Package Details: sc-im 0.8.2-1

Git Clone URL: (read-only, click to copy)
Package Base: sc-im
Description: A spreadsheet program based on SC
Upstream URL:
Licenses: BSD
Conflicts: sc-im-git, scim-spreadsheet
Submitter: Rhinoceros
Maintainer: Rhinoceros (andmarti1424)
Last Packager: Rhinoceros
Votes: 54
Popularity: 1.42
First Submitted: 2015-09-05 01:44 (UTC)
Last Updated: 2021-11-05 00:36 (UTC)

Dependencies (4)

Required by (0)

Sources (2)

Pinned Comments

Latest Comments

Rhinoceros commented on 2021-11-05 00:37 (UTC)

@nesk_aur base-devel, which includes bison and patch, is assumed to be installed when building packages. As per the wiki, this should not be in the makedepends array.

Thanks for the information on aarch64. I've added it to the PKGBUILD.

nesk_aur commented on 2021-11-04 20:31 (UTC) (edited on 2021-11-04 20:31 (UTC) by nesk_aur)

Additional build-time deps: bison, patch. This pkg builds fine on aarch64.

Strider85 commented on 2021-09-06 23:06 (UTC)

@Rhinoceros Thanks!

Rhinoceros commented on 2021-09-06 22:58 (UTC)

@Strider85 You probably need to look upstream. It's likely not a packaging error. I just searched and found this.

Strider85 commented on 2021-09-06 18:34 (UTC)

Hi. I was getting the faulty segmentation error for many random cases (for external functions, mainly), even with the "external_functions" option set. Worked fine after downgrade to

Rhinoceros commented on 2021-04-19 11:10 (UTC)

@GregTheMadMonk No worries at all. Easy error to make, and you aren't the first person to get confused by this. And I appreciate you posting here and attempting to improve the package too!

GregTheMadMonk commented on 2021-04-19 11:07 (UTC)

@Rhinoceros, oh, I see... My bad, sorry...

Rhinoceros commented on 2021-04-19 10:05 (UTC)

@GregTheMadMonk base-devel, which includes bison, is assumed to be installed when building packages. As per the wiki, this should not be in the makedepends array.

GregTheMadMonk commented on 2021-04-19 09:49 (UTC) (edited on 2021-04-19 09:49 (UTC) by GregTheMadMonk)

You should probably add bison to make dependencies. yacc (which is provided by bison package) seems to be required during build (or, at least, it didn't build for me unless I've installed bison)

luukvbaal commented on 2021-03-19 15:56 (UTC)

Right you are, my bad.

Rhinoceros commented on 2021-03-19 08:21 (UTC)

@luukvbaal Works for me in a clean directory. Upstream changed their versioning, but I reflected the change in the new checksum. You probably have the old one cached by your AUR helper.

luukvbaal commented on 2021-03-19 08:03 (UTC) (edited on 2021-03-19 08:04 (UTC) by luukvbaal)

Getting sc-im-0.8.0.tar.gz ... FAILED arch.0.8.0.patch ... Passed ==> ERROR: One or more files did not pass the validity check! please check.

Rhinoceros commented on 2021-02-26 04:08 (UTC)

@crossgear Sorry, I didn't mean the Markdown, but thanks for fixing up the formatting anyway. I actually meant the Warning: sc-im is only available from AUR parts, etc., which I didn't recognise. The "basic" way to install is manually doing makepkg -s, which excludes all the additional things that can go wrong when using an AUR helper to install packages. It's a good fallback test when things go wrong.

Having said that, I don't know anything about Manjaro really, so the issues might be related to different version or defaults of patch? Or more likely the paths where it builds packages? If you can't get it to build, it might be more helpful to ask at a Manjaro forum.

s2ohp4rg commented on 2021-02-26 03:45 (UTC)

@Rhinoceros No, I've just never used Markdown before. I was redirected from Github to comment here. Never tried makepkg -s, but I can try. Maybe worth noting that I'm running Manjaro.

Rhinoceros commented on 2021-02-26 03:33 (UTC)

@crossgear I just tried again and it works for me. What is all that extra text you posted? Are you using some AUR helper? Can you test with just a manual makepkg -s?

s2ohp4rg commented on 2021-02-26 03:25 (UTC) (edited on 2021-02-26 03:41 (UTC) by s2ohp4rg)

When attempting to install, I get this error output:

Warning: sc-im is only available from AUR


Cloning sc-im build files...

Checking sc-im dependencies...

Resolving dependencies...

Checking inter-conflicts...

To build (1):

sc-im 0.7.0-2 AUR

Edit build files : [e]

Apply transaction ? [e/y/N] y

Building sc-im...

