Repo Administration and Housekeeping

[slook]

Matters regarding the running and maintenance of the main repository

• This is not a conversation thread, and as such, please keep replies in-line and on-topic.

[slook]

DONE

What to do about the Italian translation mixup?

My apologies for this chaos, I have suggested edits in #1518 but now I think they are blocking 401f760 or some other issue not sure.

I made a pickle here I think, sorry for the chaos, I didn't realize that the #1518 would be versioning dependent when I requested that @M0Rf30 make edits to it. There are 4 items that require minor alteration. The timing is not ideal due to the significant UI changes.

I cant edit the file directly, so I commented it to make sure the Italian users dont have to put up with a broken interface until who knows when, but I didn't realize this would cause a block, I'm very sorry about any inconvenience this has causes anybody.

Unfortunately, this can't be merged until it's edited and approved, but I am unable to do this.

The whole this started with the balls up in # where I made a school boy error and suggested Alt+F for a button, but I later realised thats always the _File menu.

In future, I shall learn from this mistake and suggest future key bindings with the header bar disabled.

In the meantime @mathiascode I think you might need to step in to edit the .po so that it is correct with the changes you have made.

[mathiascode]

Don't worry. Your suggested edits are only suggestions (code review), and are not causing any blockage. @M0Rf30 can make changes as usual when they have time.

[slook]

I see, thank you. I am wondering that many changes will be needed for 3.2, do you get pull requests for translations often?

Perhaps I should review the whole thing again, perhaps many strings are changed now. Will this be okay?

So then why is Result _Filters still appearing on my local copy of 3.2.0.dev1 ?

[mathiascode]

Perhaps I should review the whole thing again, perhaps many strings are changed now. Will this be okay?

I'm not sure if this is necessary effort. The .pot translation template (list of original strings) is updated automatically when running python setup.py update_pot, and missing strings in the Italian translation will then be pointed out to the translator.

[slook]

Right I see, so the whole thing will just shake itself out then. Phew.

I am not familiar with this kinds of versioning scenarios, especially where one mistake (like with the Whole Alf+F up) leads to another mistake that crosses, sorry about that. I appreciate your guidance on this.

[slook]

It looks like 234d081 has been updated and is ready to go for now.

DONE

[slook]

FIXED

Unstable Windows Installer / Package is difficult to obtain

In reference to README.md, the link provided (64/32-bit Installers / Packages) does not easily lead to the current Artifact.

Windows testers are much less likely to be familiar with Git, and Win testing seems to be limited.

[mathiascode]

8aefea2

[slook]

FIXED

Incorrect project title summary description on Launchpad repo

The words "file" and "sharing" are still present from the previous project title summary string (changed in #1481 and #1482)

https://launchpad.net/nicotine+
https://launchpad.net/~nicotine-team
https://code.launchpad.net/~nicotine-team/+recipe/nicotine+-daily

[slook]

RESOLVED Nightly.link update interval is after every commit

I am unsure if the following text is correct @mathiascode the timestamps inside the latest package are few hours ago, yet your last commit was a few minutes ago (even if there is a timezone issue, the minutes are nearly an hour ago).

Unstable

Unstable Windows packages are generated after every commit to the master branch, and should only be used for testing.

I wonder if these packages are infact only updated "Nightly"? In which case, the above line of text in README.md might be incorrect. (Edit: no, it is correct)

[mathiascode]

The nightly link gave me the latest build when I tried a few mins ago.

[slook]

DONE 3.2.x branch - It would be good if there were Labels or Tags for the issues in the tracker, to make it easier to sort between Stable and Unstable, or version numbers such as [3.1.1] and [3.2.0.dev1] for example.

This would help to prevent activity from developers and testers from mingling with Stable user submitted bug reports, which makes the project appear as if it is plagued with problems at first glance.

Perhaps there is some GitHub repo feature that can be implemented to serve this purpose?

I don't know if it is worth creating 2 separate branches instead of master, such as stable and unstable, whilst this might enable urgent/minor revision patches to be more easily released for Stable, perhaps it would create undesired work.

[slook]

NOT REQUIRED - Consider creating a repo mirror on GitLab

'Import your project from GitHub to GitLab' https://docs.gitlab.com/ee/user/project/import/github.html

[mathiascode]

Any specific reason? I don't use gitlab.com personally, and I'm not sure if I want more mirrors that potentially pollute search results, considering the past of this project with Sourceforge and several outdated websites.

