specdb Xt defintions confused
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
lsb |
In Progress
|
Medium
|
Unassigned | ||
Mandriva |
In Progress
|
Medium
|
Bug Description
There's probably more than this mess, but this has been specifically
causing problems with devchk. Recorded here so there's a record of it
somwhere.
INSERT INTO `Type` VALUES
(9576,'
INSERT INTO `Type` VALUES (9577,'_WidgetRec
*','Pointer'
INSERT INTO `Type` VALUES
(9581,'
INSERT INTO `Type` VALUES (9582,'
*','Pointer'
INSERT INTO `Type` VALUES
(31896,
INSERT INTO `Type` VALUES (31897,'_WidgetRec
*','Pointer'
INSERT INTO `Type` VALUES
(31946,
INSERT INTO `Type` VALUES (31947,
*','Pointer'
INSERT INTO `Type` VALUES
(35938,
INSERT INTO `Type` VALUES (35940,'_WidgetRec
*','Pointer'
INSERT INTO `Type` VALUES
(35937,
INSERT INTO `Type` VALUES (35939,
*','Pointer'
The widget-related base definitions are repeated three times in the
database. Usage is scattered all over the place. For example, the
derived types WidgetRec, WidgetClassRec, CoreRec, CoreClassRec were,
based on assigned type numbers, added about the time of the third
set, but their ArchType entries make them derived from the first set.
Meanwhile, it's the second set (structs _WidgetRec and _WidgetClassRec)
that are marked as included. Only the first set has struct members
defined, so we're getting "can't take sizeof" errors from devchk from
some of the derived types which reference other than the first set.
Not quite sure how to clean up this mess yet.
Changed in mandriva: | |
importance: | Unknown → Medium |
status: | Unknown → In Progress |
tags: | added: spec |
tags: | added: zpatch |
tags: |
added: zclose removed: zpatch |