On 24/02/15 17:12, Alexander Surkov wrote: > Thanks, Pete. > > Hi, Dominic. Are there any consumers for relation change events? From > what I can tell I was never requested for it by AT vendor iirc. > > ATK has generic property changed event and I find it nice,
FWIW, this generic property changed event doesn't include relations. In fact there is a bug open on atk to add a signal when the relation set (and the attribute set) changes: https://bugzilla.gnome.org/show_bug.cgi?id=649771 That has a half-made patch, included on the experimental atk3 patch. But probably it could be added on the current atk2/at-spi2. > but MSAA has separate events for most of properties already [1], and I > don't see a good way to implement it in IAccessible2 because of > Windows event model. IAccessible2 has text changed event by > newText/oldText hack but it doesn't really work for out process AT. > > We might have OBJECT_CHANGED event, but SHOW/HIDE bundle looks like > appropriate workaround. I'm curious though to hear AT vendors on it. > > Thanks. > Alex. > > [1] > https://msdn.microsoft.com/en-us/library/windows/desktop/dd318066%28v=vs.85%29.aspx > > > On Mon, Feb 23, 2015 at 11:51 AM, Pete Brunet <[email protected] > <mailto:[email protected]>> wrote: > > Alex, et al. FYI > > On 2/23/15 10:29 AM, [email protected] > <mailto:[email protected]> wrote: >> https://bugs.linuxfoundation.org/show_bug.cgi?id=1263 >> >> Bug #: 1263 >> Summary: Need event to fire when a relation is added/removed >> Product: Accessibility >> Version: rc >> Platform: All >> OS/Version: All >> Status: NEW >> Severity: normal >> Priority: P2 >> Component: IAccessible2 >> AssignedTo: [email protected] <mailto:[email protected]> >> ReportedBy: [email protected] <mailto:[email protected]> >> Classification: Unclassified >> >> >> I work on Google Chrome's accessibility support. There doesn't seem to >> be a >> good event to fire on Windows when an IAccessible2 object's relations >> change, >> i.e. when IAccessible2::relations() would return a different answer than >> before >> - assuming that no other property of the object happens to change. >> >> There might be a tiny handful of other object properties that could >> change that >> don't have a specific associated event, it might be worth auditing the >> API to >> see. >> >> The remaining cases are somewhat obscure, so I wonder if instead or in >> addition >> it might make sense to have an event meaning that *any* property of an >> object >> changed without specifying which one. As another use for this, if lots of >> properties of an object change at once (name, description, role, etc.) a >> server >> could fire just one event instead of multiple. >> > > -- > *Pete Brunet* > > a11ysoft - Accessibility Architecture and Development > (512) 467-4706 <tel:%28512%29%20467-4706> (work), (512) 689-4155 > <tel:%28512%29%20689-4155> (cell) > Skype: pete.brunet > IM: ptbrunet (AOL, Google), [email protected] > <mailto:[email protected]> (MSN) > http://www.a11ysoft.com/about/ > Ionosphere: WS4G > > _______________________________________________ > Accessibility-ia2 mailing list > [email protected] > <mailto:[email protected]> > https://lists.linuxfoundation.org/mailman/listinfo/accessibility-ia2 > > > > > _______________________________________________ > Accessibility-ia2 mailing list > [email protected] > https://lists.linuxfoundation.org/mailman/listinfo/accessibility-ia2 -- Alejandro Piñeiro ([email protected])
_______________________________________________ Accessibility-ia2 mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/accessibility-ia2
