lzma implementation #5717
lzma implementation #5717
Conversation
|
This turned out surprisingly effective, about 2/3s of the tests pass, which is enough for me at the moment. The main missing parts involve a bunch of unsafe things that As a side note I also abstracted out |
c09442c to
2bad5d2
Compare
|
Lots of tests are failing because they were previously blocked due to no implementation at all. |
|
Would you be able to squash these commits down? Ideally into one for copying the files from cpython, and one for all the rest? |
|
Sure |
19a43f5 to
ca2dc44
Compare
|
It's at 3 now, I committed things in a weird fashion so I can't get it below that. |
Lib/tarfile.py
Outdated
| @@ -392,7 +392,8 @@ def __init__(self, name, mode, comptype, fileobj, bufsize, | |||
|
|
|||
| elif comptype == "xz": | |||
| try: | |||
| import lzma | |||
| # TODO: RUSTPYTHON remove underscore | |||
There was a problem hiding this comment.
Is this name changed to raise ImportError? What happens if it is lzma?
There was a problem hiding this comment.
Yes, if left as is everything breaks because lzma is expected to be unimportable.
There was a problem hiding this comment.
xz seems to be not support. Added reason and avoided to use fake module name
|
oh no.. I thought #5728 and this one had shared work of the new compression module |
|
@arihant2math What will be the best way? Is it worth to keep #5728? |
|
Probably easiest to revert, yes. |
…tPython#5728)" This reverts commit 9c88475.
Signed-off-by: Ashwin Naren <arihant2math@gmail.com>
Signed-off-by: Ashwin Naren <arihant2math@gmail.com>
Signed-off-by: Ashwin Naren <arihant2math@gmail.com>
Based on #5709 because of the zlib abstraction.