98704 - Node::getXPathParent is needed all over the place, ContextState::getParentNode should die

Status: VERIFIED FIXED

(Core :: XSLT, defect, P1)

Description

[Axel Hecht]

This bug is a split off from bug 76070.
The upcoming patch implements Node::getXPathParent, does the right thing to
source/xml/util/DOMHelper, and fixes the whitespace mess in
NodeDefinition::getBaseURI (just as I go).

Axel

Comment 1

[Axel Hecht]

Attachment: getXPathParent.patch

Comment 2

[Peter Van der Beken [:peterv]]

Comment on attachment 48577 [details] [diff] [review]
getXPathParent.patch

r=peterv, now on to killing ProcessorState::getParentNode

Comment 3

[Brendan Eich [:brendan]]

Comment on attachment 48577 [details] [diff] [review]
getXPathParent.patch

rs=[e-mail]

Comment 4

[Axel Hecht]

patch 48577 is in, leaving this bug open and morping it to kill
ContextState::getParentNode (which is ProcessorState::getParentNode).

Axel

Comment 5

[Axel Hecht]

Attachment: second patch, kill ProcessorState::getParentNode alltogether

Comment 6

[Axel Hecht]

added
 MBool LocationStep::matches(Node* node, Node* context, ContextState* cs) {
 
-    if ( !nodeExpr ) return MB_FALSE;
+    if (!nodeExpr || !node)
+        return MB_FALSE;
 
-    if ( !nodeExpr->matches(node, context, cs) ) return MB_FALSE;
+    if (!nodeExpr->matches(node, context, cs))
+        return MB_FALSE;

to make ::matches 0 crash safe. Thanx to sicking for catching this.
(the second one is just whity fun)

Axel

Comment 7

[Jonas Sicking (:sicking) No longer reading bugmail consistently]

Comment on attachment 49080 [details] [diff] [review]
second patch, kill ProcessorState::getParentNode alltogether

r=sicking

Comment 8

[Peter Van der Beken [:peterv]]

Comment on attachment 49080 [details] [diff] [review]
second patch, kill ProcessorState::getParentNode alltogether

r=peterv

Comment 9

[Johnny Stenback (:jst)]

Comment on attachment 49080 [details] [diff] [review]
second patch, kill ProcessorState::getParentNode alltogether

sr=jst

Comment 10

[Axel Hecht]

I forgot to note in the first landing that this is code by sicking, sorry for
that.

second stage landed, marking fixed.

Axel

Comment 11

[Jonas Sicking (:sicking) No longer reading bugmail consistently]

verified