[slook]

Purely for backup and redundancy purposes, in case of a takedown or gh service outage. Perhaps a robot.txt file and an alternate README.md which redirects to the main repo might be used to prevent such a mirror from being indexed on search engines.

[mathiascode]

We have a mirror at https://code.launchpad.net/nicotine+

[slook]

DONE - The website testing page needs pointing to the master branch

https://nicotine-plus.org/doc/TESTING.html is still on the 3.2.x branch at 3.2.6

[mathiascode]

Thanks for the reminder

[slook]

@mathiascode ^^

[slook]

RESOLVED - Windows machines are slow and unreliable when running GitHub actions

Investigate exactly why it takes so long. If we can't do anything to fix the problems with it, then perhaps there are some redundant actions that can be avoided. Maybe the service host might appreciate the reduced load on their servers, it feels like we often create a lot of unnecessary work for the machines to do.

It might be nice to have a alternate installer script that is more rapid for use during day-to-day development tasks, as opposed to a more through CI setup that needs to be used for preparing releases.

[mathiascode]

Fixed in astroid 2.13.2

[slook]

https://github.com/nicotine-plus/nicotine-plus/actions/runs/3972109783/jobs/6809679795#step:4:42

curl: (77) error setting certificate file: C:/msys64/mingw32/ssl/certs/ca-bundle.crt
subprocess.CalledProcessError: Command '['curl', '-O', 'https://repo.msys2.org/mingw/mingw32/mingw-w64-i686-cairo-1.17.4-4-any.pkg.tar.zst']' returned non-zero exit status 77.

Does the cached certificate need updating or something?

[mathiascode]

There was an update that changed the certificate path from /ssl to /etc/ssl, but some packages still use the old path. Should be fixed in a couple of days.

[slook]

What about putting /etc/ssl in the PATH, or make a symlink?

[mathiascode]

d0d6e32 takes care of the remaining issue. Everything should be back to normal now.

[slook]

RESOLVED - macOS machine DistutilsOptionError: error in setup script

Guess this means these runners are obsoleted then?

distutils.errors.DistutilsOptionError: error in setup script: command 'build' has no such option 'build_exe'

https://github.com/nicotine-plus/nicotine-plus/actions/runs/4010007829/jobs/6886016813#step:5:29

[mathiascode]

Fixed in a39cc2c. Reported the other regression here: marcelotduarte/cx_Freeze#1767

[slook]

Your report might be a duplicate of marcelotduarte/cx_Freeze#1765, or very closely related.

[slook]

Fixed in cx_Freeze 6.14.1 by marcelotduarte/cx_Freeze#1768

[mathiascode]

RESOLVED - Fix fatal exception in Windows integration test

test_startup (test.integration.test_startup.StartupTest)
Verify that regular startup works ... Windows fatal exception: code 0xc0000139
[2023-01-30 20:50:46] Loading Python 3.10.9
[2023-01-30 20:50:46] Loading Nicotine+ 3.3.0.dev4

Thread 0x00001318 (most recent call first):
[2023-01-30 20:50:46] Translation files (.mo) are unavailable, using default English strings
[2023-01-30 20:50:46] Loading GTK 4.8.3
  File "D:/a\nicotine-plus\nicotine-plus/pynicotine/events.py", line 262 in _run_scheduler
  File "C:/msys64/mingw64/lib/python3.10/threading.py", line 953 in run
  File "C:/msys64/mingw64/lib/python3.10/threading.py", line 1016 in _bootstrap_inner
  File "C:/msys64/mingw64/lib/python3.10/threading.py", line 973 in _bootstrap

