I guess it's now ready for testing
- Queries
- All Stories
- Search
- Advanced Search
- Transactions
- Transaction Logs
Advanced Search
Nov 13 2024
All changes have been backported for VSD 3.3
I added this to gpg4win master. I assume that this branch will be used for VSD 3.3 and Gpg4win 4.4.
All missing translations mentioned in the original description should now be there (once a new kleopatra snapshot is created).
That's a bug in Firefox. Chromium doesn't exhibit the problem. I have problems with fonts in Firefox on many different websites.
Forget the explanation. msgcat --use-first works correctly. We didn't use it correctly for our purposes. msgcat --use-first took the first translation of the context-less messages. This happened to be marked-as-obsolete messages in the main po file (because KDE keeps obsolete messages for easier recovery). And since they were marked as obsolete they were dropped when the po file were compiled. The script now removes obsolete messages before merging in additional po files. This way non-obsolete messages in the additional po files don't get shadowed anymore by obsolete messages in the main po files.
Nov 12 2024
Andre's comment was "Use use first to avoid double translations that confuse it.".
I have removed the --use-first from the msgcat command that merges the combined messages from KDE's svn and the po file in our repo with the local l10n files for libkleo and kleopatra in gpg4win/packages.
The problem seems to be that the msgcat commands in gen-tarball.sh (gpg4win repo) are used with --use-first and that seems to ignore msgctxt when looking for duplicate msgids.
- "GnuPG VS-Desktop Audit Log Viewer" (nur Fenstertitel unübersetzt)
- "Valid until (between %1 and %2):" (Schlüsselgenerierung, Erweiterte Einstellungen)
- Unter Extras -> Zwischenablage: 4 von 5 Unterpunkten nicht übersetzt
- Lookup on server (nur im Fenstertitel unübersetzt)
- "Publish on keyserver afterwards" (im Fenster Beglaubigung)
- "Enter a message to encrypt or decrypt…" Notepad-Placeholdertext
Should be ready for testing
Backported for VSD 3.3
This bugfix shall be backported for VSD 3.3 as discussed with ebo
Latest changes backported for VSD 3.3
Backported for VSD 3.3
And backported for VSD 3.3.
Fixed.
Nov 11 2024
The QProcesses we start should be configured with closed stdin so that the started process don't hang waiting for input, e.g. by calling closeWriteChannel() immediately after start(). Or by calling start() with QIODeviceBase::ReadOnly, but, sadly, the documentation of QProcess doesn't document what effect that has.
Nov 8 2024
We shouldn't have to add explicit line breaks. Explicit line breaks will lead to bad formatting (German: Flattersatz) for people who use huge font sizes (a11y!) and Qt will wrap long tool tips so that they won't become wider than the screen. Alternatively, we wrap the tool tips (in the code that sets the tool tips) into <html> tags. Then (for whatever reason) Qt will make the tool tips much narrower. (Some people think too narrow, but we have to die one death.)
Nov 7 2024
You mean that a disabled certificate with secret key isn't listed with bold font? That's probably because we have an appearance filter for disabled certificates which takes precedence.
I agree that it doesn't make sense anymore because we never show disabled and not-disabled certificates next to each other.
I have updated the translations of the filters defined in the libkleopatrarc*.desktop files.
I assume by "do such things in the background" you mean that GnuPG should do this automatically in the background.
Nov 6 2024
I cannot reproduce the crash anymore. I guess this was fixed with the fix of T7372: Kleopatra: Crash when unplugging smartcard while operation is in progress. I'll close this ticket as duplicate of T7372.
@ebo I suspect that we want to fix this crash also for VSD 3.3 (if it is still reproducible). I found this ticket by accident while searching for READKEY.
I haven't added any project tags because I'm not sure for which projects this is relevant. Since GnuPG 2.2 doesn't support multiple smartcards it's likely not relevant for VSD 3.3.
@gniibe It seems that a keylisting (with gpg and gpgsm) interferes with a READKEY --card --no-data -- NKS-NKS3.4571 gpg-agent command and makes it hang until scdaemon is killed.
It looks as if a keylisting interfered with a gpg-agent/scdaemon command.
