• Home
  • Features
  • Pricing
  • Docs
  • Announcements
  • Sign In

IJHack / QtPass / 43741034965
96%

Build:
DEFAULT BRANCH: main
Ran 15 Nov 2026 05:46PM UTC
Jobs 1
Files 87
Run time 1min
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst

15 Nov 2026 05:25PM UTC coverage: 96.321% (+0.01%) from 96.307%
43741034965

push

github

web-flow
Keep the bytes of an entry that is not UTF-8 (#2195)

* Keep the bytes of an entry that is not UTF-8

An entry whose content is not UTF-8 (a Latin-1 password, a key file
stored with pass insert -m) reached QtPass as text with U+FFFD for each
such byte, and two paths wrote that text back with success reported:
native re-encryption of a folder, and OK in the edit dialog, even with
nothing changed.

Executor now says, for the output it emits, whether the text encodes back
to the bytes the process wrote (lastOutputLossless()); Pass passes it on
as a third, defaulted argument of finishedShow(), and PasswordDialog does
not load an entry that would not, as it already does for one with too
many lines. A literal U+FFFD, as older versions left behind, encodes back
and stays editable.

Re-encryption carries the plaintext as bytes end to end: a byte variant of
the blocking run (Executor::executeBlockingRaw(), ImitatePass::
execBlockingRaw()), and the check after encrypting compares bytes, not two
strings damaged alike. It no longer appends a newline to an entry without
one, and every encrypt passes --no-textmode, so a gpg.conf textmode line
cannot drop the entry's carriage returns.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Signed-off-by: Anne Jan Brouwer <brouwer@annejan.com>

* tests: avoid 'caf' in executor test data (codespell)

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Signed-off-by: Anne Jan Brouwer <brouwer@annejan.com>

* PasswordDialog: a refused answer takes back an earlier load

A later answer for the same entry that the dialog refuses (not UTF-8,
too many lines) left what an earlier one loaded in place with OK
enabled, so OK wrote it over the entry. Both refusals now reset the
loaded state and lock the editor and OK (CodeRabbit).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Signed-off-by: Anne Jan Brouwer <brouwer@annejan.com>

---------

Signed-off-by: Anne Jan Brouwer <brouwer@annejan.com>
Co-authored-by: Clau... (continued)

59 of 59 new or added lines in 5 files covered. (100.0%)

1 existing line in 1 file now uncovered.

8665 of 8996 relevant lines covered (96.32%)

605.72 hits per line

Coverage Regressions

Lines Coverage ∆ File
1
97.56
0.15% src/executor.cpp
Jobs
ID Job ID Ran Files Coverage
1 43741034965.1 15 Nov 2026 05:46PM UTC 87
96.32
GitHub Action Run
Source Files on build 43741034965
  • Tree
  • List 87
  • Changed 8
  • Source Changed 8
  • Coverage Changed 5
Coverage ∆ File Lines Relevant Covered Missed Hits/Line
  • Back to Repo
  • Github Actions Build #43741034965
  • 7527a437 on github
  • Prev Build on main (#43739213287)
  • Next Build on main (#43743282531)
STATUS · Troubleshooting · Open an Issue · Sales · Support · CAREERS · ENTERPRISE · START FREE TRIAL · SCHEDULE DEMO
ANNOUNCEMENTS · TWITTER · TOS & SLA · Supported CI Services · What's a CI service? · Automated Testing

© 2026 Coveralls, Inc