Tree-sitter rolling fixes, 1.123 edition #1118
Conversation
…where it got flummoxed by URLs with subdomains.
There was a problem hiding this comment.
Pretty much a rubber-stamping this PR.
If I knew a lot more I could be more insightful. But speaking as a reviewer: Short and small change sets are generally good!
Didn't see anything spooky or objectionable, not that I would know, TBH, but there's less diff to inspect and in which to miss anything, so I feel maybe more confident in that than usual. At least.
Anyway, caveats aside, thanks for keeping up maintenance of these grammars and such!
There was a problem hiding this comment.
Seems legit. I do a bit of parameter expansion here and there myself.
Docs for reference: https://www.gnu.org/software/bash/manual/html_node/Shell-Parameter-Expansion.html
There was a problem hiding this comment.
Bumping the parser seems probably desirable I assume 👍
Linking the diff in the actual parser package repo, in case anyone better at reading this stuff (or just curious) wants to read through it: savetheclocktower/tree-sitter-hyperlink@04c3a66...0704b3e
| ; The "foo" in `int &foo` with in a parameter list. | ||
| (parameter_declaration | ||
| declarator: (reference_declarator | ||
| (identifier) @variable.parameter.cpp)) | ||
| ; The "foo" in `const char *foo` within a parameter list. | ||
| ; (Should work no matter how many pointers deep we are.) | ||
| (reference_declarator (identifier) @variable.parameter.cpp | ||
| (#is? test.descendantOfType "parameter_declaration")) |
There was a problem hiding this comment.
Wish I understood the syntax better but haven't achieved that enlightenment yet. (Both for C itself and for tree-sitter grammars.)
But... presumably this is good.
There was a problem hiding this comment.
“When a reference_declarator node has a child named identifier, apply the variable.parameter.cpp scope name to the identifier node… but only if identifier is a descendant of parameter_declaration.”
Tree-sitter has an annoying oversight in its query syntax: you can't directly query ancestor/descendant relationships — only parent/child relationships. A parameter_declaration like int foo can have an identifier as a child…
int main (int foo) {…or it can have a reference_declarator as a child, which itself has an identifier child…
int main (int *foo) {…or it can be doubly-nested in reference_declarators and be an extra layer deep…
int main (int **foo) {…and, though this is probably as deep as it gets, at this point one looks around for a way to match all identifiers anywhere underneath all parameter_declaration nodes — instead of having to handle these three cases in three separate queries.
The test.descendantOfType query predicate allows us to do that. It traverses upward from identifier and returns true as soon as it finds a node of type parameter_declaration, or false if it gets to the root without finding one. This would be pretty costly — executing against all identifiers, and failing most of the time! — except that a reference_declarator node is pretty much guaranteed to exist inside of a parameter_declaration, so the initial query winnows out nearly all of the identifiers that would fail this test.
|
Indeed the one "Editor Tests" failing spec is known problematic across multiple unrelated PRs at the moment, so it's not this PR's fault, so safe to ignore with regard to merging this particular PR or not. |
Diligence in bug reporting means I have two new small fixes even though we just released! More to come, I'm sure.