From The Linux Foundation
Revision as of 20:41, 6 August 2007 by Ptbrunet
Agenda for 2007/08/07
- Introductions if needed
- LF: Open A11y news - Janina
- Version 2 of text attributes
- Please review version 2 of the attribute spec. There may be some attributes that can be eliminated thus cutting down on the workload for implementers.
- The suggestions to date are:
- Remove line-height-at-least.
- Remove num-format, num-prefix, and num-suffix and replace with a new attribute named "prefix" where the value is the actual number (with any leading or trailing punctuation such as a trailing period and/or leading and/or trailing parens), or a bullet character, or the word "graphic".
- Addition of IAText::defaultAttributes - raised by Aaron
- In the process of adding text attribute support for FF3, the FF3 a11y team noticed that ATK has a method for returning default text attributes but IA2 does not.
- ATK behavior:
- IAText::attributes returns all the attributes.
- It has been suggested that similar methods should be implemented in IA2.
- It should be noted that this issue was not raised by the AT or app developers during the implementation of IA2 in Notes 8. Was the Linux design driven by the performance issue related to Linux using an out of process architecture?
- Note that ATK also has an in parameter on the getter for attributes and the getter for default attributes which allows the caller to specify a subset of the attributes so the app doesn't have to gather up and present all the attributes each time.
- Since IAText::attributes is already in use its behavior can't be changed. For the next release of IA2 should it be deprecated in favor of IAText::defaultAttributes and IAText::changedAttributes?
- Inspect tool - Mike Squillace
- Firefox - Aaron Leventhal
No progress on the following
- IDL submission to BZR.
- There have been no changes since the March 26 version.
- Prior to publishing it on the main page the 3/26 version is available via bug 111.
- Eclipse - Barry Feigenbaum
- Check back in August
- Python wrappers
- Waiting for IBM legal approval to release as BSD.
- Documentation of memory management issues for [out] parameters - the following need to be understood and documented.
- Here is a rough first draft:
- BSTRs need to be SysFreeString'd at end of life; ditto for when they are in structs or arrays.
- COM interfaces need to be AddRef'd when used and Release'd at end of life.
- Single longs, HWNDs, booleans, and structs are allocated by the caller and passed by reference. The marshaller does all the memory management. (I don't believe there are any structs with BSTRs or interface pointers.)
- Arrays of longs - the server allocates the top level array and anything it points to; nothing needs to be free'd by the client. The marshalling code handles freeing anything on the client side.
- TBD: How are VARIANTs handled? Like interfaces? These methods return VARIANTs: IAHyperlink::anchor/anchorTarget, IAValue::current/maximum/minimumValue
- TBD: Are arrays of IUnknowns, e.g. IARelation::targets, correctly specified as **IUnknown?
- This Don Box article can be used for understanding: http://www.microsoft.com/msj/1196/activex1196.aspx
- Here is a rough first draft:
- Please contribute and/or review.
- Developer Guide
- No new status on when it will be back in the SDK.
- BZR patch submission process
- Enter defects into Bugzilla
- What else needs to be discussed?
- What IA2 information do you need?
- What IA2 issues need resolution soon? Target date?
- What would you like to see done?