This reverts commit 0d9dcb161f.
This turns out to introduce a subtle bug related to destruction order
between `Log::instance()` and the `Logger` instances.
New `DIAGNOSE_ERROR` and `DIAGNOSE_CRITICAL` macros are added.
These accept an ID which should indicate a diagnostic source via
a function definition in `codeql::diagnostics`, together with the usual
format + arguments accepted by other `LOG_*` macros.
When the log is flushed, these special logs will result in an error JSON
diagnostic entry in the database.
* visiting now happens in a later stage than fetching labels. While
fetching a list of entities to be visited is created, and then acted
upon in actual extraction. This partially flattens the recursive
nature of `fetchLabel` into a loop inside `SwiftVisitor::extract`.
Recursion in `fetchLabel` will only happen on labels fetched while
naming an entity (calling into `SwiftMangler`).
* The choice whether to name a declaration or type has been moved from
the translators to `SwiftMangler`. Acting on this choice is contained
in `SwiftDispatcher::createLabel`.
* The choice whether to emit a body of a declaration has been moved from
`DeclTranslator` to the dispatcher. This choice is also contained in
`SwiftDispatcher::createLabel`.
* The simple functionality of the `LabelStore` has been moved to the
`SwiftDispatcher` as well.
This also removes the TODO about using `absl::StrJoin` to dump the environment because we can't easily get a range from a null-terminated `envp`. It also doesn't suffer from the usual awkwardness around inserting a separator *between* elements but not after the last one, so a for loop is clear enough.
It's not much cleaner due to arithmetic to convert truncating division to a ceiling, but has two advantages:
1. It doesn't suffer from rounding issues with large TRAP labels. This is largely theoretical, but does let us handle `undefined` uniformly.
2. It should be much faster (using LZCNT/BSR instead of floating point arithmetic). This is probably not a performance bottleneck, so *shrug*.
It turns out mangled names can sometimes be too long. While this code
will eventually be replaced by our own mangling, we need to use hashing
to cut down the names.
Module and decl names are preserved in the trap file names for
debuggability.
In case the extractor is run in isolation for debugging/testing, this
will avoid littering the current working directory with artifacts, and
instead having a single `extractor-out` directory to inspect or clean
up.
Also extractor logs have been nested into a `swift` directory, as the
log directory provided by the `codeql` cli is actually shared between
languages.
Assertions before fetching a non optional label are not needed as
the dispatcher will replace those with unspecified elements (and
properly log those instances).
