Documentation fixes #4954
Documentation fixes #4954
Conversation
| @@ -59,7 +59,7 @@ typedef __haiku_std_int64 git_time_t; | |||
| * app, even though /we/ define _FILE_OFFSET_BITS=64. | |||
| */ | |||
| typedef int64_t git_off_t; | |||
| typedef int64_t git_time_t; | |||
| typedef int64_t git_time_t; /**< time in seconds from epoch */ | |||
There was a problem hiding this comment.
Didn't we have some problems correctly rendering those inline comments? Not quite sure whether I remember correctly
There was a problem hiding this comment.
I think I had this fixed on my tweaked docurium, but it's currently complaing (loudly) about /var/folders/f4/b3wl780s12gfc25kpvl1rhk40000gn/T/HEAD-20190128-84175-yftmsd/git2/common.h:10:10: fatal error: 'time.h' file not found, which I'll have to investigate.
There was a problem hiding this comment.
So, I can confirm that vanilla & tweaked docurium completely ignores "primitive" typedefs. I can tweak it so it doesn't (modulo more fixes to grab the correct comment), but it means I'll pickup all "primitive" typedefs (like our size_t DOCURIUM override, or the typedef goop inside our Win32 inttypes.h)
|
Two questions about our documentation that ends up rendered on our website:
|
|
AFAIU, https://libgit2.org/libgit2/ comes from docurium (aka. this repo's gh-pages), which I run sometimes and was planning to do soon™. The rest of https://libgit2.org comes from https://libgit2.github.com/libgit2.github.com (which has a few PR). Right now it's completely manual, and though I'm making steps teaching docurium NOT to regenerate every version for every run, I don't think it's easily automatable (at least for the time being). I do have a local Rakefile to quickly spin up updated docs (ie. |
fcd7a2e to
690e55e
Compare
|
Woo. You can now download a zip of the docs from the PR validation build to check it out locally. 👀😀 |
| * @param time time when the action happened | ||
| * @param offset timezone offset in minutes for the time | ||
| * @param time time (in seconds from epoch) when the action happened | ||
| * @param offset timezone offset (in minutes) for the time |
There was a problem hiding this comment.
Hmm, these aren't minutes exactly, this is minutes and hours in ISO 8601 format. (An hour is 100 not 60.) Would you mind fixing that up while you're here?
There was a problem hiding this comment.
Sorry, invoking ISO8601 just made me 😨. You mean that it's represented as a 4-digit hhmm number ?
There was a problem hiding this comment.
I might be misunderstanding, but it looks like the hhmm decimal format only relates to parsing signatures, not the public API. My "minutes" link above uses difftime (hence seconds) to calculate the offset in minutes, and there's no 100 anywhere.
There was a problem hiding this comment.
Ha, yes, of course you're right.
|
I had a few comments, otherwise 👍 |
|
Thanks @tiennou! |
This should close #4908, attempts at #4783 and #4711, and some API comment touch-ups.