gh-76534: Added docs about Embedding with an frozen module limitation. #4910
Conversation
|
@brettcannon, @rhettinger, @warsaw any of you would like to review these docs? |
|
Alright added the news entry as well. |
|
@brettcannon Mind finding someone to review this please. |
|
@AraHaan the problem is I have no embedding experience so I'm not in a good position to know if the documented solution is accurate enough. I would ask on python-dev for a reviewer. |
|
@birkenfeld any status yet? |
|
This PR is stale because it has been open for 30 days with no activity. |
|
Any new inputs to this? |
| ============================================ | ||
|
|
||
| While it is possible to use frozen modules in embedded python if the main module | ||
| is also frozen. It seems people are bitten when they don't have their main module |
There was a problem hiding this comment.
What do you mean by 'bitten'? I think it's better to use direct language and be clear on what the problem might be.
There was a problem hiding this comment.
As in when they try to run it, it won't work properly.
There was a problem hiding this comment.
Is there a specific exception/error that appears, or does it do something like hang?
There was a problem hiding this comment.
I believe it crashes or errors.
|
This PR is stale because it has been open for 30 days with no activity. |
|
Anything else I have to add to this? |
Co-authored-by: Stanley <46876382+slateny@users.noreply.github.com>
|
I wonder if the logic I wrote before could apply to cases where you might want to freeze the entire standard library into an embedded interpreter. I could see some use cases for that as well. Edit: Now that I think about it, why can't the entire standard library be frozen by default as well (the ci running the freezing steps on all of the standard library before it compiles |
|
This PR is stale because it has been open for 30 days with no activity. |
|
@erlend-aasland @gvanrossum @kumaraditya303 (as the ones related to |
There was a problem hiding this comment.
This reads more like a blog post than a piece of official documentation (even if some other sections in the same file also are written in a pretty informal style). I don't think it is ready to be included.
Alas, I don't have the time to research the root cause of the problem for which it is proposing a solution, so I can't really say whether I even approve of the proposed code snippets. Therefore, I don't want to create the impression that if the language could be cleaned up it would be acceptable. Sorry, but this just needs a lot more work.
|
This PR is stale because it has been open for 30 days with no activity. |
This should help fix some of the documentation by adding an section for some common mistakes in embedding type situations.
https://bugs.python.org/issue32353