659743 - Remove nsImageMapUtils

Status: VERIFIED FIXED

(Core :: DOM: Core & HTML, defect)

Description

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

Attachment: Patch v1

Comment 1

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

Attachment: Part c: Cleanup in nsImageMap

And while here, I noticed some not-so-useful QI'ing

Comment 2

[Olli Pettay [:smaug][[e-mail]]]

Comment on attachment 535181 [details] [diff] [review]
Part c: Cleanup in nsImageMap

>-class nsImageMap : public nsStubMutationObserver, public nsIDOMFocusListener
>+class nsImageMap : public nsStubMutationObserver
>+                 , public nsIDOMFocusListener
, should be in the previous line

Comment 3

[Olli Pettay [:smaug][[e-mail]]]

Comment on attachment 535164 [details] [diff] [review]
Patch v1

>+Element*
>+nsDocument::FindImageMap(const nsAString& aUsemap)
>+{
>+  // We used to strip the whitespace from the usemap value as a Quirk,
>+  // but it was too quirky and didn't really work correctly - see bug
>+  // 87050
>+
>+  if (aUsemap.IsEmpty())
>+    return NULL;

if (aUsemap.IsEmpty()) {
  return nsnull;
}

>+
>+  nsAString::const_iterator start, end;
>+  aUsemap.BeginReading(start);
>+  aUsemap.EndReading(end);
>+
>+  PRInt32 hash = aUsemap.FindChar('#');
>+  if (hash < 0) {
>+    return NULL;
nsnull

>+  }
>+  // aUsemap contains a '#', set start to point right after the '#'
>+  start.advance(hash + 1);
>+
>+  if (start == end) {
>+    return NULL; // aUsemap == "#"
nsnull


>+  }
>+
>+  const nsAString& mapName = Substring(start, end);
>+
>+  if (!mImageMaps) {
>+    mImageMaps = new nsContentList(this, kNameSpaceID_XHTML, nsGkAtoms::map, nsGkAtoms::map);
>+  }
>+
>+  PRUint32 i, n = mImageMaps->Length(PR_TRUE);
>+  for (i = 0; i < n; ++i) {
>+    nsIContent* map = mImageMaps->GetNodeAt(i);
>+    if (map->AttrValueIs(kNameSpaceID_None, nsGkAtoms::id, mapName,
>+                         eCaseMatters) ||
>+        map->AttrValueIs(kNameSpaceID_None, nsGkAtoms::name, mapName,
>+                         eIgnoreCase)) {
>+      return map->AsElement();
>+    }
>+  }
>+
>+  return NULL;
>+}
In the new code all the documents are handled the same way


>-  nsCOMPtr<nsIHTMLDocument> htmlDoc(do_QueryInterface(aDocument));
>-  if (htmlDoc) {
>-    nsCOMPtr<nsIDOMHTMLMapElement> map =
>-      do_QueryInterface(htmlDoc->GetImageMap(usemap));
>-    return map.forget();
>-  } else {
>-    // For XHTML elements embedded in non-XHTML documents we get the
>-    // map by id since XHTML requires that where a "name" attribute
>-    // was used in HTML 4.01, the "id" attribute must be used in
>-    // XHTML. The attribute "name" is officially deprecated.  This
>-    // simplifies our life becase we can simply get the map with
>-    // getElementById().
>-    Element* element = aDocument->GetElementById(usemap);
>-
>-    if (element) {
>-      nsIDOMHTMLMapElement* map;
>-      CallQueryInterface(element, &map);
>-      return map;
>-    }
>-  }
But here we have different code for non-HTML documents

Why such change? The change to the behavior should be done in a different bug, and then the patch should contain
tests.

Comment 4

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

Attachment: Part a: Support image maps according to spec in non-HTMLDocuments

Comment 5

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

Attachment: Part b: Remove nsImageMapUtils

Comment 6

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

http://hg.mozilla.org/mozilla-central/rev/7a7aea19dd85
http://hg.mozilla.org/mozilla-central/rev/54eecb1a8614
http://hg.mozilla.org/mozilla-central/rev/bef2d3989289

Comment 7

[Eric Shepherd [:sheppy]]

This looks like it's really just a matter of noting that this went away for core app developers, and oughtn't to affect web developers at all. True?

Comment 8

[Eric Shepherd [:sheppy]]

This is a pretty arcane case, and after talking to Ms2ger about it in IRC, we decided it's not worth documenting.

Comment 9

[Ioana (away)]

content/html/content/test/test_bug659743.xml passed on all the OSs:
https://tbpl.mozilla.org/php/getParsedLog.php?id=6420477&full=1
https://tbpl.mozilla.org/php/getParsedLog.php?id=6420362&full=1
https://tbpl.mozilla.org/php/getParsedLog.php?id=6420285&full=1
https://tbpl.mozilla.org/php/getParsedLog.php?id=6420482&full=1