Current thread 0x000016a8 (most recent call first):
  File "<frozen importlib._bootstrap>", line 241 in _call_with_frames_removed
  File "<frozen importlib._bootstrap_external>", line 1180 in create_module
  File "<frozen importlib._bootstrap>", line 571 in module_from_spec
  File "<frozen importlib._bootstrap>", line 674 in _load_unlocked
  File "<frozen importlib._bootstrap>", line 1006 in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 1027 in _find_and_load
  File "C:/msys64/mingw64/lib/python3.10/site-packages/cairo/__init__.py", line 1 in <module>
  File "<frozen importlib._bootstrap>", line 241 in _call_with_frames_removed
  File "<frozen importlib._bootstrap_external>", line 887 in exec_module
  File "<frozen importlib._bootstrap>", line 688 in _load_unlocked
  File "<frozen importlib._bootstrap>", line 1006 in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 1027 in _find_and_load
  File "<frozen importlib._bootstrap>", line 241 in _call_with_frames_removed
  File "<frozen importlib._bootstrap_external>", line 1180 in create_module
  File "<frozen importlib._bootstrap>", line 571 in module_from_spec
  File "<frozen importlib._bootstrap>", line 674 in _load_unlocked
  File "<frozen importlib._bootstrap>", line 1006 in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 1027 in _find_and_load
  File "C:/msys64/mingw64/lib/python3.10/site-packages/gi/__init__.py", line 62 in <module>
  File "<frozen importlib._bootstrap>", line 241 in _call_with_frames_removed
  File "<frozen importlib._bootstrap_external>", line 887 in exec_module
  File "<frozen importlib._bootstrap>", line 688 in _load_unlocked
  File "<frozen importlib._bootstrap>", line 1006 in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 1027 in _find_and_load
  File "D:/a\nicotine-plus\nicotine-plus/pynicotine/gtkgui/__init__.py", line 41 in check_gtk_version
  File "D:/a\nicotine-plus\nicotine-plus/pynicotine/gtkgui/__init__.py", line 86 in run
  File "D:/a\nicotine-plus\nicotine-plus/pynicotine/__init__.py", line 215 in run
  File "D:/a\nicotine-plus\nicotine-plus/pynicotine/__main__.py", line 22 in <module>
  File "C:/msys64/mingw64/lib/python3.10/runpy.py", line 86 in _run_code
  File "C:/msys64/mingw64/lib/python3.10/runpy.py", line 196 in _run_module_as_main
[2023-01-30 20:50:48] Loaded plugin Nicotine+ Commands
[2023-01-30 20:50:48] Loading plugin system
[2023-01-30 20:50:48] You need to specify a username and password before connecting…

[slook]

https://github.com/nicotine-plus/nicotine-plus/actions/runs/4048173048/jobs/6963090493#step:7:58

  File "C:/msys64/mingw64/lib/python3.10/runpy.py", line 196 in _run_module_as_main
Gsk-Message: 21:24:35.862: Failed to realize renderer of type 'GskGLRenderer' for surface 'GdkWin32Toplevel': Compilation failure in shader.

 Error Message:
ERROR: 0:64: 'attribute' :  supported in GLSL ES 1.00 only
ERROR: 0:65: 'attribute' :  supported in GLSL ES 1.00 only
ERROR: 0:66: 'attribute' :  supported in GLSL ES 1.00 only
ERROR: 0:67: 'attribute' :  supported in GLSL ES 1.00 only
ERROR: 0:68: 'varying' :  supported in GLSL ES 1.00 only
ERROR: 0:136: 'gl_Position' : undeclared identifier
ERROR: 0:136: 'assign' : l-value required (can't modify a const)
ERROR: 0:136: '=' : dimension mismatch
ERROR: 0:136: 'assign' : cannot convert from '4-component vector of float' to 'const highp float'

 64| attribute vec2 aPosition;
 65| attribute vec2 aUv;
 66| attribute vec4 aColor;
 67| attribute vec4 aColor2;
 68| _OUT_ vec2 vUv;

  8| #if defined(GSK_GLES) || defined(GSK_LEGACY)
  9| #define _OUT_ varying

136| gl_Position = u_projection * u_modelview * vec4(aPosition, 0.0, 1.0);

Looks like something to do with window or dialog box positioning, as in #2312 (reply in thread)

[slook]

DONE - pylint disable: superfluous-parens

https://github.com/nicotine-plus/nicotine-plus/actions/runs/4070169259/jobs/7010728031#step:6:63

Unnecessary parens after '=' keyword (superfluous-parens)

Personally I find "superfluous" parenthesis around expressions is better for readability rather than using bool().

[slook]

RESOLVED - fedora-rocky CI test fail

Only the very first test run fails, all others do succeed ok.

