Add an initial implementation of xargs #121
Conversation
refi64
force-pushed
the
xargs
branch
2 times, most recently
from
6483075
to
49c962f
Compare
|
Looks like the failures from #120 were affecting here too, should be working here now as well (also fixed the clippy warnings) |
Codecov Report
@@ Coverage Diff @@
## main #121 +/- ##
==========================================
+ Coverage 47.97% 52.58% +4.60%
==========================================
Files 20 23 +3
Lines 4106 4650 +544
Branches 1431 1529 +98
==========================================
+ Hits 1970 2445 +475
- Misses 1650 1660 +10
- Partials 486 545 +59
Continue to review full report at Codecov.
|
|
looks like it is working well |
Well all of them work locally, so 🤞 |
|
i did this: |
|
could you please rebase this one too? |
This includes much of the core xargs functionality, with the following notable exceptions: - Parallel execution (`-P`): This option currently just does nothing, that way anything that passes -P can at least run without a notable behavior shift (other than simply being slower). - Replacement strings (`-I`): This can easily be worked around via an intermediate shell invocation (e.g. `xargs -L1 sh -c 'do-things-with $@' --`). - EOF strings (`-E`): I've honestly never seen this actually used, though it would not be particularly difficult to implement given the current architecture. Closes uutils#37 Signed-off-by: Ryan Gonzalez <ryan.gonzalez@collabora.com>
|
Just rebased, the GNU test suite is now running on this xargs so a few new tests will fail (since this doesn't support stuff like |
This includes much of the core xargs functionality, with the following
notable exceptions:
-P): This option currently just doesnothing, that way anything that passes -P can at least run without a
notable behavior shift (other than simply being slower).
-I): This can easily be worked around via anintermediate shell invocation (e.g.
xargs -L1 sh -c 'do-things-with $@' --).-E): I've honestly never seen this actually used,though it would not be particularly difficult to implement given the
current architecture.
Closes #37
Signed-off-by: Ryan Gonzalez ryan.gonzalez@collabora.com
This has not been tested with Windows CI yet and also depends on #120, so I've marked it as a draft.