Created attachment 127346 [details] [review]
[PATCH] Bug 569375 – g[u]intptr undocumented

 docs/reference/glib/glib-sections.txt |    4 ++++
 docs/reference/glib/tmpl/types.sgml   |   19 ++++++++++++++++++-
 2 files changed, 22 insertions(+), 1 deletions(-)

you say "any gpointer" in the gintptr doc, but "any pointer" in the guintptr doc. (the latter is probably better?)

Right. I had "gpointer" first and changed one but forgot to change the other occurrence of it.

Created attachment 134929 [details] [review]
updated patch; ok to commit?

Probably should say "C99 type" instead of "standard C"? Many platforms still use just C89 compilers. If we required C99, gintptr and guintptr would not be needed. Otherwise fine with me to commit.

Please commit with the change that Tor recommends.

Committed with s/standard C ... type/C99 type .../.

Now I wonder about the placement of g[u]intptr in this category:

<listitem><para>
Types which correspond exactly to standard C types, but are included
for completeness - #gchar, #gint, #gshort, #glong, #gfloat, #gdouble,
#gintptr, #guintptr.
</para></listitem>

Should another category be added here, for types that are in the C99 std only? (The other existing categories don't fit for g[u]intptr either.)

(In reply to Christian Persch from comment #7)
> Should another category be added here, for types that are in the C99 std
> only? (The other existing categories don't fit for g[u]intptr either.)

Done.

The following fix has been pushed:
5040de9 docs: Mention gsize, goffset, etc. are in C99

Created attachment 361038 [details] [review]
docs: Mention gsize, goffset, etc. are in C99

Signed-off-by: Philip Withnall <withnall@endlessm.com>