(org.nicotine_plus.Nicotine:444): Gtk-WARNING **: 21:08:48.473: Unable to acquire session bus: Cannot spawn a message bus without a machine-id: Invalid machine ID in /var/lib/dbus/machine-id or /etc/machine-id

https://github.com/nicotine-plus/nicotine-plus/actions/runs/5094957932/jobs/9159388030#step:7:13

test_startup (test.integration.test_startup.StartupTest.test_startup)
Verify that regular startup works ... [2023-05-26 21:08:48] Loading Python 3.11.3
[2023-05-26 21:08:48] Loading Nicotine+ 3.3.0.dev5
[2023-05-26 21:08:48] Translation files (.mo) are unavailable, using default English strings
[2023-05-26 21:08:48] Loading GTK 4.11.2

(org.nicotine_plus.Nicotine:444): Gtk-WARNING **: 21:08:48.473: Unable to acquire session bus: Cannot spawn a message bus without a machine-id: Invalid machine ID in /var/lib/dbus/machine-id or /etc/machine-id
[2023-05-26 21:08:48] Loaded plugin Nicotine+ Commands
[2023-05-26 21:08:48] Loading plugin system
[2023-05-26 21:08:48] You need to specify a username and password before connecting…

(org.nicotine_plus.Nicotine:444): Gdk-WARNING **: 21:08:48.711: The program 'org.nicotine_plus.Nicotine' received an X Window System error.
This probably reflects a bug in the program.
The error was 'GLXBadFBConfig'.
  (Details: serial 1070 error_code 167 request_code 150 (GLX) minor_code 0)
  (Note to programmers: normally, X errors are reported asynchronously;
   that is, you will receive the error a while after causing it.
   To debug your program, run it with the GDK_SYNCHRONIZE environment
   variable to change this behavior. You can then get a meaningful
   backtrace from your debugger if you break on the gdk_x_error() function.)
/usr/lib64/python3.11/multiprocessing/resource_tracker.py:224: UserWarning: resource_tracker: There appear to be 3 leaked semaphore objects to clean up at shutdown
  warnings.warn('resource_tracker: There appear to be %d '
======================================================================
ERROR: test_startup (test.integration.test_startup.StartupTest.test_startup)
Verify that regular startup works
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/__w/nicotine-plus/nicotine-plus/test/integration/test_startup.py", line 44, in test_startup
    subprocess.check_call(command, timeout=5)
  File "/usr/lib64/python3.11/subprocess.py", line 413, in check_call
    raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '['python3', '-m', 'pynicotine', '--config=/__w/nicotine-plus/nicotine-plus/test/integration/config', '--user-data=/__w/nicotine-plus/nicotine-plus/test/integration', '--ci-mode']' returned non-zero exit status 1.

----------------------------------------------------------------------
Ran 57 tests in 9.673s

FAILED (errors=1)
Error: Process completed with exit code 1.

[slook]

RESOLVED - Difficulty compiling release notes and changelog for 3.3.0

How the hell are we going to gather the list of changes and resolved issues for the release notes? There is an overlap between branches and the number of commits is massive since the last release, I literally don't even know where to start from!

Suggestions are welcome as to how this task would be best approached.

[mathiascode]

Any changes since March 5 should be fairly straightforward, the rest will be a... fun challenge. I think we could at least start with comparing the issue tracker to the issues listed in NEWS.md (before March 5, only feature requests should be missing), then I'll probably have to dig through old discussions and commits to find relevant things.

Perhaps there's some tool out there that can filter out commits with the same commit message in two branches, so we only get a list of commits exclusive to the master branch.

[mathiascode]

List of closed issues done in 23d2719

[mathiascode]

Do you think you could add some items to the Corrections section? Relevant changes should be between March (release of 3.2.9) and now.

If it's too overwhelming, I could do it eventually.

[slook]

Delete old versions of the IP country database to reduce repo size

It serves no purpose to keep old versions of the binary file, yet everytime it is updated the total repo size is increased by a few megabytes.

https://github.com/nicotine-plus/nicotine-plus/commits/master/pynicotine/external/ipcountrydb.bin

In future it might be ideal for the application to acquire the database from the remote source into local user data storage at some point after installation during runtime, for example upon first time click on an "Unknown" country flag icon or the Country search result filter text entry is focused for the first time.

Currently the total size of the repo is 105MB.

[mathiascode]

I don't think it's worth rewriting the git history. All revisions of the country database take about 15 MB of storage combined (packed/compressed).

[slook]

Okay I agree with you. Still, it would be preferable to avoid further excessive growth, if possible.