 (0x00007f56f8e51000)
libm.so.6 => /usr/lib/libm.so.6 (0x00007f56f8b50000)
libc.so.6 => /usr/lib/libc.so.6 (0x00007f56f87a8000)
libdl.so.2 => /usr/lib/libdl.so.2 (0x00007f56f85a4000)
libgoogleearth_free.so => ./libgoogleearth_free.so (0x00007f56f7f7d000)
libglobalnew.so => ./libglobalnew.so (0x00007f56f7d7b000)
libQtGui.so.4 => /usr/lib/libQtGui.so.4 (0x00007f56f70d1000)
libQtNetwork.so.4 => /usr/lib/libQtNetwork.so.4 (0x00007f56f6d90000)
libfontconfig.so.1 => /usr/lib/libfontconfig.so.1 (0x00007f56f6b54000)
libfreetype.so.6 => /usr/lib/libfreetype.so.6 (0x00007f56f68af000)
libX11.so.6 => /usr/lib/libX11.so.6 (0x00007f56f6574000)
libXrender.so.1 => /usr/lib/libXrender.so.1 (0x00007f56f636a000)
libXext.so.6 => /usr/lib/libXext.so.6 (0x00007f56f6158000)
libGL.so.1 => /usr/lib/libGL.so.1 (0x00007f56f5e25000)
libGLU.so.1 => /usr/lib/libGLU.so.1 (0x00007f56f5ba6000)
librt.so.1 => /usr/lib/librt.so.1 (0x00007f56f599e000)
libQtCore.so.4 => /usr/lib/libQtCore.so.4 (0x00007f56f54bd000)
libQtWebKit.so.4 => /usr/lib/libQtWebKit.so.4 (0x00007f56f3114000)
libstdc++.so.6 => /usr/lib/libstdc++.so.6 (0x00007f56f2e10000)
libgcc_s.so.1 => /usr/lib/libgcc_s.so.1 (0x00007f56f2bfa000)
libz.so.1 => /usr/lib/libz.so.1 (0x00007f56f29e4000)
/lib64/ld-lsb-x86-64.so.3 (0x00007f56f99b0000)
libIGCore.so => ./libIGCore.so (0x00007f56f26a7000)
libIGUtils.so => ./libIGUtils.so (0x00007f56f2472000)
libaction.so => ./libaction.so (0x00007f56f226a000)
libapiloader.so => ./libapiloader.so (0x00007f56f2066000)
libauth.so => ./libauth.so (0x00007f56f1e04000)
libbase.so => ./libbase.so (0x00007f56f1ab7000)
libcommon.so => ./libcommon.so (0x00007f56f181c000)
libcommon_gui.so => ./libcommon_gui.so (0x00007f56f159a000)
libcommon_platform.so => ./libcommon_platform.so (0x00007f56f1396000)
libcommon_webbrowser.so => ./libcommon_webbrowser.so (0x00007f56f1138000)
libcomponentframework.so => ./libcomponentframework.so (0x00007f56f0f30000)
libgeobase.so => ./libgeobase.so (0x00007f56f08fb000)
libgeobaseutils.so => ./libgeobaseutils.so (0x00007f56f06bc000)
libge_net.so => ./libge_net.so (0x00007f56f045d000)
libgdata.so => ./libgdata.so (0x00007f56f0222000)
libgoogleapi.so => ./libgoogleapi.so (0x00007f56efffe000)
libmath.so => ./libmath.so (0x00007f56efd6b000)
libmoduleframework.so => ./libmoduleframework.so (0x00007f56efb5c000)
libmaps.so => ./libmaps.so (0x00007f56ef933000)
libport.so => ./libport.so (0x00007f56ef727000)
libprintmodule.so => ./libprintmodule.so (0x00007f56ef4b1000)
libprofile.so => ./libprofile.so (0x00007f56ef2af000)
librender.so => ./librender.so (0x00007f56ef072000)
libreporting.so => ./libreporting.so (0x00007f56eee64000)
libsgutil.so => ./libsgutil.so (0x00007f56eeb22000)
libspatial.so => ./libspatial.so (0x00007f56ee728000)
libxsltransform.so => ./libxsltransform.so (0x00007f56ee3ad000)
libglib-2.0.so.0 => /usr/lib/libglib-2.0.so.0 (0x00007f56ee0ad000)
libpng16.so.16 => /usr/lib/libpng16.so.16 (0x00007f56ede78000)
libgobject-2.0.so.0 => /usr/lib/libgobject-2.0.so.0 (0x00007f56edc27000)
libSM.so.6 => /usr/lib/libSM.so.6 (0x00007f56eda1f000)
libICE.so.6 => /usr/lib/libICE.so.6 (0x00007f56ed803000)
libssl.so.1.0.0 => /usr/lib/libssl.so.1.0.0 (0x00007f56ed596000)
libcrypto.so.1.0.0 => /usr/lib/libcrypto.so.1.0.0 (0x00007f56ed190000)
libexpat.so.1 => ./libexpat.so.1 (0x00007f56ecf67000)
libbz2.so.1.0 => /usr/lib/libbz2.so.1.0 (0x00007f56ecd57000)
libxcb.so.1 => /usr/lib/libxcb.so.1 (0x00007f56ecb37000)
libnvidia-tls.so.331.38 => /usr/lib/libnvidia-tls.so.331.38 (0x00007f56ec934000)
libnvidia-glcore.so.331.38 => /usr/lib/libnvidia-glcore.so.331.38 (0x00007f56ea126000)
libjpeg.so.8 => /usr/lib/libjpeg.so.8 (0x00007f56e9ed1000)
libudev.so.1 => /usr/lib/libudev.so.1 (0x00007f56e9cbf000)
libgstapp-0.10.so.0 => /usr/lib/libgstapp-0.10.so.0 (0x00007f56e9ab2000)
libgstinterfaces-0.10.so.0 => /usr/lib/libgstinterfaces-0.10.so.0 (0x00007f56e98a0000)
libgstpbutils-0.10.so.0 => /usr/lib/libgstpbutils-0.10.so.0 (0x00007f56e967b000)
libgstvideo-0.10.so.0 => /usr/lib/libgstvideo-0.10.so.0 (0x00007f56e945f000)
libgstbase-0.10.so.0 => /usr/lib/libgstbase-0.10.so.0 (0x00007f56e920b000)
libgstreamer-0.10.so.0 => /usr/lib/libgstreamer-0.10.so.0 (0x00007f56e8f22000)
libsqlite3.so.0 => /usr/lib/libsqlite3.so.0 (0x00007f56e8c67000)
libQtXmlPatterns.so.4 => /usr/lib/libQtXmlPatterns.so.4 (0x00007f56e862a000)
libQtOpenGL.so.4 => /usr/lib/libQtOpenGL.so.4 (0x00007f56e832c000)
libIGMath.so => ./libIGMath.so (0x00007f56e80c9000)
libfusioncommon.so => ./libfusioncommon.so (0x00007f56e7ec4000)
libfilmstrip.so => ./libfilmstrip.so (0x00007f56e7c97000)
libfreebl3.so => ./libfreebl3.so (0x00007f56e7a2e000)
libnspr4.so => ./libnspr4.so (0x00007f56e77f1000)
libnss3.so => ./libnss3.so (0x00007f56e74b8000)
libnssckbi.so => ./libnssckbi.so (0x00007f56e7223000)
libnssdbm3.so => ./libnssdbm3.so (0x00007f56e6ffb000)
libnsssysinit.so => ./libnsssysinit.so (0x00007f56e6df8000)
libnssutil3.so => ./libnssutil3.so (0x00007f56e6bcc000)
libplc4.so => ./libplc4.so (0x00007f56e69c7000)
libplds4.so => ./libplds4.so (0x00007f56e67c3000)
libsmime3.so => ./libsmime3.so (0x00007f56e6596000)
libsoftokn3.so => ./libsoftokn3.so (0x00007f56e6357000)
libsqlite3.so => ./libsqlite3.so (0x00007f56e60df000)
libssl3.so => ./libssl3.so (0x00007f56e5e9e000)
libge_chrome_net.so => ./libge_chrome_net.so (0x00007f56e4781000)
liblayer.so => ./liblayer.so (0x00007f56e440d000)
libIGAttrs.so => ./libIGAttrs.so (0x00007f56e4159000)
libIGGfx.so => ./libIGGfx.so (0x00007f56e3b28000)
libIGSg.so => ./libIGSg.so (0x00007f56e37d0000)
libpcre.so.1 => /usr/lib/libpcre.so.1 (0x00007f56e3569000)
libffi.so.6 => /usr/lib/libffi.so.6 (0x00007f56e3361000)
libuuid.so.1 => /usr/lib/libuuid.so.1 (0x00007f56e315c000)
libXau.so.6 => /usr/lib/libXau.so.6 (0x00007f56e2f58000)
libXdmcp.so.6 => /usr/lib/libXdmcp.so.6 (0x00007f56e2d52000)
libgmodule-2.0.so.0 => /usr/lib/libgmodule-2.0.so.0 (0x00007f56e2b4e000)
libgthread-2.0.so.0 => /usr/lib/libgthread-2.0.so.0 (0x00007f56e294c000)
libxml2.so.2 => /usr/lib/libxml2.so.2 (0x00007f56e25e4000)
liborc-0.4.so.0 => /usr/lib/liborc-0.4.so.0 (0x00007f56e235e000)
libwmsbase.so => ./libwmsbase.so (0x00007f56e2121000)
libwebbrowser.so => ./libwebbrowser.so (0x00007f56e1f09000)
liblzma.so.5 => /usr/lib/liblzma.so.5 (0x00007f56e1ce6000)

Det commented on 2014-02-24 05:16

Removed "PKGEXT='.pkg.tar'" due to a request of having your own way through makepkg.conf.

Det commented on 2014-02-23 17:58

I tried that, but even without "-O3" Google Earth just keeps freezing.

I'm assuming you either don't have Nvidia or a 64-bit system.

Nestor_013 commented on 2014-02-10 09:40

I faced that too... but I used the solution from
http://forums.fedoraforum.org/showpost.php?p=1678303&postcount=60
and it works like a charm.

However I had to preload libpng12 or libpng15, because I had strange messages about "libpng error: bad parameters to zlib"

Hope it will help...

dax79 commented on 2014-02-09 18:00

Hi, me too
[0209/185328:ERROR:net_util.cc(2195)] Not implemented reached in bool net::HaveOnlyLoopbackAddresses()
[0209/185329:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0209/185329:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0209/185329:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0209/185329:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
System fully up to date, 64bit nvidia 331.38

Scoteanax commented on 2014-02-05 20:35

Also getting crashes on startup with the same flood of network-related errors shown by @frojnd. My system is fully up to date.

skrat commented on 2014-02-02 10:17

Same here.

frojnd commented on 2014-01-30 10:43

google-earth crashes when I start it. Here is the output from terminal:

[0130/112952:ERROR:net_util.cc(2195)] Not implemented reached in bool net::HaveOnlyLoopbackAddresses()
[0130/112952:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0130/112952:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
...
Another crash happened while handling crash!

xf86-video-intel: 2.99.907-2

Everything is up to date till this day. Any ideas how to solve this?

Det commented on 2014-01-13 02:40

Also mentioned in the Wiki.

agapito commented on 2014-01-13 02:38

Using radeonsi driver (HD7950) google earth was totally frozen my computer. With xf86-video-ati-git from aur, google-earth works perfect now.

Det commented on 2014-01-08 20:20

@darkshines, we know. It's even in the Wiki: https://wiki.archlinux.org/index.php/Google_Earth#Earth_shows_a_blue_marble_with_yellow_borders

darkshines commented on 2014-01-08 19:48

If anyone encounters a bug where there is no land, just a blue marble and black land when you zoom in, disable texture compression in the preferences.

zottelef commented on 2013-12-30 14:47

google-earth crashes with noveau driver 1.0.10-2 and xorg 1.15.0-2 on linux kernel 3.12.6-1. Architecture x86_64

1230/154828:ERROR:net_util.cc(2195)] Not implemented reached in bool net::HaveOnlyLoopbackAddresses()
[1230/154828:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[1230/154828:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
...
Another crash happened while handling crash!

hutou commented on 2013-12-15 21:34

I tried your suggestion and compilation was Ok, so second assignment was correctly done.
Then I move the "if [ "$CARCH" ... fi" block to just before the "options=('!emptydirs')" line and compilation was Ok again.
Then I successively moved this block down one line and finally found the line which caused the bug to appear : it was line 30, beginning with "[ "$CARCH" = 'x86_64' ]".
After several tests, I can conclude now that aura does not correctly interpret the old [ ... ] construct, nor the [ ... ] && [ ... ] foo="bar" trick.
Replacing all [ ... = ... ] with [[ ... == ... ]] and putting an if/fi around lines 30 and 37 solve the aura's problem.

Det commented on 2013-12-14 22:47

It seems like Aura's parser somehow halts on the first encounter of the dependencies, instead of reading all the way through. If you do what you did, but then remove the if/else/fi condition altogether, do you still get the correct dependencies?

Also, as mentioned, the ld-lsb/any thing was never an Aura issue. I was simply explaining to you why both Pacman and Yaourt were trying to install 'lib32-glibc'. It's listed as a dependency for both architectures assumably because the package is defined as 'any' and therefore built only on a 64-bit system.

hutou commented on 2013-12-14 22:22

I tried the --hotedit to edit PKGBUILD on the fly.
I first removed the [ "$_bin32" != '1' ] statement, but no luck.
I then moved the "depends" and "optdepends" first assignments to an else statement in the "if [ "$CARCH" = 'i686' ]..." statement and bingo, aura compile is Ok.
Very strange !

As for ld-lsb, even if arch=('any') is a bad idea, the 'depends' variable should remain empty if $CARCH is not equal to "x86_64" and this time, aura behaviour is just correct !

Det commented on 2013-12-14 21:40

Well, actually they both (i686 and x86_64 versions) list "lib32-glibc" as a dependency. It's a mistake in the PKGBUILD to define the package as 'any'.

Anyway, could you test what I mentioned?

hutou commented on 2013-12-14 21:34

What is very strange, is that the ld-lsb PKGBUILD files are indentical in both repositories Archlinuxfr and AUR.
and when ArlinuxFr repository is defined
"pacman -S" and "yaourt -S" complain about lib32-glibc dependency while "aura -A" just compiles without any warning.

Det commented on 2013-12-14 21:10

Mmh. It could be with the "[ "$_bin32" != '1' ]" statements or just Aura trying to fetch the first list of dependencies it can find.

The latter's a little unlikely considering Aura's age and the number of votes it's got (https://aur.archlinux.org/packages/aura/).

Anyway, could you try it out by first removing the "[ "$_bin32" != '1' ]" checks with the "--hotedit" flag?

hutou commented on 2013-12-14 20:36

Here it is !
$ cat /etc/makepkg.conf | grep CARCH
CARCH="i686"
$

While investigating this problem, I found that the ld-lsb package exists in 2 repositories (not a good idea !) :
$ LANG=C sudo aura -Ss ld-lsb
archlinuxfr/ld-lsb 3-6
Faking a Linux Standard Base runtime
$ LANG=C sudo aura -As ld-lsb
aur/ld-lsb 3-6 (731)
Faking a Linux Standard Base runtime
and that 'aura -A ld-lsb' compile is Ok while 'aura -S ld-lsb' gives the lib32-glibc dependency error.
So I commented out the [Archlinuxfr] repository in my /etc/pacman.conf and tried to install google-earth again.
"sudo aura -A google-earth" still terminated in error (lib32-fontconfig dependency), but "yaourt -S google-earth" properly installed ld-lsb, then google-earth.
So, you are right, aura fails to correctly parse your PKGBUILD.

Det commented on 2013-12-14 19:45

Could you post "cat /etc/makepkg.conf | grep CARCH", please?

hutou commented on 2013-12-14 18:23

Well, to be sure, I uninstalled google-earth, then ld-lsb and then tried to install directly google-earth with :
LANG=C yaourt -S google-earth

I got
... lot of lines missing...
==> Building and installing package
==> Install or build missing dependencies for google-earth:
resolving dependencies...
warning: cannot resolve "lib32-glibc", a dependency of "ld-lsb"
:: The following package cannot be upgraded due to unresolvable dependencies:
ld-lsb

:: Do you want to skip the above package for this upgrade? [y/N] n
error: failed to prepare transaction (could not satisfy dependencies)
:: ld-lsb: requires lib32-glibc
==> Restart building google-earth ? [y/N]

That's why I supposed a PKGBUILD error (both aura and yaourt referencing a lib32 library dependency on a 32 bits system).
Sorry if I am wrong.

Det commented on 2013-12-14 16:49

Well, again, from your bug report[1], 1) the PKGBUILD is not at fault - Aura is and 2) Yaourt didn't complain about lib32-glibc - Aura did.

[1] = https://github.com/fosskers/aura/issues/174

hutou commented on 2013-12-14 16:33

@Det
Done.

Det commented on 2013-12-13 13:22

It's another dependency for 64-bit systems so I assume it was the other way around.

Again, Aura somehow doesn't take into account the "if [ "$CARCH" == 'x86_64' ]" statement and tries to install the dependency nonetheless.

Have you mentioned this to fosskers?

hutou commented on 2013-12-13 12:51

Success finally, with
sudo aura -A ld-lsb
yaourt -S google-earth

Separate install of ld-lsb was needed because, on first try, "yaourt -S google-earth" complained about lib32-glibc, dependency of ld-lsb

Det commented on 2013-12-13 09:37

Added the compression workaround to the Wiki: https://wiki.archlinux.org/index.php/Google_Earth

Det commented on 2013-12-12 21:20

Oh. Whatever.

karol_007 commented on 2013-12-12 21:20

Works fine with cower. Maybe it's aura's fault?

@Det
Multilib on a 32-bit system?

karol_007 commented on 2013-12-12 21:19

Works fine with cower. Maybe it's aura's fault?

Det commented on 2013-12-12 21:18

Enable [multilib].

hutou commented on 2013-12-12 21:11

$ sudo aura -A google-earth
aura >>= Determining dependencies...
aura >>= Dependency checking failed for these reasons:
The dependency `lib32-fontconfig` could not be found. You may need to search for a package to satisfy it.

$ uname -a
Linux m1530al 3.12.4-1-ARCH #1 SMP PREEMPT Mon Dec 9 08:33:26 CET 2013 i686 GNU/Linux

Problem in PKGBUILD for 32 bits systems I guess ?

vitke commented on 2013-12-11 19:36

Indeed, unchecking compression does solve the problem with nvidia 331.20-2 drivers.

piquer commented on 2013-12-11 15:14

Tools->Options->3D View, Texture Colors, uncheck Compression
This helped for me with the nvidia problem (no satelite images shown).

vitke commented on 2013-12-08 19:43

It was the video driver. I replaced nvidia 331.20-2 with nvidia-304xx and now it sort of works. The error messages are still there, and Panoramio photos don't work of course.

vitke commented on 2013-12-07 13:08

I get a blue marble with no satelite images. Other features (borders etc) are present. Also error messages:

[1207/135554:ERROR:net_util.cc(2195)] Not implemented reached in bool net::HaveOnlyLoopbackAddresses()
[1207/135555:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[1207/135555:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[1207/135555:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[1207/135555:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
etc...

I searched a bit and found this:

https://productforums.google.com/forum/#!msg/earth/CkHwQHkt3Zo/pAHeZoIJF8cJ

So I should modify one of the files in the .deb file. However, if I do this then md5sums check fails and I don't know how to fix it.

luuuciano commented on 2013-11-26 01:39

Use google earth 6, until it is fixed... (the only thing I really miss, once you get used to it, is the search... for some reason do not work in GE6)

bred commented on 2013-11-25 17:19

I've always the problem with the images from panoramio !!!
http://i.imgur.com/Wycmk9i.jpg

Somebody know how to fix this ????

Det commented on 2013-10-25 18:18

1) It was downgraded by Google.

2) Version control? What we have is this, if that's what you mean: http://pkgbuild.com/git/aur-mirror.git/log/google-earth

yar commented on 2013-10-23 20:47

Why was this downgraded?

version a few weeks ago: 7.1.2.2019-1
version now: 7.1.1.1888-1

By the way, do you happen to have your PKGBUILDs in version control anywhere? Thanks.

Det commented on 2013-10-18 17:53

I doubt they'll care (even if he actually reaches them).

jstjohn commented on 2013-10-18 17:52

sl1pkn07, you should report that issue to the Google Earth team; otherwise, it won't get fixed.

Det commented on 2013-10-18 17:35

That's a developer message. You can ignore it.

sl1pkn07 commented on 2013-10-18 17:29

kbuildsycoca4(30905) KConfigGroup::readXdgListEntry: List entry Categories in "/usr/share/applications/google-earth.desktop" is not compliant with XDG standard (missing trailing semicolon).
kbuildsycoca4(30905) KConfigGroup::readXdgListEntry: List entry MimeType in "/usr/share/applications/google-earth.desktop" is not compliant with XDG standard (missing trailing semicolon).

msx commented on 2013-10-16 00:42

Sorry for the OT but I can't let this one just pass by:
"And that is simply because Google is the friend of all things and it will follow you to the end."
Beautiful phrase, I take it.

Det commented on 2013-10-13 12:18

Not really. It actually seems like you're misunderstanding pretty much everything you just brought up.

1) I hadn't seen your comment before I already posted mine and it's not one I want to remove. And that is simply because Google is the friend of all things and it will follow you to the end.

2) You may feel like this is some sort of a debate, but to me this thing is actually something constructive I wanted to keep discussing. It's got some pretty valuable points.

3) It's not offensive being asked newbie questions.

4) Having arguments with people really doesn't necessarily mean you hate your job, the people you work with and yourself. That's number #3.

And that's the _one thing_ I dislike in maintaining my beautiful packages.

E: Also, maybe it's just me but could you please write PKGBUILD in capitals, as it's meant to be? Thank-you.

Det commented on 2013-10-13 08:13

Not really. It actually seems like you're misunderstanding pretty much everything you just brought up.

1) I hadn't seen your comment before I already posted mine and it's not one I want to remove. And that is simply because Google is the friend of all things and it will follow you to the end.

2) You may feel like this is some sort of a debate, but to me this thing is actually something constructive I wanted to keep discussing. It's got some pretty valuable points.

3) It's not offending being asked newbie questions.

4) Having arguments with people really doesn't necessarily mean you hate your job, the people your work with and yourself. That's number #3.

And that's the _one thing_ I dislike in maintaining my beautiful packages.

E: Also, maybe it's just me but could you please write PKGBUILD in capitals, as it's meant to be? Thank-you.

Det commented on 2013-10-13 08:10

Not really. It actually seems like you're misunderstanding pretty much everything you just brought up.

1) I hadn't seen your comment before I already posted mine and it's not one I want to remove. And that is simply because Google is the friend of all things and it will follow you to the end.

2) You may feel like this is some sort of a debate, but to me this thing is actually something constructive I wanted to keep discussing. It's got some pretty valuable points.

3) It's not offending being asked newbie questions.

4) Having arguments with people really doesn't necessarily mean you hate your job, the people your work with and yourself. That's number #3.

And that's the _one thing_ I dislike in maintaining my beautiful packages.

TheWretched commented on 2013-10-13 01:36

You mean like my second comment minutes after the first,

"Nevermind, it turned out I was missing a simple package I overlooked. lib32-glu, works now."

You could have just added it to the pkgbuild with a simple reply and moved on rather than entering into a debate about how the audacity of my initial comment offended you.

You're welcome by the way. Looking at your ongoing argument with some of the others you got going on, maybe if you dislike maintaining packages so intensely, you should hand it off to someone else.

Det commented on 2013-10-12 07:28

Well, suggestions wouldn't really help with this particular problem, but in this case, yes, getting that dependency included was a beneficial thing to do. It's just the way you brought it up that wasn't.

"needs (lib32-)glu" instead of the error message that already tells you what the problem is again would have saved time for me, although you did actually even solve the problem yourself.

So useless? No, but annoyingly reminding me of all the other problems that _didn't_ benefit my package? Yeah.

TheWretched commented on 2013-10-11 23:24

Yeah, well in this case it's now in the pkgbuild and probably should have been in the first place as a dependency so if you feel that was a useless comment to bring it up then I'm not really sure what to suggest to you at this point.

Det commented on 2013-10-11 21:47

"[...] combined with snotty why don't you figure it out yourself answers."

I understand that and I know exactly how incredibly annoying times like these can be, but the fact is that this is simply what our own perspective has come to be. It's not a sweet fruit for us maintainers either having time and time again answer questions that had already been answered enough times before and/or that would have been so blatantly easy to just figure out yourself. The incessant flow of e-mail notifications doesn't really help either.

And I'm not from the worst end either. Though I'm not in a position in this whole Linux community to make anybody "SHUT THE FUCK UP" like Linus Torvalds[1], I do still easily descend to these snotty responses I myself used to sigh at. And I still sigh at them whenever it's me and not somebody else.

So, yes, that makes me a hypocrite, but no, it doesn't make me change my habits. I would rather have people spend 2 hours searching for a simple answer that is like shit to them rather than asking here.

Because that's me and I've reached my end for these things.

[1] = https://lkml.org/lkml/2012/12/23/75

TheWretched commented on 2013-10-11 21:06

One man's definition of easy is not necessarily equal to another. And things often seem easier in retrospect than they do when you don't know what the heck is going on yet. Also no I didn't know about the "site feature" of google until now.

TheWretched commented on 2013-10-11 21:01

Easy is a relative term where differing ability levels exist.

TheWretched commented on 2013-10-11 21:00

Easy is a subjective term where differing ability levels exist. One of the things I've always disliked about the Debian communities (ubuntu particularly) is the confusing muddle of differing and seldom working answers combined with snotty why don't you figure it out yourself answers.

I don't subscribe to the thought process of classifying any question as below oneself in a public forum. It takes more time to do that than to reply in the first place usually.

Det commented on 2013-10-11 20:35

@Superpiffer, 7.1.2.2019 is not yet for Linux.

Det commented on 2013-10-11 19:27

In such easy cases it's even more redundant to come here and ask first :).

A lot of people don't know about the "site:" feature in Google: https://www.google.fi/search?&q=site%3Aarchlinux.org+libGLU.so.1. It helps when you don't want to get flooded with e.g. Debian related suggestions.

TheWretched commented on 2013-10-11 18:14

The problem I usually have with google is getting flooded with debian related suggestions that may or may not even be helpful. Particularly with google earth it's usually pages and pages of people snarkily telling people to go install the official 32 bit .deb and shut up.

Actually in my case I figured out what I was missing faster with pacman -Ss than I did with google.

Nevertheless... the 32bit build seems to finally solve all my arch related google-earth headaches, (knock on wood). Finally nice clean fonts even with infinality and it doesn't crash 98% of the time like the 64 bit.

Really google earth is the bane of my existence, because I need to include it for our sales staff machines, otherwise I would cast it off to hell where it belongs and never look at it again.

Det commented on 2013-10-11 15:48

You need lib32-glu (Google is your friend too). Added to the PKGBUILD.

rtfreedman can start posting his diffs again.

TheWretched commented on 2013-10-11 15:42

Nevermind, it turned out I was missing a simple package I overlooked. lib32-glu, works now.

TheWretched commented on 2013-10-11 15:17

I can't seem to get the 32 bit version working via the flag in the packagebuild. I enabled it and everything went through. I have lib32-nvidia-utils for my graphics card.

When I launch I get the following error though, "./googleearth-bin: error while loading shared libraries: libGLU.so.1: cannot open shared object file: No such file or directory"

Det commented on 2013-10-06 22:00

You can decide for it in any way you like. But I'm deciding it for everybody else.

You are not one of those 10% I was comparing to because you are one of those 0,1% who do the recompile that doesn't benefit them, but who even come complaining about not getting nagged to do so.

And that's why you have no say in this.

rtfreedman commented on 2013-10-06 21:31

Well, I'm one of those 10% ... and I really don't care of your abilities/skills
to maintain a binary package, shit happens & errors are normal
- just notify me by way of pkgrel and let *me* decide of it's worthiness.

Det commented on 2013-10-06 18:58

Seriously, stop:

1) whining about my pkgrel changes (or the lack of) and

2) spamming these diffs nobody gives a sh*t about.

Those changes _still_ don't change _absolutely anything_ unless you change that flag and 90% of users _won't_, so why should I bump the pkgrel for the sake of individuals such as you? If you care about my changes so much, enable the comment notifications, and you'll get an email every time I add something that won't change anything but that you still want to have.

If you don't understand that, that's okay. I'm the maintainer. I know these things better than you do.

I'm seriously starting to lose my composure.

rtfreedman commented on 2013-10-06 18:53

And this diff still doesn't deserve a new pkgrel?
Do you really expect us to poll your AUR page and remember 'Last updated'?
You clearly missed the point of having pkgrel and something like 'aurcheck'

BTW. your latest update: tar tf google-earth.tar.gz
google-earth/
google-earth/google-earth.install
google-earth/.tar.gz
google-earth/google-earth-mimetypes.xml
google-earth/PKGBUILD
google-earth/googleearth

Do you see it?

