The Documentation Workgroup will be holding its next meeting on Monday, November 18th, 2024 from 8:00am to 9:00am PT (or see the time in your own timezone).
This meeting will be open to anyone who wishes to contribute. If you wish to participate, please reach out to @swift-documentation-workgroup in forums via DM for a link to the WebEx meeting.
Meeting notes for this meeting will posted in this thread shortly after.
Please propose agenda items in this thread.
Joe: symbolgraph expectations - naming of the files, deriving the module name, and the idea of primary vs. extension symbol graphs
(Sven) Docc6 slow rendering follow-up: New DocC Renderer environment variable hack for SPI is working nicely. Would be nice to know when we can clean that up (fixed release)
(Joe) Is there a naming convention for the symbol graphs as they are created? Is the content for symbolgraph untrustyworthy such that the module name should come from the name of the graph rather than the contents.
(Franklin) Main - swift compiler generates the symbols there, and the extensions came through from ObjC.
(David) The primary symbol graph file is always readable by itself, where an extension may not have all the pieces it needs (relationships referring to target symbols from the extended module)
(Franklin) Perhaps you can call out the core issue you're hitting and we can talk it through?
(Joe) The code in symbolkit has a current expectation that it's pulling the name from the file name of the graph in some cases, but not always, and I wasn't sure what the background was there.
( referring to proposed changes in this PR for SymbolKit )
(David) I recall that there's some quirk with the module name, but not what the quirk was. Perhaps that the module name data isn't the name you'd expect.
(Joe) Sounds good - probably best to wait for the formal review of the PR with a fix.
(David) Just did a quick check of the module name field - In an extension symbolgraph file the module name is the extending module instead of the extended module. In other words; First@Second.symbols.json has "First" as the module name in the symbol graph data.