FrameTreeNodeID outside content: The name of the getter

[Rob Wu]

The value of FrameTreeNodeID will be exposed outside of content [1], because it is the only stable ID that can identify a frame.

Charlie & Nasko requested to expose this ID via a name that is different from FrameTreeNodeID (FTN ID), because FTNs are a concept that is internal to content/ [2].
I considered different names (e.g. UniqueFrameId, ContentFrameId, GlobalFrameId, StableFrameId), but ended up with sticking to FrameTreeNodeID because:

- FTN ID is already leaked outside of content/ (e.g. [3])

- Having another name that means exactly the same is confusing.

- If properly documented, the name does not matter that much, as long as the meaning is unambiguous and obvious. My alternative names are not better in this regard.

- I'm using "FrameTreeNodeID" instead of "FrameTreeNodeId" (camelCase is preferred these days) because all IDs in RenderFrameHost end with "ID" instead of "Id", and I preferred local consistency over the global coding standard.

content/ and extensions/ interfaces via the FTN ID through a single file, so if something better comes up than a FTN ID, it can easily be swapped.

Do you all agree with exposing the ID as GetFrameTreeNodeID, or do you have a strong argument for something else?

Kind regards,

 Rob

 https://robwu.nl

[1] https://codereview.chromium.org/1413543005/#msg28
[2] https://codereview.chromium.org/1413543005/diff/80001/content/public/browser/render_frame_host.h#newcode67

[3] https://cs.chromium.org/-file%3Acontent%20frame_tree_node_id
[4] https://codereview.chromium.org/1413543005/#ps180001

[Nick Carter]

Go with GetFrameTreeNodeId

All your arguments are pretty convincing. And I polled Nasko and Charlie, and nobody has a better idea.

Whether to capitalize the terminal 'D': the rule is to use Id style in "new or modified" code; I think this counts as a sufficient modification to use the new style, so go with the lower case approach. If you wish to draft a patch updating the existing capitalized FrameTreeNodeID cases so that we're consistent, I would lgtm it -- there are not that many: https://code.google.com/p/chromium/codesearch#search/&q=file:content%20FrameTreeNodeID%20case:yes&sq=package:chromium&type=cs.