--- google-earth.old1/PKGBUILD 2013-09-27 22:40:26.000000000 +0200
+++ google-earth/PKGBUILD 2013-10-04 23:22:00.000000000 +0200
@@ -2,6 +2,10 @@
# Contributors: 458italia, Madek, Berseker, Syr
# Check the latest version with: $ curl -s https://dl.google.com/dl/earth/client/current/google-earth-stable_current_x86_64.rpm | head -c96 | strings | cut -d "-" -f4

+# Use 32-bit binaries on x86_64. Should improve stability.
+# "1" to enable.
+_bin32="0"
+
pkgname=google-earth
pkgver=7.1.2.2019
pkgrel=1
@@ -9,28 +13,46 @@
arch=('i686' 'x86_64')
url="https://www.google.com/earth/index.html"
license=('custom:earth')
-depends=('desktop-file-utils' 'fontconfig' 'hicolor-icon-theme' 'ld-lsb>=3-5' 'libgl' 'libsm' 'libxrender' 'mesa' 'shared-mime-info' 'xdg-utils')
-optdepends=('ati-dri: For the open source AMD/ATI Radeon driver'
- 'catalyst-utils: For AMD Catalyst'
- 'gtk2: SCIM support'
- 'intel-dri: For the open source Intel driver'
- 'nouveau-dri: For the open source Nouveau driver'
- 'nss-mdns: In case the application fails to contact the servers'
- 'nvidia-utils: For the NVIDIA driver'
- 'nvidia-utils-bumblebee: For the NVIDIA driver + Bumblebee setups'
+depends=('desktop-file-utils' 'hicolor-icon-theme' 'ld-lsb>=3-5' 'lib32-fontconfig' 'lib32-gcc-libs' 'lib32-libgl' 'lib32-libsm' 'lib32-libxrender' 'lib32-mesa' 'shared-mime-info' 'xdg-utils')
+optdepends=('lib32-ati-dri: For the open source AMD/ATI Radeon driver'
+ 'lib32-catalyst-utils: For AMD Catalyst'
+ 'lib32-gtk2: SCIM support'
+ 'lib32-intel-dri: For the open source Intel driver'
+ 'lib32-nouveau-dri: For the open source Nouveau driver'
+ 'lib32-nss-mdns: In case the application fails to contact the servers'
+ 'lib32-nvidia-utils: For the NVIDIA driver'
+ 'lib32-nvidia-utils-bumblebee: For the NVIDIA driver + Bumblebee setups'
'qt4: For changing the font size with qtconfig'
'ttf-ms-fonts: Fonts')
options=('!emptydirs')
install=$pkgname.install
_arch=i386 # Workaround for the AUR Web interface Source parser
-[ "$CARCH" = 'x86_64' ] && _arch=amd64
+if [ "$CARCH" = 'x86_64' ] && [ "$_bin32" != '1' ]; then
+ _arch=amd64
+fi
source=("${pkgname}-stable_${pkgver}_${_arch}.deb::https://dl.google.com/earth/client/current/$pkgname-stable_current_$_arch.deb"
'googleearth'
"$pkgname-mimetypes.xml")
md5sums=('fd4c6bcc1f2fd3d3611bfad2efdd5e6e' # google-earth-stable_$pkgver_i386.deb
'c49e41e2c45802392cebdcb03633fd28' # googleearth
'e3c67b8d05c3de50535bd7e45eee728e') # google-earth-mimetypes.xml
-[ "$CARCH" = 'x86_64' ] && md5sums[0]='53536d6287dd759e41f25ead4f7a2ab9' # google-earth-stable_$pkgver_amd64.deb
+if [ "$_arch" = 'amd64' ]; then
+ md5sums[0]='53536d6287dd759e41f25ead4f7a2ab9' # google-earth-stable_$pkgver_amd64.deb
+fi
+
+if [ "$CARCH" = 'i686' ] || [ "$_bin32" != '1' ]; then
+ depends=('desktop-file-utils' 'fontconfig' 'hicolor-icon-theme' 'ld-lsb>=3-5' 'libgl' 'libsm' 'libxrender' 'mesa' 'shared-mime-info' 'xdg-utils')
+ optdepends=('ati-dri: For the open source AMD/ATI Radeon driver'
+ 'catalyst-utils: For AMD Catalyst'
+ 'gtk2: SCIM support'
+ 'intel-dri: For the open source Intel driver'
+ 'nouveau-dri: For the open source Nouveau driver'
+ 'nss-mdns: In case the application fails to contact the servers'
+ 'nvidia-utils: For the NVIDIA driver'
+ 'nvidia-utils-bumblebee: For the NVIDIA driver + Bumblebee setups'
+ 'qt4: For changing the font size with qtconfig'
+ 'ttf-ms-fonts: Fonts')
+fi

_instdir=/opt/google/earth/free/
PKGEXT='.pkg.tar'
@@ -56,8 +78,15 @@
curl -s ${url/i*}/license.html -o "$pkgdir"/usr/share/licenses/$pkgname/license.html

msg2 "Removing the Debian-intended cron job and duplicated images"
- rm "$pkgdir"/etc/cron.daily/$pkgname "$pkgdir"/$_instdir/product_logo_*.png
+ rm -r "$pkgdir"/etc/cron.daily/ "$pkgdir"/$_instdir/product_logo_*.png

# msg2 "Fixing the coordinates regression"
# install -m755 googleearth "$pkgdir"/$_instdir/
+
+ # Write the _bin32 value to .install
+ if [ "$_bin32" != '1' ]; then
+ sed -i "s/_bin32=.*/_bin32=\"0\"/" ../$pkgname.install
+ else
+ sed -i "s/_bin32=.*/_bin32=\"1\"/" ../$pkgname.install
+ fi
}
\ No newline at end of file
diff -rNau google-earth.old1/google-earth.install google-earth/google-earth.install
--- google-earth.old1/google-earth.install 2013-09-08 17:22:11.000000000 +0200
+++ google-earth/google-earth.install 2013-09-29 18:24:48.000000000 +0200
@@ -1,11 +1,20 @@
+# Whether to mention about the -bin32 build.
+_bin32="0"
+
# Some colored makepkg-like functions
msg_blue() {
printf "${BLUE}==>${ALL_OFF}${BOLD} ${1}${ALL_OFF}\n"
}

-note() {
- printf "${BLUE}==>${ALL_OFF}${BOLD} NOTE:${ALL_OFF} ${1}\n"
-}
+if [ "$_bin32" != "1" ]; then
+ note() {
+ printf "${BLUE}==>${ALL_OFF}${BOLD} NOTE:${ALL_OFF} ${1}\n"
+ }
+else
+ note() {
+ printf "${BLUE}==>${ALL_OFF}${BOLD} NOTE${1}:${ALL_OFF} ${2}\n"
+ }
+fi

ALL_OFF="$(tput sgr0)"
BOLD="$(tput bold)"
@@ -24,7 +33,13 @@
post_install() {
_update
echo
- note "The binary is called 'google-earth'."
+ if [ "$_bin32" != "1" ]; then
+ note "The binary is called 'google-earth'."
+ else
+ note 1 "The binary is called 'google-earth'."
+ note 2 "Using a 32-bit version. It is recommended to remove ~/.googleearth/Cache/
+ for correct rendering."
+ fi
echo
}

rtfreedman commented on 2013-10-06 18:51

And this diff still doesn't deserve a new pkgrel?
Do you really expect us to poll your AUR page and remember 'Last updated'?
You clearly missed to point of having pkgrel and something like 'aurcheck'

BTW. your latest update: tar tf google-earth.tar.gz
google-earth/
google-earth/google-earth.install
google-earth/.tar.gz
google-earth/google-earth-mimetypes.xml
google-earth/PKGBUILD
google-earth/googleearth

Do you see it?

--- google-earth.old1/PKGBUILD 2013-09-27 22:40:26.000000000 +0200
+++ google-earth/PKGBUILD 2013-10-04 23:22:00.000000000 +0200
@@ -2,6 +2,10 @@
# Contributors: 458italia, Madek, Berseker, Syr
# Check the latest version with: $ curl -s https://dl.google.com/dl/earth/client/current/google-earth-stable_current_x86_64.rpm | head -c96 | strings | cut -d "-" -f4

+# Use 32-bit binaries on x86_64. Should improve stability.
+# "1" to enable.
+_bin32="0"
+
pkgname=google-earth
pkgver=7.1.2.2019
pkgrel=1
@@ -9,28 +13,46 @@
arch=('i686' 'x86_64')
url="https://www.google.com/earth/index.html"
license=('custom:earth')
-depends=('desktop-file-utils' 'fontconfig' 'hicolor-icon-theme' 'ld-lsb>=3-5' 'libgl' 'libsm' 'libxrender' 'mesa' 'shared-mime-info' 'xdg-utils')
-optdepends=('ati-dri: For the open source AMD/ATI Radeon driver'
- 'catalyst-utils: For AMD Catalyst'
- 'gtk2: SCIM support'
- 'intel-dri: For the open source Intel driver'
- 'nouveau-dri: For the open source Nouveau driver'
- 'nss-mdns: In case the application fails to contact the servers'
- 'nvidia-utils: For the NVIDIA driver'
- 'nvidia-utils-bumblebee: For the NVIDIA driver + Bumblebee setups'
+depends=('desktop-file-utils' 'hicolor-icon-theme' 'ld-lsb>=3-5' 'lib32-fontconfig' 'lib32-gcc-libs' 'lib32-libgl' 'lib32-libsm' 'lib32-libxrender' 'lib32-mesa' 'shared-mime-info' 'xdg-utils')
+optdepends=('lib32-ati-dri: For the open source AMD/ATI Radeon driver'
+ 'lib32-catalyst-utils: For AMD Catalyst'
+ 'lib32-gtk2: SCIM support'
+ 'lib32-intel-dri: For the open source Intel driver'
+ 'lib32-nouveau-dri: For the open source Nouveau driver'
+ 'lib32-nss-mdns: In case the application fails to contact the servers'
+ 'lib32-nvidia-utils: For the NVIDIA driver'
+ 'lib32-nvidia-utils-bumblebee: For the NVIDIA driver + Bumblebee setups'
'qt4: For changing the font size with qtconfig'
'ttf-ms-fonts: Fonts')
options=('!emptydirs')
install=$pkgname.install
_arch=i386 # Workaround for the AUR Web interface Source parser
-[ "$CARCH" = 'x86_64' ] && _arch=amd64
+if [ "$CARCH" = 'x86_64' ] && [ "$_bin32" != '1' ]; then
+ _arch=amd64
+fi
source=("${pkgname}-stable_${pkgver}_${_arch}.deb::https://dl.google.com/earth/client/current/$pkgname-stable_current_$_arch.deb"
'googleearth'
"$pkgname-mimetypes.xml")
md5sums=('fd4c6bcc1f2fd3d3611bfad2efdd5e6e' # google-earth-stable_$pkgver_i386.deb
'c49e41e2c45802392cebdcb03633fd28' # googleearth
'e3c67b8d05c3de50535bd7e45eee728e') # google-earth-mimetypes.xml
-[ "$CARCH" = 'x86_64' ] && md5sums[0]='53536d6287dd759e41f25ead4f7a2ab9' # google-earth-stable_$pkgver_amd64.deb
+if [ "$_arch" = 'amd64' ]; then
+ md5sums[0]='53536d6287dd759e41f25ead4f7a2ab9' # google-earth-stable_$pkgver_amd64.deb
+fi
+
+if [ "$CARCH" = 'i686' ] || [ "$_bin32" != '1' ]; then
+ depends=('desktop-file-utils' 'fontconfig' 'hicolor-icon-theme' 'ld-lsb>=3-5' 'libgl' 'libsm' 'libxrender' 'mesa' 'shared-mime-info' 'xdg-utils')
+ optdepends=('ati-dri: For the open source AMD/ATI Radeon driver'
+ 'catalyst-utils: For AMD Catalyst'
+ 'gtk2: SCIM support'
+ 'intel-dri: For the open source Intel driver'
+ 'nouveau-dri: For the open source Nouveau driver'
+ 'nss-mdns: In case the application fails to contact the servers'
+ 'nvidia-utils: For the NVIDIA driver'
+ 'nvidia-utils-bumblebee: For the NVIDIA driver + Bumblebee setups'
+ 'qt4: For changing the font size with qtconfig'
+ 'ttf-ms-fonts: Fonts')
+fi

_instdir=/opt/google/earth/free/
PKGEXT='.pkg.tar'
@@ -56,8 +78,15 @@
curl -s ${url/i*}/license.html -o "$pkgdir"/usr/share/licenses/$pkgname/license.html

msg2 "Removing the Debian-intended cron job and duplicated images"
- rm "$pkgdir"/etc/cron.daily/$pkgname "$pkgdir"/$_instdir/product_logo_*.png
+ rm -r "$pkgdir"/etc/cron.daily/ "$pkgdir"/$_instdir/product_logo_*.png

# msg2 "Fixing the coordinates regression"
# install -m755 googleearth "$pkgdir"/$_instdir/
+
+ # Write the _bin32 value to .install
+ if [ "$_bin32" != '1' ]; then
+ sed -i "s/_bin32=.*/_bin32=\"0\"/" ../$pkgname.install
+ else
+ sed -i "s/_bin32=.*/_bin32=\"1\"/" ../$pkgname.install
+ fi
}
\ No newline at end of file
diff -rNau google-earth.old1/google-earth.install google-earth/google-earth.install
--- google-earth.old1/google-earth.install 2013-09-08 17:22:11.000000000 +0200
+++ google-earth/google-earth.install 2013-09-29 18:24:48.000000000 +0200
@@ -1,11 +1,20 @@
+# Whether to mention about the -bin32 build.
+_bin32="0"
+
# Some colored makepkg-like functions
msg_blue() {
printf "${BLUE}==>${ALL_OFF}${BOLD} ${1}${ALL_OFF}\n"
}

-note() {
- printf "${BLUE}==>${ALL_OFF}${BOLD} NOTE:${ALL_OFF} ${1}\n"
-}
+if [ "$_bin32" != "1" ]; then
+ note() {
+ printf "${BLUE}==>${ALL_OFF}${BOLD} NOTE:${ALL_OFF} ${1}\n"
+ }
+else
+ note() {
+ printf "${BLUE}==>${ALL_OFF}${BOLD} NOTE${1}:${ALL_OFF} ${2}\n"
+ }
+fi

ALL_OFF="$(tput sgr0)"
BOLD="$(tput bold)"
@@ -24,7 +33,13 @@
post_install() {
_update
echo
- note "The binary is called 'google-earth'."
+ if [ "$_bin32" != "1" ]; then
+ note "The binary is called 'google-earth'."
+ else
+ note 1 "The binary is called 'google-earth'."
+ note 2 "Using a 32-bit version. It is recommended to remove ~/.googleearth/Cache/
+ for correct rendering."
+ fi
echo
}

Det commented on 2013-10-06 18:14

Meaning it does now?

qqqqqqqqq9 commented on 2013-10-06 17:16

Hi Det,
the relative path in this line:

sed -i "s/_bin32=.*/_bin32=\"0\"/" ../$pkgname.install

does not work when $BUILDDIR is set in /etc/makepkg.conf.

test0 commented on 2013-10-03 12:22

@HighValueWarrior: yes, it's a long-running issue (32-bit systems too).

test0 commented on 2013-10-03 12:22

@HighValueWarrior: yes, it's a long-running issue.

HighValueWarrior commented on 2013-10-01 05:30

Panoramio pictures not displaying, only a white box showing.
64 bit Arch

HighValueWarrior commented on 2013-10-01 05:27

Panoramio pictures not displaying .... only white box.

Det commented on 2013-09-29 16:32

Someone mentioned a while back about the new 64-bit builds being a lot more unstable, and that the i386 deb works better.

Well, before I start splitting this to a 'bin32-google-earth' we already got rid of so long ago, for now there's a "_bin32" flag at the top of the PKGBUILD to try it out.

I also noticed the Earth getting completely garbled after the switch, so removing ~/.googleearth/Cache/ is required (also mentioned in the .install).

Det commented on 2013-09-28 18:54

It's not "fixing" to change something that you won't even see.

But if it really brings you that much piece of mind, then go ahead, rebuild the whole thing.

Shall I personally notify you of changed comments and empty spaces in the PKGBUILD as well?

rtfreedman commented on 2013-09-28 17:52

>They don't have to...
.. don't have to what? Notifying users of a fixed PKGBUILD?

diff -Naur google-earth.old/PKGBUILD google-earth/PKGBUILD
--- google-earth.old/PKGBUILD 2013-09-25 03:01:16.000000000 +0200
+++ google-earth/PKGBUILD 2013-09-27 22:40:26.000000000 +0200
@@ -56,8 +56,8 @@
curl -s ${url/i*}/license.html -o "$pkgdir"/usr/share/licenses/$pkgname/license.html

msg2 "Removing the Debian-intended cron job and duplicated images"
- rm "$pkgdir"/etc/cron.daily/$pkgname "$pkgdir"/$_instdir/product_logo_*
+ rm "$pkgdir"/etc/cron.daily/$pkgname "$pkgdir"/$_instdir/product_logo_*.png

- msg2 "Fixing the coordinates regression"
- install -m755 googleearth "$pkgdir"/$_instdir/
+# msg2 "Fixing the coordinates regression"
+# install -m755 googleearth "$pkgdir"/$_instdir/

Det commented on 2013-09-28 16:20

They don't have to, if it's not gonna change anything :D.

Don't you worry about that.

rtfreedman commented on 2013-09-28 16:07

Please always bump pkgrel unless pkgver has been changed!
How do you think users get notified of a changed PKGBUILD?

wifi commented on 2013-09-25 08:42

Update to the last version and earth doesn't start anymore.
This is the verbose if started from terminal:

[0925/093759:ERROR:net_util.cc(2195)] Not implemented reached in bool net::HaveOnlyLoopbackAddresses()
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093800:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093800:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093800:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093800:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093800:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093800:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093800:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.


Another crash happened while handling crash!

wifi commented on 2013-09-25 08:40

My system update earth to the new version an it's not working anymore.
This is the verbose if started from terminal:

[0925/093759:ERROR:net_util.cc(2195)] Not implemented reached in bool net::HaveOnlyLoopbackAddresses()
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093800:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093800:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093800:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093800:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093800:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093800:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093800:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.


Another crash happened while handling crash!
[0925/093759:ERROR:net_util.cc(2195)] Not implemented reached in bool net::HaveOnlyLoopbackAddresses()
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093759:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093800:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093800:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093800:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093800:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093800:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093800:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093800:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
[0925/093800:WARNING:backend_impl.cc(1875)] Destroying invalid entry.


Another crash happened while handling crash!

Det commented on 2013-09-02 05:30

Btw. Linux Mint Lisa is of course version 12 and not 10 as I said a long way back there in the comments.

Just mentioning this unnecessity because I hate to spread bullsh*t.

Det commented on 2013-09-01 09:12

It clearly states "ld-lsb" anyway.

justinzane commented on 2013-09-01 01:52

Doh! :)

justinzane commented on 2013-09-01 01:52

Cannot install because of missing dependency:

==> Building and installing package
==> Install or build missing dependencies for ld-lsb:
error: target not found: lib32-glibc

thestinger commented on 2013-08-31 22:47

lib32-glibc is in [multilib], there's no problem with the package

justinzane commented on 2013-08-31 22:44

Cannot install because of missing dependency:

==> Building and installing package
==> Install or build missing dependencies for ld-lsb:
error: target not found: lib32-glibc

hamelg commented on 2013-08-25 13:06

about the Panoramio issue :
https://productforums.google.com/d/msg/earth/_h4t6SpY_II/6O_DTry49pgJ

olive commented on 2013-08-25 07:31

This version crash in my system (even after removing the 65-font-persian file) (my graphic card is a Radeon 6570 with xorg 1.13 and catalyst). I am on a 64 bits system. I have also tried it on the latest ubuntu and I have the same crash. However I have tried the 32 bits version directly downloaded from Google and it works properly (I have the multilib repository enabled and several lib32 packages). Has anyone succeeded to run the 64 bits version? It seems that it would be an idea to package the 32 bits version even for 64 bits system (unless I am an isolated case of course).

pataloca commented on 2013-08-19 12:11

After this new update I removed orphans and it removed all 32 bit multilib deps....is it finally all 64?

lks commented on 2013-08-19 06:53

I am sorry for all my comments I posted to this package. I wasn't thinking and they are simply false.
lks

Det commented on 2013-08-09 16:31

> Well 1882 means 38% of all registered aur-users.
Nno, we have over 49 _thousand_ registered accounts: https://aur.archlinux.org/

But it doesn't mean all of those are for different people.

lks commented on 2013-08-06 20:00

Thank you for all the responses.
I feel sorry for that. I was really not thinking and I was a bit angry that 1881 (now 1882) votes go for that non-free package that I can't even install on my archlinuxarm box.
Well 1882 means 38% of all registered aur-users. That means much of them have not registered to submit packages but instead to vote for that package.
lks

Det commented on 2013-08-04 14:50

It's prohibited, unless permitted by Google (http://earth.google.com/intl/en-US/license.html):

[...]
2. Restrictions on Use. Unless you have received prior written authorization from Google (or, as applicable, from the provider of particular Content), you must not:
[...]
(c) redistribute, sublicense, rent, publish, sell [...]

The 'googleearth-package' community app is assumably for Mint Isadora (9; 2010-05-18) and Lisa (10; 2011-11-26), even though they've had it in the repos since Daryna (4.0; 2007-10-15): http://www.linuxmint.com/rel_daryna.php

Det commented on 2013-08-04 14:48

It's prohibited, unless permitted by Google (http://www.google.com/earth/license.html):
[...]
2. Restrictions on Use. Unless you have received prior written authorization from Google (or, as applicable, from the provider of particular Content), you must not:
[...]
(c) redistribute, sublicense, rent, publish, sell [...]

The 'googleearth-package' community app is assumably for Mint Isadora (9; 2010-05-18) and Lisa (10; 2011-11-26), even though they've had it in the repos since Daryna (4.0; 2007-10-15): http://www.linuxmint.com/rel_daryna.php

karol_007 commented on 2013-08-04 14:31

The license prohibits redistribution: https://wiki.archlinux.org/index.php/Community_repo_candidates

Linux Mint: http://community.linuxmint.com/software/view/googleearth-package

Det commented on 2013-08-04 12:51

Oh, and the 'non-free' thing means nothing. We got Skype on our repo and Linux Mint got Google Earth in theirs: http://packages.linuxmint.com/pool/import/g/googleearth/, http://packages.linuxmint.com/pool/import/g/google-earth-stable/

karol_007 commented on 2013-08-04 04:01

@lks
archstuff unofficial repo provides google-earth 5.2.1.1588-1 - is that what you mean by 'binary repository with pre-build packages'? The package was build by using a PKGBUILD, as every other Arch Linux package.

Det commented on 2013-08-02 20:16

No, nowhere is the right place.

First of all, you mean repackaging? I personally am very content not having some Launchpad-like separate repo system in here for every single package that doesn't use build().

You're even suggesting that one of the 1881 voters (why are they more capable than anybody else, by the way?) should start maintaining their _own_ repo, tell about it in wherever and then have this package removed. And what then when the maintainer gets bored and decides, "Who gives a shit?" I'm assuming we'll create a second user repo and again inform the world just how magnificent the idea is. And you not only want this mess to be done for this package but for everybody else that use binaries? Did you actually think this through before you said it?

And the thing about Google creating its own repo for Arch Linux, I don't know where did you get this idea from, but it will probably not happen in the next 50 years.

It just doesn't work.

lks commented on 2013-08-02 19:36

I don't know if that's the right place but google earth is the google related packages with the most votes. I think these packages shouldn't be in the aur because the aur and the PKGBUILDs are wrappers for building packages - and not download them and do some sort of reverse-engineering.
Google or one of the 1881 should install it's own binary repository with pre-build packages, I think. If there is a binary repository the aur is not needed. This stuff shouldn't be in 'community' since it's nonfree.
lks

Det commented on 2013-07-26 15:28

The "-i" flag installs the package for makepkg and "-U" for pacman. That's pretty basic stuff: https://wiki.archlinux.org/index.php/Pacman#Additional_commands, https://wiki.archlinux.org/index.php/Makepkg#Usage

The result is a simple uncompressed .pkg.tar (PKGEXT='.pkg.tar') because it's just so damn faster to build.

E: And this release crashes a lot for me too, so you might wanna either stick to the previous 7.1 or even 6.2: https://aur.archlinux.org/packages/google-earth6/

Det commented on 2013-07-26 15:27

The "-i" flag installs the package for makepkg and "-U" for pacman. That's pretty basic stuff: https://wiki.archlinux.org/index.php/Pacman#Additional_commands, https://wiki.archlinux.org/index.php/Makepkg#Usage

The result is a simple uncompressed .pkg.tar (PKGEXT='.pkg.tar') because it's just so damn faster to build.

frojnd commented on 2013-07-26 10:13

Hi there,

I also use 64bit arch and for the first time I ran googleearth it was all fine.

Now for every time I started it it crashes:

[0726/120418:ERROR:net_util.cc(2195)] Not implemented reached in bool net::HaveOnlyLoopbackAddresses()
[0726/120418:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0726/120418:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0726/120419:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0726/120419:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0726/120420:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0726/120420:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.


Another crash happened while handling crash!

Also after makepkg -s there is no .xz file. How do I install google-earth with pcman. Now I have to run it manually: /home/frojnd/aur/google-earth/google-earth/pkg/google-earth/opt/google/earth/free/googleearth

ivictbor commented on 2013-07-25 12:50

Hi! I use 64bit arch. Panoramio photos are blank. Here is console output:
/opt/google/earth/free $ google-earth
[0725/154912:ERROR:net_util.cc(2195)] Not implemented reached in bool net::HaveOnlyLoopbackAddresses()
[0725/154912:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0725/154912:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0725/154912:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0725/154912:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0725/154913:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0725/154913:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0725/154914:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0725/154914:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.
[0725/154914:ERROR:nss_ocsp.cc(581)] No URLRequestContext for OCSP handler.

rtfreedman commented on 2013-07-19 15:06

==> Validating source files with md5sums...
google-earth-stable_7.1.1.1888_i386.deb ... FAILED
googleearth ... Passed
google-earth-mimetypes.xml ... Passed
==> ERROR: One or more files did not pass the validity check!

The old/previous PKGBUILD got it right
85655417a249712a9ccd452733d8234c google-earth-stable_7.1.1.1888_i386.deb

Alister.Hood commented on 2013-07-19 03:03

Oh, I see. It is because yaourt was trying to install the binary ld-lsb package from archlinuxfr, which is only for x86_64. I just had to install it from the aur separately, and that fixes the problem.

Alister.Hood commented on 2013-07-19 02:58

Is that anything to do with this?:

# google-earth
/usr/bin/google-earth: line 17: ./googleearth-bin: No such file or directory

Alister.Hood commented on 2013-07-19 02:47

On i686 a dependency on ls-lsb doesn't work:

warning: cannot resolve "lib32-glibc", a dependency of "ld-lsb"
:: The following package cannot be upgraded due to unresolvable dependencies:
ld-lsb

Kompilator commented on 2013-07-18 10:43

This are my lines 30 - 33 in my PKGBUILD:

md5sums=('82b179c2d4ff0f2eb40ebd5cfea27a1f' # google-earth-stable_$pkgver_i386.deb
'c49e41e2c45802392cebdcb03633fd28' # googleearth
'e3c67b8d05c3de50535bd7e45eee728e') # google-earth-mimetypes.xml
#[ "$CARCH" = 'x86_64' ] && md5sums[0]='' # google-earth-stable_$pkgver_amd64.deb

einseenai commented on 2013-07-17 23:07

% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 44.2M 100 44.2M 0 0 521k 0 0:01:26 0:01:26 --:--:-- 596k
-> Found googleearth
-> Found google-earth-mimetypes.xml
==> ERROR: Integrity checks (md5) differ in size from the source array.
==> ERROR: Makepkg was unable to build google-earth.
==> Restart building google-earth ? [y/N]
==> -------------------------------------
==>
==> ERROR: unable to update

slav commented on 2013-07-17 14:07

@Kompilator: thank you very much for your help

so it should be change line in PKGBUILD for:

[ "$CARCH" = 'x86_64' ] && md5sums[0]='82b179c2d4ff0f2eb40ebd5cfea27a1f' # google-earth-stable_$pkgver_amd64.deb

Kompilator commented on 2013-07-17 13:23

Maybe, the line

"$CARCH" = 'x86_64' ] && md5sums[0]='' # google-earth-stable_$pkgver_amd64.deb

is redundant.

slav commented on 2013-07-17 11:44

guys I don't know what I'm doing wrong, can someone tell me please what should be change in PKGBUILD to run google-earth-stable_7.1.1.1888_amd64.deb package?

md5sums=('' # google-earth-stable_$pkgver_i386.deb
'c49e41e2c45802392cebdcb03633fd28' # googleearth
'e3c67b8d05c3de50535bd7e45eee728e') # google-earth-mimetypes.xml

eazar001 commented on 2013-07-16 16:26

@scorici: No problem, I tried inspecting the panoramio object and there is no data availabe, no errors or warnings in the console as well. When trying to reload the contents in the object/widget, it appears as if google default dummy-reports this as "error 404 not found." I am curious as to what the problem is myself. I personally have never heard of Panoramio until you brought up the problem, but maybe I will find out soon [=

eazar001 commented on 2013-07-16 16:25

@scorici: No problem, I tried inspecting the panoramio object and there is no data availabe, no errors or warnings in the console as well. When trying to reload the contents in the objec/widget, it appears as if google default dummy-reports this as "error 404 not found." I am curious as to what the problem is myself. I personally have never heard of Panoramio until you brought up the problem, but maybe I will find out soon [=

Ralf_Mardorf commented on 2013-07-16 12:51

$ yaourt -Syua
[snip]
==> Software upgrade (new version) :
aur/google-earth 7.1.1.1871-1 -> 7.1.1.1888-1
[snip]
-> Found googleearth
-> Found google-earth-mimetypes.xml
==> ERROR: Integrity checks (md5) differ in size from the source array.
==> ERROR: Makepkg was unable to build google-earth.
[snip]

scorici commented on 2013-07-16 12:32

@eazar001 Thanks!
But sadly Panoramio still not working =(

grayich commented on 2013-07-16 11:38

==> ERROR: Integrity checks (md5) differ in size from the source array.

eazar001 commented on 2013-07-16 11:08

For anyone having problems (as well as package maintainer) with md5 64-bit md5sum is: '82b179c2d4ff0f2eb40ebd5cfea27a1', and i386 is: '85655417a249712a9ccd452733d8234c'.

eazar001 commented on 2013-07-16 10:57

@agapito: md5sum is not wrong. Md5 is missing for #google-earth-stable_$pkgver_i386.deb. The i385 md5 is '85655417a249712a9ccd452733d8234c'. For now you can download the tarball, and makepkg -i after extracting. Edit the build and locate that line which has the comment and use that for md5sum if you are i386, otherwise run md5sum on the deb package.

eazar001 commented on 2013-07-16 10:52

@agapito: md5sum is not wrong. Md5 is missing for #google-earth-stable_$pkgver_i386.deb. The i385 md5 is '85655417a249712a9ccd452733d8234c'. For now you can download the tarball, and makepkg -i after extracting. Edit the build and locate that line which has the comment and use that for md5sum if you are i386, otherwise run md5sum on the deb package.

agapito commented on 2013-07-16 07:30

md5sum is wrong.

hutou commented on 2013-07-01 20:16

Just installed GE 7.1.1.1871 today (but first had to install manually ld-lsb) : GE seems to work fine **except** that Panoramio photos are not showing (360 cities are ok) !
Search on Google shows this is a problem potentially related to Qt libraries. Any solution ?

hutou commented on 2013-07-01 20:04

Just installed GE 7.1.1.1871 today (but first had to install manually ld-lsb) : GE seems to work fine **except** that photos are not loading !
Search on Google shows this is a known problem which seems related to Qt libraries. Any solution ?

hutou commented on 2013-07-01 19:59

Just installed GE 7.1.1.1871 today (but first had to install manually ld-lsb) : GE seems to work fine **except** that photos are not loading !
Search on Google shows this is a known problem which seems related to Qt libraries. Any solution ?

Det commented on 2013-07-01 15:15

No, you're using the wrong 'executable'. "/opt/google/earth/free/googleearth" is just a script and "/usr/bin/google-earth" and "/opt/google/earth/free/google-earth" both symlinks to it.

GDB: $ LD_LIBRARY_PATH=/opt/google/earth/free/ gdb /opt/google/earth/free/googleearth-bin
(gdb) run: http://pastebin.com/qbCTznWt

Strace: $ LD_LIBRARY_PATH=/opt/google/earth/free/ strace -f /opt/google/earth/free/googleearth-bin: http://pastebin.com/SYcJA3d6

crashlog-*.txt's are naturally empty.

There's just something wrong with loading the websites (i.e. the renderer or something).

