chore: Move bundling from CMake to Python #508
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
WillAyd
reviewed
Jun 10, 2024
| import re | ||
|
|
||
|
|
||
| def read_content(path_or_content): |
There was a problem hiding this comment.
Of course up to you but I find the behavior of this function surprising. More often than not, when you accept a path argument then accepting a string usually still represents the filepath/name, not the contents itself.
If you wanted you could refactor this to just be:
def read_content(path_or_content: io.StringIO | pathlib.Path | str):
with open(path_or_content) as f:
return f.read()For the caller this would require them to pass io.StringIO("some_contents") instead of just "some_contents", but I think makes it less ambiguous in case someone expects a string argument here to still represent a path
There was a problem hiding this comment.
I agree it's a little confusing...if this utility gets more widespread use than just this bundler it should definitely get revisited!
paleolimbot
force-pushed
the
bundle-using-python
branch
from
149e6bc
to
bab1182
Compare
assignUser
approved these changes
Jun 18, 2024
paleolimbot
force-pushed
the
bundle-using-python
branch
from
525e7d2
to
e126672
Compare
paleolimbot
pushed a commit
that referenced
this pull request
Jun 21, 2024
Quick follow up to #508 - the array.c includes stopped working without this ``` FAILED: libnanoarrow.so.p/src_nanoarrow_array.c.o ccache cc -Ilibnanoarrow.so.p -I. -I.. -fdiagnostics-color=always -D_FILE_OFFSET_BITS=64 -Wall -Winvalid-pch -Wextra -std=c99 -O0 -g -fPIC -MD -MQ libnanoarrow.so.p/src_nanoarrow_array.c.o -MF libnanoarrow.so.p/src_nanoarrow_array.c.o.d -o libnanoarrow.so.p/src_nanoarrow_array.c.o -c ../src/nanoarrow/array.c ../src/nanoarrow/array.c:23:10: fatal error: nanoarrow/nanoarrow.h: No such file or directory 23 | #include "nanoarrow/nanoarrow.h" | ^~~~~~~~~~~~~~~~~~~~~~~ ```
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Some of the work with the Meson build system has highlighted that the bundling is currently very inflexible because it's implemented in CMake. This was initially done because it made it easy to use
configure_file()in CMake to keep the version information consistent, and because it allowed a straightforward path to testing the bundled files in the same way as the normal build.This had the unfortunate side effect of requiring CMake to "build" the R or Python packages as is generally confusing, and used the CMake "install" in a confusing way (it installed sources). Instead, we can do this with a Python script (and call that from CMake, R, and Python). The testing of the namespaced builds is still done in CMake, which is essential to ensure that both build types are tested in the same way.
This also closes #485 by prefixing all nanoarrow includes with "nanoarrow/" in the source files. Many of them had to not use this prefix because when the files were bundled they weren't always bundled into a "nanoarrow/" directory. This feature has been moved to the bundling script which can now namespace or un-namespace the nanoarrow includes.