Summary: | RotateProperties has an array of 'atoms' where the spec says 'properties' | ||
---|---|---|---|
Product: | XCB | Reporter: | Robert Ancell <robert.ancell> |
Component: | Protocol | Assignee: | xcb mailing list dummy <xcb> |
Status: | RESOLVED WONTFIX | QA Contact: | xcb mailing list dummy <xcb> |
Severity: | normal | ||
Priority: | medium | ||
Version: | unspecified | ||
Hardware: | Other | ||
OS: | All | ||
Whiteboard: | |||
i915 platform: | i915 features: |
Description
Robert Ancell
2011-12-07 15:06:11 UTC
> --- a/src/xproto.xml
> +++ b/src/xproto.xml
> @@ -2351,9 +2354,9 @@ authorization from the authors.
> <request name="RotateProperties" opcode="114">
> <pad bytes="1" />
> <field type="WINDOW" name="window" />
> - <field type="CARD16" name="atoms_len" />
> + <field type="CARD16" name="properties_len" />
> <field type="INT16" name="delta" />
> - <list type="ATOM" name="atoms">
> + <list type="ATOM" name="properties">
> <fieldref>atoms_len</fieldref>
> </list>
> </request>
>
does the name actually matter for anything?
the change breaks struct xcb_rotate_properties_request_t API, for no
significant gain afaict.
It doesn't gain anything really. I was just comparing some existing code I had to XCB generated code and filed bugs for all the differences that seemed valid. Feel free to close WONTFIX. On 12/08/11 17:37, Josh Triplett wrote: > On Thu, Dec 08, 2011 at 04:18:23PM -0800, Bart Massey wrote: >> One of the things we do try to enforce is that the names in XCB match >> the names in the spec (where there is one :-) to support just the kind >> of case you describe. On the other hand, I hate to break existing >> code. >> >> Suggestions appreciated; I hate to just let this one die if we could >> figure out something better. > > I can think of one possible solution, if we don't mind using anonymous > union support. Just define the relevant field as an anonymous union, > containing both field names. That would preserve both API and ABI. Unfortunately, anonymous unions are a gcc extension that are not universally supported yet (the next generation ISO C standard, C1X, is supposed to include them, so someday we'll get there). Closing a WONTFIX since this would break compatibility for no gain. |
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.