Det commented on 2013-07-01 15:11

Actually it does, since 7.1.1, but it's not very useful:

GDB: $ LD_LIBRARY_PATH=/opt/google/earth/free/ gdb /opt/google/earth/free/googleearth-bin
(gdb) run: http://pastebin.com/qbCTznWt

Strace: $ LD_LIBRARY_PATH=/opt/google/earth/free/ strace -f /opt/google/earth/free/googleearth-bin: http://pastebin.com/SYcJA3d6

crashlog-*.txt's are naturally empty.

There's just something wrong with loading the websites (i.e. the renderer or something).

Anonymous comment on 2013-07-01 10:26

Anyone else having issues with almost everything under the About menu (things like Keyboard Shortcuts and Report Feedback in particular) causing a crash? This is with official Nvidia drivers and a system last updated about a week ago. Even the built-in crash handler crashes; I can't even get a crash trace, and gdb doesn't understand the executable format...

Det commented on 2013-06-29 15:49

Seems like the fontconfig bug is no longer there, but the coordinates regression is.

Det commented on 2013-06-26 09:22

But ld-lsb isn't and I'm not the maintainer of that thing.

thesame commented on 2013-06-26 07:55

Dependency google-earth->ld-lsb->lib32-glibc is still redundant

Det commented on 2013-06-23 13:44

Seems like this thing is actually 64-bit now:

└┌(%:~)┌- objdump -f /opt/google/earth/free/googleearth-bin
/opt/google/earth/free/googleearth-bin: file format elf64-x86-64

Didn't see anything in the changelog: http://support.google.com/earth/bin/answer.py?hl=en&answer=40901

E: By the way, the orphans (unless you use things like Wine or the previous Google Earth) can be removed as shown in the Wiki: https://wiki.archlinux.org/index.php/Pacman_Tips#Removing_orphaned_packages

Det commented on 2013-06-23 13:34

Seems like they're actually releasing 64-bit binaries now:

└┌(%:~)┌- objdump -f /opt/google/earth/free/googleearth-bin
/opt/google/earth/free/googleearth-bin: file format elf64-x86-64

Didn't see anything in the changelog: http://support.google.com/earth/bin/answer.py?hl=en&answer=40901

thesame commented on 2013-06-22 17:03

I've modified ld-lsb PKGBUILD this way:
b
if [ "$CARCH" == 'x86_64' ]; then
ln -sf "ld-linux-x86-64.so.2" "$pkgdir/usr/lib/ld-lsb-x86-64.so.2"
ln -sf "ld-linux-x86-64.so.2" "$pkgdir/usr/lib/ld-lsb-x86-64.so.3"
fi
ln -sf "ld-linux-x86-64.so.2" "$pkgdir/usr/lib/ld-lsb.so.2"
ln -sf "ld-linux-x86-64.so.2" "$pkgdir/usr/lib/ld-lsb.so.3"

And this allowed me to run google earth without lib32-* packages. They might be redundant for this package.

snoxu commented on 2013-06-06 17:28

Getting pixelization in maps and street view.

No such issues with GE 6. Meh.

Intel integrated graphics

Det commented on 2013-05-28 05:34

@a1russell, and what's your beautiful reason to flag this thing?

Det commented on 2013-05-21 23:42

I added the version in the dependency, as too many guys aren't updating their AUR packages, not Googling for the solution and then asking here.

Det commented on 2013-05-21 23:39

It is the ld-lsb, my dear, which you either don't have or haven't updated: https://aur.archlinux.org/packages/ld-lsb/

Nothing at all to do with the Google Earth installation itself.

Firefishe commented on 2013-05-21 20:49

Removed and Built Fresh from AUR, then reinstalled. Detailed Process Here: http://pastie.org/7940739

Firefishe commented on 2013-05-21 20:46

Removed and Build from AUR, then reinstalled. Detailed Process Here: http://pastie.org/7940739

Firefishe commented on 2013-05-21 19:55

I just did a system update on 5-21 (pacman -Syu) and did a complete reinstall of google-earth. When I try to start it I get this output:

$ google-earth
/usr/bin/google-earth: line 18: ./googleearth-bin: No such file or directory

Launching it from /opt/google/earth/free via the command: ./google-earth --- yields the same output as above.

ldd output yields the following:

$ ldd /opt/google/earth/free/googleearth-bin
/usr/bin/ldd: line 116: /opt/google/earth/free/googleearth-bin: No such file or directory

All files in /opt/google/earth/free seem to be present.

Det commented on 2013-05-21 11:06

Yeah, we know..

bred commented on 2013-05-21 10:21

Embedded images are not visible:

http://i.imgur.com/Wycmk9i.jpg
I'm using the x86-64 version

The ldd result is:
http://pastebin.com/xGLPnwnb

Det commented on 2013-05-15 09:06

@tayhe, the right _comment_ should have been: "why yes, Det, your PKGBUILD DOES INDEED include the 64-bit link, which cannot be shown in the front page due to the restrictions of AUR that I didn't know about. I will be off now to file a feature request in the AUR Bug Tracker.

Please note that your PKGBUILD file is fabulous."

tayhe commented on 2013-05-15 02:05

the right downloads url should be :
https://dl.google.com/dl/earth/client/current/google-earth-stable_current_amd64.deb

please fix the pkgbuild file.

dcbdbis commented on 2013-05-15 01:57

Please disregard my previous comment.

I found the error I was receiving at the cli posted on a Google forum, and it's a known issue that is supposed to be fixed in the next release.

FWIW

dcbdbis commented on 2013-05-14 22:01

Help request stabilizing this package?

Ref: https://bbs.archlinux.org/viewtopic.php?id=163328

Det commented on 2013-05-08 11:11

No for two reasons. First of because: no, and secondly because that's the default anyway.

scan commented on 2013-05-08 09:24

-PKGEXT=".pkg.tar"
+PKGEXT=".pkg.tar.xz"

scan commented on 2013-05-08 08:46

-PKGEXT=".pkg.tar"
+PKGEXT=".pkg.tar.xz"

scan commented on 2013-05-08 08:45

-PKGEXT=".pkg.tar"
+PKGEXT=".pkg.tar.xz"

Det commented on 2013-05-03 22:06

That's just the "fake" one, pointing to the actual libs owned by glibc. LSB 4.0 was released back in May 2009.

RaphAstronome commented on 2013-05-03 21:50

On the Google's sites it is writed "Linux : Bibliothèques LSB 4.0 (Linux Standard Base)"
Now the package ld-lsb seems to be in version 3.

If you want I have the google-earth-7.0.3.8542 package (with the PKGBUILD and others) but the URL points only on last version of the software so it doesn't work.

hamelg commented on 2013-05-03 12:09

@luuuciano
Panaramio not showing is a known issue :
http://productforums.google.com/d/topic/earth/4Or9Gleq5BU/discussion

hamelg commented on 2013-05-03 12:08

@luuuciano
Panaramio not showing is a known issue :
http://productforums.google.com/forum/#!searchin/earth/panoramio|sort:date/earth/4Or9Gleq5BU/pAiyw72YW_8J

Det commented on 2013-05-03 01:44

Yeah, that's the one I meant. Visible on the "Sources" section there. And since it has "stable" in its name it's kind of obvious :).

A .bin's fine but it's 5.2. Google still hasn't updated it.

luuuciano commented on 2013-05-03 01:31

Hover it where?
I mean that http://dl.google.com/earth/client/current/$pkgname-stable_current_$_arch.deb always points to the latest version...:
http://dl.google.com/earth/client/current/google-earth-stable_current_i386.deb

OH, btw, just found the advanced setup options on http://www.google.com/earth/download/ge/agree.html
Will try usgin "Previous version (7.0)"

^- Forget that... it downloads a .bin file...

luuuciano commented on 2013-05-03 01:31

Hover it where?
I mean that http://dl.google.com/earth/client/current/$pkgname-stable_current_$_arch.deb always points to the latest version...:
http://dl.google.com/earth/client/current/google-earth-stable_current_i386.deb

OH, btw, just found the advanced setup options on http://www.google.com/earth/download/ge/agree.html
Will try usgin "Previous version (7.0)"

^- Forget that... it downloads a .bin file...

luuuciano commented on 2013-05-03 01:30

Hover it where?
I mean that http://dl.google.com/earth/client/current/$pkgname-stable_current_$_arch.deb always points to the latest version...:
http://dl.google.com/earth/client/current/google-earth-stable_current_i386.deb

OH, btw, just found the advanced setup options on http://www.google.com/earth/download/ge/agree.html
Will try usgin "Previous version (7.0)"

Det commented on 2013-05-03 01:08

Well, you can just hover your mouse over the Source link.

luuuciano commented on 2013-05-03 00:50

But you were right, google releases the "stable" version, without versioning... I could not find a way to find the 7.0.3.8542 source file...

Det commented on 2013-05-03 00:35

Lol, that's so funny, I can see my PKGBUILD changes in the AUR Git clone logs: http://pkgbuild.com/git/aur-mirror.git/log/google-earth/PKGBUILD?id=8a7b59d6f1dc079e5489a63b780f2544527914d1

Totally forgot about this thing.

Det commented on 2013-05-02 23:05

Hmm, yes, that is all I can see. Mount Rushmore sure looks good on the 3D Terrain view, though, lol.

Didn't know it was such a nice app that it can actually show pictures taken by people who were actually there.

luuuciano commented on 2013-05-02 22:45

In the Layers tab (on the left, you have Search, Places, Layers), you have to put ON Photographs/Panoramio (just guessing, I have it on spanish)
Then you will see spots on the ground, of pictures taken by people http://maxcdn.googletutor.netdna-cdn.com/wp-content/uploads/2006/12/google-earth-rushmore1.jpg
You clic over each one, and see the picture, description, etc...

I have uninstalled it and using 6 right now, lol...

I miss a few features... but it works

Det commented on 2013-05-02 22:27

At least yaourt has an option to move finished packages there ("EXPORT=2" in /etc/yaourtrc).

I don't even know what the Panoramio window is that you're talking about :D. If it's the "Earth Gallery >>" thing, then yes, it crashes.

I'm pretty sure it was due to 7.1, which is a beta (but the only 7.x one available for Linux, using even the "stable" URL: http://dl.google.com/earth/client/current/google-earth-stable_current_amd64.deb).

E: Just to add, on 6.2 it works just fine.

Det commented on 2013-05-02 22:15

At least yaourt has an option to move finished packages there ("EXPORT=2" in /etc/yaourtrc).

I don't even know what the Panoramio window is that you're talking about :D. If it's the "Earth gallery >>" thing, then yes, it crashes.

I'm pretty sure it was due to 7.1, which is a beta (but the only 7.x one available for Linux, using even the "stable" URL: http://dl.google.com/earth/client/current/google-earth-stable_current_amd64.deb).

luuuciano commented on 2013-05-02 22:07

Mmmm, but I think that /var/cache/pacman/pkg/ is only official packages, no AUR ones... (I do not see it there, maybe I did a -Scc, do not recall)

The panoramio window works ok on your google-earth build? maybe the problem was another package updated...

Det commented on 2013-05-02 16:11

No, I don't have the previous PKGBUILD for you either and even if I did it'd just download the same thing.

You could either get it from /var/cache/pacman/pkg/, if it's in there or use google-earth6: https://aur.archlinux.org/packages/google-earth6/

luuuciano commented on 2013-05-02 15:07

The panoramio window (where you see pictures after clic on it) is broken on the latest upgrade... :(
It shows a title about the picture location, and nothing else...

BTW, do you have the previous pkgbuild to revert it?

luuuciano commented on 2013-05-02 15:06

The panoramio window (where you see pictures after clic on it) is broken on the latest upgrade... :(
It shows a title about the picture location, and nothing else...

solstice commented on 2013-05-01 13:46

I got it but only ld-lsb 3-4. I upgraded to ld-lsb-3-5 and it worked.
thank you

Det commented on 2013-05-01 10:34

Sure you got ld-lsb?

solstice commented on 2013-04-30 18:59

@Det: Yes that's the same. I have already tried something like that. that's really weird

RaphAstronome commented on 2013-04-29 20:32

No I don't have it in the cache.

I tried to delete the ~/.drirc file (it was empty) like described in https://bbs.archlinux.org/viewtopic.php?id=161677 but it don't works.

I also tried to rename the /etc/fonts/conf.d/65-fonts-persian.conf but without surprise (your package handle this problem) it is not better.

Det commented on 2013-04-29 19:24

Don't you have it in /var/cache/pacman/pkg/?

RaphAstronome commented on 2013-04-29 19:19

It is possible to downgrade to 7.0 ?
The old packages don't work because it download the 7.1 version and detects a checksum error.

Det commented on 2013-04-29 12:03

@RaphAstronome, it's a known bug with 7.1: http://productforums.google.com/forum/#!topic/earth/HU5sNXOU9r8

@solstice, does it work with this?: http://pastebin.com/MxzFeQQ7

solstice commented on 2013-04-28 20:31

I have this rather odd error:
/usr/bin/google-earth: line 26: ./googleearth-bin: No such file or directory

but there IS such a file so I don't understand why I got this problem

RaphAstronome commented on 2013-04-28 16:43

After upgrade, Google earth refuses to start and I get this error message :

[0428/184023:ERROR:net_util.cc(2195)] Not implemented reached in bool net::HaveOnlyLoopbackAddresses()
[0428/184023:WARNING:backend_impl.cc(1875)] Destroying invalid entry.
Google Earth has caught signal 11.

We apologize for the inconvenience, but Google Earth has crashed.
This is a bug in the program, and should never happen under normal
circumstances. A bug report and debugging data have been written
to this text file:

$HOME/.googleearth/crashlogs/crashlog-517d50f7.txt

Please include this file if you submit a bug report to Google.


The crashlog-517d50f7.txt :
Major Version 7
Minor Version 1
Build Number 0001
Build Date Apr 19 2013
Build Time 19:38:48
OS Type 3
OS Major Version 3
OS Minor Version 8
OS Build Version 10
OS Patch Version 0
Crash Signal 11
Crash Time 1367167223
Up Time 0.779381

Stacktrace from glibc:
./libgoogleearth_free.so(+0x23876c)[0x7f2eb474676c]
./libgoogleearth_free.so(+0x2389cd)[0x7f2eb47469cd]
/usr/lib/libpthread.so.0(+0xf0e0)[0x7f2eb53f20e0]
./libbase.so(_ZN5earth15GfxCardInfoUnix25GetGraphicsCardMemoryInMBERi+0xe)[0x7f2eae9ac6fe]
/opt/google/earth/free/libevll.so(_ZN5earth4evll13VisualContext26DetectHardwareCapabilitiesEv+0x450)[0x7f2ea2ec2490]
/opt/google/earth/free/libevll.so(_ZN5earth4evll13VisualContext11OpenContextEN3Gap3Gfx25igRenderDestinationFormatERKNS0_8InitInfoE+0x1f4)[0x7f2ea2ec28d4]
/opt/google/earth/free/libevll.so(_ZN5earth4evll13VisualContext4initERKNS0_8InitInfoE+0x123)[0x7f2ea2ec2a33]
/opt/google/earth/free/libevll.so(_ZN5earth4evll17RenderContextImpl4initERKNS0_8InitInfoE+0xeb)[0x7f2ea2d704db]
./librender.so(_ZN12RenderWidget6SetApiEPN5earth4evll3APIE+0x7a)[0x7f2eabeb1a6a]
./librender.so(_ZN5earth6render12RenderWindow12createWidgetEv+0x17c)[0x7f2eabea0d6c]
./libgoogleearth_free.so(_ZN5earth6client12ModuleWidget9showEventEP10QShowEvent+0x86)[0x7f2eb4726fc6]
./libQtGui.so.4(_ZN7QWidget5eventEP6QEvent+0x599)[0x7f2eb384e299]
./libQtGui.so.4(_ZN19QApplicationPrivate13notify_helperEP7QObjectP6QEvent+0xac)[0x7f2eb37f749c]
./libQtGui.so.4(_ZN12QApplication6notifyEP7QObjectP6QEvent+0x20f)[0x7f2eb37ff73f]
./libQtCore.so.4(_ZN16QCoreApplication14notifyInternalEP7QObjectP6QEvent+0x8c)[0x7f2eb1b21a6c]
./libQtGui.so.4(_ZN14QWidgetPrivate11show_helperEv+0x12a)[0x7f2eb38502ca]
./libQtGui.so.4(_ZN14QWidgetPrivate12showChildrenEb+0xb1)[0x7f2eb38500e1]
./libQtGui.so.4(_ZN14QWidgetPrivate11show_helperEv+0x3f)[0x7f2eb38501df]
./libQtGui.so.4(_ZN7QWidget10setVisibleEb+0x1aa)[0x7f2eb38505da]
./libQtGui.so.4(_ZN14QWidgetPrivate12showChildrenEb+0x146)[0x7f2eb3850176]
./libQtGui.so.4(_ZN14QWidgetPrivate11show_helperEv+0x3f)[0x7f2eb38501df]
./libQtGui.so.4(_ZN14QWidgetPrivate12showChildrenEb+0xb1)[0x7f2eb38500e1]
./libQtGui.so.4(_ZN14QWidgetPrivate11show_helperEv+0x3f)[0x7f2eb38501df]
./libQtGui.so.4(_ZN7QWidget10setVisibleEb+0x1aa)[0x7f2eb38505da]
./libQtGui.so.4(_ZN14QWidgetPrivate12showChildrenEb+0x146)[0x7f2eb3850176]
./libQtGui.so.4(_ZN14QWidgetPrivate11show_helperEv+0x3f)[0x7f2eb38501df]
./libQtGui.so.4(_ZN14QWidgetPrivate12showChildrenEb+0xb1)[0x7f2eb38500e1]
./libQtGui.so.4(_ZN14QWidgetPrivate11show_helperEv+0x3f)[0x7f2eb38501df]
./libQtGui.so.4(_ZN7QWidget10setVisibleEb+0x1aa)[0x7f2eb38505da]
./libQtGui.so.4(_ZN14QWidgetPrivate12showChildrenEb+0x146)[0x7f2eb3850176]
./libQtGui.so.4(_ZN14QWidgetPrivate11show_helperEv+0x3f)[0x7f2eb38501df]
./libQtGui.so.4(_ZN7QWidget10setVisibleEb+0x1aa)[0x7f2eb38505da]
./libQtGui.so.4(_ZN14QWidgetPrivate12showChildrenEb+0x146)[0x7f2eb3850176]
./libQtGui.so.4(_ZN14QWidgetPrivate11show_helperEv+0x3f)[0x7f2eb38501df]
./libQtGui.so.4(_ZN7QWidget10setVisibleEb+0x1aa)[0x7f2eb38505da]
./libQtGui.so.4(_ZN14QWidgetPrivate12showChildrenEb+0x146)[0x7f2eb3850176]
./libQtGui.so.4(_ZN14QWidgetPrivate11show_helperEv+0x3f)[0x7f2eb38501df]
./libQtGui.so.4(_ZN7QWidget10setVisibleEb+0x1aa)[0x7f2eb38505da]
./libQtGui.so.4(_ZN7QWidget10showNormalEv+0x3d)[0x7f2eb3841c0d]
./libgoogleearth_free.so(_ZN10MainWindow18readScreensizeInfoEv+0xbc5)[0x7f2eb4723855]
./libgoogleearth_free.so(_ZN5earth6client11Application12SetupMainWinENS0_3Kvw7ProductEb+0x323)[0x7f2eb47cfc03]
./libgoogleearth_free.so(_ZN5earth6client11Application3runEv+0x700)[0x7f2eb47d66d0]
./libgoogleearth_free.so(+0x238182)[0x7f2eb4746182]
./libgoogleearth_free.so(earthmain+0x1dc)[0x7f2eb474638c]
/usr/lib/libc.so.6(__libc_start_main+0xf5)[0x7f2eb4d59a15]
./googleearth-bin[0x4008b9]

Det commented on 2013-04-27 02:48

Yes.

Anonymous comment on 2013-04-27 02:28

Do you know if Google Earth's 64-bit version require 32-bit libs?

I'd like to install this but am really straining to avoid cluttering up this system with multi-arch dependencies. Thanks.

Det commented on 2013-04-07 19:43

Changing to PKGEXT=".pkg.tar.lzo". I'll be adding the 'pacman>=4.1' dependency, if there's too many complaints.

E: Here's some statistics:

.tar: 117M/0.14s
.tar.lrzip: 28M/0.15s
.tar.lzop: 59M/0.54s
.tar.gz: 42M/6.0s
.tar.xz 28M/51.1s

E2: Oh, well, no, I can't do that, if lzop's provided by [extra] and not even required by pacman (and if I made it a makedep I might as well use lrzip).

Det commented on 2013-04-07 19:39

Changing to PKGEXT=".pkg.tar.lzo". I'll be adding the 'pacman>=4.1' dependency, if there's too many complaints.

E: Here's some statistics:

.tar: 117M/0.14s
.tar.lrzip: 28M/0.15s
.tar.lzop: 59M/0.54s
.tar.gz: 42M/6.0s
.tar.xz 28M/51.1s

E2: Oh, well, no, I can't change it to .tar.lzop, if lzop's in [extra] and not required by pacman (and if I made it a makedep I might as well use lrzip).

Det commented on 2013-04-07 18:10

Changing to PKGEXT=".pkg.tar.lzo". I'll be adding the 'pacman>=4.1' dependency, if there's too many complaints.

Det commented on 2013-04-03 08:22

No, because that would mean you're downloading both.

The AUR Sources parser doesn't understand "$CARCH" statements or alternative links. Makepkg does. That's why the right one will be downloaded anyway (from the PKGBUILD):

_arch=i386 # Workaround for the AUR Web interface Source parser
[ "$CARCH" = 'x86_64' ] && _arch=amd64
source=("$pkgname-stable_$pkgver_$_arch.deb"::"http://dl.google.com/earth/client/current/$pkgname-stable_current_$_arch.deb"

anadon commented on 2013-04-03 04:10

Strike that, it just needs to be listed in the sources.

anadon commented on 2013-04-03 04:04

Shouldn't the 64 bit version be an option?

http://dl.google.com/earth/client/current/google-earth-stable_current_AMD64.deb

czubek commented on 2013-03-20 21:28

Recent updates to various lib32 files prompted me to try google-earth 7 to see if it worked. The program launched but I only got the yellow lines again. Just wanted to test it. GE6 works fine.

Det commented on 2013-03-20 12:36

@eworm, no, my dear, I've responded _overall_ three times. It's just that apparently if I don't include that "@eworm" in the beginning or some other incredibly visible way try to get your attention you miss it altogether.

So here goes one more time:

"The problem is that it's a _Git_ version. If this was just a bad commit it'd be stupid to actually do something about it. I like it the way it is and unless you can _prove_ through an actual released version that I need to have 'em: I won't."

And just so you notice once again that this a post and this post is directed at you:

################### @EWORM, THIS MESSAGE IS DIRECTED AT YOU! ####################

eworm commented on 2013-03-20 12:28

I try to report problems (and even give a solution) and hope to get them changed by the maintainer. I do not like to be ignored. And I feel ignored in three ways:

* You do not include a change for a real problem.
* You do not give a good reason why the change is not done.
* You do not respond at all! (Well, you did after I posted this the third time... But all you said is "No." with a lot of unfriendly stuff around.)

I will shut up now and you should rethink the situation.

Det commented on 2013-03-19 07:23

Lovely guy.

################### @EWORM, THIS MESSAGE IS DIRECTED AT YOU! ####################

No.

#################################################################################

eworm commented on 2013-03-18 19:37

And again...
makepkg (at least the one from pacman git master) has problems with the source line, probably caused by underscores. Please add some brackets and change to:

"${pkgname}-stable_${pkgver}_${_arch}.deb"::"http://dl.google.com/earth/client/current/${pkgname}-stable_current_${_arch}.deb"

Thanks!

Scimmia commented on 2013-03-15 19:54

it's a virtual dependency provided by the lib32 versions of mesa-libgl, nvidia-utils, or catalyst-utils. makepkg -s should have given you those options to satisfy the dep.

Anonymous comment on 2013-03-15 19:50

Hi,
"lib32-libGL" is still as dependency to "google-earth", but does not exist anymore. I'm not able to install.

Det commented on 2013-03-15 12:33

Seems like it alredy is: https://wiki.archlinux.org/index.php/Bumblebee#ERROR:_ld.so:_object_.27libdlfaker.so.27_from_LD_PRELOAD_cannot_be_preloaded:_ignored.

E: Actually it's mentioned overall 3 times in the Wiki: https://wiki.archlinux.org/index.php/Bumblebee#Installation

You could ask the 'lib32-nvidia-utils-bumblebee' maintainer to include those as optional dependencies, though: https://aur.archlinux.org/packages/lib32-nvidia-utils-bumblebee/

Det commented on 2013-03-15 12:05

Seems like it alredy is: https://wiki.archlinux.org/index.php/Bumblebee#ERROR:_ld.so:_object_.27libdlfaker.so.27_from_LD_PRELOAD_cannot_be_preloaded:_ignored.

Det commented on 2013-03-14 09:48

Yeah, already included those in the PKGBUILD.

oliv commented on 2013-03-14 09:44

Hi Det,

Yeah sorry for vmware-patch, I don't have vmware anymore because I use now archlinux as a virtualized system, so it is quite hard to test (I don't even use archlinux kernel because of xen pv thing).

About my comment, the thing is that 32bits gl library is required, but I'm not sure 32bits libraries are installed by default on 64 bits systems (at least not for libgl).

Det commented on 2013-03-11 12:06

Saw it, responded to it, will do so again.

The problem is that it's a _Git_ version. If this was just a bad commit it'd be stupid to actually do something about it. I like it the way it is and unless you can _prove_ through an actual released version that I need to have 'em: I won't.

As a matter of fact, rc1 was just tagged and I see no indication of deprecating variables that aren't separated from the surrounding text with curly brackets: https://mailman.archlinux.org/pipermail/pacman-dev/2013-March/016750.html.

The commit "makepkg: fix incorrect bracket usage" is something totally different: http://gitorious.org/pacman/mainline/commit/2abe1f16461133c0dc386f8ecb8b06865c22940b.

eworm commented on 2013-03-11 08:47

Probably you misses my comment...
makepkg (at least the one from pacman git master) has problems with the source line, probably caused by underscores. Please add some brackets and change to:

"${pkgname}-stable_${pkgver}_${_arch}.deb"::"http://dl.google.com/earth/client/current/${pkgname}-stable_current_${_arch}.deb"

Thanks!

Det commented on 2013-03-09 09:48

I'm just messing with you :D. Of course it's the better way.

I updated google-earth6 as well.

Det commented on 2013-03-08 18:11

There is a downside with automatic installations but I'm just trying to think this through here.

I will let you know when I have something.

Scimmia commented on 2013-03-08 17:52

Why? There's no down sides. That is the correct way of doing it if the program needs libgl. You really think it's better to leave out a required dep entirely?

Det commented on 2013-03-08 17:50

I just said that and I don't have any plans on doing it here.

Scimmia commented on 2013-03-08 16:25

The best way is to include the virtual deps "libgl" and "lib32-libgl", which will prompt the user to pick one if they don't have any installed. Even if it's from the AUR, anything that actually provides libgl should have it in the "provides" array, which would satisfy the dep if it's already installed.

Scimmia commented on 2013-03-08 16:22

The best way is to include the virtual deps "libgl" and "lib32-libgl", which will prompt the user to pick one if they don't have any installed.

Det commented on 2013-03-08 12:47

Oh hey oliv, you're alive. Mind updating 'vmware-patch', by any chance?: https://aur.archlinux.org/packages/vmware-patch/

Anyway, the thing is that, first of all including '(lib32-)mesa-libgl' in the dependencies would mean that everybody would have to install it, even if they already had a proprietary driver (meaning nvidia or catalyst), second, even if I was to include '(lib32-)libgl', provided by (lib32-)nvidia-/catalyst-utils, and let the user choose between the 3, there's still the issue of people wanting to use one from the AUR.

This is why it's easier to just include all in the optdepends=() and let the user decide what he wants to do.

oliv commented on 2013-03-08 10:27

There seems to be a missing dependency for 64 bits systems:

lib32-mesa-libgl

czubek commented on 2013-03-07 22:50

My Intel Corporation Atom Processor D4xx/D5xx/N4xx/N5xx Integrated Graphics Controller fell prey to the hardware limitation. Reverted to the google-earth6 version. All works. Thanks for making it available. I'll keep track of comments to see if version 7 becomes cooperative.

Det commented on 2013-03-07 10:49

And so you didn't figure this was just a bad commit?

eworm commented on 2013-03-07 08:08

makepkg (at least the one from pacman git master) has problems with the source line, probably caused by underscores. Please add some brackets and change to:

"${pkgname}-stable_${pkgver}_${_arch}.deb"::"http://dl.google.com/earth/client/current/${pkgname}-stable_current_${_arch}.deb"

Det commented on 2013-03-06 14:07

@olive, yeah, we all get that the latest Google Earth for Linux is not to your liking. No need to repeat that on every comment.

@Syr, well, you have the same file in there. It's supposed to crash it for everybody.

The 'separate config' idea is pretty neat of course and makes me wonder why I never had it, though, you don't want to be installing in $HOME in case there are multiple users (even if pretty rare for Arch).

I also removed the user configuration lines because same files there can have the same effect. We'll see how it goes. Pkgrel won't be bumped.

Also I've said this a million times but please use Pastebin.

olive commented on 2013-03-06 08:02

The application says that my graphic card is unsupported and I need a 3D graphic card with 16Mb of vram; when I click OK it crashes (same behaviour if I remove the 65-fonts-persian.conf or with Syr trick).

My graphic card (and my computer) is old:

00:02.0 VGA compatible controller: Intel Corporation Mobile 945GM/GMS, 943/940GML Express Integrated Graphics Controller (rev 03)

but I do have 3D and 16Mb of graphic card (the memory is shared and I have 2Gb total). All works fine with google-earth6.

I get tired of this; clearly Google neglects this software (or the Linus version of it). If Google want to be synonymous of quality, it has failed.

Updated: The application starts if I put
enableTips=false
in $HOME/.config/Google/GoogleEarthPlus.conf (and applying Syr trick), but the yellow line issue remains the same as before.

olive commented on 2013-03-06 07:47

The application says that my graphic card is unsupported and I need a 3D graphic card with 16Mb of vram; when I click OK it crashes (same behaviour if I remove the 65-fonts-persian.conf or with Syr trick).

My graphic card (and my computer) is old:

00:02.0 VGA compatible controller: Intel Corporation Mobile 945GM/GMS, 943/940GML Express Integrated Graphics Controller (rev 03)

but I do have 3D and 16Mb of graphic card (the memory is shared and I have 2Gb total). All works fine with google-earth6.

I get tired of this; clearly Google neglects this software (or the Linus version of it). If Google want to be synonymous of quality, it has failed.

Anonymous comment on 2013-03-06 06:02

Having the same crash-on-startup-font-related-problems as everyone, but solved them using a dummy config, with no need to rename or delete any system files.

