Package Details: lastpass

Git Clone URL: (read-only)
Package Base: lastpass
Description: The Universal LastPass installer for Firefox, Chrome, and Opera
Upstream URL:
Licenses: custom
Submitter: Det
Maintainer: eschwartz
Last Packager: eschwartz
Votes: 88
Popularity: 1.470235
First Submitted: 2013-06-02 17:50
Last Updated: 2018-10-19 19:45

Latest Comments

1 2 3 4 5 6 ... Next › Last »

eschwartz commented on 2018-09-02 02:50

Hmm, they just removed their own hosted copy in favor of downloading one from the Chrome webstore. Problem is, the only thing that crx is needed for, is the opera plugin, which is not contained in the chrome webstore version of the crx... which is logical because why was it ever there in the one hosted on

I'll note their linux download instructions and is therefore incorrect and not working.

I'm not really sure what the story is supposed to be here. But I don't use opera. So, I guess I will drop this, and use the new lplinux.tar.bz2 location for the chrome/firefox native messaging binary. Either that or someone who uses Opera can discuss with Lastpass support/tell me how to re-enable it.

Patricol commented on 2018-09-01 18:20 is a 404 for me.

eschwartz commented on 2018-07-13 15:34

Yeah, I don't usually see extensions pulled like that. :/

Updated again.

I've been moving things to nvchecker so I should see updates faster anyway.

becatlibra commented on 2018-07-13 12:51

Currently getting a 404. It appears that 4.15.07 is no longer available

zman0900 commented on 2017-12-19 04:45

Binary component doesn't appear to be working with firefox. I tried deleting ~/.mozilla/firefox/<profile>/extensions/ just to be sure I am using the package provided extension, but the options to copy username or password are still missing. </profile>

In a windows VM using the same versions of firefox and the lastpass extension, installed with the lastpass provided universal installer, the copy options are there, so it seems like this should be working.

eschwartz commented on 2017-10-20 16:05

Oh, uh, whoops.

[eschwartz@arch ~/git/pkgbuilds/lastpass]$ /usr/bin/diff -ur src/lastpass-4.2.0a pkg/lastpass/usr/lib/firefox/browser/extensions/
Binary files src/lastpass-4.2.0a/data/ and pkg/lastpass/usr/lib/firefox/browser/extensions/ differ

So why did this all of a sudden start getting modified during tidy_strip? And I never noticed it when updating this, because my Firefox build has extension signing verification disabled.

Try it now.

jskier commented on 2017-10-20 14:58

This version 4.2.0a-1 appears to have an unverified Firefox extension with it. I put in a support ticket and will follow up if necessary.

eschwartz commented on 2017-08-13 03:20

Whoops, I seem to have forgotten to add that when I switched the packaging logic to what I use in my other Firefox extensions. Fixed now.

nocturnus commented on 2017-08-13 03:09

'unzip' from the Extra Repository should be a dependency of the package, otherwise it will fail to build, due to unzip command usage.

eschwartz commented on 2017-07-17 23:44

I doubt it...

the pkgver is not actually used in the download urls at all, so changing the pkgver and the _chromever accomplishes nothing other than giving the source files a unique local name. So your suggestion is a no-op.
If the files were updated on the Lastpass server, then it would redownload the same *unversioned* URLs under an updated local name, and update the checksums if relevant... but that is a no-op at the moment as well.

Additionally, the chrome browser extension is downloaded and updated by Chrome itself, since Chrome doesn't do locally-installed global extensions like Firefox does -- all that this package does is install a policy file telling Chrome to install lastpass on its own.

The native messaging binary is installed, as per the official Lastpass universal installer, from their universal installer tarball (and the webstore doesn't contain at all). The opera plugin is installed, again as per the official installer script, from the unversioned CRX hosted on their website, which does have a and is also behind the Chrome webstore version -- I have no idea why.
(But then again, the release notes say the latest Chrome webstore version is 4.1.55 and the latest Firefox version is 4.1.53 and apparently neither are true.)

I have just double-checked and none of the (unversioned) source files have been updated, so this package should still be valid. If there was an update, then either the checksums would fail, or I would change the _amo_file ID in order to download the new version of the Firefox extension.
(I have a convenience script for the Firefox extension, see ./


So given that nothing which you described should have fixed anything, I am not sure what the problem is or how you expect me to fix it.
Maybe restarting Chrome caused it to fix itself?