==> Making package: sc-im 0.7.0-2 (Wed 24 Feb 2021 07:25:11 PM CST)

==> Checking runtime dependencies...

==> Checking buildtime dependencies...

==> Retrieving sources...

-> Found sc-im-0.7.0.tar.gz

-> Found arch.patch

-> Found 258.patch

==> Validating source files with sha256sums...

sc-im-0.7.0.tar.gz ... Passed

arch.patch ... Passed

258.patch ... Passed

==> Removing existing $srcdir/ directory...

==> Extracting sources...

-> Extracting sc-im-0.7.0.tar.gz with bsdtar

==> Starting prepare()...

can't find file to patch at input line 3

Perhaps you should have used the -p or --strip option?

The text leading up to this was:


|--- Makefile.orig 2017-12-14 04:48:59.000000000 +1100

|+++ Makefile 2017-12-15 09:23:07.776195283 +1100


File to patch:

Skip this patch? [y] y

Skipping patch.

2 out of 2 hunks ignored

==> ERROR: A failure occurred in prepare().


Rhinoceros commented on 2020-06-04 22:59 (UTC)

Thank you @cloudhead. That works perfectly. Package updated.

cloudhead commented on 2020-06-04 13:51 (UTC)

gcc-8 is not needed if the -fcommon flag is passed to gcc. Editing the PKGBUILD build() phase to have:

build() {
  cd "$pkgname-$pkgver/src"
  make CC='gcc -fcommon'

Solves the problem for me.

Rhinoceros commented on 2020-05-27 12:22 (UTC) (edited on 2020-05-27 12:22 (UTC) by Rhinoceros)

Thanks @teleportex for all that information! I've updated the PKGBUILD. A few notes:

  • I didn't need to add the -C flag with a clean directory.
  • gcc8 was added to makedepends
  • gnuplot was added to optdepends, because the main functionality of sc-im works without it

Thanks again.

teleportex commented on 2020-05-27 09:44 (UTC) (edited on 2020-05-27 09:46 (UTC) by teleportex)

I have ecnountered 2 problems when building:

  • On issues of the repository i saw there are problems building with gcc 9 it is recomended to build with gcc-8, that worked to me.
  • A problem on locating some file, I compiled it from the package root using -C flag and worked.
  • Gnuplot not found.

Changes to the PKGBUILD:

  • Add gcc8 and gnuplot to dependencies
  • Build stage changed to:
    cd "$pkgname-$pkgver"
    make -C src CC=gcc-8

seifferth commented on 2020-01-20 12:51 (UTC)

@Tymekm: That's true. yacc is provided by bison, which is part of base-devel. base-devel is assumed as an implicit dependency for building any package (see <>). This is why bison is not included in the makedepends array. Seeing that you don't seem to have bison installed on your system, you might want to run pacman -S base-devel.

[Sorry for cross-posting from sc-im-git, but this may be useful to know here as well.]

AnarchoJohn commented on 2020-01-20 10:52 (UTC) (edited on 2020-01-20 11:06 (UTC) by AnarchoJohn)

The build will fail without yacc.

yacc -d gram.y
make: yacc: Command not found
make: *** [Makefile:167: gram.c] Error 127

snake commented on 2019-11-11 23:25 (UTC)

@Rhinoceros It does indeed look like it. Last comment suggests that trouble comes from "shared formulas", but these are so common in speadsheets!

Rhinoceros commented on 2019-11-11 22:54 (UTC)

@snake Looks like this?

snake commented on 2019-11-11 22:49 (UTC)

Did anyone manage to make the --xlsx_readformulas option work? On my system it fails with a segmentation fault.

Rhinoceros commented on 2018-08-16 03:04 (UTC)

@terinjokes That's pretty straightforward for me to do, but I'm surprised that would be reachable and wouldn't be?

terinjokes commented on 2018-08-15 20:27 (UTC)

Do you mind including the patch in the Git repo, rather than fetching it from GitHub? I was having issues fetching it this morning.

Rhinoceros commented on 2018-05-25 10:09 (UTC)

Thanks @tealeaf. Good call. I've patched with the pull request, changing the executable name. Please let me know if it works (I don't use the scim package).

tealeaf commented on 2018-05-25 07:53 (UTC)

Ah, looks like an upstream change:

tealeaf commented on 2018-05-25 07:46 (UTC)

I notice that installing sc-im-git creates a binary at /usr/bin/sc-im, whilst this PKGBUILD produces /usr/bin/scim. Is that a naming change upstream, or has the scim name been implemented here?

