1758753 - Migrate XPCOM hashtable documentation from MDN archive to Firefox Source Docs

Status: RESOLVED FIXED

(Core :: XPCOM, task)

Description

[Mike Conley (:mconley) (:⚙️)]

The using_cxx_in_firefox_code.rst documentation links to the now broken MDN documentation for XPCOM hashtables in multiple places.

I retrieved the hashtable documents from the MDN archive, used pandoc to convert them to rst, and tried to scrub out outdated information that I was aware of. Hopefully someone who uses our hashtables more often and knows about how they work deeply under the hood can take a look and make sure I didn't miss anything.

Original documents:

• https://raw.githubusercontent.com/mdn/archived-content/main/files/en-us/mozilla/tech/xpcom/guide/hashtables/raw.html
• https://raw.githubusercontent.com/mdn/archived-content/main/files/en-us/mozilla/tech/xpcom/guide/hashtables_detailed/raw.html

Comment 1

[Mike Conley (:mconley) (:⚙️)]

Attachment: Bug 1758753 - Migrate XPCOM hashtable documentation from MDN to Firefox Source Docs. r?nika!

Comment 2

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/452423a65ba6
Migrate XPCOM hashtable documentation from MDN to Firefox Source Docs. r=nika

Comment 3

[Sandor Molnar[:smolnar]]

https://hg.mozilla.org/mozilla-central/rev/452423a65ba6

Comment 4

[Mathew Hodson]

It looks like you accidentally introduced a change to Document.cpp when you landed this.

--- a/dom/base/Document.cpp
+++ b/dom/base/Document.cpp
@@ -14585,17 +14585,17 @@ void Document::GetWireframeWithoutFlushi
   }

   auto& wireframe = aWireframe.SetValue();
   wireframe.mCanvasBackground = shell->GetCanvasBackground();

   FrameForPointOptions options;
   options.mBits += FrameForPointOption::IgnoreCrossDoc;
   options.mBits += FrameForPointOption::IgnorePaintSuppression;
-  options.mBits += FrameForPointOption::OnlyVisible;
+  // options.mBits += FrameForPointOption::OnlyVisible;

   AutoTArray<nsIFrame*, 32> frames;
   const RelativeTo relativeTo{rootFrame, mozilla::ViewportType::Layout};
   nsLayoutUtils::GetFramesForArea(relativeTo, pc->GetVisibleArea(), frames,
                                   options);

   // TODO(emilio): We could rewrite hit testing to return nsDisplayItem*s or
   // something perhaps, but seems hard / like it'd involve at least some extra

Comment 5

[Mike Conley (:mconley) (:⚙️)]

Hi! Wow, yes, good catch! Thank you!

Comment 6

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/mozilla-central/rev/432bf7d2d4ec
reverted commented line a=backout