48458.670390 2024/11/06 12:43:04.238 5772 kleopatra.exe org.kde.pim.kleopatra: ReaderStatusThread[GUI]::ping()
^ update of the smart cards is requested
48458.670695 2024/11/06 12:43:04.238 5772 kleopatra.exe org.kde.pim.kleopatra: ReaderStatusThread[2nd]: new iteration command= "__update__" ; nullSlot= true
^ background thread starts update of the smart cards
48459.147743 2024/11/06 12:43:04.728 5772 kleopatra.exe org.kde.pim.kleopatra: ReaderStatusThread[GUI]::ping()
^ another update of the smart cards is requested (the request is queued)
48464.804883 2024/11/06 12:43:10.393 5772 kleopatra.exe org.kde.pim.kleopatra: ReaderStatusThread: Card "89490171500022806460" with app "nks" was added 48464.805095 2024/11/06 12:43:10.393 5772 kleopatra.exe org.kde.pim.kleopatra: ReaderStatusThread: Card "D2760001240100000006154932910000" with app "openpgp" was added 48464.807483 2024/11/06 12:43:10.393 5772 kleopatra.exe org.kde.pim.kleopatra: ReaderStatusThread: Card "D2760001240100000006154932910000" with app "piv" was added
^ the background thread completed the update of the smart cards and found three card apps
48464.811286 2024/11/06 12:43:10.393 5772 kleopatra.exe org.kde.pim.kleopatra: ReaderStatusThread[2nd]: new iteration command= "__update__" ; nullSlot= true
^ background thread starts another update of the smart cards
48464.924701 2024/11/06 12:43:10.492 5772 kleopatra.exe org.kde.pim.kleopatra: ReaderStatusThread[GUI]::learnCardsCMS()
^ learn cards is requested (and queued) -> Kleopatra shows the progress overlay
48465.796319 2024/11/06 12:43:11.291 5772 kleopatra.exe org.kde.pim.libkleo: KeyCache::RefreshKeysJob start
^ a keylisting is started (OpenPGP and S/MIME)
48467.549251 2024/11/06 12:43:12.874 5772 kleopatra.exe org.kde.pim.libkleo: sendStatusLinesCommand "SCD LEARN --force" : got ( status( "READER" ) = "SCM Microsystems Inc. SPRx32 USB Smart Card Reader 0" [...] 48467.550423 2024/11/06 12:43:12.875 5772 kleopatra.exe org.kde.pim.libkleo: sendCommand "READKEY --card --no-data -- NKS-NKS3.4531" 48467.895485 2024/11/06 12:43:13.187 5772 kleopatra.exe org.kde.pim.libkleo: sendStatusLinesCommand "READKEY --card --no-data -- NKS-NKS3.4531" : got ( ) 48467.896400 2024/11/06 12:43:13.188 5772 kleopatra.exe org.kde.pim.libkleo: sendCommand "READKEY --card --no-data -- NKS-NKS3.45B1" 48468.209551 2024/11/06 12:43:13.471 5772 kleopatra.exe org.kde.pim.libkleo: sendStatusLinesCommand "READKEY --card --no-data -- NKS-NKS3.45B1" : got ( ) 48468.209660 2024/11/06 12:43:13.471 5772 kleopatra.exe org.kde.pim.libkleo: sendCommand "READKEY --card --no-data -- NKS-NKS3.4571"
^ the background thread sends multiple commands to gpg-agent to gather information about the smart cards
^ the last READKEY command seems to hang
48468.598283 2024/11/06 12:43:13.822 5772 kleopatra.exe org.kde.pim.libkleo: Kleo::KeyCache::RefreshKeysJob(0x63ccba8) RefreshKeysJob::done
^ the keylisting is done
Backported for VSD 3.3
Canceling the password prompt is now handled correctly, i.e. the operation is aborted without further feedback.
Nov 5 2024
This has also been reported at https://bugs.kde.org/show_bug.cgi?id=477798 (although there a crash occurs). Porting the command to gpgme didn't help, but the remaining problems are in gpg and/or gpgme.
This problem was also reported at https://bugs.kde.org/show_bug.cgi?id=479567#c1
Fixed and backported for VSD 3.3
I'm now using the name "Compliance Check" for the test if no compliance is active/has been configured. I have also checked all other usages of DeVSCompliance::name() in libkleo and kleopatra to make sure it's only used if compliance is active.
I have reverted the commit mentioned by Carl and another text codec related commit for the Qt 5 builds. This will hopefully fix the broken umlauts in the progress messages.
Fixed and backported for VSD 3.3
Nov 4 2024
Looking at the Windows "Problem reports" I saw that it lists many crashes of Kleopatra since 2024-06-19. An older Kleopatra (gpg4win-4.3.2-beta15, built on 2024-04-16) does not crash on exit. The next Kleopatra (gpg4win-4.3.2-beta23, built on 2024-07-05) does crash on exit. The reports point to the libkleo DLL.