The reason I ask is that /usr/bin/scim is a file conflict with the scim input method, which prevented me from installing both.

Looking at github, it does seem that the binary should be sc-im.

Rhinoceros commented on 2018-01-28 05:05 (UTC)

@armanleftarm Thanks for the feedback. Done.

armanleftarm commented on 2018-01-28 03:35 (UTC)

Add armv7h to the arch, builds proper on Samsung XE303C12

Rhinoceros commented on 2018-01-18 01:04 (UTC)

Good catch @aplund. Thank you and fixed.

aplund commented on 2018-01-18 00:42 (UTC)

Needs the following patch if BUILDDIR is set:

diff --git a/PKGBUILD b/PKGBUILD index 800dc83..3a25dda 100644 --- a/PKGBUILD +++ b/PKGBUILD @@ -20,7 +20,7 @@ MAKEFLAGS='-j1' prepare() { cd "$pkgname-$pkgver/src" # install things in the correct place for package managers - patch <../../../arch.patch + patch < "$srcdir/arch.patch" }

build() {

Rhinoceros commented on 2017-11-16 23:57 (UTC)

@Hund bison is part of `base-devel` [1], which is assumed to be installed and should not be in makedepends [2]. [1] [2]

commented on 2017-11-16 18:06 (UTC)

I got this: ``` ==> Starting build()... which: no gnuplot in (/usr/local/sbin:/usr/local/bin:/usr/bin:/usr/lib/jvm/default/bin:/usr/bin/site_perl:/usr/bin/vendor_perl:/usr/bin/core_perl) yacc -d gram.y make: yacc: Command not found make: *** [Makefile:160: gram.c] Error 127 ==> ERROR: A failure occurred in build(). Aborting... ``` But I solved it by installing the package `bison`. It should be added to the makedepends.

Rhinoceros commented on 2017-05-09 11:47 (UTC)

Oh oops, I missed the difference between writer and reader. And xls and xlsx. Thanks for that information. I'm not sure how most people do it, but I only look at the optdepends after installation is complete, so I'm unsure how useful it'd be. Nevertheless, I'll add it to optdepends with the next version. Cheers.

seifferth commented on 2017-05-09 10:27 (UTC)

That might be true for _xls import_. I was talking about _xlsx export_ however, which is enabled by satisfying the optional dependency libxlsxwriter at build time. This works great with my setup and doesn't require changing the makefile. It does require rebuilding sc-im after installing libxlsxwriter though.

Rhinoceros commented on 2017-03-30 01:46 (UTC)

@ber532k Thank you for the heads up. Nope, I'm totally happy for you to maintain the -git version. Thanks also for the information about libxlsxwriter. I'll add that to the next version. Also, for your two -git PKGBUILDs, you might want to have a read of how the pkgver function should work for VCS files [1]. This will keeps the installed version correct, and prevents unnecessary upgrades. Cheers. [1]

seifferth commented on 2017-03-29 15:31 (UTC)

I added an sc-im-git package to the AUR. Together with the libxlsxwriter package, which I also added to the AUR, this makes the new xlsx-export option available, if libxlsxwriter is already installed when sc-im-git is compiled. @Rhinoceros: Since I basically just copied your PKGBUILD and changed the source to the git repo the credit for bringing this to Arch probably still belongs to you. If you want to assume responsibility for sc-im-git just drop me a message, I will be glad to hand it over.

andmarti1424 commented on 2017-02-20 23:48 (UTC)

@Rhinoceros, I see them now. Thanks!

Rhinoceros commented on 2017-02-20 20:36 (UTC)

@emartine Yes, I noticed that the backspace key now works. Thank you! The new arch.patch is to install the files in /usr/bin, etc. rather that /usr/local/bin, since the files are installed by a package manager, not manually. I documented the changes here.

andmarti1424 commented on 2017-02-20 18:57 (UTC) (edited on 2017-02-20 18:57 (UTC) by andmarti1424)

@Rhinoceros, I have noticed you still added the arch.patch file. It shouldn't be neccesary in v0.5.0.. The BS key should work out of the box.

Rhinoceros commented on 2017-01-01 22:45 (UTC)

$ sudo showkey --keycodes kb mode was ?UNKNOWN? [ if you are trying this under X, it might not work since the X server is also reading /dev/console ] press any key (program terminates 10s after last keypress)... keycode 14 press keycode 14 release

andmarti1424 commented on 2016-12-29 18:58 (UTC)

@Rhinoceros, could you please run showkey --keycodes as root or with sudo, and press the backspace key? Thanks!

