[Github-comments] [geany] doxygen: various doxygen-related fixes in preparation for gtkdoc generation (302b40e)

Thomas Martitz notifications at xxxxx
Fri Dec 18 23:49:46 UTC 2015


Am 18. Dezember 2015 16:49:17 MEZ, schrieb Matthew Brush <notifications at github.com>:
>> That doesn't automatically make everything else strictly private,
>though. [...]. So these "private" members are in fact part of the ABI
>and in practice public.
>
>You keep confusing compiler with convention. When we say "public" or
>"private" we mean "plugins should/shouldn't use this stuff", not
>"plugins can/can't technically use this stuff". Once you grok the
>difference, it will become clear. Basically if you're thinking in terms
>of ABI, you're not getting it :) A better example @b4n eluded to is
>"private" fields/methods in Python. To make something "private" in a
>class, you append/prepend an underscore or some other convention. The
>members are still technically public, but the convention is that if
>there's an underscore like that, they're private and should not be used
>outside of the class, and there's no guarantee they won't change.
>

I'm not confusing these things. My point is that even if no plugin uses the fields you still can't just go and change them without consideration. The members are part of the ABI, that's a fact and can't be argue away with conventions. That is, even well-behaving plugins (following your convention) can be negatively affected if such members are removed or changed in an incompatible way.

>> I didn't actually start to document the members, their status is
>unchanged. I merely documented the *type*
>
>You added new types to the "public" API (remember "public" by
>convention, nothing to do with compiler/ABI).
>

How's this a problem?





---
Reply to this email directly or view it on GitHub:
https://github.com/geany/geany/commit/302b40e9778ff49407902f3ef9f272ed202f4547#commitcomment-15077372
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.geany.org/pipermail/github-comments/attachments/20151218/16ddc6c9/attachment.html>


More information about the Github-comments mailing list