feat(gdscript): add .gd extraction support (#697) - #1836
Open
TPAteeq wants to merge 1 commit into
Open
Conversation
|
Hey, could you also include godot resource parsing ( |
ozdemirsarman
added a commit
to ozdemirsarman/graphify
that referenced
this pull request
Jul 26, 2026
Address review (Graphify-Labs#1929): depend on tree-sitter-language-pack>=1.12.5,<1.13 (as Graphify-Labs#1836 does) instead of the unpublished tree-sitter-gdscript wheel. The .gd extractor loads the bundled gdscript grammar; the scene/resource extractor now uses the bundled godot_resource grammar, falling back to a dependency-free line parser when the extra is absent.
luevano
added a commit
to luevano/graphify
that referenced
this pull request
Jul 27, 2026
luevano
pushed a commit
to luevano/graphify
that referenced
this pull request
Jul 27, 2026
Address review (Graphify-Labs#1929): depend on tree-sitter-language-pack>=1.12.5,<1.13 (as Graphify-Labs#1836 does) instead of the unpublished tree-sitter-gdscript wheel. The .gd extractor loads the bundled gdscript grammar; the scene/resource extractor now uses the bundled godot_resource grammar, falling back to a dependency-free line parser when the extra is absent.
luevano
added a commit
to luevano/graphify
that referenced
this pull request
Jul 27, 2026
luevano
added a commit
to luevano/graphify
that referenced
this pull request
Jul 27, 2026
…namespace Complete the godot_scene -> godot_resource rename across the module, its test and the registry (Graphify-Labs#1836 parity), and fix three id-correctness bugs in the Godot extractors. res:// resolution ended in .resolve(), so cross-file targets - autoload calls, preload/extends, and every ext_resource in a .tscn - were built from the on-disk path while every other id derives from the scan-relative path the walker passes in. extract()'s Graphify-Labs#502 remap cannot repair those: it re-derives ids from the emitting file's source_file, which for a cross-file target is the wrong file, so the edge dangles for good. The resolved path is only ever used to build ids and labels, never for filesystem access, so it now stays relative. Neither extractor filtered engine API, so Godot types, global functions and Object/Node methods became graph nodes accumulating edges from every call site. Bare calls need the method set too, since implicit-self calls like add_child(x) are idiomatic GDScript. extends edges to engine types are deliberately kept: one per script rather than one per call site, so they cannot accumulate into a god node, and which scripts are CharacterBody3D vs Control is real structure. handle_attribute_call took the first identifier of an attribute chain as the receiver, so PlayerProfile.library.set_active_skin() resolved into player_profile.gd, which defines no such function. Only a direct Autoload.method() resolves now. On a 77-file Godot project this takes AST dangling edge endpoints from 118 to 0, and the graph from 2026 nodes / 3611 edges to 1484 / 2460. Both test modules built their expected ids with .resolve(), the same absolutizing step at fault, so they would have gone green over a broken extractor; they now derive expectations from the extractor's own output. Adds regression tests for absolute-path leaks, engine-method phantoms and chained attribute calls.
luevano
pushed a commit
to luevano/graphify
that referenced
this pull request
Jul 27, 2026
Address review (Graphify-Labs#1929): depend on tree-sitter-language-pack>=1.12.5,<1.13 (as Graphify-Labs#1836 does) instead of the unpublished tree-sitter-gdscript wheel. The .gd extractor loads the bundled gdscript grammar; the scene/resource extractor now uses the bundled godot_resource grammar, falling back to a dependency-free line parser when the extra is absent.
luevano
added a commit
to luevano/graphify
that referenced
this pull request
Jul 27, 2026
…namespace Complete the godot_scene -> godot_resource rename across the module, its test and the registry (Graphify-Labs#1836 parity), and fix three id-correctness bugs in the Godot extractors. res:// resolution ended in .resolve(), so cross-file targets - autoload calls, preload/extends, and every ext_resource in a .tscn - were built from the on-disk path while every other id derives from the scan-relative path the walker passes in. extract()'s Graphify-Labs#502 remap cannot repair those: it re-derives ids from the emitting file's source_file, which for a cross-file target is the wrong file, so the edge dangles for good. The resolved path is only ever used to build ids and labels, never for filesystem access, so it now stays relative. Neither extractor filtered engine API, so Godot types, global functions and Object/Node methods became graph nodes accumulating edges from every call site. Bare calls need the method set too, since implicit-self calls like add_child(x) are idiomatic GDScript. extends edges to engine types are deliberately kept: one per script rather than one per call site, so they cannot accumulate into a god node, and which scripts are CharacterBody3D vs Control is real structure. handle_attribute_call took the first identifier of an attribute chain as the receiver, so PlayerProfile.library.set_active_skin() resolved into player_profile.gd, which defines no such function. Only a direct Autoload.method() resolves now. On a 77-file Godot project this takes AST dangling edge endpoints from 118 to 0, and the graph from 2026 nodes / 3611 edges to 1484 / 2460. Both test modules built their expected ids with .resolve(), the same absolutizing step at fault, so they would have gone green over a broken extractor; they now derive expectations from the extractor's own output. Adds regression tests for absolute-path leaks, engine-method phantoms and chained attribute calls.
ozdemirsarman
added a commit
to ozdemirsarman/graphify
that referenced
this pull request
Jul 27, 2026
…yze/build/MIGRATION, README + fixtures (Graphify-Labs#1929) - Register .gd (gdscript) and .tscn/.tres/project.godot (godot_resource) in analyze.py, build.py and extractors/MIGRATION.md, matching Graphify-Labs#1836. - Rename the scene extractor to godot_resource (module, extract_godot_resource, registry key, wiring) per review. - Shrink the pyproject godot comment to one line. - README: add a godot extras row and split the godot file types into their own GDScript / Godot resources rows. - Move the inline gdscript/godot test bodies into tests/fixtures/godot/ and rename the test file to test_godot_resource.py.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Closes #697
Summary
.gdfiles.graphifyy[gdscript]dependency using a tree-sitter-compatible language-pack release.Extraction coverage
res://inheritanceType.new()instantiationspreload()andload()dependenciesGodot built-ins are filtered to avoid noisy graph hubs, ambiguous receiver calls are not guessed, and unresolved type stubs use a reserved namespace to prevent
collisions with file nodes.
Validation
uv run pytest -q— 3,119 passed, 30 skippedgraphify update .completed successfully