path sep and windows related changes

[ThomasWaldmann]

No description provided.

[ThomasWaldmann]

also discussed os.path.normpath/abspath/realpath usage on windows, #7120.

[ThomasWaldmann]

@jedenastka do you have feedback about the contents of the plan?

[jedenastka]

My concern is that it is AI generated. To me, this signals laziness and lack of thought, which is not something I want to see in a backup tool.

A plan means you're actually planning something. If AI does it, it is not planning, because AI cannot think.

Feel free to disagree. I do not plan to engage in further conversation.

[ThomasWaldmann]

@jedenastka Well, I understand your concerns. I also have quite some concerns about AI usage, but (as so often), it depends on how you use it.

The planning one can see here is the result of an hours long "discussion" between me and the AI, so maybe that was not what you meant with "lazy" and "lack of thought". It helped not to overlook aspects or places of the needed changes.

Just wanted to clarify that. You don't need to reply if you do not have more specific feedback about the plan contents.

[RonnyPfannschmidt]

wrt drive letters and extraction - how about storing the "extraction root" in the archive in a way that all items in a archive are relative paths to the extraction root

this would come with some perks like backing up the same file tree in a different location would generate the same archive content including path names for all but the archive metadata

wrt forbidden path names on windows - would using unc full paths work - i dislike schemes that may trigger a overlap - and statefull overlap prevention is unreasonably expensive in terms of implementation complexity

[ThomasWaldmann]

@RonnyPfannschmidt one root would only be the simplest use case, but would not be enough for the general case. You can give multiple paths to borg as cli params or via the patterns file. You can even feed a list of arbitrary paths to borg.

[codecov]

Codecov Report

❌ Patch coverage is 92.00000% with 4 lines in your changes missing coverage. Please review.
✅ Project coverage is 75.85%. Comparing base (d602cf2) to head (6c1fa5f).
⚠️ Report is 8 commits behind head on master.

Files with missing lines	Patch %	Lines
src/borg/helpers/fs.py	77.77%	1 Missing and 1 partial ⚠️
src/borg/helpers/parseformat.py	75.00%	1 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #9264      +/-   ##
==========================================
+ Coverage   75.84%   75.85%   +0.01%     
==========================================
  Files          86       86              
  Lines       14742    14754      +12     
  Branches     2194     2195       +1     
==========================================
+ Hits        11181    11192      +11     
  Misses       2888     2888              
- Partials      673      674       +1     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.