Skip to content

fix(parsers): capture PHP imports instead of trait uses #1563

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Shashankss1205 merged 1 commit into from
Aug 2, 2026
Merged

fix(parsers): capture PHP imports instead of trait uses #1563

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
110 changes: 80 additions & 30 deletions src/codegraphcontext/tools/languages/php.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,8 +31,16 @@
name: (name) @name
) @trait
""",
# A top-level `use App\Models\User;` is a `namespace_use_declaration`.
# `use_declaration` is the *trait* use inside a class body (relied on at
# ~line 309), so capturing it here dropped every real import and labelled
# trait uses as imports instead.
#
# Capture the clause rather than the declaration: it is the level that
# exists once per imported symbol, so the braced group form
# `use Foo\{A, B as C};` yields one row per symbol for free.
"imports": """
(use_declaration) @import
(namespace_use_clause) @import
""",
"calls": """
(function_call_expression
Expand Down Expand Up @@ -394,40 +402,82 @@ def _parse_imports(self, captures: list, source_code: str) -> list[dict]:
for node, capture_name in captures:
if capture_name == "import":
try:
import_text = self._get_node_text(node)
# use Foo\Bar as Baz;
# Node usually has children: name (qualified_name), optional alias

name_node = None
alias_node = None

# `namespace_use_clause` shapes, in tree-sitter-php:
# App\Models\User -> qualified_name
# App\Models\Post as Blog -> qualified_name, name(alias)
# A -> name (group member)
# B as C -> name, name(alias) (group member)
# function Bar\helper -> qualified_name, with a
# leading `function`/`const` token
#
# The alias, when present, is always the final `name` child.
path_node = None
alias = None
name_children = []

for child in node.children:
if child.type == "qualified_name" or child.type == "name":
name_node = child
# Alias in PHP: use X as Y; The 'as' is usually implicit structure or explicit?
# Tree sitter grammar: use_declaration -> use_clause -> (use_as_clause (qualified_name) (name))

# Assuming simple handling for now, extracting string from text
# Regex might be safer given tree complexity for `use`
import_match = re.search(r'use\s+([\w\\]+)(?:\s+as\s+(\w+))?', import_text)
if import_match:
import_path = import_match.group(1).strip()
alias = import_match.group(2).strip() if import_match.group(2) else None

import_data = {
"name": import_path,
"full_import_name": import_text,
"line_number": node.start_point[0] + 1,
"alias": alias,
"context": (None, None),
"lang": self.language_name,
"is_dependency": False,
}
imports.append(import_data)
if child.type == "qualified_name":
path_node = child
elif child.type == "name":
name_children.append(child)

if path_node is not None:
# qualified_name carries the path; a trailing bare name
# is therefore the alias.
import_path = self._get_node_text(path_node)
if name_children:
alias = self._get_node_text(name_children[-1])
elif name_children:
# Group member: first name is the symbol, second (if
# any) is its alias.
import_path = self._get_node_text(name_children[0])
if len(name_children) > 1:
alias = self._get_node_text(name_children[-1])
else:
continue

# A braced group prefixes its members: `use Foo\{A, B};`
# means Foo\A and Foo\B. Walk up to recover the prefix so
# the recorded path is the real fully-qualified name.
group = node.parent
if group is not None and group.type == "namespace_use_group":
decl = group.parent
if decl is not None:
# The prefix is a `namespace_name` sibling of the
# group (`use Foo\Bar\{A, B};` -> "Foo\Bar").
prefix_parts = [
self._get_node_text(c)
for c in decl.children
if c.type in ("namespace_name", "qualified_name", "name")
]
if prefix_parts:
import_path = (
prefix_parts[0].rstrip("\\") + "\\" + import_path
)

import_path = import_path.strip()
if not import_path:
continue

import_data = {
# `name` is the symbol as referenced in code — the last
# segment, or the alias when one is given.
"name": alias or import_path.split("\\")[-1],
"full_import_name": import_path,
"line_number": node.start_point[0] + 1,
"alias": alias,
"context": (None, None),
"lang": self.language_name,
"is_dependency": False,
}
imports.append(import_data)
except Exception as e:
error_logger(f"Error parsing import: {e}")
continue

# Query captures arrive in match order, not source order; sort so the
# emitted rows are deterministic across runs.
imports.sort(key=lambda i: (i["line_number"], i["full_import_name"]))
return imports

def _parse_calls(self, captures: list, source_code: str, var_type_map: dict) -> list[dict]:
Expand Down
Loading
Loading