Rhinoceros commented on 2016-12-28 09:20 (UTC)

@emartine No, I'm not using tmux at all.

andmarti1424 commented on 2016-12-27 12:34 (UTC) (edited on 2016-12-27 14:09 (UTC) by andmarti1424)

@Rhinoceros, I did some tests regarding the Backspace issue, and noticed that I can reproduced the problem in Arch if using tmux. Without it, Backspace key works just fine. Did you use tmux in your tests ??

Rhinoceros commented on 2016-12-23 21:51 (UTC)

@derrickcope yacc is installed by bison [a]. This is in base-devel [b], and assumed to be installed by default [c]. [a] [b] [c]

Rhinoceros commented on 2016-12-23 21:49 (UTC)

@emartine Sorry, I only just read your edit to your comment… I don't get email notifications of edits! I'm not sure how to use that command. $ showkey --keycodes Couldn't get a file descriptor referring to the console

derrickcope commented on 2016-12-21 03:54 (UTC)

I am getting an error on build Starting build()... yacc -d gram.y make: yacc: Command not found make: *** [Makefile:181: gram.c] Error 127 ==> ERROR: A failure occurred in build(). Aborting... sorry for my ignorance, what is yacc and how do i install it?

andmarti1424 commented on 2016-12-01 12:46 (UTC) (edited on 2016-12-01 12:47 (UTC) by andmarti1424)

What I do not get is why you have this problem in your Arch installation and I don't in my Arch installation. I will do more tests.. Whats is the output of showkey --keycodes in your system, when you press BackSpace?

Rhinoceros commented on 2016-12-01 00:33 (UTC)

The patch changes that line to: #define OKEY_BS 0x7f This is required for me to get it to work. I don't really know on what level backspace is defined, but I presumed that it would be consistent in Arch Linux, if not all of Linux?

andmarti1424 commented on 2016-12-01 00:01 (UTC)

Hello. Thanks. If I can be any help.. What was the bug? As my backspace key is working OK, and I have it defined like this in macros.h: #define OKEY_BS 0x107

Rhinoceros commented on 2016-11-30 23:48 (UTC)

Welcome to Arch! I've uploaded the new version and added you as a co-maintainter for this package. I'm happy to let you take full control if you like. Also, FWIW, 0.4.0 still has the backspace bug, so this PKGBUILD still applies the fix_backspace.patch.

andmarti1424 commented on 2016-11-30 15:37 (UTC) (edited on 2016-11-30 18:35 (UTC) by andmarti1424)

Just upload a new release: v0.4.0 to git. Arch Linux is my main OS right now.

andmarti1424 commented on 2016-05-11 01:34 (UTC) (edited on 2016-05-11 01:36 (UTC) by andmarti1424)

Hello again. I'm Andmarti1424. The one who is developing it. Could you please open a new issue in github and we continue there? I did replace the ncurses includes to. . see if that fix the issue in Arch Linux without being necessary to create symlinks ..

Rhinoceros commented on 2016-05-10 22:11 (UTC)

@emartine: No. As per my last comment, it won't build. Read the link for more information and a workaround, but I'm not sure how to implement it in terms of this PKGBUILD.

andmarti1424 commented on 2016-05-10 16:08 (UTC)

Hello Rhinoceros. Could you manage to build it?

Rhinoceros commented on 2016-05-07 04:22 (UTC)

This has been updated to 0.3.0, but I can't get it to build. Related discussion here: I'm not sure how to resolve this without creating a whole bunch of system-level symlinks, which seems suboptimal, so I'm open to suggestions.

Rhinoceros commented on 2016-05-06 23:34 (UTC)

@eXeC64 git shouldn't be needed, as this PKGBUILD pulls the tar.gz release, not git release. yacc is provided by bison, which is assumed to be installed [a], since it's in base-devel [b]. [a] [b]

eXeC64 commented on 2016-05-06 23:08 (UTC)

The PKGBUILD is missing make dependencies such as git, whatever package provides yacc, etc.

Rhinoceros commented on 2016-04-08 04:10 (UTC)

@riX2000 Hmm, I changed the sums in the last commit, which worked on my system, so I wonder if upstream tagged a different commit. Anyway, it should be fixed now. Thanks for letting me know!

riX2000 commented on 2016-04-07 22:09 (UTC)

Seems like the sha256 signature for is wrong in the PKGBUILD.

Rhinoceros commented on 2015-09-05 02:17 (UTC)

Thanks for the comment, dalcde. I've filed for deletion and uploaded a new PKGBUILD at

dalcde commented on 2015-09-05 00:23 (UTC)

The repository has been moved

andmarti1424 commented on 2015-04-18 19:03 (UTC)

License changed to BSD License in v0.1.9.