Create a file in ~/.config/fontconfig/ which may be named fonts.conf.google-earth and put the following inside:

<?xml version='1.0'?>
<!DOCTYPE fontconfig SYSTEM 'fonts.dtd'>
<fontconfig>

<dir>/usr/share/fonts</dir>
<dir>/usr/share/fonts</dir>
<dir prefix="xdg">fonts</dir>
<!-- the following element will be removed in the future -->
<dir>~/.fonts</dir>

<cachedir>/var/cache/fontconfig</cachedir>
<cachedir prefix="xdg">fontconfig</cachedir>
<!-- the following element will be removed in the future -->
<cachedir>~/.fontconfig</cachedir>

</fontconfig>

This is taken from the default /etc/fonts/fonts.conf but omits all the custom conf.d stuff that causes trouble.

Now, modify the /usr/bin/google-earth script and add the following before the actual run:

export FONTCONFIG_FILE=$HOME/.config/fontconfig/fonts.conf.google-earth

This can be easily implemented from the PKGBUILD.

Det commented on 2013-03-04 14:47

Any Intel people noticing improvements besides the yellow borders?: http://support.google.com/earth/bin/answer.py?hl=en&answer=40901

"[...]
Imagery now displays for Linux users running specific families of Intel GPUs.
[...]"

Det commented on 2013-03-04 10:31

That's because it's _out-of-date_.

jpkotta commented on 2013-03-03 20:34

google-earth-stable_amd64.deb md5sum changed to d34b798b2ff2435b431935a134ad35e8

jpkotta commented on 2013-03-03 20:33

.deb md5sum changed to d34b798b2ff2435b431935a134ad35e8

Det commented on 2013-02-26 11:25

http://pastebin.com/raw.php?i=e7b3Bdvw

Anonymous comment on 2013-02-26 10:52

I am running google-earth in a 64bit system and it fails to launch:

Here is the crash log:

========================================================================

Major Version 7
Minor Version 0
Build Number 0002
Build Date Dec 13 2012
Build Time 17:54:43
OS Type 3
OS Major Version 3
OS Minor Version 7
OS Build Version 9
OS Patch Version 0
Crash Signal 11
Crash Time 1361875451
Up Time 0.008445

Stacktrace from glibc:
./libgoogleearth_free.so(+0x1e9fab)[0xf7577fab]
./libgoogleearth_free.so(+0x1ea1f3)[0xf75781f3]
linux-gate.so.1(__kernel_sigreturn+0x0)[0xf7735400]

Det commented on 2013-02-22 19:36

Well it gets too hard. Doesn't seem to get past "21-aliases-wine-win7-inf.conf" without emptying the whole file.

Det commented on 2013-02-21 15:58

K, that makes more sense then. I'll have a look at it some time soon.

You were just repeating so much that was already mentioned in the .install that I thought you had no idea it was already covered.

Btw. You can't just ./ the binary without actually cd'ing to the directory in question beforehand. That's why I just use the one line with the full paths.

Det commented on 2013-02-21 15:48

K, that makes more sense then. I'll have a look at it some time soon.

You were just repeating so much that was already mentioned in the .install that I thought you had no idea it was already covered.

Anonymous comment on 2013-02-21 15:42

The instructions in .install instructed people to remove the links to the offending configuration files. My advice was to edit, rather than to delete. Both 60-group-non-tt-fonts.conf and 60-group-tt-fonts.conf contain lists of fonts that are compatible or not compatible with Infinality; deleting those two configuration files (or the links to them) effectively disables, across the board, most of the improvements Infinality provides. Surgically editing the configuration files to comment out the few fonts that cause problems with Google Earth is preferable to deleting the entire thing.

Det commented on 2013-02-21 13:19

Thank-you for this informative comment.

Except, wasn't this just something I included in the .install LIKE OVER 2 MONTHS AGO?

Anonymous comment on 2013-02-21 13:08

It appears that Google Earth and Infinality do not play nice with each other. There are several fonts in 60-group-tt-fonts.conf and 60-group-non-tt-fonts.conf that, if not commented out, will cause Google Earth to crash on startup.

In particular, I had to comment out :

DokChampa
Dorsa
Sansita One

You should be able to find the offending fonts by doing the following :
LD_LIBRARY_PATH=.:$LD_LIBRARY_PATH strace ./googleearth-bin 2>&1|grep -B5 string

Det commented on 2013-02-08 00:29

Who would've thought. I should probably add this to the package and everything!

JohnnyDeacon commented on 2013-02-07 23:51

@lineage, Sorry for double post, this line in the comment you suggested did the trick! :

"For *ME* removing/renaming /etc/fonts/conf.d/65-fonts-persian.conf solved the crashes. thanks guys!"

JohnnyDeacon commented on 2013-02-07 23:46

@lineage it crash everytime i try to start the program and it's repeatable. I haven't could use the program.



Det commented on 2013-02-07 01:02

It's the same old fontconfig bug already stated a million times - including in the post_install/upgrade.

Some even go as far as stating things like that in the pkgdesc because they're annoyed by people repeatedly complaining about the same thing. I'm just not quite there yet because I think it's even stupider.

lineage commented on 2013-02-06 23:54

Hi JonnyDeacon,
Give us a bit more to go on.. Did it crash before it started-up, or had you been running? Repeatable?

If its start-up, did you follow the instructions when you installed it?
They are down lower in this page on a post with date 2013-01-04 15:42.

Det commented on 2013-02-06 23:46

Where do you guys keep coming from..

JohnnyDeacon commented on 2013-02-06 23:08

Hi All, I'm getting this error while executing google-earth:

Google Earth has caught signal 11.



We apologize for the inconvenience, but Google Earth has crashed.
This is a bug in the program, and should never happen under normal
circumstances. A bug report and debugging data have been written
to this text file:

/home/jairo/.googleearth/crashlogs/crashlog-511072ae.txt

Please include this file if you submit a bug report to Google.

Maybe a bug?

Det commented on 2013-02-06 20:22

Yeah, well, that's why I said the thing about you not being a mobile browser because I just figured you meant the former one when you asked me who has it directed to.

lineage commented on 2013-02-06 20:17

Ha ha ha... Now I see.. FWIW a 'No' at 10:25 would have saved a lot of bandwidth. Languages can be so much fun :-)

Det commented on 2013-02-06 19:55

:D. I meant that back when I posted that link about Google forums I did it with a _mobile browser_ and hit refresh, which resent my post.

Are we clear on that now?

lineage commented on 2013-02-06 19:29

Mobile in the sense that _my_ location changes - true, in the sense that I may use different pc's - true - but only in the sense of _mobile_device_ if you include a conventional i7 laptop in that classification.

Sorry - I thought you were still talking about my GE problems and commenting on the h/ware I was running GE on...

So did I send the same comment twice? - if so its not visible to me, and if I did it was just vanilla firefox and probably me hiting reload.

Det commented on 2013-02-06 18:36

Well, I just thought you didn't think you were a mobile browser.

lineage commented on 2013-02-06 18:36

Det, I really have no idea what you are talking about.

Det commented on 2013-02-06 17:37

My view is that a mobile browser (which is exactly what it says) shouldn't resend your post by refreshing the page without asking you.

What better place in the vast amount of AUR packages' home pages to share that than one of my own.

lineage commented on 2013-02-06 12:19

Not sure what you mean by "mobile browser". This is an ASUS laptop/netbook. I presume you think either ASUS Laptops/netbooks are crap or perhaps the Intel Atom is crap? Either way, they are both a fact of life - and fairly mainstream. However if you hold that view, there are probably better places on the net to share it :-)

Det commented on 2013-02-06 10:25

Yeah.

lineage commented on 2013-02-06 09:36

Hi Det - was that directed at me?

Det commented on 2013-02-06 08:34

Piece of crap mobile browser.

Det commented on 2013-02-06 08:33

They're at http://productforums.google.com/forum/#!categories/earth/Linux where they wouldn't respond to anybody either so doubt this place will make a difference.

Det commented on 2013-02-06 08:31

They're at http://productforums.google.com/forum/#!categories/earth/Linux where they wouldn't respond to anybody either so doubt this place will make a difference.

lineage commented on 2013-02-06 00:34

On the off chance that there is a google-earth lurker here.. gdb suggests that it died doing a dlsym() at the beginning of earthmain. Possibly something to do with "backtrace". Note this is with glibc-2.17 on an Intel Atom CPU N570. But the issue is not present with glibc-2.17 on an Intel Core i7-2640M.

lineage commented on 2013-02-05 22:08

Hi, Tried enableTips in the general section - no change. Tried removing .config - no change - thanks anyway.

olive commented on 2013-02-05 09:57

@lineage For some version of Google earth, you need to disable the Show tips at startup (~/.config/Google/GoogleEarthPlus.conf ; put enableTips=false). Also I have sometimes seen that deleting the Google earth configuration directories (rm -rf ~/.googleearth/ ; rm -rf .config/Google) solves some crashes. I do not think it is Arch fault. Google Earth may rely in some specific version of specific libraries. But all of that is undocumented and the software is closed source. What can Arch do? In my case google-earth6 works fine with the current version of Arch.

olive commented on 2013-02-05 09:56

@lineage For some version of Google earth, you need to disable the Show tips at startup (~/.config/Google/GoogleEarthPlus.conf ; put enableTips=false). Also I have sometimes seen that deleting the Google earth configuration directories (rm -rf ~/.googleearth/ ; rm -rf .config/Google) solves some crashes. I do not think it is arch fault. Google Earth rely may rely ion some specific version of specific libraries. But all of that is undocumented and the software is closed source. What can Arch do?

olive commented on 2013-02-05 09:54

@lineage For some version of Google earth, you need to disable the Show tips at startup (~/.config/Google/GoogleEarthPlus.conf ; put enableTips=false). Also I have sometimes seen that deleting the Google earth configuration directories (rm -rf ~/.googleearth/ ; rm -rf .config/Google) solves some crashes.

lineage commented on 2013-02-05 09:36

Hi Olive, As I said, but maybe not so clearly, google-earth 6 stopped working when I upgraded arch. So neither 6 or 7 work for me. They both crash in what appears from the strace to be the SAME way.

olive commented on 2013-02-05 07:31

All of my problem were magically solved with google-earth6 that Det has kindly packaged. Does the 7 version have something so essential? Google earth is a closed source program and I think the only way it can be fixed is by waiting an updated version from Google. I don't understand though why Google does not take all these problems more seriously, they have a reputation of being very reliable, at least for their services (Gmail, etc.). Releasing a software that's so buggy definitively do not go in that direction.

lineage commented on 2013-02-05 00:27

I'm inclined to agree that it could well be a display driver/mesa issue but without an unstripped version of the executables we can only guess. Unless someone who has got the code wants to offer an opinion :-)
I don't think I can realistically back out the 'X' part of the arch update. So the only option is to wait a couple of weeks and upgrade arch again and hope that fixes it and doesn't break anything else.
Thanks for your help anyway.

Det commented on 2013-02-04 23:23

Handler for SIGSEGV before the futex? Futex_wake_private is a system call for tasks waiting on that specific address so I don't think the SIGSEGV is happening _before_ futex() is actually waking the task to start reading it. That's also the reason I cut it from the second gettimeofday() onward because it or any of those other calls might have something to do with it.

Also as I said I have a feeling that this is just a driver/mesa issue so you installing "handlers" like nvidia-utils wouldn't make a difference.

lineage commented on 2013-02-04 21:25

Still, it looks like there are things happening in your paste with rt_sigaction that are not in mine.. And I think your #-# line should be 12 lines further down after the futex - since that completed (return stat = 0) in mine.. So in your system you had installed a handler for SIGSEGV before the futex - in my case it hadn't.

So there is certainly something wierdly different.

I tried GE6 - exactly the same! Which is very odd.

@olive - In this case its not google - this is Arch that has screwed something up (yet again). GE6 worked before I upgraded arch.

Det commented on 2013-02-04 14:46

Yeah, no, because the app is i386:

└┌(%:~/Desktop)┌- objdump -f /opt/google/earth/free/googleearth-bin

/opt/google/earth/free/googleearth-bin: file format elf32-i386
architecture: i386, flags 0x00000112:
EXEC_P, HAS_SYMS, D_PAGED
start address 0x080486c0

lineage commented on 2013-02-04 13:40

Hi Det, No, I didn't have time to download and build a GE6 pkg this morning. Will try to do it later. Thanks for the paste. I wondered if the 32 [i686 ;-)] vs 64 bit would make any difference.

kolombo commented on 2013-02-04 11:29

I have fixed my crush installing https://launchpadlibrarian.net/7037027/libGL.so.1.
But earth view is completelly grey :)

Det commented on 2013-02-04 11:25

Yes, I have an x86 machine, but my system is 64-bit, if that's what you meant :). Not that it matters, since the app is 32-bit anyway.

I had a look at it yesterday when I compared the two and I can't say it helps (the one with 65-fonts-persian.conf): http://pastebin.com/B7FwBbxV

The hyphen/hash break at line 861 signals the differences after the 3rd gettimeofday() (which in my case is called twice in a row).

E: You had a look at the GE6 log yet?

Det commented on 2013-02-04 11:22

Yes, I have an x86 machine, but my system is 64-bit, if that's what you meant :). Not that it matters, since the app is 32-bit anyway.

I had a look at it yesterday when I compared the two and I can't say it helps (the one with 65-fonts-persian.conf): http://pastebin.com/B7FwBbxV

The hyphen/hash break at line 861 signals the differences after the 3rd gettimeofday() (which in my case is called twice in a row).

lineage commented on 2013-02-04 09:06

Det,
libGL didn't help. Two more datapoints.
Google earth doesn't even get as far as creating a crashlog.
If I run onto a remote X display on another machine I get the same behavior.
If I run the googleearth (6.2) from the remote machine with its display on this machine, it does a lot better, we get spash screen and eventually a crashlog.
If you have this v7 installed on an x86 machine, can you tell me what happens on your system after the futex call? Might be a useful clue.

lineage commented on 2013-02-04 08:45

Thanks Det,

I'll try the libGL. The X driver for the graphics on this CPU has caused me obscure problems before.

If all else fails I'll try v6 too, but I have a bad feeling it won't help.

Det commented on 2013-02-04 00:05

No, it has nothing at all to do with /etc/localtime. There's also the business interest with Windows and keeping your source closed. That's why other people can't fix the black screen problems (which are Intel/Mesa related anyway).

olive commented on 2013-02-03 22:46

Just my thought. How does Google manage to make a so buggy software? It crashes because we have a cyrillic font, it does not show the earth just black instead, and now /etc/localtime??? We could expect more from Google

Det commented on 2013-02-03 22:27

Yeah, that's a really obscure error (posting in here for everyone to see: http://pastebin.com/jgEYLStx) and I'm afraid I've got nothing to go on.

You could try installing https://launchpadlibrarian.net/7037027/libGL.so.1 to /opt/google/earth/free/ and see if it makes a difference (an old workaround for Intel cards).

But I doubt it will.

Det commented on 2013-02-03 22:27

Yeah, that's a really obscure error (posting in here for everyone to see: http://pastebin.com/jgEYLStx) and I'm afraid I've got nothing to go on.

You could try installing http://librarian.launchpad.net/7037027/libGL.so.1 to /opt/google/earth/free/ and see if it makes a difference (an old workaround for Intel cards).

But I doubt it will.

Det commented on 2013-02-03 21:41

No I meant the actual log with /etc/localtime and that pastebin's the tool you should be using for them anyway.

lineage commented on 2013-02-03 21:35

Tried running under gdb to see if that would generate any more clues.. Not too useful, but here it is anyway.

Program received signal SIGSEGV, Segmentation fault.
0x00000000 in ?? ()
(gdb) bt
#0 0x00000000 in ?? ()
#1 0xb7fff4c4 in _rtld_local () from /lib/ld-lsb.so.3
#2 0xb7f98ed4 in dlsym () from /usr/lib/libdl.so.2
#3 0xb7e2ac55 in earthmain ()
from /opt/google/earth/free/libgoogleearth_free.so
#4 0x0804879b in ?? ()
#5 0xb5148825 in __libc_start_main () from /usr/lib/libc.so.6
#6 0x080486e1 in ?? ()

lineage commented on 2013-02-03 21:10

Ah sorry.. I had forgotten I tried removing localtime - but no change. Same SEGFAULT.
It looks like something related to futex.
the AUR google-earth has worked on here before, but I recently upgraded arch and google-earth. I have a feeling that the arch upgrade may have broken the old version but I'm not 100% sure.

Not sure where else I would have posted it for you to see :-)

Det commented on 2013-02-03 20:51

Huh. And _with_ /etc/localtime? Does the previous one work? Did they used to?

I was never meaning for you to post it in here, by the way :D.

lineage commented on 2013-02-03 19:13

Try http://pastebin.com/Em5L2H74

lineage commented on 2013-02-03 18:58

About 70k - you looking for something specific?

Det commented on 2013-02-03 17:53

And the full output? Of "LD_LIBRARY_PATH=/opt/google/earth/free/:$LD_LIBRARY_PATH strace /opt/google/earth/free/googleearth-bin" that is.

There's also the previous one here for those unlucky ones who have no other choice: https://aur.archlinux.org/packages/google-earth6/

lineage commented on 2013-02-03 16:07

Not for me.
Removed 65-fonts-persian.conf - no change
Tried the 'further debug' option and the last thing it did was open /etc/localtime then closed it then did
gettimeofday({1359907137, 574974}, NULL) = 0
futex(0xb7726058, FUTEX_WAKE_PRIVATE, 2147483647) = 0
--- SIGSEGV {si_signo=SIGSEGV, si_code=SEGV_MAPERR, si_addr=0} ---

Det commented on 2013-02-01 21:52

and me

pavelchavyr commented on 2013-02-01 16:13

me too

mrbit commented on 2013-02-01 15:49

For *ME* removing/renaming /etc/fonts/conf.d/65-fonts-persian.conf solved the crashes.

Det commented on 2013-01-30 06:02

Just going out on a limb here but reading the instructions would probably help.

katafractari commented on 2013-01-29 23:17

I also have the same problem as canurabus and testdude281:

Crash log:
Major Version 7
Minor Version 0
Build Number 0002
Build Date Dec 13 2012
Build Time 17:54:43
OS Type 3
OS Major Version 3
OS Minor Version 7
OS Build Version 4
OS Patch Version 0
Crash Signal 11
Crash Time 1359501189
Up Time 0.004529

Stacktrace from glibc:
./libgoogleearth_free.so(+0x1e9fab)[0xf75dffab]
./libgoogleearth_free.so(+0x1ea1f3)[0xf75e01f3]
linux-gate.so.1(__kernel_sigreturn+0x0)[0xf779d400]

I've deleted .googleearth directory but it's still crashing.

katafractari commented on 2013-01-29 23:16

I'm also have the same problem as canurabus and testdude281:

Crash log:
Major Version 7
Minor Version 0
Build Number 0002
Build Date Dec 13 2012
Build Time 17:54:43
OS Type 3
OS Major Version 3
OS Minor Version 7
OS Build Version 4
OS Patch Version 0
Crash Signal 11
Crash Time 1359501189
Up Time 0.004529

Stacktrace from glibc:
./libgoogleearth_free.so(+0x1e9fab)[0xf75dffab]
./libgoogleearth_free.so(+0x1ea1f3)[0xf75e01f3]
linux-gate.so.1(__kernel_sigreturn+0x0)[0xf779d400]

I've deleted .googleearth directory but it's still crashing.

Det commented on 2013-01-27 22:39

No you didn't.

canurabus commented on 2013-01-27 22:38

I have the same issue as testdude281, with the same crash log. The google-earth6 package doesn't work for me either... I posted a comment on that package's page.

Det commented on 2013-01-27 21:43

What did I just say?

luismy commented on 2013-01-27 21:24

I fear that these options do not work 100%, the differences are minimal between them and the default.

Det commented on 2013-01-25 19:24

Yeah, the "-style GTK|plastique|cleanlooks" flags have some minor effect but I don't know to what extent does it follow your current theme.

luismy commented on 2013-01-25 00:39

Hi Det.

Is possible make who Google-Earth use a Gtk theme?

I saw an option in ubuntu that seems to work, at least in older versions but does not work in Arch:

http://www.omgubuntu.co.uk/2009/04/make-google-earth-use-gtk

Det commented on 2013-01-24 06:33

I could.

But that would defeat the purporse of not doing it, of course.

testdude281 commented on 2013-01-23 23:27

google earth just crashed on me (Signal 11). can you take a look at the crash log?

http://pastebin.com/sQVB9hiQ

I had this problem previously when i upgraded but i solved it from the forums. So i dont know why it has come back.

bananabrain commented on 2013-01-22 03:12

Clever boy.
Works a peach.
Thank you.

Det commented on 2013-01-20 21:46

Oh, aren't you sweet.

Well. I'll either have it downloaded manually or just link it in the license.

bananabrain commented on 2013-01-20 20:56

Hard URLs, I guess...

http://earth.google.com/intl/en-US/license.html
http://www.google.com/intl/en_uk/earth/license.html

Det commented on 2013-01-20 20:51

Seems that way.

bananabrain commented on 2013-01-20 20:43

Not at all. I've put this on a few machines over the last couple of years and never a hitch with checksums or licenses until now. Maybe google have changed their approach?

Det commented on 2013-01-20 14:11

Wow. That's so different from mine.

This the first time you're installing Google Earth, then?

bananabrain commented on 2013-01-19 21:20

Cheers Det. I'm in England. Your first sentence goes without saying, but I'm embarassed to say I hadn't considered the obvious - different licences for different countries.

Here's the text:
http://pastebin.com/wVDdb3aM

Det commented on 2013-01-19 19:44

Thank-you, for the pastebin, but I still hate you.

marcinfa commented on 2013-01-19 15:51

When I run the program always end this:
Google Earth has caught signal 11.- terminal info
log:
http://pastebin.com/CNYHa1sK

marcinfa commented on 2013-01-19 15:45

Major Version 7
Minor Version 0
Build Number 0002
Build Date Dec 13 2012
Build Time 17:54:43
OS Type 3
OS Major Version 3
OS Minor Version 6
OS Build Version 11
OS Patch Version 0
Crash Signal 11
Crash Time 1358610230
Up Time 0,005714

Stacktrace from glibc:
./libgoogleearth_free.so(+0x1e9fab)[0xf7606fab]
./libgoogleearth_free.so(+0x1ea1f3)[0xf76071f3]
linux-gate.so.1(__kernel_sigreturn+0x0)[0xf77d7400]

Det commented on 2013-01-19 08:21

Well, installing a license, even with a bad md5sum, doesn't do squat to the actual application so you're kind of free to install it even if it doesn't match.

The problem is that you're downloading it from a different server, which gives you a different license than mine. I've included the license in both packages for now.

You mind uploading yours and telling me the country you're getting it from?

bananabrain commented on 2013-01-19 02:26

It matches just fine?

License md5 in PKGBUILD for v6 and v7 says 7363c6144ebb298b1d7aec713ca8a82a

I get d414602116908d737207d905cff261b2 no matter how many times I download it.

I could edit your script, but I'd rather you told me the error of my ways...

Det commented on 2013-01-18 13:10

Try to keep comments about separate packages in their equilevant pages. As to what comes to your error, just redownload the license. It matches just fine.

lemmy999 commented on 2013-01-18 12:17

Install failed with
"Validating source files with md5sums...
googleearth_6.2.2.6613-r0_amd64.deb ... Passed
license.html ... FAILED
google-earth6-mimetypes.xml ... Passed
==> ERROR: One or more files did not pass the validity check!"

Det commented on 2013-01-18 07:00

Nothing?

luismy commented on 2013-01-18 01:21

The package "google-eart6" work OK for me too.

What happens to the current version???

olive commented on 2013-01-17 14:27

Thanks a lot. The 6 version works great for me.

Det commented on 2013-01-17 14:11

https://aur.archlinux.org/packages/google-earth6/

Det commented on 2013-01-17 13:10

No.

olive commented on 2013-01-17 08:15

I see that I am not the only one with problems with this version.
I have tried to download the previous version (6.x) of google-earth (which did work properly) from Google, but although they say the previous version is 6.2, what you got is actually 5.x ?! Is 6.2 available somewhere? Would Det be so kind as to provide a package for it.

olive commented on 2013-01-17 08:08

I see that I am not the only one with problems with this version.
I have tried to download the previous version (6.x) of google-earth (which did work properly) from Google, but although they say the previous version is 6.2, what you got is actually 5.x ?! Is 6.2 available somewhere? Would Det be so kind as to provide a package for it.

luismy commented on 2013-01-11 13:33

Same problem than Olive in three computers:

After remove /etc/fonts/conf.d/65-fonts-persian.conf I only can see continents lines and other information layers, but I can't see the planet in background.

My computer's video cards are: Intel 945GM, NVidia ION and NVidia Geforce 6200

All NVidia cards are with propietary drivers working ok in another applications.


-------------

Sorry for my bad English.

mrbit commented on 2013-01-11 10:32

For *ME* removing/renaming /etc/fonts/conf.d/65-fonts-persian.conf solved the crashes.

Det commented on 2013-01-07 15:46

Of COURSE you did, stop SAYING it!

Anonymous comment on 2013-01-07 14:27

In my case, I also had to rename:
/etc/fonts/conf.avail/65-fonts-persian.conf

olive commented on 2013-01-06 11:15

After having removed /etc/fonts/conf.d/65-fonts-persian.conf, it starts but the earth remains black (with only the continents in yellow) (Intel Corporation Mobile 945GM/GMS, 943/940GML with the intel Xorg driver).

Det commented on 2013-01-04 15:55

Lol. Yeah. You should do that.

drdrewdown commented on 2013-01-04 15:42

*TO ANYONE HAVING PROBLEMS*

Fought this for a while, be sure to read the output after install.

[code]
NOTE: The binary is called 'google-earth'
NOTE2: To prevent a crash on startup with -dri drivers you may need to add a new line to ~/.drirc with:
$ echo >> ~/.drirc
NOTE3: Certain font confs are known to crash Google Earth at startup. These include the following:
* /etc/fonts/conf.d/65-fonts-persian.conf (owned by 'fontconfig')
* /etc/fonts/infinality/styles.conf.avail/infinality/21-aliases-wine-win7-inf.conf
* /etc/fonts/infinality/styles.conf.avail/infinality/60-group-non-tt-fonts.conf
* /etc/fonts/infinality/styles.conf.avail/infinality/60-group-tt-fonts.conf
(all three owned by 'fontconfig-infinality')
Please either remove or rename any of these to prevent them from being parsed by expat.
NOTE4: Other such confs may be found one at a time by debugging (requires 'strace'):
$ echo; LD_LIBRARY_PATH=/opt/google/earth/free/:$LD_LIBRARY_PATH strace /opt/google/earth/free/googleearth-bin 2>&1 | grep -B10 SEGV_MAPERR | grep open | cut -d '"' -f2; echo

Optional dependencies for google-earth
lib32-ati-dri: For the open source ATI driver
lib32-catalyst-utils: For AMD Catalyst
lib32-gtk2: SCIM support
lib32-intel-dri: For the open source Intel driver
lib32-nouveau-dri: For the open source Nouveau driver
lib32-nss-mdns: In case the application fails to contact the servers
lib32-nvidia-utils: For the NVIDIA driver
lib32-nvidia-utils-bumblebee: For the NVIDIA driver + Bumblebee users
qt: For changing the font size with qtconfig
ttf-ms-fonts: Fonts
[/code]

For *ME* removing/renaming /etc/fonts/conf.d/65-fonts-persian.conf solved the crashes. thanks guys!

Anonymous comment on 2013-01-03 14:09

Also crashed here (ATI card, Arch x64), but after removing

/etc/fonts/conf.d/65-fonts-persian.conf

it works. Could be faster but its usable.

Det commented on 2012-12-23 14:03

And now I've fixed the coordinates regression.

rukolonist commented on 2012-12-22 15:12

Now I disabled discrete ATI adapter in BIOS, removed ATI drivers and configured Xorg as I had only Intel video adapter. But it still crashes the same way.

Det commented on 2012-12-22 11:43

Yeah, I saw that (http://productforums.google.com/forum/#!msg/earth/dlzBfGl4eKM/723naNYBo30J) but didn't care enough to do anything about it. I'll play around with the LC's and the -style's once I get back.

Berseker commented on 2012-12-22 10:38

DEt, only yo highlight that We have again the old problem about wrong coordinates for some "locales" see here http://productforums.google.com/forum/#!topic/earth/h4H2KR5i3qE

Berseker commented on 2012-12-22 10:25

det, thank you for your tips. Now GE 7 works on my system (had to remove all 4 .confs mentioned in post_install()

luspi commented on 2012-12-22 08:45

I have the exact same problems as rukolonist. Also have an integrated Intel video adapter (in use) and a discrete ATI video adapter. Everything else (3d-wise, etc.) is working just fine...

Det commented on 2012-12-20 20:53

K. Good luck.

rukolonist commented on 2012-12-20 20:43

First I wrote to Google forums and only after that I started to write here. Anyway thanks for your advice.

I use standard kernel from "core", "lib32-libx11" from "multilib", "google-earth" from your package. No modifications or non-standard AUR-packages.

Det commented on 2012-12-20 20:35

Of which the strace log doesn't say anything useful of course..

I'm kind of stumped at this point. Your best bet would probably be to ask directly in the Google forums: http://productforums.google.com/forum/#!categories/earth/linux. I doubt you'll get responded but I don't think I can help you any further either.

I would probably try reinstalling some of my packages. You're not using a custom kernel or modifying the "lib32-libx11" package nor the 'google-earth' main script either, right?

Det commented on 2012-12-20 20:34

Of which the strace log doesn't say anything useful of course..

I'm kind of stumped at this point. Your best bet would probably be to ask directly in the Google forums: http://productforums.google.com/forum/#!categories/earth/linux. I doubt you'll get responded but I don't think I can help you any further either.

You're not using a custom kernel or modifying the "lib32-libx11" package nor the 'google-earth' main script either, right? I would probably try reinstalling some of my packages.

rukolonist commented on 2012-12-20 20:16

I don't know other 3D applications to test, but glxgears works fine at about 60 FPS. OpenGL screensavers work fine too.

I tried installing "lib32-nvidia-cg-toolkit" but it does not help. This is strace with "lib32-nvidia-cg-toolkit" installed: http://pastebin.com/raw.php?i=DKHRek99

This is gdb log: http://pastebin.com/HdfTc8cT and it seems that segmentation fault appears in /usr/lib32/libX11.so.6. File exists and it links to libX11.so.6.3.0. After installing "lib32-nvidia-cg-toolkit" gdb output is the same.

Det commented on 2012-12-20 19:58

Oh, so it says xD. "(No such file or directory)".

Well, what does the GDB output tell us?:
$ LD_LIBRARY_PATH=/opt/google/earth/free/ gdb /opt/google/earth/free/googleearth-bin
(gdb) run

Also are you sure you have the required drivers installed? Does any other 3D stuff work as intended? You could even try going so far as to actually installing that "lib32-nvidia-cg-toolkit" if it's really insisting on it and then reposting the strace log.

rukolonist commented on 2012-12-20 19:49

Intresting is that I don't have "/usr/lib32/libCg.so" because package "lib32-nvidia-cg-toolkit" is not installed.

