rfc:include_cleanup
Differences
This shows you the differences between two versions of the page.
| Next revision | Previous revision | ||
| rfc:include_cleanup [2023/01/18 14:52] – created maxk | rfc:include_cleanup [2023/02/15 15:11] (current) – declined maxk | ||
|---|---|---|---|
| Line 3: | Line 3: | ||
| * Date: 2023-01-18 | * Date: 2023-01-18 | ||
| * Author: Max Kellermann, max.kellermann@ionos.com | * Author: Max Kellermann, max.kellermann@ionos.com | ||
| - | * Status: | + | * Status: Declined |
| * First Published at: https:// | * First Published at: https:// | ||
| Line 55: | Line 55: | ||
| * later Dmitry Stogov changed his mind, saying "This is just a useless permutation. [...] How is this clearly?" | * later Dmitry Stogov changed his mind, saying "This is just a useless permutation. [...] How is this clearly?" | ||
| * Dmitry Stogov rejected the idea of forward declarations because " | * Dmitry Stogov rejected the idea of forward declarations because " | ||
| + | * Derick Rethans believes that " | ||
| + | * Dmitry Stogov and Derick Rethans worried that these changes would make merging branches harder, due to possible merge conflicts or build failures because the source branch does not have correct #includes and now uses a new symbol from a header which was not included (though the same can be said about all code changes, and changing #includes is not very intrusive, because these rarely ever change in stable branches, and there' | ||
| + | * There was some opposition against struct forward declarations; | ||
| ===== Backward Incompatible Changes ===== | ===== Backward Incompatible Changes ===== | ||
| Line 99: | Line 101: | ||
| ===== Proposed Voting Choices ===== | ===== Proposed Voting Choices ===== | ||
| - | * Yes (clean | + | * Clean up #include directives? (Yes/No) |
| - | * No (do not clean up headers and #include directives) | + | |
| - | + | ||
| - | Other questions that could possibly be decided on: | + | |
| * Is it allowed to document an #include line with a code comment? (Yes/No) | * Is it allowed to document an #include line with a code comment? (Yes/No) | ||
| + | * Is it allowed to forward-declare structs/ | ||
| * Is it allowed to split a large header, e.g. move zend_result to a separate header, to reduce dependencies on catch-all headers such as zend_types.h? | * Is it allowed to split a large header, e.g. move zend_result to a separate header, to reduce dependencies on catch-all headers such as zend_types.h? | ||
| + | |||
| + | ===== Vote ===== | ||
| + | |||
| + | Voting started on 2023-02-01 (two weeks after the RFC was posted to php-internals) and will end on 2023-02-15 at 15:00 UTC (another two weeks). | ||
| + | |||
| + | Primary vote: | ||
| + | |||
| + | <doodle title=" | ||
| + | * Yes | ||
| + | * No | ||
| + | </ | ||
| + | |||
| + | Secondary (clarifications on how to clean up): | ||
| + | |||
| + | <doodle title=" | ||
| + | * Yes | ||
| + | * No | ||
| + | </ | ||
| + | |||
| + | <doodle title=" | ||
| + | * Yes | ||
| + | * No | ||
| + | </ | ||
| + | |||
| + | <doodle title=" | ||
| + | * Yes | ||
| + | * No | ||
| + | </ | ||
| ===== Patches and Tests ===== | ===== Patches and Tests ===== | ||
| Line 111: | Line 139: | ||
| https:// | https:// | ||
| https:// | https:// | ||
| + | |||
| + | https:// | ||
| + | |||
| + | https:// | ||
| + | |||
| + | If voters decide that #include directives should not have code comments, then existing comments should be removed to reduce " | ||
| ===== References ===== | ===== References ===== | ||
rfc/include_cleanup.1674053541.txt.gz · Last modified: 2023/01/18 14:52 by maxk