Timeline for How are specification issues / defects of POSIX collected and how can I participate?
Current License: CC BY-SA 4.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| May 5 at 12:18 | comment | added | ilkkachu |
Even if it says "equivalent to unlink()", I don't think that forbids an atomic replacement. I mean, in practice it likely means that whatever happens, it must do the same normal bookkeeping unlink() does. And if you do a normal unlink(), and then "really quickly" (enforced with a lock) create the symlink without letting anything mess in between, well, how could anyone blame you? :)
|
|
| May 4 at 21:31 | history | became hot network question | |||
| May 4 at 14:42 | vote | accept | Marcus Müller | ||
| May 4 at 13:42 | answer | added | Chester Gillon | timeline score: 18 | |
| May 4 at 13:42 | comment | added | Marcus Müller | @Stewart while that is true, some committees within IEEE do actually offer an open ear to external parties. In this case, I'm not even sure IEEE is actually the party that standardizes. Compare C++: while C++23 is standardized ISO/IEC 14882:2024, C++'s evolution happens in the C++ WGs, their processes are quite out there in the open. | |
| May 4 at 13:27 | history | asked | Marcus Müller | CC BY-SA 4.0 |