I have integrated Intel video adapter (active) and discrete ATI video adapter:
$ lspci | grep VGA
00:02.0 VGA compatible controller: Intel Corporation 2nd Generation Core Processor Family Integrated Graphics Controller (rev 09)
01:00.0 VGA compatible controller: Advanced Micro Devices [AMD] nee ATI Caicos [Radeon HD 6400M/7400M Series]

Det commented on 2012-12-20 19:35

Yeah. Well, this is clearly not a font conf crash, as already proven by the fact that you say this used to happen with 6.2 as well.

But on the other hand I'm a bit worried by your "Resource temporarily unavailable" messages :D. I don't know if they have anything to do with the crash but it's the first time I've seen such notices in a GE strace output.

Could you try to rename "/usr/lib32/libCg.so" (owned by 'lib32-nvidia-cg-toolkit'), as it seems to be the one where the seg fault happens? I tried to install the package myself but GE just started up without never getting as far as trying to load it (though neither did it go looking for anything in "/var/cache/fontconfig/" or "/usr/share/fonts/").

rukolonist commented on 2012-12-20 18:48

Sorry, sorry :) This is the full output: http://pastebin.com/raw.php?i=hhamNqdA

Det commented on 2012-12-20 18:38

I meant the full output of "LD_LIBRARY_PATH=/opt/google/earth/free/:$LD_LIBRARY_PATH strace /opt/google/earth/free/googleearth-bin" (I don't think you want to remove ld.so.cache anyway xD).

rukolonist commented on 2012-12-20 18:27

Det > And do you 'cd' to the actual directory first? What's the full output?
Of course. This is commands I run:
$ cd /opt/google/earth/free/
$ LD_LIBRARY_PATH=.:$LD_LIBRARY_PATH strace ./googleearth-bin 2>&1 | grep -B10 SEGV_MAPERR | cut -d '"' -f2 | grep /etc/ | tail -1

The output is:
/etc/ld.so.cache

When I remove /etc/ld.so.cache, then the command outputs nothing. But it still crashes during splash screen.

Content of crashlog-50d358bc.txt: http://pastebin.com/kbD4jUvd

msx commented on 2012-12-20 11:56

Runs smooth here, many thanks @Det.
SysSpecs: x86_64 fully updated (what else) / Intel HD Graphics Arrendale (8086:0046)

Det commented on 2012-12-19 22:08

And do you 'cd' to the actual directory first? What's the full output?

rukolonist commented on 2012-12-19 21:00

I removed fonts links, but it still crashes just after or during splash screen.

When I use:
LD_LIBRARY_PATH=.:$LD_LIBRARY_PATH strace ./googleearth-bin 2>&1 | grep -B10 SEGV_MAPERR | cut -d '"' -f2 | grep /etc/ | tail -1

then it returns nothing.

Crashlog contains:
Stacktrace from glibc:
./libgoogleearth_free.so(+0x1e9fab)[0xf761dfab]
./libgoogleearth_free.so(+0x1ea1f3)[0xf761e1f3]
linux-gate.so.1(__kernel_sigreturn+0x0)[0xf77ab400]


I use uptodated Arch x86_64. The same problem was with the 6th version of GoogleEarth.

Det commented on 2012-12-19 20:33

Well, if the following doesn't work I don't know what will:

$ cd /opt/google/earth/free/
$ LD_LIBRARY_PATH=.:$LD_LIBRARY_PATH strace ./googleearth-bin 2>&1 | grep -B10 SEGV_MAPERR | cut -d '"' -f2 | grep /etc/ | tail -1

I modified some other stuff there as well. Just let me know if the output is getting too wide.

Det commented on 2012-12-19 20:33

2012-12-19 20:28
Well, if the following doesn't work I don't know what will:

$ cd /opt/google/earth/free/
$ LD_LIBRARY_PATH=.:$LD_LIBRARY_PATH strace ./googleearth-bin 2>&1 | grep -B10 SEGV_MAPERR | cut -d '"' -f2 | grep /etc/ | tail -1

I modified some other stuff there as well. Just let me know if the output is getting too wide.

kerneliq commented on 2012-12-19 20:31

@Peace4all Thank you! Everything was more difficult, but your way to help.

Det commented on 2012-12-19 20:28

Well, if the following doesn't work I don't know what will:

$ cd /opt/google/earth/free/
$ LD_LIBRARY_PATH=.: strace ./googleearth-bin 2>&1 | grep -B10 SEGV_MAPERR | cut -d '"' -f2 | grep /etc/ | tail -1

I modified some other stuff there as well. Just let me know if the output is getting too wide.

Peace4all commented on 2012-12-19 17:09

I found the following font confs. caused problems for me: http://pastebin.com/NK2jpgAR.
@ kerneliq, you can you try using 'LD_LIBRARY_PATH=.:$LD_LIBRARY_PATH strace ./googleearth-bin 2>&1 | grep -B10 SEGV_MAPERR | grep fonts', I used this to find mine, as in this example: http://pastebin.com/RvcUXpK4. keep running that line, and renaming the conf. files, until google-earth starts.

Det commented on 2012-12-19 12:31

The hell?

kerneliq commented on 2012-12-19 07:12

@Det \t\t<test name=\

Det commented on 2012-12-19 01:59

@kerneliq, lol. Well it's the "/etc/fonts/infinality/conf.d/60-group-non-tt-fonts.conf" in your case. I included the last 5 lines before the SEGV_MAPERR in the strace script again. It should be working now.

But so for anybody who doesn't wanna rebuild (or 're-repackage') this thing, the full command "sequence" is now:
$ cd /opt/google/earth/free/
$ LD_LIBRARY_PATH=.:$LD_LIBRARY_PATH strace ./googleearth-bin 2>&1 | grep -B5 SEGV_MAPERR | cut -d '"' -f2 | head -1

@bred, 1) _read_ the post_upgrade and 2) _don't_ paste such long logs here. Use pastebin or github.

E: Uhh.. and of course it's '"' instead of '\"', since we're not echoing anything.

Det commented on 2012-12-18 22:29

@kerneliq, lol. Well it's the "/etc/fonts/infinality/conf.d/60-group-non-tt-fonts.conf" in your case. I included the last 5 lines before the SEGV_MAPERR in the strace script again. It should be working now.

But so for anybody who doesn't wanna rebuild (or 're-repackage') this thing, the full command "sequence" is now:
$ cd /opt/google/earth/free/
$ LD_LIBRARY_PATH=.:$LD_LIBRARY_PATH strace ./googleearth-bin 2>&1 | grep -B 5 SEGV_MAPERR | cut -d '\"' -f2 | head -1

@bred, 1) _read_ the post_upgrade and 2) _don't_ paste such long logs here. Use pastebin or github.

bred commented on 2012-12-18 21:13

It crash!

Google Earth has caught signal 11.



We apologize for the inconvenience, but Google Earth has crashed.
This is a bug in the program, and should never happen under normal
circumstances. A bug report and debugging data have been written
to this text file:

/home/myself/.googleearth/crashlogs/crashlog-50d0dc08.txt

Please include this file if you submit a bug report to Google.


---------------------------------------------------------------------

Major Version 7
Minor Version 0
Build Number 0002
Build Date Dec 13 2012
Build Time 17:54:43
OS Type 3
OS Major Version 3
OS Minor Version 6
OS Build Version 10
OS Patch Version 0
Crash Signal 11
Crash Time 1355865096
Up Time 0,004915

Stacktrace from glibc:
./libgoogleearth_free.so(+0x1e9fab)[0xf7597fab]
./libgoogleearth_free.so(+0x1ea1f3)[0xf75981f3]
linux-gate.so.1(__kernel_sigreturn+0x0)[0xf775d400]

kerneliq commented on 2012-12-18 20:59

Det!
http://pastebin.com/Tmdzeejs

kerneliq commented on 2012-12-18 20:54

Det!
http://pastebin.com/Tmdzeejs

Det commented on 2012-12-18 18:18

Updated to 7.0.2. The strace command is also mentioned in the .install.

I decided not to include it in the main script nor run it automatically for anyone.

Det commented on 2012-12-17 13:29

Actually the following should print the segfaulting conf for everybody (the echos are just for clarification):

echo; cd /opt/google/earth/free/ && export LD_LIBRARY_PATH=. && strace ./googleearth-bin &> /tmp/strace; grep -B 4 SEGV_MAPERR /tmp/strace | cut -d '"' -f2 | head -1; echo

(E: Obviously, you need to have 'strace' installed.)

Det commented on 2012-12-17 13:25

Actually the following should print the segfaulting conf for everybody:

echo; cd /opt/google/earth/free/ && export LD_LIBRARY_PATH=. && strace ./googleearth-bin &> /tmp/strace; grep -B 4 SEGV_MAPERR /tmp/strace | cut -d '"' -f2 | head -1; echo

I added the echos because it's just clearer that way.

Det commented on 2012-12-16 16:26

It's the "/etc/fonts/infinality/conf.d/21-aliases-wine-win7-inf.conf" (SIGSEGV with "/usr/lib/libexpat.so.1" when parsing it).

Remove/rename it and see what happens.

E: Also I actually already uploaded the new GE7 tarball (.8283) for you earlier this week: http://www25.zippyshare.com/v/79875837/file.html
E2: Oh, you already said you tried it. Well it's the same deb being fetched there so you don't need to do that yourself.

Det commented on 2012-12-16 16:08

It's the "/etc/fonts/infinality/conf.d/21-aliases-wine-win7-inf.conf" (SIGSEGV with "/usr/lib/libexpat.so.1" when parsing it).

Remove/rename it and see what happens.

E: Also I actually already uploaded the new GE7 tarball (.8283) for you earlier this week: http://www25.zippyshare.com/v/79875837/file.html

Berseker commented on 2012-12-16 08:06

thank you for your help:

installed 7.0.1.8283 using this deb -> http://ubuntuone.com/6uRHYNDWYonQPbY7Fs3GVY

and this PKGBUILD (I moved manually the downloaded .deb into the PKGBUILD dir
https://gist.github.com/4304135

after that,

GDB: http://pastebin.com/iDZ6S22X
Strace: http://pastebin.com/T70WGSHG
crashlog https://gist.github.com/4304164

Det commented on 2012-12-15 17:25

Uhhh.. I might not be able to help you but could you post the results of:

1) GDB: cd /opt/google/earth/free/ && export LD_LIBRARY_PATH=. && gdb ./googleearth-bin
(gdb) run

2) strace: cd /opt/google/earth/free/ && export LD_LIBRARY_PATH=. && strace ./googleearth-bin

3) the GE crash log (~/.googleearth/crashlogs/crashlog-XXXXXXXX.txt)

..in _PasteBin_?: http://pastebin.com/

And yeah, install 'gdb' and 'strace' first, of course.

Berseker commented on 2012-12-15 06:43

yes :(

Det commented on 2012-12-12 11:14

You actually tried what was suggested in the post upgrade?

Berseker commented on 2012-12-11 19:44

so there is no solution yet for ge7 on arch? I also tried the other day to use that tarball, but without success (still signal 11)

Det commented on 2012-12-11 16:25

Apparently somebody was kind enough to redistribute the .8283 in Ubuntu One: http://www.webupd8.org/2012/12/download-google-earth-7-with-fixed.html

You can get the tarball from here: http://www25.zippyshare.com/v/79875837/file.html

Still crashes on the persian conf, though.

msx commented on 2012-12-08 05:28

Hey @Det, thanks a lot for packaging GE!
Everyone else: be patient, "Google Earth for Linux Fix Coming ‘Shortly’"
http://www.omgubuntu.co.uk/2012/12/google-earth-for-linux-signal-11-fix-coming

msx commented on 2012-12-08 05:28

Hey @Det, thanks a lot for packaging GE!
Everybody else: be patient, "Google Earth for Linux Fix Coming ‘Shortly’"
http://www.omgubuntu.co.uk/2012/12/google-earth-for-linux-signal-11-fix-coming

Det commented on 2012-12-07 12:33

It's not out of date, my dear. The optional slow, buggy 7.0 beta is still here: http://www29.zippyshare.com/v/24125743/file.html

Det commented on 2012-11-29 10:02

You either don't have the same problem or you're not using 6.2.

kolombo commented on 2012-11-29 09:54

Same problem with `6.2.2.6613-1`, I believe it worked well before.
65-fonts-persian.conf is removed.

Det commented on 2012-11-29 09:49

That _is_ the same one.. which is sincerely quite interesting since the "65-fonts-persian.conf" should be the only one expat is failing to parse.

You could go thourgh some other suggestions in here. See if it makes a difference (or just stick to 6.2): http://productforums.google.com/forum/#!topic/earth/dlzBfGl4eKM/discussion

mrechte commented on 2012-11-29 09:42

Well, sort of:

Stacktrace from glibc:
./libgoogleearth_free.so(+0x1e9cfb)[0xf75e0cfb]
./libgoogleearth_free.so(+0x1e9f43)[0xf75e0f43]
linux-gate.so.1(__kernel_sigreturn+0x0)[0xf7796400]

Det commented on 2012-11-29 09:36

I just re-uploaded 6.2 using the Mint repositories as that's apparently the only place in the world I could find it.

The tarball for the new laggy 7.0 beta is still here, in case someone insists on it: http://www29.zippyshare.com/v/24125743/file.html

Det commented on 2012-11-29 09:20

The _same_ one? You sure of that?

mrechte commented on 2012-11-29 07:17

I removed the /etc/fonts/conf.d/65-fonts-persian.conf link. Now I just get the GE splash screen followed by the same error.

TheWretched commented on 2012-11-28 17:24

I don't always stab myself, but when I do I do it with a need for more than 4Gb of ram.

TheWretched commented on 2012-11-28 17:11

I don't always stab myself, but when I do I do it with a need for more than 4Gb of ram.

Det commented on 2012-11-28 17:08

Stab yourself inside it? That would be quite a trick.

But thanks for the suggestion. I might just try these out one day.

TheWretched commented on 2012-11-28 17:03

I thought it was a shame sharpfonts is down, so here's the sharpfonts files in the interim.

http://support.mcsnet.ca/sharpfonts/sharpfonts.tar

In case anyone want's to try what I mentioned below.

TheWretched commented on 2012-11-28 16:55

For what it's worth.

I was just loading a new system and came across this package and it's problems.

It seems to be working fine for me (knock on wood) with my typical setup and I thought I would mention it just for the sake of it, in case it helps anyone.

I am using KDE and the latest nvidia proprietary drivers 8xxx series hardware. My font configuration is set up in the way that I install the Windows 7 fonts, and use the sharpfonts.co.cc fontconfig files.

Since sharpfonts.co.cc is down, and the aur package for sharpfonts also doesn't work because of this it's a bit of a pickle, I had them saved on my ftp however. I modified the ms rules.conf to one I saw linked in the aur sharpfonts package as well, and placed them all under the /etc/fonts/conf.d/ folder as 99-foo.conf.

I don't really know if using the MS fonts with rules has anything to do with it running seemingly quite well. But it does work and is the only way I have ever been able to get fonts that don't look like they make me want to stab myself in 64 bit arch.

Det commented on 2012-11-28 15:22

You've motherf*cking got to be kidding me. For f*ck's sake [...].

E: Ok, credit where it's due, the fontconfig package was updated just the other day in addition to neither of the edit hacks described here actually working anymore.

So it's either:
1) Remove _all_ the <alias> tags from under "<!-- Name changes and spelling variant aliases -->" OR
2) Remove the symlink "/etc/fonts/conf.d/65-fonts-persian.conf" -> "/etc/fonts/conf.avail/65-fonts-persian.conf" (or just the file being pointed to itself).

I'm too lazy and still so delightfully and yet quite so stilishly, while at the same time piteously pissed off about this to go through any more edit hack searches.. at this point.

Det commented on 2012-11-28 15:01

You've motherf*cking got to be kidding me. For f*ck's sake:

"So, instead of breaking it for everyone, while at the same time offering something for those who didn't have 6.2 to smile about I decided that the 'split package trick' is realistically the only sane choice here.

So what this means is that the package still shows up as 6.2 but actually downloads 7.0, when manually done so.

The only problem is still the segfault on startup to which somebody already found maybe a bit ugly fix in the Support Forums by simply removing the font .conf symlink in question but it's the _only_ solution we have so let's just work with it.

So after the install, as already pointed out in the .install, you need to remove or rename the symlink '/etc/fonts/conf.d/65-fonts-persian.conf' -> '/etc/fonts/conf.avail/65-fonts-persian.conf' (owned by 'fontconfig').

When renaming it has to be something other than '[0-9]*.conf' (e.g. 'fonts-persian.conf' or '65-fonts-persian.confa').

You can play around with the file too of course and I've already noticed that it's enough, if you either remove:
1) all the <alias> tags from under "<!-- Name changes and spelling variant aliases -->" except for 'Fantezi' (the last one) OR
2) all the <match> tags from under "<!-- Use Titr in titles -->".

So yeah. The performance is slow as hell no matter what you disable (e.g. even the new 3D imagery) but at least my most important package won't be an insult to all the other ones I have by carrying a sign that says it's out-of-date."

In short: READ the post install output.

mrechte commented on 2012-11-28 12:33

Hello. When starting google-earth, I get the following:

Google Earth has caught signal 11.
....
$ cat .googleearth/crashlogs/crashlog-50b6022f.txt
Major Version 7
Minor Version 0
Build Number 0001
Build Date Oct 29 2012
Build Time 19:13:39
OS Type 3
OS Major Version 3
OS Minor Version 6
OS Build Version 7
OS Patch Version 0
Crash Signal 11
Crash Time 1354105391
Up Time 0,005812

Stacktrace from glibc:
./libgoogleearth_free.so(+0x1e9cfb)[0xf7586cfb]
./libgoogleearth_free.so(+0x1e9f43)[0xf7586f43]
linux-gate.so.1(__kernel_sigreturn+0x0)[0xf773c400]

I use the Intel driver (intel-dri 9.0.1-1, lib32-intel-dri 9.0.1-1, xf86-video-intel 2.20.13-1) on a x86_64 platform.

Det commented on 2012-11-26 12:53

You don't say!

echinotrix commented on 2012-11-26 09:38

I've got the same crash message as technic93.

Det commented on 2012-11-25 10:29

Really..?

technic93 commented on 2012-11-24 11:23

I have the following crash with ati r600 open source driver.
Anybody knows how to fix this

Major Version 7
Minor Version 0
Build Number 0001
Build Date Oct 29 2012
Build Time 19:13:39
OS Type 3
OS Major Version 3
OS Minor Version 6
OS Build Version 6
OS Patch Version 0
Crash Signal 11
Crash Time 1353756076
Up Time 6.16881

Stacktrace from glibc:
./libgoogleearth_free.so(+0x1e9cfb)[0xb756ccfb]
./libgoogleearth_free.so(+0x1e9f43)[0xb756cf43]
linux-gate.so.1(__kernel_sigreturn+0x0)[0xb7725400]

Det commented on 2012-11-16 13:03

@canurabus, then it's a different crash (e.g. if you didn't see fit to read the opt depends).

canurabus commented on 2012-11-16 11:47

Changing/removing the symlink doesn't work for me -- still crashes.

Det commented on 2012-11-07 13:10

So, instead of breaking it for everyone, while at the same time offering something for those who didn't have 6.2 to smile about I decided that the 'split package trick' is realistically the only sane choice here.

So what this means is that the package still shows up as 6.2 but actually downloads 7.0, when manually done so.

The only problem is still the segfault on startup to which somebody already found maybe a bit ugly fix in the Support Forums by simply removing the font .conf symlink in question but it's the _only_ solution we have so let's just work with it.

So after the install, as already pointed out in the .install, you need to remove or rename the symlink '/etc/fonts/conf.d/65-fonts-persian.conf' -> '/etc/fonts/conf.avail/65-fonts-persian.conf' (owned by 'fontconfig').

When renaming it has to be something other than '[0-9]*.conf' (e.g. 'fonts-persian.conf' or '65-fonts-persian.confa').

You can play around with the file too of course and I've already noticed that it's enough, if you either remove:
1) all the <alias> tags from under "<!-- Name changes and spelling variant aliases -->" except for 'Fantezi' (the last one) OR
2) all the <match> tags from under "<!-- Use Titr in titles -->".

So yeah. The performance is slow as hell no matter what you disable (e.g. even the new 3D imagery) but at least my most important package won't be an insult to all the other ones I have by carrying a sign that says it's out-of-date.

Det commented on 2012-10-31 22:44

Seems like the Ubuntu people have no better luck than us: http://productforums.google.com/forum/#!category-topic/earth/linux/dlzBfGl4eKM

The previous URL still points to the old 5.2, even though it says "6.2" (http://www.google.com/earth/download/ge/agree.html -> advanced setup, or http://dl.google.com/earth/client/advanced/previous/GoogleEarthLinux.bin).

kevincox commented on 2012-10-31 20:08

The .deb doesn't pass validity check. The new md5sum is 'cc953c54fb687785028db10d951aa3c6'.

I don't know if this is related but it keeps segfaulting for me. Maybe the file just got corrupted.

Anonymous comment on 2012-10-21 13:46

error: '/tmp/yaourt-tmp-x/PKGDEST.KXc/*.pkg.tar.xz': could not find or read package

Det commented on 2012-10-13 09:48

Yeah. Added.

thestinger commented on 2012-10-12 22:56

It runs very well with Mesa 9 on Ivy Bridge.

By the way, this should depend on lib32-libsm and lib32-gcc-libs too.

thestinger commented on 2012-10-12 22:55

It runs very well with Mesa 9 on Ivy Bridge.

By the way, this should depend on lib32-sm and lib32-gcc-libs too.

Det commented on 2012-10-11 16:29

Btw. how is the new Mesa?

Any of you guys tested how well it goes along with your various Intels?

Det commented on 2012-10-05 23:46

FS#472863 - [google-earth] should use PKGEXT=".pkg.tar"

Opened by R00KIE - Friday, 05 October 2012, 19:49 GMT-1
Last edited by Det - Friday, 05 October 2012, 23:46 GMT-1

Task Type | General Gripe
Category | Packages: AUR
Status | Closed
Assigned To | Det
Architecture | All
Severity | Low
Priority | Normal
Reported Version | 6.2.2.6613-1
Due in Version | Undecided
Due Date | Undecided
Percent Complete | ##########
Votes | 1 (you voted for this task)
Private | No
Watching | Yes (stop watching)

Closed by Det
Friday, 05 October 2012, 23:46 GMT-1
Reason for closing: Implemented

R00KIE commented on 2012-10-05 19:49

Could you consider adding PKGEXT=".pkg.tar" to the PKGBUILD?
The reason is there is no benefit in compressing a big package like this and wasting cpu cycles when it is going to be installed locally and not transferred over the internet.

SanskritFritz commented on 2012-10-03 20:40

helmuthdu that didn't help me. Can you tell us more about this? Maybe I need another locale or font. Where can I read about those switches?

helmuthdu commented on 2012-10-03 14:07

I have fixed the ugly font display with this command:
google-earth -fn -xos4-terminus-medium-r-normal--12-120-72-72-c-60-iso8859-9

srl commented on 2012-09-10 01:34

Now I feel stupid. Thanks - multilib it is.

dgbaley27 commented on 2012-09-10 01:32

The 64-bit packages provided by google contain 32-bit binaries.

srl commented on 2012-09-10 01:28

There does seem to be a 64-bit deb available - should this package not use that instead when appropriate? (That would eliminate the need for [multilib].)

dront78 commented on 2012-09-08 18:46

Have anyone tryed
http://swyear.blogspot.com/2011/05/run-google-earth-6-on-opensuse.html
to fix antialiasing?

Det commented on 2012-08-28 13:37

It's in [multilib]: https://wiki.archlinux.org/index.php/Multilib

smu commented on 2012-08-28 07:38

# packer -S google-earth
Dependency `lib32-fontconfig' of `google-earth' does not exist.

Det commented on 2012-08-26 11:42

Glad we got that figured out.

DevRandom commented on 2012-08-26 02:09

To people who still have trouble with:

./googleearth: line 43: ./googleearth-bin: No such file or directory

You are missing /lib/ld-lsb.so.3, which is a symlink installed by ld-lsb. Check the following:

- Multilib repo active in pacman.conf
- lib32-glibc installed

The recent switch from /lib to a symlink blew up both lib32-glibc and ld-lsb on my setup, and I had accidentally deactivated the multilib repository in pacman.conf before updating the packages. This led to exactly this error.

Det commented on 2012-08-19 18:36

That's something you raarely hear from a fellow user.

You are welcome.

jhernberg commented on 2012-08-19 16:13

You are right it doesn't render correctly when adding that lib. Can't seem to find another way of making it work. So I guess it's pacman -R time :) Thanks for your time and best of luck with the package though!

Det commented on 2012-08-19 15:20

Some people say removing or adding a new line to .drirc helps (also mentioned in the post install/upgrade).

Then there's other discoveries in different sites as well, including disabling the Start-Up Tips: http://www.webupd8.org/2010/10/fix-google-earth-crashing-on-startup-in.html

You're not alone. For some people replacing intel-dri with the launchpad libGL helps, for some it's the othe way around. I've just heard that using the launchpad libGL there's some flickering and/or missing polygons. You could try cycling through these to see, if it's of any assistance: http://productforums.google.com/forum/#!searchin/earth/signal$2011

jhernberg commented on 2012-08-19 15:02

FWIW, it crashes with a "Google Earth has caught signal 11." until I install http://librarian.launchpad.net/7037027/libGL.so.1 into /opt/google/earth/free. i686 system with intel-dri 8.0.4-2.

jhernberg commented on 2012-08-19 13:42

Fwiw, license.html looks ok when opened in a browser. Indeed I am getting served by another google server: Connecting to earth.google.com|74.125.230.232|:80... connected. Hopefully I am not getting my DNS spoofed :) Am online via umts dialup in Spain (telefonica). For me personally not a big problem, but I thought I'd let you know that there is a potential problem with the script. Repeating the process in /tmp yields the same result and the same checksum.
Running the script on a machine in Lxembourg, yields the expected checksum of '7363c6144ebb298b1d7aec713ca8a82a'.

Det commented on 2012-08-19 12:49

I'm assuming you either still got the same corrupt license there or you're getting it from a different server with a different version. We can confirm this by you telling me the ip you're getting with wget:

Connecting to earth.google.com|173.194.32.2|:80... connected.

But you can either just re-generate the md5sums like that, skip them (--skipinteg) or just omit the license (the install line as well).

I'm a little reluctant to do that in the PKGBUILD since this is the first time we've had this happen. If it won't go away I can include the license in the tarball.

Also you wouldn't need the 'z' with tar. Tar automatically reads the format of the archive (even regardless of the name).

E: It _could_ also be that the sector in the disk used by the license is corrupt (even after redownloading it, if that's what you did). If it's still '3dda103111cf68bdd2eff2485456bd98' even in /tmp (mounted as tmpfs, located in RAM) then it's just the server giving you a different license.

Det commented on 2012-08-19 12:45

I'm assuming you either still got the same corrupt license there or you're getting it from a different server with a different version. We can confirm this by you telling me the ip of the server you're getting with wget:

Connecting to earth.google.com|173.194.32.2|:80... connected.

But you can either just re-generate the md5sums like that, skip them (--skipinteg) or just omit the license (the install line as well).

I'm a little reluctant to do that in the PKGBUILD since this is the first time we've had this happen. If it won't go away I can include the license in the tarball.

Also you wouldn't need the 'z' with tar. Tar automatically reads the format of the archive (even regardless of the name).

jhernberg commented on 2012-08-19 12:29

Running wget http://earth.google.com/intl/en/license.html and then running makepkg -g also shows a checksum of '3dda103111cf68bdd2eff2485456bd98' :( I guess some kind of google magic then. Does the license.html file really have to be there? Maybe the solution would be to just remove it, or possibly to download it separately without having it in the source array.

jhernberg commented on 2012-08-19 12:19

Just to be sure I downloaded the tarball again, unpacked it with tar xzvf google-earth and ran makepkg again. Same error, my md5 seems to be '3dda103111cf68bdd2eff2485456bd98'. makepkg -g shows again that the checksum should be '3dda103111cf68bdd2eff2485456bd98'... Wonder if I'm getting a different license.html somehow from google then?

Det commented on 2012-08-19 12:04

I could. But it wouldn't make any difference.

jhernberg commented on 2012-08-19 11:38

@Det, did you try on i686?

jhernberg commented on 2012-08-19 11:37

hmm license.html fails the md5 validity check (on i686)?

Det commented on 2012-08-19 09:46

It doesn't. Yours is corrupt.

jhernberg commented on 2012-08-19 09:22

hmm license.html fails the md5 validity check (on i686)?

jhernberg commented on 2012-08-19 09:04

hmm license.html fails the md5 validity check?

rtfreedman commented on 2012-08-17 23:21

[q]You really getting the same error? [/q]
Yep - I had the same errors for the last days I've tried: /usr/bin/google-earth: line 43: ./googleearth-bin: No such file or directory
I've tried to 'ldd googleearth' to no avail - how does it happen?
... I just rebooted to arch and tried again - successful ?!? WTF!
I don't know what have changed - something with the last update?
On a side note - I had to boot to linuxmint to be able to post
- archlinux (firefox, midori et al) doesn't connect to https://aur... but that's another (unsolved) story.
For now it works again and looks pretty - unlike with linuxmint where I had to fix the font size with qtconfig :(

Det commented on 2012-08-17 22:09

Explaining what exactly is wrong with your Google Earth. I see nothing wrong there.

You really getting the same error?

rtfreedman commented on 2012-08-17 14:57

[quote]Well duh, you need to actually have the symlink to /usr/lib/ (glibc 2.16.0-2) before you start using it.[/quote]
Which 'symlink' do you mean?

I have a fully updated system, /lib -> usr/lib and all; links are setup:
lrwxrwxrwx 1 root root 10 Aug 11 09:56 ld-linux.so.2 -> ld-2.16.so
lrwxrwxrwx 1 root root 18 Aug 17 15:31 ld-lsb.so.3 -> /lib/ld-linux.so.2

What do I miss?

grayich commented on 2012-07-27 21:40

~> google-earth
/usr/bin/google-earth: line 43: ./googleearth-bin: No such file or directory

Det commented on 2012-07-22 12:40

This is already done through ld-lsb.

Anonymous comment on 2012-07-22 12:32

Try this: sudo ln -s /lib/ld-linux.so.2 /lib/ld-lsb.so.3
Then launch google-earth.

Det commented on 2012-07-19 15:10

Well duh, you need to actually have the symlink to /usr/lib/ (glibc 2.16.0-2) before you start using it. No reinstall would've been required.

