Joey Hess [Thu, 30 Jan 2025 18:34:21 +0000 (14:34 -0400)]
bring back OsPath changes
I hope that the windows test suite failure on appveyor was fixed by
updating to a newer windows there. I have not been able to reproduce
that failure in a windows 11 VM run locally.
Joey Hess [Wed, 29 Jan 2025 22:59:09 +0000 (18:59 -0400)]
throwing away a week's work
This was an attempt at continuing the RawFilePath transition into an
OsPath transition. Unforunately, it had multiple test suite failures on
windows. Debugging those seems that it would be more work than the
minor optimisation value of this.
Joey Hess [Wed, 29 Jan 2025 21:22:21 +0000 (17:22 -0400)]
disable OsPath build flag on windows for now
Test suite failure looks like this:
fatal: could not open '\\?\C:\Users\appveyor\AppData\Local\Temp\1\hash-cc81b41d-dfda-4ae8-904b-b531742443cc' for reading: No such file or directory
fatal: could not open '\\?\C:\Users\appveyor\AppData\Local\Temp\1\hash-cc81b41d-dfda-4ae8-904b-b531742443cc' for reading: No such file or directory
fatal: could not open '\\?\C:\Users\appveyor\AppData\Local\Temp\1\hash-cc81b41d-dfda-4ae8-904b-b531742443cc' for reading: No such file or directory
fatal: could not open '\\?\C:\Users\appveyor\AppData\Local\Temp\1\hash-cc81b41d-dfda-4ae8-904b-b531742443cc' for reading: No such file or directory
fatal: could not open '\\?\C:\Users\appveyor\AppData\Local\Temp\1\hash-cc81b41d-dfda-4ae8-904b-b531742443cc' for reading: No such file or directory
fatal: could not open '\\?\C:\Users\appveyor\AppData\Local\Temp\1\hash-cc81b41d-dfda-4ae8-904b-b531742443cc' for reading: No such file or directory
fatal: could not open '\\?\C:\Users\appveyor\AppData\Local\Temp\1\hash-cc81b41d-dfda-4ae8-904b-b531742443cc' for reading: No such file or directory
fatal: could not open '\\?\C:\Users\appveyor\AppData\Local\Temp\1\hash-cc81b41d-dfda-4ae8-904b-b531742443cc' for reading: No such file or directory
fatal: could not open '\\?\C:\Users\appveyor\AppData\Local\Temp\1\hash-cc81b41d-dfda-4ae8-904b-b531742443cc' for reading: No such file or directory
fatal: could not open '\\?\C:\Users\appveyor\AppData\Local\Temp\1\hash-cc81b41d-dfda-4ae8-904b-b531742443cc' for reading: No such file or directory
fatal: could not open '\\?\C:\Users\appveyor\AppData\Local\Temp\1\hash-cc81b41d-dfda-4ae8-904b-b531742443cc' for reading: No such file or directory
git-annex: fd:4: Data.ByteString.hGetLine: end of file
git-annex: user error (git ["--git-dir=.git","--work-tree=.","--literal-pathspecs","hash-object","-w","--no-filters","--stdin-paths"] exited 128)
This is apparently in Git.HashObject, and probably in hashBlob, which uses a
temp file with a name starting with "hash", but I have not been able to tell
what's wrong.
I don't understand where the "\\?" path prefix (windows UNC-style path)
is coming from in the path that gets fed into git hash-file, or why git
hash-file can't open the file.
luciusf [Sun, 26 Jan 2025 11:19:53 +0000 (11:19 +0000)]
rename bugs/rsyncurl_without___34____58____34___creates_local_folder_as_repo.mdwn to bugs/rsyncurl_without___34____58____34___creates_local_folder_as_remote.mdwn
Joey Hess [Wed, 22 Jan 2025 17:22:51 +0000 (13:22 -0400)]
simplify replaceFile using relatedTemplate
Now that truncateFilePath and relatedTemplate have both been optimised,
may as well use them in replaceFile, rather than the custom hack it
used.
Removed the windows-specific ifdef as well, because on Windows long
filepaths no longer really a problem, since ghc and git-annex use UNC
converted paths.
replaceFile no longer checks fileNameLengthLimit. That took a syscall,
and since we have an existing file, we know filenames of its length are
supported by the filesystem. Assuming that the withOtherTmp directory is
on the same filesystem as the file replaceFile is being called on, which
I believe it is.
Joey Hess [Wed, 22 Jan 2025 16:34:54 +0000 (12:34 -0400)]
optimize truncateFilePath
Often the filepath will be all ascii, or mostly so, and this
optimisation makes a file that has an ascii suffix of sufficient length
be roundtrip converted between String and ByteString only once, rather
than once per character.
Joey Hess [Tue, 21 Jan 2025 21:00:37 +0000 (17:00 -0400)]
use openTempFile from file-io
And follow-on changes.
Note that relatedTemplate was changed to operate on a RawFilePath, and
so when it counts the length, it is now the number of bytes, not the
number of code points. This will just make it truncate shorter strings
in some cases, the truncation is still unicode aware.
When not building with the OsPath flag, toOsPath . fromRawFilePath and
fromRawFilePath . fromOsPath do extra conversions back and forth between
String and ByteString. That overhead could be avoided, but that's the
non-optimised build mode, so didn't bother.