DarioP commented on 2012-07-19 09:06

From a fresh instal same problem here:

/usr/bin/google-earth: line 43: ./googleearth-bin: No such file or directory

Reinstalled ld-lsb: nothing appened.
Uninstalled both google-earth and ld-lsb, reinstalled both: still same problem.
Fully update of the whole system (including packages from multilib) and reinstalled both, now it works :)

Det commented on 2012-07-15 13:05

Or just update?

kotyz commented on 2012-07-15 12:51

@gborzi
Problem solved, thank you ;-)

gborzi commented on 2012-07-15 11:08

@kotyz
Try to uninstall and reinstall ld-lsb, it worked for me.

dgbaley27 commented on 2012-07-15 02:15

Ditto, I think since the glibc move to /usr/lib

kotyz commented on 2012-07-15 01:18

[kotyz @ behemot : ~] $ google-earth
/usr/bin/google-earth: line 43: ./googleearth-bin: No such file or directory

Det commented on 2012-06-05 16:24

Wouldn't be there otherwise, would they? They both use the same 32-bit binaries.

ayr0 commented on 2012-06-05 15:31

Does the 64bit version require 32bit libraries?

Det commented on 2012-05-15 09:07

That's provided by libgl. There's been complaints about mesa 8.0 but there's always someone not checking them out.

Anonymous comment on 2012-05-14 20:47

The extra/intel-dri 8.0.2-1 has been already installed. But It can't work without ligGL.so.1

Det commented on 2012-05-13 20:04

(lib32-)intel-dri should work even better.

Anonymous comment on 2012-05-13 17:30

It doesn't work for me without this modification:
https://bbs.archlinux.org/viewtopic.php?pid=942307#p942307

Det commented on 2012-05-11 12:12

Well, then I can't help you taken I don't even know what your problem is. Here it works just fine with nouveau too.

You could try running this thing under valgrind to see what's it failing at.

Anonymous comment on 2012-05-10 18:38

Yes. nouveau-dri was installed as version 7.11.2-1 on 2012-01-05, upgraded to version 8.0.1-2 on 2012-03-17 and again upgraded to version 8.0.2-1 on 2012-03-24, all according to /var/log/pacman.log.

Det commented on 2012-05-09 17:54

So.. did you or did you not install (lib32-)nouveau-dri? Or lib32-nvidia-utils, in case you switched to the nvidia driver.

Det commented on 2012-05-09 17:53

So.. did you or did you not install (lib32-)nouveau-dri? Or (lib32-)nvidia-utils, in case you switched to the nvidia driver.

Det commented on 2012-05-09 17:51

So.. did you or did you not install (lib32-)nouveau-dri? Or (lib32-)nvidia-dri, in case you switched to the nvidia driver.

Anonymous comment on 2012-05-09 17:47

No I am not. The error is still there, in spite of my having an empty ~/.drirc. What do you suggest I do now - other than complaining ;-)

Det commented on 2012-05-09 09:32

Are you kidding me?

Anonymous comment on 2012-05-08 15:58

What is the status on the segfault problem?

Det commented on 2012-05-06 08:32

It is because those optdeps also include (lib32-)nvidia-utils.

vanadium50 commented on 2012-05-06 04:48

Solved problem with NVidia GeForce 9600: blank (black) main window, otherwise seemed to work.

Installed lib32-nvidia-utils (replacing lib32-libgl). Similar lib32-xxxx for other video brands are listed in optdepends. Sorry if this is obvious to everyone - got the idea from http://forums.debian.net/viewtopic.php?f=6&t=54394

Det commented on 2012-04-24 01:33

The software itself was. The URLs just never change.

Anonymous comment on 2012-04-23 23:41

It looks like the md5sum was updated. Thanks, I appreciate it :)

Anonymous comment on 2012-04-22 14:04

@Det: Many thanks to you for maintaining this package, I do appreciate it =)

dgbaley27 commented on 2012-04-22 01:03

I've had this package for months (maybe even a year now) _without problems_

Det commented on 2012-04-21 23:22

Come on man, I'm trying my best here.

Anonymous comment on 2012-04-21 20:41

The md5sum check fails when building the package with makepkg -s. Is this expected?

==> Validating source files with md5sums...
google-earth-stable-6.2.1.6014-amd64.deb ... FAILED
license.html ... Passed
google-earth-mimetypes.xml ... Passed
==> ERROR: One or more files did not pass the validity check!
==> ERROR: Makepkg was unable to build google-earth.

Anonymous comment on 2012-04-18 13:32

Confirming problem that Dragetz mentions. My google earth with intel graphics locks xserver when zooming in to the surface. Downgrading packages works as mentioned as well.

Det commented on 2012-04-17 14:30

I think you do.

Anonymous comment on 2012-04-16 13:38

Does anybody have problem with new mesa 8.0 on intel graphics. It causes segmentation fault somewhere in X. It locks up Xsever and Xserver stops processing events. So I have to do hard reset or ssh from another machine and restart Xserver. It's works at first i can rotate the globe, but when i try to zoom in everything freezes.

When i downgrade lib32-intel-dri lib32-libgl to 7.11 version everything seems to be OK.

indianahorst commented on 2012-04-14 21:00

Since some days (maybe since the last update of the nvidia drivers?) I noticed Google Earth to be veeeeery slow. After calling "google-earth" in terminal it takes 4-5 minutes to show the whole earth clean and sharp. After loading this start position it remains unusable, it takes again several minutes to show one of "my places" or something like that. While it runs, it uses 100% cpu. No changes to any configuration has been made before this.

Anybody here who has the same problem?

Det commented on 2012-04-13 19:28

2 things. First off, I don't think I've actually ever seen anybody trying to use 'locate' when searching for a package in the repos. You look for a package, you use the package manager: https://wiki.archlinux.org/index.php/Pacman

Secondly, 'nouveau-dri', along with all the other *-dri stuff, is mentioned as an optional dependency. If not right after the installation, then at least have a look at those when your application doesn't even start up.

Anonymous comment on 2012-04-13 17:16

No, I'm not. Linux Mint doesn't even now it and on my Arch Linux-box I only see 3 lines from /var/lib/pacman/local/ when I search for nouveau-dri with 'locate'

Det commented on 2012-04-13 16:59

You even using (lib32-)nouveau-dri then?

Anonymous comment on 2012-04-13 15:30

You have got me confused on a higher level. What driver are you talking about.

I'm concerned about a segmentation fault when I try to launch Google Earth, and I am pretty convinced that it is a packaging error for Arch Linux. I have taken the Debian package being downloaded as part of the installation and installed it on my laptop which runs Linux Mint. It works fine on that machine. Both machines have Nvidia graphics hardware and run the Nouveau driver.

Please enlighten me.

Bent

Det commented on 2012-04-13 12:36

In a week? It takes a bit longer to develop a driver.

Anonymous comment on 2012-04-13 10:58

Any news on the signal 11 (segfault) problem?

Det commented on 2012-04-11 23:19

Because my Nvidia _needs_ an Intel DRI driver?

tancrackers commented on 2012-04-11 21:16

lib32-intel-dri should be listed as a necessity for 64bit oses.

giddie commented on 2012-04-07 22:43

@akspecs Thanks for the info. Yeah, it's clearly a driver issue. I think there's still some room for improvement for SandyBridge support, to be honest, although it's *so* much better than a year ago. I have the same issue if I try to "Experience MapsGL" in Google Maps. I get a few seconds of 3D, and then a lockup, although when that happened I did get some crippled activity following that, but it was pretty clear that 3D acceleration had gone out the window.

akspecs commented on 2012-04-07 22:22

@giddle I run on HD3000 graphics as well and have similar behavior as do you. If running Gnome, it can handle the crash by killing X, but then when restarting X it will restart in fallback mode. On a lower end system, running i686 nvidia nvs quadro 135m graphics it runs fine. So it my have to do with xf86-video-intel or lib32-intel-dri or something :/

Det commented on 2012-04-07 19:34

Not gonna help you two but out of curiosity I tested which browsers asked whether to resend the data.

It turns out both Firefox and Chrome have started doing this since their first public releases (0.8 and 0.2.149.27 Beta, respectively). IE also does this at least since 7 (nothing lower works with Windows 7), while Opera doesn't with neither the latest 12.00 alpha (1351) or the stable 11.62.

giddie commented on 2012-04-07 15:32

Any idea why I get a complete system freeze after a minute or so when using Google Earth with HD3000 Intel graphics?

akspecs commented on 2012-04-07 12:10

I still get crash signal 11 after installing lib32-intel-dri - just 30 seconds after I start rather than immediately after startup.

Anonymous comment on 2012-04-07 11:19

T'was an error 40. :-( My browser is Firefox 11.0

Det commented on 2012-04-06 20:39

Upgrade to a better browser then. Modern firefox and chrome versions both ask first whether to resend the same thing.

dgbaley27 commented on 2012-04-06 20:33

Det: He posted, then hit refresh awaiting your reply and the browser resent the data.

Det commented on 2012-04-06 18:19

Because there was no new release in between.

BTW. How come did you post the same thing twice?

Anonymous comment on 2012-04-06 17:42

I did. The file ~/dridc did not exist. After the echo command I had an empty file, but that didn't prevent google-earth from segfaulting. I have the package 'nouveau-dri' installed.

BTW. How come the PKBUILD was updated without a version number increase?

Det commented on 2012-04-06 16:14

No. Rebuild it and have a look.

Anonymous comment on 2012-04-06 15:43

If this is what you mean by 'post install':

NOTE: The binary is called 'google-earth'

then it doesn't tell me anything I didn't know.

Det commented on 2012-04-05 21:44

Uhh.. you still getting it after doing what's said in the post install?

Anonymous comment on 2012-04-05 21:41

When will the crash with signal 11 (segfault) be corrected??

Det commented on 2012-03-30 12:16

More of a Wiki thing but whatever.

TheJJ commented on 2012-03-29 15:52

crashed on startup with signal 11, i had to add a new line in ~/.drirc with

echo "" >> ~/.drirc

maybe you could integrate this command in the post_install/upgrade scripts

Det commented on 2012-03-13 16:30

I _could_. But that would defeat the purpose of not doing it.

xzy3186 commented on 2012-03-13 15:33

Wow, there is no more black pieces of blocks on the earth map. Thank you very much. BTW, could you please change the version from 6.2.1.6014-1 to 6.2.1.6014-2 or something different?

Det commented on 2012-03-13 14:56

Wow. It took _ages_ to run that thing on the .pkg.tar.xz so congrats for your patience.

There was a lot of redundant/misleading stuff of course but I've fixed the rest (nothing too serious, though, so still not bumping the pkgrel).

Didn't see your Qt thing, though. You probably just deleted those or something.

ridikulusrat commented on 2012-03-13 14:02

@Det: namcap report

google-earth E: Dependency lib32-qt detected and not included (libraries ['usr/lib32/libQtNetwork.so.4', 'usr/lib32/libQtCore.so.4', 'usr/lib32/libQtGui.so.4'] needed in files ['opt/google/earth/free/plugins/imageformats/libqjpeg.so', 'opt/google/earth/free/libQtWebKit.so.4', 'opt/google/earth/free/libcommon.so', 'opt/google/earth/free/libcomponentframework.so' and many more files

google-earth E: Files in /usr/share/icons/hicolor but no call to gtk-update-icon-cache or xdg-icon-resource to update the icon cache

google-earth W: Directory (etc/cron.daily) is empty

google-earth W: ELF file ('opt/google/earth/free/libbase.so') has text relocations.
google-earth W: ELF file ('opt/google/earth/free/libwebbrowser.so') has text relocations.
google-earth W: ELF file ('opt/google/earth/free/libsgutil.so') has text relocations.
google-earth W: ELF file ('opt/google/earth/free/libgps.so') has text relocations.
and many other ELF files (almost all fo them)

google-earth W: Dependency included and not needed ('ld-lsb')

Det commented on 2012-03-13 13:35

@f45, true. Added 'nouveau-dri' too.

@ridikulus_rat, that's funny.

ridikulusrat commented on 2012-03-13 06:34

lib32-qt should be added as dependency.

Anonymous comment on 2012-03-13 02:59

lib32-ati-dri should be added as an optional dependency for those using the open source radeon drivers.

Det commented on 2012-03-07 09:37

K. Could be explained by the file being a decade old anyway.

dgbaley27 commented on 2012-03-07 09:31

I haven't really been following this, but with my i915, it works a lot smoother with no missing polygons if I remove the libGL in /opt/google

Det commented on 2012-03-07 09:04

Or is everybody else dead?

Det commented on 2012-03-03 00:54

I guess I could add that to the optional dependencies.

Does it work better than the Launchpad libGL.so.1 solution with anybody else?

Anonymous comment on 2012-03-02 19:05

I have the same problem with flickering and missing polygons with my Intel graphcs. If I turn ON desktop effects, the flickering stops but there are still missing polygons.

I just deleted libGL.so from /opt/google/earth/free, i belive that one is from launchpad link mentioned below.

And installed package lib32-intel-dri, now it works without any problems.

Det commented on 2012-02-23 23:13

Not by any chance caused by you not checking out the optional dependencies?

Spheerys commented on 2012-02-22 17:36

I can't use Google Earth.
When I launch it, I have this error : Google Earth has caught signal 11.

In the crashlog, I have this :

Major Version 6
Minor Version 2
Build Number 0001
Build Date Feb 3 2012
Build Time 15:28:07
OS Type 3
OS Major Version 3
OS Minor Version 2
OS Build Version 6
OS Patch Version 0
Crash Signal 11
Crash Time 1329932069
Up Time 6,27007

Stacktrace from glibc:
./libgoogleearth_free.so(+0xc1583)[0xf76ae583]
./libgoogleearth_free.so(+0xc1703)[0xf76ae703]
[0xf7743400]
/usr/lib32/xorg/modules/dri/swrast_dri.so(+0x3fd96c)[0xe84e996c]

Det commented on 2012-02-08 08:50

You and your Intels.

goetzc commented on 2012-02-08 04:23

If I don't remove the libGL.so.1 file, GE crashes the xserver, just after loading the interface.
But when it loads, it is transparent. http://i.imgur.com/H0MNc.png When I quit, GE crashes http://pastebin.com/TpekLTxv ...Another crash happened while handling crash! he he!

00:02.0 VGA compatible controller: Intel Corporation 82865G Integrated Graphics Controller (rev 02)

goetzc commented on 2012-02-08 04:22

If I don't remove the libGL.so.1 file, GE crashes the xserver, just after loading the interface.
But when it loads, it is transparent. http://i.imgur.com/H0MNc.png When I quit, GE crashes http://pastebin.com/TpekLTxv ...Another crash happened while handling crash! he he!

blkqi commented on 2012-02-06 19:51

had darose's flickering issues on a intel GPU in i3 wm. I can confirm the flcikering goes away with xcompmgr running

Det commented on 2012-02-06 09:51

It's not. I'd assume it's just for GPUs with a bad driver.

darose commented on 2012-02-06 04:42

Hmmm ... I think I have my answer: turn on compositing. When I turn that on, all the flicker goes away.

Odd. I didn't realize compositing was a requirement for GE.

Thanks -- DR

darose commented on 2012-02-06 04:27

DL0 - great tip! That worked fine ... under KDE4.8. Only problem is, I normally use XFCE. :-( Any idea how to duplicate that setting under that DE?

DL0 commented on 2012-02-05 22:00

After installing Bumblebee I got many error messages, no navigation controls, missing tiles and poor performance running Google-Earth 6.2 with optirun on my Thinkpad T520.

I was able to improve it substantially by renaming (or removing) the libGL.so.1 and libGLU.so.1 files in /opt/google/earth/free and creating a symlink to the libGL.so.1 and libGLU.so.1 files in /usr/lib/

I'm getting a "./googleearth.bin: corrupted double linked list" message that needs to be resolved but at least Google Earth now works!

...Doug

DL0 commented on 2012-02-05 04:01

darose -

Try using vsync. It eliminated the flickering for me using KDE 4.8. (For KDE, in the "Advanced" tab under Desktop Effects in System Settings/Workspace Appearance).

...Doug


bananaoomarang commented on 2012-01-28 11:38

Looks like it is now!

Sorry though. I just assumed it was. Wrong thing to do!

Det commented on 2012-01-27 13:34

But not for linux.

bananaoomarang commented on 2012-01-27 10:51

6.2 is out.

Det commented on 2012-01-27 07:10

And I wasn't suggesting that you were suggesting. I was suggesting that the one who flagged this thing was.

darose commented on 2012-01-27 04:08

I wasn't suggesting the package wasn't up to date. Was just wondering if anybody had any suggestions for how I might address the flicker. It's pretty bad. The earth flickers off and then back pretty much every time I move the mouse.

Det commented on 2012-01-26 23:19

Uhhhh.. this _is_ up-to-date. The flicker thing is not my problem.

darose commented on 2012-01-24 04:21

So then ... back to the flicker problems. Anybody have any ideas as to cause/workaround?

Det commented on 2012-01-21 20:10

Same what? The crash? Because we know that already.

Anonymous comment on 2012-01-21 18:06

Same..

kolombo commented on 2012-01-18 15:32

It is expected to crash without that :)

Det commented on 2012-01-18 15:30

Lol.

darose commented on 2012-01-18 15:28

It crashes when I move that file away.

Det commented on 2012-01-18 11:34

_Might_ be caused by the Intel-intended 'libGL.so.1', which is supposed to make GE even start up. You can try removing it from "/opt/google/earth/free/" and see if it helps.

If it does, I'm going to sigh - and not for relief.

darose commented on 2012-01-18 09:39

Just updated to GE 6.1.0.5001-3, and I'm having really bad flicker problems. Anyone have any idea for cause/fix?

I'm running XFCE, using an Intel graphics card. (Xorg.0.log reports it as "Intel(R) Sandybridge Mobile (GT2+)".)

Det commented on 2011-12-30 12:31

Can't because it's not 64-bit. Just bin-32.

dgbaley27 commented on 2011-12-29 23:56

Are you going to switch over to the 64bit binary which is now available?

kolombo commented on 2011-12-27 15:14

:)
In my case yes.

Det commented on 2011-12-27 15:13

You guys make my life so hard. Does this work?

kolombo commented on 2011-12-27 13:38

Det, Could you please include this `http://librarian.launchpad.net/7037027/libGL.so.1` lib in your package ?
This will make Intel card users life much easier.
lib32-libgl (7.11.2-2) didn't work for me.

kolombo commented on 2011-12-27 13:23

Det, Could you please include this `http://librarian.launchpad.net/7037027/libGL.so.1` lib in your package ?
This will make Intel card users life much easier.

SanskritFritz commented on 2011-12-19 21:50

Yeah. I already tried all these. Back in the GE5 days it helped, but not now with GE6 :(

Det commented on 2011-12-19 20:04

Sure: https://groups.google.com/a/googleproductforums.com/forum/#!categories/earth/problems-and-errors

Just by searching for "font" I got 119 results. Here's one of them: https://groups.google.com/a/googleproductforums.com/forum/?hl=en#!searchin/earth/font/earth/BT5kyvlwFME/uWZd3SZAEu8J

Most people say that either installing 'ttf-ms-fonts' or increasing the Qt applications' fonts with 'qtconfig' helps.

SanskritFritz commented on 2011-12-19 14:44

What I find strange is that I just couldnt find any upstream help about this problem. Is there a useful forum or help for GE on linux?

Det commented on 2011-12-19 14:36

Dunno. It's fine by me. If I had some anti-aliasing it'd be perfect.

Actually, I tried with 'ttf-win7-fonts' (the-oh-so-successor of the apparently now deprecated 'ttf-ms-fonts') and I can't even tell the difference. This is a bit illogical since GE's now using the same fonts as my Win 7 one.

Berseker commented on 2011-12-18 17:52

sad comments below :) We have to find out someohow how to fix this situation

SanskritFritz commented on 2011-12-18 07:44

No comment :p

Det commented on 2011-12-17 23:15

You're probably not using Windows. This one's from Arch: http://i.imgur.com/5h7CE.jpg

SanskritFritz commented on 2011-12-17 22:01

Incredible, why, oh why can't I have this beautiful Google Earth.

Det commented on 2011-12-17 21:24

http://i.imgur.com/u1imh.jpg

Berseker commented on 2011-12-17 11:58

det, can you post a screenshot of your google earth?

SanskritFritz commented on 2011-12-10 16:57

I really envy you, I have this problem for over a year now :(

Det commented on 2011-12-10 16:37

Well that's interesting because it helped with me. In fact, whenever I change the font with 'qtconfig' it changes in Google Earth as well, without even re-logging.

Det commented on 2011-12-10 16:36

Well that's interesting because it helped with me. In fact, whenever I change the font with 'qtconfig' they change in Google Earth as well, without even re-logging.

SanskritFritz commented on 2011-12-10 14:27

Det, I did all that, tried many things, for the life of me, I can't figure it out. Fonts just look ugly in GE here. At least I can use it, I just gave up on that problem :( This is how it looks: http://content.wuala.com/contents/SanskritFritz/Public/2011-01-04-203520_1440x900_scrot.png?dl=1
This screenshot was taken at least a year ago, I used KDE back then, now I use Openbox or Compiz, GE look has not changed since.

Det commented on 2011-12-10 10:49

You try installing 'ttf-ms-fonts' or messing around with 'qtconfig' (you might need to re-login)?

SanskritFritz commented on 2011-12-10 08:52

YW. Does it look ok now, or are the fonts still somewhat distorted? I cannot figure out how to solve that fonts problem.

silvik commented on 2011-12-09 16:21

man, i couldn't have figured that in a million years.
thanks SanskritFritz.

SanskritFritz commented on 2011-12-09 14:16

silvik you have the Ahem font installed that comes with the ttf-droid package. Remove it manually. The problem of weird fonts will not be solved, but at least you can read whats on the screen :)

silvik commented on 2011-12-09 13:56

I have some weird font problem - x86_64 system
here's a screenshots, it's hard to explain: http://imgur.com/Y3YBq
any ideas what's wrong?

silvik commented on 2011-12-09 13:56

I have some weird font problem - x86_64 system
here's a screenshots, it's hard to explain: http://imgur.com/Y3YBq
any ideas what's wrong?

Det commented on 2011-12-01 11:21

Oh, all these contributions. Makes me wanna cry.

eworm commented on 2011-11-30 09:08

Oh, same for x86_64 of course. ;)

eworm commented on 2011-11-30 09:07

Can you please change

source=(http://dl.google.com/dl/earth/client/current/$pkgname-stable_current_i386.deb)

to

source=(${pkgname}-stable_${pkgver}_i386.deb::http://dl.google.com/dl/earth/client/current/${pkgname}-stable_current_i386.deb)

This way you do not get failed source file validation errors with old tarballs (or deb packages) in source destination.

Det commented on 2011-11-29 23:03

Uh, I wouldn't recommend that...

You either let this thing install (lib32-)libgl or use the proprietary stuff, meaning (lib32-)catalyst-/nvidia-utils.

I don't understand why would you want a Launchpad library file to be used with _Arch_ anyway.

Anonymous comment on 2011-11-29 18:34

Solution for error with file libgoogleearth_free.so

download the file libGL.so.1 with wget https://launchpadlibrarian.net/7037027/libGL.so.1

and copy it to /opt/google/earth/free with cp libGL.so.1 /opt/google/earth/free/

And that's all ;)

Det commented on 2011-11-29 15:45

Hehe, seems the lib32-* dependencies are still needed after all - the amd64 .deb package has 'ia32-libs' as its dependency (and they didn't just forget it there - 'lsof' confirms this).

In any case, there's no real reason to have 2 separate packages for Google Earth when the other one can be both at the same time. Whenever Google starts offering real 64-bit binaries for Google Earth, could be a week from now, could be a year, this thing can start using them.

I also added 'ttf-ms-fonts' as an optional dependency, as it seems to solve the ugly fonts for me and some random Ubuntu users.

Det commented on 2011-11-29 12:58

Removed the unnecessary cron job. Thanks for pointing the unnecessity of it out, raqua.

@Berseker, the fonts can be changed with 'qtconfig'. I also added 'qt' to optional dependencies.

Berseker commented on 2011-11-28 18:51

I see that thw different versions of this PKGBUILD have been merged.

I am now asking to all: We have to find a way to make Google Earth look good on our desktop, as it is now it seems a Windows95 app (really unwatchable)
Have anyone done some hacking about this?

raqua commented on 2011-11-28 18:06

Hello. I just discovered google-earth script in /etc/cron.daily. Seems to be some Debian related stuff, which I believe is not needed.

Det commented on 2011-11-28 12:22

http://mailman.archlinux.org/pipermail/aur-general/2011-November/016717.html

Det commented on 2011-11-28 12:22

http://mailman.archlinux.org/pipermail/aur-general/2011-November/016717.html

Det commented on 2011-11-26 15:17

By the way, you are defining all the 64-bit stuff the wrong way:

[ $CARCH = 'x86_64' ] && _a="amd64"

This line doesn't mean that "everything else after this line will also pass said 64-bit check" - it only means that the "_a" variable is given the value of "amd64". You are supposed to use the if-else logic for that.

Your current implementation gives everyone a 64-bit version of Google Earth and its dependencies.

Det commented on 2011-11-23 10:50

Wait for those 6 days :). Then I can either give you the link to the discussion I create or you can start it yourself.

You can do this simply by sending an e-mail to "aur-general@archlinux.org". If you don't want to wait, you can do it already. I'll send the request in 6 days nontheless.

kiodo1981 commented on 2011-11-22 19:57

I'd like to adopt the package google-earth.
What should I do?

Det commented on 2011-11-22 19:22

Hey. As I'm sure you know, the maintainer of google-earth (458italia) has been unresponsive to e-mail and to the flagging of his package. In 6 more days I am going to make a request to orphan his package in the AUR Mailing List (AML: http://mailman.archlinux.org/pipermail/aur-general/2011-November/thread.html) and this is where you step in.

Would it be ok for you to become the new maintainer?

Det commented on 2011-11-22 19:10

6 days and I'll request this to be orphaned. Kiodo1981 might as well start maintaining it.

kiodo1981 commented on 2011-11-16 09:17

Version name update.
MD5 license.html update.

martvefun commented on 2011-11-16 08:12

The md5sum of license.html has change, new one : '7363c6144ebb298b1d7aec713ca8a82a'

Anonymous comment on 2011-11-14 21:38

There was no "lastTip=NN" in my configuration and i tried adding "enableTips=false" but it did not help. I have all the optional dependencies except for "lib32-nvidia-utils" because i am running the rt kernel with the special options compiling my nvidia driver. Does that have anything to do with the problems?

Berseker commented on 2011-11-14 19:36

i've just sent him an email, let's wait.

Det commented on 2011-11-13 18:45

Yeah, I'd be glad if 458italia disowned this one (or some TU did it for him), since I'm not seeing any progress here.

Det commented on 2011-11-13 18:44

Please just request for orphaning of the original 'google-earth' package and put the contents of this one there. After that you should remove this one.

Also, I think the pkgver should be "6.1.0.5001".

Berseker commented on 2011-11-13 18:39

please cooperate with the author of this
https://aur.archlinux.org/packages.php?ID=53869

and merge the two packages
We do not need two different google-earth

Anonymous comment on 2011-11-12 21:43

I also had a problem with libgoogleearth_free.so on Archlinux x86_64 and a Nvidia GPU. I solved this problem by installing the optional dependencies:

ttf-bitstream-vera: Fonts
lib32-nss-mdns: In case the application fails to contact the servers
lib32-gtk2: SCIM support
lib32-nvidia-utils: 3D support for Nvidia cards
lib32-catalyst-utils: 3D support for ATI/AMD cards

kiodo1981 commented on 2011-11-12 17:30

Try this.
Open ~/.config/Google/GoogleEarthPlus.conf
Overwrite line " lastTip=NN " whit " enableTips=false "

Anonymous comment on 2011-11-12 17:08

I get these crashes and I don't know what it means:
Crash Signal 11
Crash Time 1321117626
Up Time 0.522335

Stacktrace from glibc:
./libgoogleearth_free.so(+0xbcd53)[0xf7746d53]
./libgoogleearth_free.so(+0xbced3)[0xf7746ed3]
[0xf77c8400]

Det commented on 2011-10-27 10:44

@R00KIE, yyyes, that was kind of already noticed before..

R00KIE commented on 2011-10-25 19:19

I'm not sure but it seems that this may be released only in .deb and .rpm formats as it is stated that only Ubuntu and RHEL Linux are officially supported [1].

You could always extract from the .deb with bsdtar or ar which should be installed already as they are dependencies of other system packages. You can specify which part to extract (only the data part is needed) and then pick only the stuff needed.

[1] http://earth.google.com/support/bin/answer.py?hl=en&answer=20701

Anonymous comment on 2011-10-24 15:47

It's required by libgl. That's why it wasn't there before.

Btw. this isn't bash. You can just type (lib32-)libxxf86vm.

ilpianista commented on 2011-10-24 12:42

{lib32-,}libxxf86vm seems to be needed

$ googleearth
./googleearth-bin: error while loading shared libraries: libXxf86vm.so.1: cannot open shared object file: No such file or directory

Det commented on 2011-10-22 19:24

Just use the .debs as with google-chrome. It's only 2 lines anyway (https://aur.archlinux.org/packages/go/google-chrome/PKGBUILD).

Anonymous comment on 2011-10-04 23:50

Does anyone know of a stable URL to retrieve the .bin file Google Earth 6.1? It seems that they've only posted RPM and deb packages, so I would love it if anyone knew where to find the generic binary distribution from their servers to avoid having to use something to convert the rpm or deb binary package.

Anonymous comment on 2011-09-30 17:20

Google Earth 6.1 for GNU/Linux is out.

lagagnon commented on 2011-09-25 14:05

Build crashes for me with this error:

==> Starting package()...
tar: google-earth/googleearth-linux-x86.tar: Cannot open: No such file or directory
tar: Error is not recoverable: exiting now
==> ERROR: A failure occurred in package().
Aborting...

buergi commented on 2011-09-24 22:29

Again wanted to start google-earth, and as always it crashs, same problem as zyghom.
Hope 6.1 will fix this bug

Det commented on 2011-09-22 14:35

..but not for linux (yet).

alexcortes commented on 2011-09-21 18:32

Google Earth 6.1 is out.

Anonymous comment on 2011-09-14 17:22

@zyghom

Were you able to reproduce the crash with the older version of glibc? Did you update anything Xorg related before this crash occurred?

zyghom commented on 2011-09-08 04:47

Major Version 6
Minor Version 0
Build Number 0003
Build Date May 17 2011
Build Time 00:40:40
OS Type 3
OS Major Version 3
OS Minor Version 0
OS Build Version 0
OS Patch Version 0
Crash Signal 11
Crash Time 1315456967
Up Time 2.40408

Stacktrace from glibc:
./libgoogleearth_free.so(+0xab953)[0xb766f953]
./libgoogleearth_free.so(+0xabad3)[0xb766fad3]
[0xb76f7400]
/usr/lib/libexpat.so.1(XML_ParserFree+0x204)[0xb466e3d4]
/usr/lib/xorg/modules/dri/r600_dri.so(driParseConfigFiles+0x206)[0xaaef9e96]

lgomes commented on 2011-08-21 20:53

@Det,

Sorry.

You right !!!

But is an older version of GoogleEarthLinux in cache.

Thanks.

Det commented on 2011-08-21 18:23

@Igomes, then your download is corrupt because I still get '6808153226b99f21a66f5f615fcf7051'.

lgomes commented on 2011-08-21 17:59

Please update PKGBUILD for correct md5sums.

The GoogleEarthLinux.bin fail the md5sum check.

The md5sum now is 737743e5ce9f464d555a2a829baebea4

Thanks !!!

andrej84 commented on 2011-08-20 13:01

Ok no it didn't solve the probelm (I also installed sharpfonts).
Any other ideas?

andrej84 commented on 2011-08-20 12:59

I'm installing now ttf-ms-fonts to see if I solve the problem of the black boxes.
But shouldn't it be marked as a dependency of the packet if it's needed to work correctly?

fettouhi commented on 2011-08-18 17:07

Can anyone get anti-alised working with google-earth yet? I still have ugly fonts (readable but ugly).

hasufell commented on 2011-08-04 21:32

regarding the ugly fonts stuff I did 3 things
1. installed ttf-ms-fonts
2. installed sharpfonts (may not be necessary)
3. defined the correct DPI for my screen via xorg.conf (nvidia-driver did not correctly recognize DisplaySize, so DPI was not set properly)

after doing nr3 the fonts were finally readable

Madek commented on 2011-07-09 17:34

package orphaned, no time for maintain this

olive commented on 2011-07-08 11:53

@fowler I have had similar crash problems; I believe after upgrading some libraries. Removing the google-earth configuration files usually fix the issues.

rm -rf ~/.googleearth/ ~/.config/Google/GoogleEarthPlus.conf

Anonymous comment on 2011-07-06 00:18

If you are seeing a bunch of black boxes instead of text, install the package ttf-ms-fonts from AUR. That alone solved the problem for me.

Anonymous comment on 2011-06-16 17:11

I followed the instructions but it didn't work.

$ googleearth
ERROR: ld.so: object 'libfreeimage.so.3' from LD_PRELOAD cannot be preloaded: ignored.
Google Earth has caught signal 11.



We apologize for the inconvenience, but Google Earth has crashed.
This is a bug in the program, and should never happen under normal
circumstances. A bug report and debugging data have been written
to this text file:

/home/rick/.googleearth/crashlogs/crashlog-4dfa36f7.txt

Please include this file if you submit a bug report to Google.


Here is what is in the crash log... I don't know what to make of it...


$ less .googleearth/crashlogs/crashlog-4dfa36f7.txt
Major Version 6
Minor Version 0
Build Number 0003
Build Date May 17 2011
Build Time 00:40:40
OS Type 3
OS Major Version 2
OS Minor Version 6
OS Build Version 38
OS Patch Version 0
Crash Signal 11
Crash Time 1308243703
Up Time 0.010294

Stacktrace from glibc:
/opt/google-earth/libgoogleearth_free.so(+0xab953)[0xf7498953]
/opt/google-earth/libgoogleearth_free.so(+0xabad3)[0xf7498ad3]
[0xf77b2400]
/usr/lib32/libexpat.so.1(+0x2dfe)[0xf3ed8dfe]
/usr/lib32/libexpat.so.1(+0x2e30)[0xf3ed8e30]
/usr/lib32/libexpat.so.1(+0x58db)[0xf3edb8db]
/usr/lib32/libexpat.so.1(+0x896d)[0xf3ede96d]
/usr/lib32/libexpat.so.1(+0x9b51)[0xf3edfb51]
/usr/lib32/libexpat.so.1(XML_ParseBuffer+0x7c)[0xf3ee191c]
/usr/lib32/libfontconfig.so.1(FcConfigParseAndLoad+0x1eb)[0xf65a12bb]
/usr/lib32/libfontconfig.so.1(+0x1d9ab)[0xf65a19ab]
/usr/lib32/libexpat.so.1(+0x8c6d)[0xf3edec6d]
/usr/lib32/libexpat.so.1(+0x9b51)[0xf3edfb51]
/usr/lib32/libexpat.so.1(+0x6a8e)[0xf3edca8e]
/usr/lib32/libexpat.so.1(+0x826a)[0xf3ede26a]
/usr/lib32/libexpat.so.1(XML_ParseBuffer+0x7c)[0xf3ee191c]
/usr/lib32/libfontconfig.so.1(FcConfigParseAndLoad+0x1eb)[0xf65a12bb]
/usr/lib32/libfontconfig.so.1(FcConfigParseAndLoad+0x544)[0xf65a1614]
/usr/lib32/libfontconfig.so.1(+0x1d9ab)[0xf65a19ab]
/usr/lib32/libexpat.so.1(+0x8c6d)[0xf3edec6d]
/usr/lib32/libexpat.so.1(+0x9b51)[0xf3edfb51]
/usr/lib32/libexpat.so.1(+0x6a8e)[0xf3edca8e]
/usr/lib32/libexpat.so.1(+0x826a)[0xf3ede26a]
/usr/lib32/libexpat.so.1(XML_ParseBuffer+0x7c)[0xf3ee191c]
/usr/lib32/libfontconfig.so.1(FcConfigParseAndLoad+0x1eb)[0xf65a12bb]
/usr/lib32/libfontconfig.so.1(FcInitLoadConfig+0x42)[0xf6596062]
/usr/lib32/libfontconfig.so.1(FcInitLoadConfigAndFonts+0xc)[0xf659617c]
/usr/lib32/libfontconfig.so.1(FcInit+0x2d)[0xf65963dd]
/usr/lib32/libQtGui.so.4(+0x1b4cef)[0xf6894cef]
~

Berseker commented on 2011-06-12 08:04

guys let's try this fix for the ugly GUI
http://ubuntuforums.org/showpost.php?p=10898398&postcount=10

found also here some other ways, let0s find out what's the useful one for ArchLinux
http://www.google.com/support/forum/p/earth/thread?tid=3fe67ea84f63bcd8

Det commented on 2011-06-05 15:08

Welcome to the club.

Anonymous comment on 2011-06-05 10:57

Argh, fonts are really really ugly under kde 4.6. How can I fix this? I've read previous messages but I didn't find a solution. Thanks.

SanskritFritz commented on 2011-05-23 22:12

When I start GE, I get this message:
Another crash happened while handling crash!
lol

Det commented on 2011-05-21 21:26

You could try the basic, like updating your system, rebooting it, and finally, moving out the settings folder (~/.config/Google/ and ~/.googleearth) and let Google Earth create new vanilla configs.

Anonymous comment on 2011-05-21 19:41

After the last updates, all the characters in the interface are squares. Am I missing a font? Which one?

Anonymous comment on 2011-05-20 13:20

==> Validating source files with md5sums...
GoogleEarthLinux.bin ... FAILED
license.html ... Passed
googleearth-i686 ... Passed
googleearth-x86_64 ... Passed
googleearth.desktop ... Passed
googleearth-mimetypes.xml ... Passed
pangorc ... Passed
qt.conf ... Passed
==> ERROR: One or more files did not pass the validity check!
==> ERROR: Makepkg was unable to build google-earth.

Det commented on 2011-04-19 15:23

@akuschki, he probably doesn't have that problem anymore :).

Anonymous comment on 2011-04-19 13:18

@david.runge, try deleting the cached GoogleEarthLinux.bin file in your SRCDEST.

Madek commented on 2011-04-07 13:49

@det: sorry, update now
@david.runge, the md5sum works fine for me, you can try to use --skipinteg option in makepkg

Det commented on 2011-04-07 13:37

Btw. Madek, why didn't you bump the pkgver when you updated this thing?

dvzrv commented on 2011-04-07 12:38

Hmm, for some reason the md5sum keeps showing up as being invalid for GoogleEarthLinux.bin.
I've downloaded it separately and checked, but it's correct (as mentioned by impherring13). It's really awkward... keep getting the same issue, when trying to install bin32-google-earth.
Any clue on how to resolve this issue?

dvzrv commented on 2011-04-07 12:21

Hmm, for some reason the md5sum keeps showing up as being invalid for GoogleEarthLinux.bin.
I've downloaded it separately and checked, but it's correct (as mentioned by impherring13). It's really awkward... keep getting the same issue, when trying to install bin32-google-earth.
Any clue on how to resolve this issue?

Anonymous comment on 2011-03-30 09:40

Seal, you missed a character on the md5sum. Correct md5sum is '0f700978030040502dcf3daa5bdf3d14' version 6.0.2.2074

seal20 commented on 2011-03-30 08:20

The current version is 6.0.2.2074, build 3/10/2011. I could installed it by modyfing the PKGBUILD, md5sum for the new version is f700978030040502dcf3daa5bdf3d14 .

seal20 commented on 2011-03-30 08:18

The current version is 6.0.2.2074, I installed it by modyfing the PKGBUILD, md5sum for the new version is f700978030040502dcf3daa5bdf3d14 .

Cheers

Madek commented on 2011-03-28 12:32

lib32-catalyst-utils actually is a optdepend, because not all the people have a ati card

Anonymous comment on 2011-03-28 10:22

lib32-catalyst-utils is needed install of lib32-libgl if someone's using the catalyst drivers. It would be nice to include this.

Berseker commented on 2011-03-20 07:50

that's really bad, I feel I'm using windows 3.1 when GE is opened..

Madek commented on 2011-03-19 16:02

nop, this issue is because the different version of qt libs that GE uses
i didn't know how resolve it

Berseker commented on 2011-03-19 14:38

no solutions yet for the "look & feel" issue?

Fabioamd87 commented on 2011-03-08 16:18

binary shoudn't called google-earth (as google-chrome for example)?

Maxr commented on 2011-02-05 14:42

Ah ... alright. Aehm ;) Thanks, over and out.

Det commented on 2011-02-05 13:13

@Maxr, no (:D), they are _not_ required if you're on i686 Arch. That's what the "if [[ "$CARCH" == 'i686' ]]; then" check is there for.

So you're either using a bad AUR tool or screwed up the PKGBUILD yourself.

Maxr commented on 2011-02-05 13:07

Can't get this to work!

All those lib32-* lib dependencies are required, and can be obtained only from x64 multilib repository (working on i686). What to do?

Maxr commented on 2011-02-05 13:05

Can't get this to work!

All those lib32-* lib dependencies are required, and can be obtained only from x64 multilib repository. What to do?

SanskritFritz commented on 2011-01-22 21:28

I searched the google forums previously, people with similar problems got solutions, which didnt work for me. I'll ask though, you're right.

Det commented on 2011-01-22 21:24

Why don't you ask in the Google forums?

SanskritFritz commented on 2011-01-22 19:54

So, I tried everything I could find and think of: system libs on and off, install ms core fonts, editing the qt.conf file, setting the font in qtconf, deleting libs from /opt/google-earth, deleting ~/.config/Google, but nothing helped. I'm desperate now. What else can I do to have normal fonts back in GE 6? (In GE 5.2, the fonts look good)

Det commented on 2011-01-19 07:53

@supernatendo, please read the pacman output when installing/updating stuff - that's what it's there for. In this particular case you were adviced to install either (lib32-)catalyst-utils or (lib32-)nvidia-utils to actually get some 3D acceleration. In addition what you should please-not-do is that you should please-not post such long build logs in _here_ because it really fills up the comment space. Use pastebin as adviced for everything else.

@1LordAnubis, then don't keep your quality settings/the memory cache size limit so high.

@Madek, it seems this kinda crash log is about _use_system_libs: http://aur.pastebin.com/tMAyUN1C - and this kinda about not having (lib32-)nvidia/-catalyst-utils: http://aur.pastebin.com/WmzuUX9Y

1LordAnubis commented on 2011-01-19 02:52

seems to be using WAY too much ram

Madek commented on 2011-01-19 00:31

did you try to change _use_system_libs in PKGBUILD

supernatendo commented on 2011-01-18 23:44

Shows the splash screen, then crashes and creates a crash log.

Major Version 6
Minor Version 0
Build Number 0001
Build Date Dec 10 2010
Build Time 22:42:27
OS Type 3
OS Major Version 2
OS Minor Version 6
OS Build Version 36
OS Patch Version 0
Crash Signal 11
Crash Time 1295394176
Up Time 0.769111

Stacktrace from glibc:
/opt/google-earth/libgoogleearth_free.so(+0xac2a3)[0xf769f2a3]
/opt/google-earth/libgoogleearth_free.so(+0xac423)[0xf769f423]
[0xf76f0400]
/usr/lib32/libGL.so.1(glXMakeCurrent+0x23)[0xf650cf93]
/opt/google-earth/libIGGfx.so(_ZN3Gap3Gfx18igOglVisualContext19internalMakeCurrentEv+0x33)[0xf4351fc3]
/opt/google-earth/libIGGfx.so(_ZN3Gap3Gfx18igOglVisualContext21internalCreateContextEv+0x1af)[0xf435431f]
/opt/google-earth/libIGGfx.so(_ZN3Gap3Gfx18igOglVisualContext4openEv+0x93)[0xf435d303]
/opt/google-earth/libevll.so(_ZN5earth4evll13VisualContext11OpenContextEN3Gap3Gfx25igRenderDestinationFormatERKNS0_8InitInfoE+0xff)[0xf15d8d8f]
/opt/google-earth/libevll.so(_ZN5earth4evll13VisualContext4initERKNS0_8InitInfoE+0x18e)[0xf15db75e]
/opt/google-earth/libevll.so(_ZN5earth4evll17RenderContextImpl4initERKNS0_8InitInfoE+0x7e)[0xf153a28e]
/opt/google-earth/librender.so(_ZN12RenderWidget6SetApiEPN5earth4evll3APIE+0x47)[0xf4a9f8c7]
/opt/google-earth/librender.so(_ZN5earth6render12RenderWindow12createWidgetEv+0x16a)[0xf4a8430a]
/opt/google-earth/libgoogleearth_free.so(_ZN5earth6client12ModuleWidget9showEventEP10QShowEvent+0x8d)[0xf7679b9d]
/opt/google-earth/libQtGui.so.4(_ZN7QWidget5eventEP6QEvent+0x770)[0xf6a2da60]
/opt/google-earth/libQtGui.so.4(_ZN19QApplicationPrivate13notify_helperEP7QObjectP6QEvent+0xac)[0xf69ca69c]
/opt/google-earth/libQtGui.so.4(_ZN12QApplication6notifyEP7QObjectP6QEvent+0x484)[0xf69d56c4]
/opt/google-earth/libQtCore.so.4(_ZN16QCoreApplication14notifyInternalEP7QObjectP6QEvent+0x78)[0xf7461ba8]
/opt/google-earth/libQtGui.so.4(_ZN14QWidgetPrivate11show_helperEv+0x145)[0xf6a2f6d5]
/opt/google-earth/libQtGui.so.4(_ZN14QWidgetPrivate14show_recursiveEv+0x7d)[0xf6a2f3cd]
/opt/google-earth/libQtGui.so.4(_ZN14QWidgetPrivate12showChildrenEb+0xc5)[0xf6a2f4b5]
/opt/google-earth/libQtGui.so.4(_ZN14QWidgetPrivate11show_helperEv+0x61)[0xf6a2f5f1]
/opt/google-earth/libQtGui.so.4(_ZN7QWidget10setVisibleEb+0x1db)[0xf6a2fa2b]
/opt/google-earth/libQtGui.so.4(_ZN14QWidgetPrivate12showChildrenEb+0x170)[0xf6a2f560]
/opt/google-earth/libQtGui.so.4(_ZN14QWidgetPrivate11show_helperEv+0x61)[0xf6a2f5f1]
/opt/google-earth/libQtGui.so.4(_ZN14QWidgetPrivate14show_recursiveEv+0x7d)[0xf6a2f3cd]
/opt/google-earth/libQtGui.so.4(_ZN14QWidgetPrivate12showChildrenEb+0xc5)[0xf6a2f4b5]
/opt/google-earth/libQtGui.so.4(_ZN14QWidgetPrivate11show_helperEv+0x61)[0xf6a2f5f1]
/opt/google-earth/libQtGui.so.4(_ZN7QWidget10setVisibleEb+0x1db)[0xf6a2fa2b]
/opt/google-earth/libQtGui.so.4(_ZN14QWidgetPrivate12showChildrenEb+0x170)[0xf6a2f560]
/opt/google-earth/libQtGui.so.4(_ZN14QWidgetPrivate11show_helperEv+0x61)[0xf6a2f5f1]
/opt/google-earth/libQtGui.so.4(_ZN14QWidgetPrivate14show_recursiveEv+0x7d)[0xf6a2f3cd]
/opt/google-earth/libQtGui.so.4(_ZN14QWidgetPrivate12showChildrenEb+0xc5)[0xf6a2f4b5]
/opt/google-earth/libQtGui.so.4(_ZN14QWidgetPrivate11show_helperEv+0x61)[0xf6a2f5f1]
/opt/google-earth/libQtGui.so.4(_ZN7QWidget10setVisibleEb+0x1db)[0xf6a2fa2b]
/opt/google-earth/libQtGui.so.4(_ZN14QWidgetPrivate12showChildrenEb+0x170)[0xf6a2f560]
/opt/google-earth/libQtGui.so.4(_ZN14QWidgetPrivate11show_helperEv+0x61)[0xf6a2f5f1]
/opt/google-earth/libQtGui.so.4(_ZN7QWidget10setVisibleEb+0x1db)[0xf6a2fa2b]
/opt/google-earth/libQtGui.so.4(_ZN14QWidgetPrivate12showChildrenEb+0x170)[0xf6a2f560]
/opt/google-earth/libQtGui.so.4(_ZN14QWidgetPrivate11show_helperEv+0x61)[0xf6a2f5f1]
/opt/google-earth/libQtGui.so.4(_ZN7QWidget10setVisibleEb+0x1db)[0xf6a2fa2b]
/opt/google-earth/libQtGui.so.4(_ZN14QWidgetPrivate12showChildrenEb+0x170)[0xf6a2f560]
/opt/google-earth/libQtGui.so.4(_ZN14QWidgetPrivate11show_helperEv+0x61)[0xf6a2f5f1]
/opt/google-earth/libQtGui.so.4(_ZN7QWidget10setVisibleEb+0x1db)[0xf6a2fa2b]
/opt/google-earth/libQtGui.so.4(_ZN7QWidget10showNormalEv+0x5c)[0xf6a1d16c]
/opt/google-earth/libgoogleearth_free.so(_ZN10MainWindow18readScreensizeInfoEv+0xd2f)[0xf766c08f]
/opt/google-earth/libgoogleearth_free.so(_ZN5earth6client11Application12SetupMainWinENS0_3Kvw7ProductEb+0x31c)[0xf76a625c]
/opt/google-earth/libgoogleearth_free.so(_ZN5earth6client11Application3runEv+0x4d6)[0xf76a9406]
/opt/google-earth/libgoogleearth_free.so(+0xaad1b)[0xf769dd1b]
/opt/google-earth/libgoogleearth_free.so(earthmain+0x247)[0xf769eed7]
./googleearth-bin[0x804871b]
/usr/lib32/libc.so.6(__libc_start_main+0xe6)[0xf5220c66]
./googleearth-bin[0x8048661]

Det commented on 2011-01-13 19:20

You and your segfaults. I couldn't reproduce that even by using _use_system_libs="yes", changing the ownership of "~/.config/Google/" and "~/.googleearth" to 'root' or replacing lib32-nvidia-utils(-beta) with lib32-libgl.

Did you screw up your installation files' permissions somehow?

Huulivoide commented on 2011-01-13 18:56

This segfaults unless its run as root

Anonymous comment on 2011-01-09 02:09

@dront78 Try using 'shake' for a good 15 mins, it usually works out seg faults...
p.s. it's in the repos

Det commented on 2011-01-07 16:06

@SanskritFritz, I was talking about this: http://www.google.com/support/forum/p/earth/thread?tid=71e5e545d1ec17fd&hl=en (the second last post. Those other links might be helpful too).

> BTW, where do I put the use_system_libs switch, I want to experiment with it.
You don't put it anywhere. It's on top of the PKGBUILD and you just change the "no" to "yes". You are free to experiment with it all you like, of course, but it just won't work. The repos' libraries are too new for GE.

SanskritFritz commented on 2011-01-07 15:02

Det, I cannot remember what qtconfig trick you are referring to. In GE5 I deleted the local qt libs, that helped. BTW, where do I put the use_system_libs switch, I want to experiment with it.

Madek commented on 2011-01-07 14:39

and the segfault is???

Det commented on 2011-01-07 10:08

_This_? That's kinda helpful. Well, did you remove lib32-nvidia/-catalyst-utils in the meantime?

dront78 commented on 2011-01-07 08:30

I've got this segfault, while bin32-google-earth works fine.
x86_64, use_system_libs=no

donvla commented on 2011-01-05 17:51

Hi all, I get this error:
"$ googleearth
/usr/bin/googleearth: line 16: /opt/google-earth/googleearth-bin: No such file or directory
/usr/bin/googleearth: line 16: /opt/google-earth/googleearth-bin: Success
"
I remember I had this also with my x86_64 installation and disappeared after an update.

Det commented on 2011-01-04 22:16

Did you try reverting the qtconfig trick you did with GE5?

SanskritFritz commented on 2011-01-04 19:46

Updated to latest GE. Here is a screenshot:
http://www.wuala.com/SanskritFritz/Public/2011-01-04-203520_1440x900_scrot.png/?mode=download
Note that the fonts in GE are way too wide and looking strange.

Det commented on 2011-01-04 13:46

Great =). Just as a note there's no need to bump the pkgrel every time. Nothing changed in the package itself.

Madek commented on 2011-01-04 12:54

@det
i didn't notice your message about libxrender, fixed it now
the think about ld-lsb is that this package is only a symlink to ld-lsb.so.2 to ld-lsb.so.3, this is the reason to =3 in pkgbuild. i think is better than theese way, to put directly in the pkgbuild the symlink

the only think is pending is a use_system_libs runtime for i686

Det commented on 2011-01-04 10:34

Lol, wasn't said only but three times already.

silvik commented on 2011-01-04 09:57

GoogleEarthLinux.bin md5sum problem

Det commented on 2011-01-04 06:30

@Madek, Actually, the md5sum of the Google Earth binary truly is wrong :). It should be the same old '737743e5ce9f464d555a2a829baebea4'.

I assume you didn't notice my '(lib32-)libxrender' dependency comment either. It's not actually the (lib32-)libx_randr_ Google Earth needs. It's the _render_ (a dependency of libxrandr).

Then there's the x86_64 dependency 'lib32-gcc-libs' too (unlike 'gcc-libs' the lib32-one is in [extra]). It's a dependency of 'lib32-pcre', I know, but it's still required to be listed as a dependency.

Another thing is this 'ld-lsb=3' dependency. The package hasn't been updated since it was released so what's the version number doing there? And why don't you have the number in the x86_64 dependencies too?

In addition I liked the way you implemented the "$CARCH stuff" (and this is kinda going to nitpicking) but you could do some indenting there, like this (has these other changes too): http://aur.pastebin.com/z6fYSrKN

Lastly, there _still_ slipped some nonsense of mine to this line:
[[ "${_use_system_libs}" == "yes" ]] && depends+=('curl' 'mesa' 'qt')
..which should (of course) be:
[[ "${CARCH}" == "i686" ]] && [[ "${_use_system_libs}" == "yes" ]] && depends+=('curl' 'mesa' 'qt')
..so that the i686 dependencies won't mess up the x86_64 ones.

Everything else looks great even on my standards (can't imagine why) :).

Det commented on 2011-01-04 06:28

@Madek, Actually, the md5sum of the Google Earth binary truly is wrong :). It should be the same old '737743e5ce9f464d555a2a829baebea4'.

I assume you didn't notice my '(lib32-)libxrender' dependency comment either. It's not actually the (lib32-)libx_randr_ Google Earth needs. It's the _render_ (a dependency of libxrandr).

Then there's the x86_64 dependency 'lib32-gcc-libs' too (unlike 'gcc-libs' the lib32-one is in [extra]). It's a dependency of 'lib32-pcre', I know, but it's still required to be listed as a dependency.

Another thing is this 'ld-lsb=3' dependency. The package hasn't been updated since it was released so what's the version number doing there? And why don't you have the number in the x86_64 dependencies too?

In addition I liked the way you implemented the "$CARCH stuff" (and this is kinda going to nitpicking) but you could do some indenting there, like this:

Lastly, there _still_ slipped some nonsense of mine to this line:
[[ "${_use_system_libs}" == "yes" ]] && depends+=('curl' 'mesa' 'qt')
..which should (of course) be:
[[ "${CARCH}" == "i686" ]] && [[ "${_use_system_libs}" == "yes" ]] && depends+=('curl' 'mesa' 'qt')
..so that the i686 dependencies won't mess up the x86_64 ones.

Everything else looks great even on my standards (can't imagine why) :).

Anonymous comment on 2011-01-04 06:17

I have the same problem with md5sum.
$ md5sum GoogleEarthLinux.bin
737743e5ce9f464d555a2a829baebea4 GoogleEarthLinux.bin

Madek commented on 2011-01-04 04:34

update: now this package support i686 and x86_64, thanks to det for the great pkgbuild
please test and notify any problem with this

Madek commented on 2011-01-04 04:12

md5sum is fine
@SanskritFritz can you share a screenshot with your problem?

lowfatcomputing commented on 2011-01-04 02:53

The md5sums are wrong.

SanskritFritz commented on 2011-01-03 21:45

I already tried all that, none helped. Sigh...

Det commented on 2011-01-03 21:38

Well can't you just _choose_ a better font in: 'Options' -> '3D View' tab (the first one) -> 'Choose 3D Font'? Or just try changing the system Qt font in KDE System Settings?

SanskritFritz commented on 2011-01-03 21:23

Deleting all config files didnt help either. :(

Det commented on 2011-01-03 17:11

~/.config/Google/GoogleEarthPlus.conf

E: @Madek, (lib32-)libxrender is also a dependency of Google Earth. You should include that too.

Det commented on 2011-01-03 17:06

~/.config/Google/GoogleEarthPlus.conf

SanskritFritz commented on 2011-01-03 14:29

I'll try to delete all configs. What directories should be deleted besides ~/.googleearth ?

SanskritFritz commented on 2011-01-03 14:23

@Det, I already have it installed. GE5 looked ok, I think it used the same font?

Det commented on 2011-01-03 14:00

@SanskritFritz, did you try installing 'ttf-bitstream-vera'?

SanskritFritz commented on 2011-01-02 23:47

The fonts are ugly. In GE5 there was a solution with the qt libs, that doesnt seem to work in GE6. Any solution?

Det commented on 2010-12-30 15:55

Oh, and 'lib32-nvidia-utils' and 'lib32-catalyst-utils' should indeed be mentioned in the optional dependencies (as done with bin32-google-earth) since majority of the users require them. Additionally I had a typo there that the file "googleearth-x86-64" should actually be "googleearth-x86_64" ("$CARCH" returns either "i686" or "x86_64"). There was also some tweaking with the "_use_system_libs" variable and 'lib32-gcc-libs' was missing from the i686 dependencies ('gcc-libs' is included in 'base-devel').

Here's the whole thing as a single tarball: http://www49.zippyshare.com/v/88017609/file.html

Det commented on 2010-12-29 21:56

:D.

Madek commented on 2010-12-29 21:17

yep, something like this

Madek commented on 2010-12-29 21:14

yep. something like this???

Det commented on 2010-12-29 20:24

Something like this? (the .install and .desktop files are already included with this thing):

PKGBUILD: http://aur.pastebin.com/iskqnc2p
googleearth-i686: http://aur.pastebin.com/gQkbXNqL
googleearth-x86-64: http://aur.pastebin.com/uHRMGJ4J
googleearth-mimetypes.xml: http://aur.pastebin.com/eB80ZRVp
pangorc: http://aur.archlinux.org/packages/bin32-google-earth/bin32-google-earth/pangorc
qt.conf: http://aur.archlinux.org/packages/bin32-google-earth/bin32-google-earth/qt.conf

You can (of course) change all that "[[ "$CARCH" == "x86_64" ]] &&" stuff in the PKGBUILD to that "if lol; then" stuff, in case it looks ugly to you the way it is.

Madek commented on 2010-12-28 15:18

may be, in the next days (january) i ve start this

kevku commented on 2010-12-27 15:43

could this PKGBUILD be made like acroread to support both x86_64 and i686?

Anonymous comment on 2010-12-27 11:20

Why in my case it keeps failing to start since some point of 5.x...

edacval commented on 2010-12-26 19:52

PKGBUILD cleanup patch:
http://aur.pastebin.com/NY1NJXq6

Berseker commented on 2010-12-18 13:13

gt_swagger, please update the package. Already a second version of Google Earth 6 is out.

If you don't have time for mantain a package, please orphan it so someone else could take care of it

fettouhi commented on 2010-12-07 19:33

Is there going to be an update any time soon?

fettouhi commented on 2010-12-07 19:33

Is there going to be an update any time soon?

Berseker commented on 2010-12-04 11:14

If someone finds a way to beutify the UI please post here. I have done various experiments but I'm still unsuccessful to find a way to do this.

OttoA commented on 2010-12-01 18:51

solstice: I don't know that for sure and I found it cleaner than to do it that way than to do it in the install script. I wouldn't be surprised if google used this in some other products.

solstice commented on 2010-12-01 18:49

@0ttoA: I find that a package just for a symlink is a little too much. Especially if the only package using it is and will be google-earth.

OttoA commented on 2010-11-30 19:47

You can try this version: https://github.com/downloads/OttoAllmendinger/aur-packages/google-earth-6.0.0.1735-1.src.tar.gz

it depends on the ld-lsb package here:
http://aur.archlinux.org/packages.php?ID=44043

OttoA commented on 2010-11-30 18:17

@zephiros: not really google earth 6 then, is it?

Anonymous comment on 2010-11-30 18:14

Copy old (5.2) googleearth-bin to /opt/google-earth/ and google earth 6 work fine

OttoA commented on 2010-11-30 17:28

@banana See my comments. This seems to allow it to start at least:

sudo ln -s /lib/ld-linux.so.2 /lib/ld-lsb.so.3

Which is some linux-standard-base stuff. I wouldn't put it in the package just yet.

The browser-related crashes seem to be gone, but it is still unstable and doesn't display panoramino pictures (for me at least)

machoo02 commented on 2010-11-30 17:27

I think the answer lies here: http://earth.google.com/support/bin/answer.py?hl=en&answer=40901

Linux Builds:Google Earth Linux binaries are now based on the Linux Standard Base. This improves the compatibility of Google Earth on different Linux distributions, but requires users to have Linux Standard Base installed on their system. If your Linux system does not have LSB, please install it using apt, yum, smart or up2date. As a side effect of moving to LSB, joysticks are currently unsupported.

bananaoomarang commented on 2010-11-30 17:19

What the hell's up with Google earth on Linux. First it won't install normally so you have to extract it, rename a file and run setup.sh manually. Then it just says this:



/usr/bin/googleearth: line 26: /opt/google-earth/googleearth-bin: No such file or directory
/usr/bin/googleearth: line 26: /opt/google-earth/googleearth-bin: Success

But it DOES exist, as shown in the comment bellow (and my machine). It's like this in 6.0 too, is it brokenon all distros, or just arch?

Huulivoide commented on 2010-11-30 13:20

Eould it be possible to intergate the 32bit version tothis PKGBUILD
It would be nice as i think the bin32 apps are ugly and it would
not be hard to intergtae it too only couple of if [ "$CARCH....
lines and its redy

OttoA commented on 2010-11-30 07:07

Please star this bug: http://code.google.com/p/earth-issues/issues/detail?id=1068

OttoA commented on 2010-11-30 06:50

Version 6 is out but is still FAIL

# ./googleearth-bin
zsh: no such file or directory: ./googleearth-bin

# ldd googleearth-bin
/usr/bin/ldd: line 126: ./googleearth-bin: No such file or directory

# /lib/ld-linux.so.2 --verify --list ./googleearth-bin
./googleearth-bin: error while loading shared libraries: libgoogleearth_free.so: cannot open shared object file: No such file or directory

# file libgoogleearth_free.so
libgoogleearth_free.so: ELF 32-bit LSB shared object, Intel 80386, version 1 (SYSV), dynamically linked, stripped

OttoA commented on 2010-11-20 15:38

tthere is a patch on the debian bug tracker http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=596423

Anonymous comment on 2010-11-16 13:59

To disable tips you don't need to be the fastest clicker in the whole whole west.
Just edit the file $HOME/.config/Google/GoogleEarthPlus.conf. Under the [General] section write "enableTips=false".

Anonymous comment on 2010-11-07 07:05

patch works for me.
thanks OttoA.

Berseker commented on 2010-10-12 12:38

Mre you've to install bin32-google-earth on a 64bit system
http://aur.archlinux.org/packages.php?ID=12890

Aerion commented on 2010-10-11 13:14

I'm getting the following error:

==> ERROR: is not available for the 'x86_64' architecture.
Note that many packages may need a line added to their PKGBUILD
such as arch=('x86_64').

Berseker commented on 2010-10-06 09:52

i forgot to mention that this beaviour is with the "use arch libs" var set to yes

Berseker commented on 2010-10-06 08:44

I made a big error some time ago when i posted the qt.conf.. please correct it because it refers to lib32 libs. here is the correct one

[Paths]
Documentation=/usr/share/doc
Libraries=/usr/lib
Plugins=/usr/lib/qt/plugins
Translations=/usr/share/qt/translations

with this GE starts without errors, but the old "white square instead of map controls" bug appears

Anonymous comment on 2010-10-02 18:33

Confirmed working after updating.

TheChessPlayer commented on 2010-10-02 12:14

Apparently Google Earth crash is suddenly a thing of the past for me. I suspect that it was solved by the QT update to version 4.7.

hamelg commented on 2010-10-01 19:35

I used kdemod and have switched back to the Arch Linux repos.
It fixes the crash at startup :)
Now it crashes at exit ...
*** glibc detected *** ./googleearth-bin: corrupted double-linked list: 0xb1158330 ***
======= Backtrace: =========
/lib/libc.so.6(+0x6c7b1)[0xb5a3e7b1]
/lib/libc.so.6(+0x6cc16)[0xb5a3ec16]
/lib/libc.so.6(+0x6d4ba)[0xb5a3f4ba]
/lib/libc.so.6(cfree+0x6d)[0xb5a431cd]
/usr/lib/libnvidia-glcore.so.256.53(+0xd854bd)[0xb3f0c4bd]

hamelg commented on 2010-10-01 19:34

I used kdemod and have switched switch back to the Arch Linux repos.
It fixes the crash at startup :)
Now it crashes at exit ...
*** glibc detected *** ./googleearth-bin: corrupted double-linked list: 0xb1158330 ***
======= Backtrace: =========
/lib/libc.so.6(+0x6c7b1)[0xb5a3e7b1]
/lib/libc.so.6(+0x6cc16)[0xb5a3ec16]
/lib/libc.so.6(+0x6d4ba)[0xb5a3f4ba]
/lib/libc.so.6(cfree+0x6d)[0xb5a431cd]
/usr/lib/libnvidia-glcore.so.256.53(+0xd854bd)[0xb3f0c4bd]


hamelg commented on 2010-10-01 19:32

I used kdemod and have switched switch back to the Arch Linux repos.
It fixes the crash at startup :)
Now it crashes at exit ...
*** glibc detected *** ./googleearth-bin: corrupted double-linked list: 0xb1158330 ***
======= Backtrace: =========
/lib/libc.so.6(+0x6c7b1)[0xb5a3e7b1]
/lib/libc.so.6(+0x6cc16)[0xb5a3ec16]
/lib/libc.so.6(+0x6d4ba)[0xb5a3f4ba]
/lib/libc.so.6(cfree+0x6d)[0xb5a431cd]
/usr/lib/libnvidia-glcore.so.256.53(+0xd854bd)[0xb3f0c4bd]


luuuciano commented on 2010-09-29 20:55

@luspi I used to do that, but anyway some things in the GE interface never showed up, and other really annoying problems... I have downgrade it to 5.0.11337.1968-1, it works awesome

luspi commented on 2010-09-29 17:01

It works here (Intel Graphic Chip)!! You just have to be quick in disabling the startup tips, then it doesn't crush.

hamelg commented on 2010-09-28 19:12

applied the patch submitted by OttoA, but it continues to crash at startup

Stacktrace from glibc:
./libgoogleearth_free.so(+0xd090b)[0xb768c90b]
[0xb76e3400]
/lib/libc.so.6(abort+0x17e)[0xb59ba5ee]
./libgoogleearth_free.so(+0xcf83e)[0xb768b83e]
./libQtCore.so.4(_Z17qt_message_output9QtMsgTypePKc+0x2c)[0xb72781e4]
./libQtCore.so.4(+0x5b2f8)[0xb72782f8]
./libQtCore.so.4(_Z6qFatalPKcz+0x19)[0xb7278359]
./libQtCore.so.4(_ZN14QObjectPrivateC2Ei+0x9c)[0xb7366e00]
/usr/lib/libQtDBus.so.4(+0x2e1f7)[0xb21331f7]
/usr/lib/libQtDBus.so.4(_ZN22QDBusAbstractInterfaceC1ERK7QStringS2_PKcRK15QDBusConnectionP7QObject+0x74)[0xb21367a4]
/usr/lib/libQtDBus.so.4(_ZN24QDBusConnectionInterfaceC1ERK15QDBusConnectionP7QObject+0x7a)[0xb211737a]
/usr/lib/libQtDBus.so.4(+0xf5de)[0xb21145de]
/usr/lib/libQtDBus.so.4(_ZN15QDBusConnection12connectToBusENS_7BusTypeERK7QString+0x1af)[0xb211583f]
/usr/lib/libQtDBus.so.4(_ZN15QDBusConnection10sessionBusEv+0xa7)[0xb2115aa7]
/usr/lib/libkdeui.so.5(_ZN15KGlobalSettingsC1Ev+0xac)[0xb27f3abc]
/usr/lib/libkdeui.so.5(_ZN15KGlobalSettings4selfEv+0x57)[0xb27f3bc7]
/usr/lib/kde4/plugins/gui_platform/libkde.so(qt_plugin_instance+0xa8)[0xb2c0a618]
./libQtCore.so.4(_ZNK14QFactoryLoader8instanceERK7QString+0x240)[0xb7342044]
./libQtGui.so.4(+0x1a6115)[0xb6b3c115]
./libQtGui.so.4(_ZN19QApplicationPrivate18x11_apply_settingsEv+0x1cfa)[0xb6b4373e]
./libQtGui.so.4(+0x1ae8c7)[0xb6b448c7]
./libQtGui.so.4(+0x1b2d99)[0xb6b48d99]
./libQtGui.so.4(_ZN19QApplicationPrivate9constructEP9_XDisplaymm+0x54)[0xb6ada18c]
./libQtGui.so.4(_ZN12QApplicationC1ERiPPci+0x55)[0xb6adab9d]
./libgoogleearth_free.so(_ZN5earth6client11ApplicationC1EiPPc+0x9d)[0xb7697e1d]
./libgoogleearth_free.so(earthmain+0x275)[0xb768bd35]
./googleearth-bin(_init+0x12e)[0x80486d2]
/lib/libc.so.6(__libc_start_main+0xe6)[0xb59a4c76]
./googleearth-bin(_init+0x9d)[0x8048641]

Anonymous comment on 2010-09-27 21:46

Have successfully built and launched 5.2.1.1588 on ATI x1250.
- the problem of non-reflecting images in pop-up windows is solved
- 'my places' saved from previous version were displaced somewhat southwards (LC_NUMERIC=C would not solve the matter)
- got Sig11 crash when openning pop-up windows (Xorg.0.log hasn't recorded anything)

Anonymous comment on 2010-09-27 19:20

If none of owners of ati cards eager to try the 5.2 version with a modified build I'll try to launch it tonight. I have an old x1250 with an open-source gallium driver.

Berseker commented on 2010-09-27 10:49

if someone succedes to use the arch libs (see var in the PKGBUILD) and obtaining a working google-earth, let me know :|

Anonymous comment on 2010-09-27 00:41

How about a new modified PKGBUILD?
Thanks.

tocer commented on 2010-09-26 01:04

still crash in ATI 9200se with ATI open driver

Anonymous comment on 2010-09-25 18:26

@ gt_swagger good build here with nvidia card glad its not crashing

OttoA commented on 2010-09-25 13:46

proper patch against latest PKGBUILD http://gist.github.com/596850

Berseker commented on 2010-09-25 06:59

@gt_swagger nothing ;) glad to have been useful
@ottoa I didn't understand where you have to put your "patch"

OttoA commented on 2010-09-24 18:10

To be precise, the plugins are the culprit here - libQtWebkit crashes on loading them. So here is my submission for the crudest patch of the year award:


sudo sed -bi "s:plugins:FAILins:g" libQtWebKit.so.4
export MOZ_PLUGIN_PATH=/var/empty
export QTWEBKIT_PLUGIN_PATH=/var/empty
export MOZILLA_HOME=/var/empty

Now GE starts without crashing on startup

OttoA commented on 2010-09-24 17:15

For those who experience crashes at the startup-tip and the picture display, the problem lies with libWebkit I think. Strace suggests that the .so file shipped from google is loadad, so it is difficult to pin down exactly what is going wrong.

Anonymous comment on 2010-09-24 17:13

D'oh! Sorry, I meant Berseker. I was disgusted I couldn't get it to work on Intel and in a hurry. :)

OttoA commented on 2010-09-24 17:10

@gt_swagger are you sure you're not confusing me with Berseker? ;)

Anonymous comment on 2010-09-24 16:10

New version:
- You still really don't want to upgrade.
- Somewhat functional but still very crashprone on nVidia. Your mileage may vary.
- Still pretty hopeless on Intel.
- I have no ATI test machines, but we all know that's probably the worst case scenario :)

Thanks to OttoA for a cleaned up and more versatile PKGBUILD.

OttoA commented on 2010-09-23 08:47

Done all that, still crashes.

Using i686 with latest nvidia drivers (256.53-1) and GeForce 7600 GT.

OttoA commented on 2010-09-23 08:47

Done all that, using i686 with latest nvidia drivers (256.53-1) and GeForce 7600 GT

Berseker commented on 2010-09-23 08:37

OttoA, try to erase .googleearth in your home directory, remove the previous package, be sure that /opt/googleearth isn't there anymore, and try to reinstall mine's or gt_swagger's PKGBUILD.
Also if it doesn't work, specify your system config so we can make some guessing on what's wrong in your install

OttoA commented on 2010-09-23 08:36

Closing the startup tip seems to work somewhat

- clicking on pictures doesn't always work, sometimes they are only displayed when the window is resized
- the internal browser crashes (same problem as the tooltip probably)

I propose adding these lines to the startup scripts: http://gist.github.com/593337

OttoA commented on 2010-09-23 07:59

Closing the startup tip works seems to work - after that I can even click pictures etc.

I see if I can come up with a patch that disables the start-up tip

OttoA commented on 2010-09-23 07:56

@Berserker: segfault when displaying the start-up tip

Anonymous comment on 2010-09-23 07:53

The new version at least works for my nvidia box ... when I get a moment I'll give it a go on my Intel one. Regardless I'll push an updated PKGBUILD today. It'd be nice if this new version FINALLY ... FINALLY works for at least intel and nvidia such that you can have one without breaking the other.

OttoA commented on 2010-09-23 07:51

@Berserker not for me - also sig11 on startup

Berseker commented on 2010-09-23 07:31

http://berseker86.altervista.org/google-earth_lastest.tar.gz
with this i get a working google-earth, lastest version. Do not change the "use_arch_lib" var because by using them google-earth crashes istantly.
I'm on a i6868 system with nvidia

OttoA commented on 2010-09-22 05:54

New version crashes instantly

Madek commented on 2010-09-22 02:12

5.2.1.1588 is out

kolos commented on 2010-09-20 13:23

@OttoA nope, also the external links gives me an error ($BROWSER variable Environnement), which is odd since I set chromium as default browser in ~/.bashrc and in the xdg-open script.

OttoA commented on 2010-09-20 07:23

@kolos: do pictures work for you (clicking on an image icon on the map)

KriK commented on 2010-09-20 02:12

And it's crashed =)

KriK commented on 2010-09-20 02:06

There's incorrect md5sum for GoogleEarthLinux.bin in PKGBUILD

kolos commented on 2010-09-19 19:54

ok, using the http://aur.pastebin.com/AhGWy3pv of @snowstorm still crashed when it was loading, but... when i tried to run it again i cliked over "close" button of the starup tips and now is working for me, so it was the freaking starup tip window the one that made the program crash on my pc.
sorry for my english.

kolos commented on 2010-09-19 19:53

ok, using the http://aur.pastebin.com/AhGWy3pv of @snowstorm still crashed when it was loading, but... when i tried to run it again i cliked over "close" button of the starup tips and now is working for me, so it was the freaking starup tip window the one that made the program crash on my pc.
sorry for my english.

OttoA commented on 2010-09-17 11:32

@snostorm: nope, still crashes for me

please everybody star the issues under http://bit.ly/b04zhG

snostorm commented on 2010-09-17 10:33

Here are the changes I made to get 5.2.1.1588 working under my chroot with NVIDIA drivers
althought the key is the QT libs from google need to stay as the current arch AT is 4.6.3 whcih is waht seemed to be causing my issues.
I changed the MD5Sum for the install bin and the remove section thath removed Google libs in favor of Arch provided libs


http://aur.pastebin.com/AhGWy3pv

Anonymous comment on 2010-09-16 17:37

Caught Seg11 from Xorg (ati-gallium) on 5.1.3535.3218 version. See report: https://bbs.archlinux.org/viewtopic.php?id=105026

Anonymous comment on 2010-09-06 13:58

Use previous version (5.1.3535.3218)
http://dl.google.com/earth/client/advanced/previous/GoogleEarthLinux.bin

Anonymous comment on 2010-09-06 11:20

the same problem anthonio has...
Google Earth crashes.

OttoA commented on 2010-09-05 17:27

Google Earth for linux seems to be a pile of shit: http://bit.ly/b04zhG - please star the linux crash issues so that the developers will notice

Just tried the windows version (13mb instead of 30mb!) in wine and it works perfectly.

Anonymous comment on 2010-09-05 14:25

Google updates binary to 5.2.1.1588 version but it crashes after runs.
anthonio ~ $ googleearth
Google Earth has caught signal 11.

We apologize for the inconvenience, but Google Earth has crashed.
This is a bug in the program, and should never happen under normal
circumstances. A bug report and debugging data have been written
to this text file:

/home/anthonio/.googleearth/crashlogs/crashlog-4c83a813.

Please include this file if you submit a bug report to Google.

artemklevtsov commented on 2010-09-04 15:29

Edit PKGBUILD. First raw in md5sums section.
New sum - 19ccb5dff2b1165c2a3644cbf6d680ee

kolos commented on 2010-08-30 19:58

md5sum error, i'm a noob so i don't know how to fix it :(

tocer commented on 2010-08-28 01:37

same error as OttoA

bkuri commented on 2010-08-28 01:08

Latest MD5SUM for GoogleEarthLinux.bin is 3d6cc89e17daa361f6087ea495d06a84

bkuri commented on 2010-08-28 01:08

Latest MD5SUM for GoogleEarthLinux.bin is 3d6cc89e17daa361f6087ea495d06a84

bkuri commented on 2010-08-28 00:56

Latest MD5SUM for GoogleEarthLinux.bin is 3d6cc89e17daa361f6087ea495d06a84

Anonymous comment on 2010-08-25 17:53

Hey dude, the md5sum of the Google Earth file isn't matching, please fix it: 3d6cc89e17daa361f6087ea495d06a84

OttoA commented on 2010-08-22 19:15

commenting out line 56 does at least enable google-earth to show the splash screen before crashing

Anonymous comment on 2010-08-22 18:37

No images in pop-up windows bug reported in freedesktop: https://bugs.freedesktop.org/show_bug.cgi?id=29732

OttoA commented on 2010-08-22 18:05

new version just crashes with signal 11

Anonymous comment on 2010-08-22 18:03

@redhack: look at comment below posted by 'hoobastank'.

Anonymous comment on 2010-08-22 17:45

IN yaourt


-> Se encontró googleearth
-> Se encontró googleearth.desktop
-> Se encontró googleearth-mimetypes.xml
-> Se encontró googleearth.install
==> ERROR: Las pruebas de integridad (md5) difieren en tamaño del arreglo fuente.
==> ERROR: Makepkg was unable to build google-earth.
==> Restart building google-earth ? [y/N]
==> -------------------------------------
==>
==> WARNING: Following packages have not been installed:
google-earth


help, please :P

Anonymous comment on 2010-08-22 14:53

@jose1711: thank you for troubling, i suppose the matter has to do with 2D graphics processing from the xorg.freedesktop.org git repository. I'll try to dig that direction.

jose1711 commented on 2010-08-22 11:12

@zandgreen: strange.. it works for me with the older version (just checked)

Anonymous comment on 2010-08-22 09:40

@jose1711: thanks, now the 'search' gives correct locations and 'my places' are stored and displayed after restart. Though panoramio pics still aren't depicted in the pop-up windows.

jose1711 commented on 2010-08-22 09:20

@zandgreen: have you tried running LC_NUMERIC=C googleearth ?

Anonymous comment on 2010-08-22 09:16

Hello, installed v.5.1.3535 proposed by jose1711 with PKGBUILD. I'm experiencing the following: -no panoramio pics; -search gives strange results(all stuck in one place somewhere in the sea). P.S. i'm using r300g driver.

Berseker commented on 2010-08-15 15:41

I also suggest to use the previous version (posted by jose1711), that works like a charm, instead of the lastest one. If you modify the md5sum section, it works too with "my" PKGBUILD with the various fixes (all the necessary is in the comment posted by trizen)

Anonymous comment on 2010-08-14 15:42

Newest Version gives some crazy integrity error in the GoogleEarth.bin itself, download the proviuos version as stated in the comments.

Anonymous comment on 2010-08-13 10:04

Hello
Would it be possible to include the changes from Bersekers PKGBUILD into this one? It is the only thing that works here on an intel system. Thanks for that!

jose1711 commented on 2010-08-11 21:08

alternate source for 5.1: http://dl.google.com/earth/client/advanced/previous/GoogleEarthLinux.bin

jose1711 commented on 2010-08-11 11:19

problem also seen on win32. downgrade to 5.1 solved the issue.

jose1711 commented on 2010-08-09 21:52

google earth stopped showing pictures (e. g. minimap in lower right corner or panoramio pics) on my box. anyone else has the same issue?

gatonero commented on 2010-08-03 21:32

lsmod | grep nvidia

Berseker commented on 2010-08-03 15:44

if you can find a way to detect if it is a Intel powered system, or a Nvidia one, maybe it is possible to download the modified libgl.so.1 only in the first case

Anonymous comment on 2010-08-02 19:55

I have some versions working on my nVidia box, and some versions working on my Intel box. If I can reconcile the two and get something that works on both I'll update the PKGBUILD.

BertiBoeller commented on 2010-08-02 16:43

Interesting googleearth works with Berseker's PKGBUILD and the libGL.so.1 mentioned below. It didn't work with the PKGBUILD from above.

trizen commented on 2010-07-31 13:23

Direct URL of src posted by Berseker on megaupload: http://trizen.go.ro/files/google-earth-5.2.1.1329-2.src.tar.gz

Berseker commented on 2010-07-31 13:18

if anyone is interested, i made what I requested in my previous post
http://www.megaupload.com/?d=YQBW0NS0

contains modified PKGBUILD and google-earth script. enjoy

Berseker commented on 2010-07-30 09:23

can you please update your PKGBUILD with some of the tricks used by srabd in his bin32-google-earth PKGBUILD?
http://aur.archlinux.org/packages.php?ID=12890

it contains fixes for some annoying bugs, plus the _use_arch_libs setting.
I think it would be a great add to this PKGBUILD

thanks

BertiBoeller commented on 2010-07-28 11:04

Perhaps you can add "libQT*" to the rm -f line. I can't test if GoogleEarth still works after that because I can't get GoogleEarth working in the first place.

Madek commented on 2010-07-28 02:10

for intel users with the famous bug at start
cd /opt/google-earth
wget http://librarian.launchpad.net/7037027/libGL.so.1 -O libGL.so.1

solved the problem for me

OttoA commented on 2010-07-22 22:10

crashed for me on startup (crashlog: http://fpaste.org/ReGA/)

commented out

rm -f lib{crypto.so.*,Qt*,GLU.so.*,curl.so.*,gcc_s.so.*,stdc++.so.*,z.so.*} &&

and got it working - there probably are version mismatches

empthollow commented on 2010-07-19 00:25

lspci | grep -i vga
01:05.0 VGA compatible controller: ATI Technologies Inc RC410 [Radeon Xpress 200M]

I linked libGL.so.1 libGLU.so.1 because googleearth complained the file didn't exist.

Now i see the splash screen for a second and it crashes with this.

Fontconfig error: line 1: unknown encoding
Fontconfig error: Cannot load default config file
Failed to load "/opt/google-earth/libevll.so" because "/opt/google-earth/libevll.so: undefined symbol: gluNewTess"

If anyone knows how to fix this or can give me a PKGBUILD that is older I would appreciate it.

buergi commented on 2010-07-16 06:30

Perhaps we should consider adding the bugfix (export LC_NUMERIC=C) for the numeric format to the 'googleearth' script.
Or at least add a note to googleearth.install that this bug exists and how to fix it if necessary.

And BTW: 5.2 doesn't work for me (ATI HD 3450) either, same error as zyghom and epsilom. i downgraded to 5.1

zyghom commented on 2010-07-13 02:50

I can confirm: newest version not working on ATI as well:

lspci | grep -i vga
01:00.0 VGA compatible controller: ATI Technologies Inc M92 [Mobility Radeon HD 4500 Series]

with the same error as below:

Fontconfig error: line 1: unknown encoding
Fontconfig error: Cannot load default config file
Fatal error in __driConfigOptions line 1, column 0: unknown encoding.
Google Earth has caught signal 6.



We apologize for the inconvenience, but Google Earth has crashed.
This is a bug in the program, and should never happen under normal
circumstances. A bug report and debugging data have been written
to this text file:

/home/papio/.googleearth/crashlogs/crashlog-4c3bd31b.txt

Please include this file if you submit a bug report to Google.

solstice commented on 2010-07-06 10:56

I have the white square problem using Gefore 8300GS and latest nvidia 256.35

solstice commented on 2010-07-06 10:55

I got a problem. google-earth 5.2 starts ok. but in the main frame, I got a a bigwhite square at the bottom right. and now navigation wheel. also it lacks a numbe of overlay graphics.
if I change the PKGBUILD to keep the library of google-earth. it works but it's ugly because it use a bad font.

I think i will downgrade

jose1711 commented on 2010-07-03 16:45

gt_swagger: please consider https://bugs.launchpad.net/medibuntu/+bug/485730. LC_NUMERIC=C fixes a bug where decimal part of latitude/longitude is cut off. after each update i need to fix it manually. thanks

partizan commented on 2010-06-29 22:23

Hi gt_swagger, i propose using different filename for different versions of google-earth(and we forget about conflicting md5-sums when new version is out).
something like:
source=(GoogleEarthLinux-${pkgver}.bin::http://dl.google.com/earth/client/current/GoogleEarthLinux.bin ...

raqua commented on 2010-06-25 18:11

I also have the same problem as epsilom

hamelg commented on 2010-06-24 19:26

same behavior as bob_art.
I have a nvidia GT9600 video card.
I've posted here :
http://www.google.com/support/forum/p/earth/thread?tid=44ca6d872c6eb838&hl=en

hamelg commented on 2010-06-24 19:26

same behavior as bob_art.
I have an nvidia GT9600 video card.
I've posted here :
http://www.google.com/support/forum/p/earth/thread?tid=44ca6d872c6eb838&hl=en

hamelg commented on 2010-06-24 17:08

same behavior as bob_art.
I have an nvidia GT9600 video card.

tocer commented on 2010-06-24 03:47

same error as epsilom

Anonymous comment on 2010-06-23 20:49

Using the AUR package got me epsilom's error.
Just download the bin package from google and use this tip:
http://richs-lxh.linux-hardcore.com/2010/04/howto-linux-install-google-earth/

bob_art commented on 2010-06-22 09:57

Hi! after the upgrade to this version (google-earth 5.2.1.1329-1) in the box disappeared navigation tools - instead see a white square in the lower right corner. We are waiting for fixes. Thank you

grayich commented on 2010-06-21 09:02

5.2.1.1329 buggy, something with the textures and the controls

Anonymous comment on 2010-06-21 02:57

Updated to 5.2.1.1329.

DO NOT UPGRADE IF YOU USE INTEL INTEGRATED GRAPHICS. IT WILL NOT WORK FOR THIS PARTICULAR VERSION. YOU HAVE BEEN WARNED.

Madek commented on 2010-06-21 02:37

thanks gborzi

gborzi commented on 2010-06-20 08:49

Hello Madek,
you can download the previous Google Earth binary from this link
http://dl.google.com/earth/client/advanced/previous/GoogleEarthLinux.bin
I downgraded to 5.1.3535.3218 on i686, strangely enough on x86_64 version 5.2.1.1329 works fine.

Madek commented on 2010-06-20 02:40

somebody have the GoogleEarthLinux.bin file v. 5.1.3535.3218 or a pkg.tar.xz from this version, share plis

XavierCLL commented on 2010-06-18 15:53

when update google-earth 5.2 not start, show:

Fontconfig error: line 1: unknown encoding
Fontconfig error: Cannot load default config file
Fatal error in __driConfigOptions line 1, column 0: unknown encoding.
Google Earth has caught signal 6.

Anonymous comment on 2010-06-18 02:32

there is version 5.2 now, update please!

Anonymous comment on 2010-06-17 20:09

[sonic@darkstar aur-google-earth]$ md5sum GoogleEarthLinux.bin
145cc4460e6ffda56828f6507e7d8a20 GoogleEarthLinux.bin

donvla commented on 2010-06-12 10:54

No install files in source array!
Please chnage this .

Anonymous comment on 2010-06-08 13:47

I think shared-mime-info and desktop-file-utils should be added to the dependencies, because googleearth.install uses update-mime-database and update-desktop-database.

Anonymous comment on 2010-06-03 21:13

блин ну в конце то концов обновите md5sum

dangole commented on 2010-05-25 10:28

updated PKGBUILD for verion 5.1.3535.3218
http://aur.pastebin.com/vHyHJZ6B

dangole commented on 2010-05-25 10:24

current version seems to be 5.1.3535.3218

md5sums=('69a52b92691f725b15e28b9fa2a123b2'
'757d9726191f365209fa13acdd0ab6e7'
'005109377cb797f6cb12a74abdd0835e'
'd819a1961382826b7f92ee32dd67f2e1'
'84e1efd656f86c52333cc629b31accb0'
'2922f8668ae19e6d9c558aaadda90af8')

dangole commented on 2010-05-25 10:17

current version seems to be 5.1.3535.3218

md5sums=('69a52b92691f725b15e28b9fa2a123b2'
'757d9726191f365209fa13acdd0ab6e7'
'005109377cb797f6cb12a74abdd0835e'
'd819a1961382826b7f92ee32dd67f2e1'
'84e1efd656f86c52333cc629b31accb0'
'2922f8668ae19e6d9c558aaadda90af8')

gborzi commented on 2010-05-23 13:53

@hoobastank
Thanks, it worked.
@hamelg
My installation is working OK with Arch QT. Perhaps it's a problem due to some specific qt configuration.

Anonymous comment on 2010-05-22 09:24

author, knock knock - wake up)))) replace md5sum for file GoogleEarthLinux.bin

hamelg commented on 2010-05-16 17:05

Hello,

Using Arch QT libs causes crashes when opening file selector.
To fix that, you must comment the line :
rm -f lib{crypto.so.*,Qt*,GLU.so.*,curl.so.*,gcc_s.so.*,stdc++.so.*,z.so.*} &&

Unfortunately, when doing that, you loose anti-aliased fonts :(

Anonymous comment on 2010-05-15 08:45

Edit PKGBUILD and delete first line in md5sum and just write
69a52b92691f725b15e28b9fa2a123b2
or learn your md5sum of your file
md5sum /tmp/yaourt-tmp-<your_user_name>/aur-google-earth/google-earth/GoogleEarthLinux.bin

gborzi commented on 2010-05-14 21:47

google-earth 5.1.3535.3218 is out, please update.

Anonymous comment on 2010-04-21 16:32

@ Szamot83 ... I think you just had a bad download. I used current md5 to build just fine.

Anonymous comment on 2010-04-18 11:12

http://earth.google.com/intl/en/license.html has wrong checksum.
If you want to install this package just edit PKGBUILD and delete line
'757d9726191f365209fa13acdd0ab6e7' in md5 sums and
http://earth.google.com/intl/en/license.html in source.

This should look like this: http://aur.pastebin.com/QjqHwCQy

Anonymous comment on 2010-03-23 20:59

Crashes due to glibc when doing File->Open. Using provided libs instead of system libs resolves the issue.

Det commented on 2007-01-01 01:39

Hehe, seems the lib32-* dependencies are still needed after all - the amd64 .deb package has 'ia32-libs' as its dependency (and they didn't just forget it there - 'lsof' confirms this).

In any case, there's no real reason to have 2 separate packages for Google Earth when the other one can be both at the same time. Whenever Google starts offering real 64-bit binaries for Google Earth, could be a week from now, could be a year, this thing can start using them.

I also added 'ttf-ms-fonts' as an optional dependency, as it seems to solve the ugly fonts for me and some random Ubuntu users.

Det commented on 2007-01-01 01:39

Hehe, seems the lib32-* dependencies are still needed after all - the amd64 .deb package has 'ia32-libs' as its dependency (and they didn't just forget it there - 'lsof' confirms this).

In any case, there's no real reason to have 2 separate packages for Google Earth when the other one can be both at the same time. Whenever Google starts offering real 64-bit binaries for Google Earth, could be a week from now, could be a year, this thing can start using them.

I also added 'ttf-ms-fonts' as an optional dependency, as it seems to solve the ugly fonts for me and some random Ubuntu users.