updated mappings of comments and filetypes

docs/source/development/change_log.rst

@@ -12,6 +12,7 @@ Fixes
 .....
 
 - 🐛 Replace absolute path with relative path to fix ``local-url`` not working on the non-local environment
+- 🐛 Add more file extensions of C/C++ for SourceDiscover
 
 .. _`release:1.0.0`:
 

src/sphinx_codelinks/analyse/oneline_parser.py

@@ -2,12 +2,7 @@
 from enum import Enum
 import logging
 
-from sphinx_codelinks.config import (
-    ESCAPE,
-    SUPPORTED_COMMENT_TYPES,
-    UNIX_NEWLINE,
-    OneLineCommentStyle,
-)
+from sphinx_codelinks.config import ESCAPE, UNIX_NEWLINE, OneLineCommentStyle
 
 # initialize logger
 logger = logging.getLogger(__name__)
@@ -204,15 +199,3 @@ def is_newline_in_field(field: str) -> bool:
     Check if the field contains a new line character.
     """
     return UNIX_NEWLINE in field
-
-
-def get_file_types(comment_type: str) -> list[str] | None:
-    """
-    Get the list of file types to be discovered.
-    """
-    file_types = (
-        list(SUPPORTED_COMMENT_TYPES)
-        if comment_type in SUPPORTED_COMMENT_TYPES
-        else None
-    )
-    return file_types

src/sphinx_codelinks/config.py

@@ -20,7 +20,6 @@
 
 COMMENT_MARKERS = {CommentType.cpp: ["//", "/*"], CommentType.python: ["#"]}
 ESCAPE = "\\"
-SUPPORTED_COMMENT_TYPES = {"c", "h", "cpp", "hpp", "py"}
 
 
 class CommentCategory(str, Enum):
@@ -47,7 +46,7 @@ def field_names(cls) -> set[str]:
     @classmethod
     def get_schema(cls, name: str) -> dict[str, Any] | None:  # type: ignore[explicit-any]
         _field = next(_field for _field in fields(cls) if _field.name is name)
-        if _field.metadata is not MISSING and "schema" in _field.metadata:
+        if _field.metadata and "schema" in _field.metadata:
             return cast(dict[str, Any], _field.metadata["schema"])  # type: ignore[explicit-any]
         return None
 
@@ -86,7 +85,7 @@ def field_names(cls) -> set[str]:
     @classmethod
     def get_schema(cls, name: str) -> dict[str, Any] | None:  # type: ignore[explicit-any]
         _field = next(_field for _field in fields(cls) if _field.name is name)
-        if _field.metadata is not MISSING and "schema" in _field.metadata:
+        if _field.metadata and "schema" in _field.metadata:
             return cast(dict[str, Any], _field.metadata["schema"])  # type: ignore[explicit-any]
         return None
 
@@ -219,14 +218,14 @@ def apply_needs_field_default(cls, given_fields: list[FieldConfig]) -> None:
     @classmethod
     def get_required_fields(cls, name: str) -> list[str] | None:
         _field = next(_field for _field in fields(cls) if _field.name is name)
-        if _field.metadata is not MISSING:
+        if _field.metadata:
             return cast(list[str], _field.metadata["required_fields"])
         return None
 
     @classmethod
     def get_schema(cls, name: str) -> dict[str, Any] | None:  # type: ignore[explicit-any]
         _field = next(_field for _field in fields(cls) if _field.name is name)
-        if _field.metadata is not MISSING and "schema" in _field.metadata:
+        if _field.metadata and "schema" in _field.metadata:
             return cast(dict[str, Any], _field.metadata["schema"])  # type: ignore[explicit-any]
         return None
 
@@ -370,7 +369,7 @@ def field_names(cls) -> set[str]:
     @classmethod
     def get_schema(cls, name: str) -> dict[str, Any] | None:  # type: ignore[explicit-any]
         _field = next(_field for _field in fields(cls) if _field.name is name)
-        if _field.metadata is not MISSING and "schema" in _field.metadata:
+        if _field.metadata and "schema" in _field.metadata:
             return cast(dict[str, Any], _field.metadata["schema"])  # type: ignore[explicit-any]
         return None
 
@@ -566,7 +565,7 @@ def field_names(cls) -> set[str]:
     def get_schema(cls, name: str) -> dict[str, Any] | None:  # type: ignore[explicit-any]
         """Get the schema for a config item."""
         _field = next(field for field in fields(cls) if field.name is name)
-        if _field.metadata is not MISSING and "schema" in _field.metadata:
+        if _field.metadata and "schema" in _field.metadata:
             return _field.metadata["schema"]  # type: ignore[no-any-return]
         return None
 

src/sphinx_codelinks/needextend_write.py

@@ -1,7 +1,7 @@
-"""Covert the generated JSON file created by CodeLinks anaylse to need-extend in RST."""
+"""Convert the generated JSON file created by CodeLinks analyse to need-extend in RST."""
 
 from collections import deque
-from dataclasses import MISSING, dataclass, field, fields
+from dataclasses import dataclass, field, fields
 from os import linesep
 from string import Template
 from typing import Any, TypedDict, cast
@@ -105,7 +105,7 @@ def field_names(cls) -> set[str]:
     @classmethod
     def get_schema(cls, name: str) -> dict[str, Any] | None:  # type: ignore[explicit-any]
         _field = next(_field for _field in fields(cls) if _field.name is name)
-        if _field.metadata is not MISSING and "schema" in _field.metadata:
+        if _field.metadata and "schema" in _field.metadata:
             return cast(dict[str, Any], _field.metadata["schema"])  # type: ignore[explicit-any]
         return None
 
@@ -165,7 +165,7 @@ def convert_marked_content(
     remote_url_field: str = "remote-url",
     title: str | None = None,
 ) -> tuple[list[str], list[str]]:
-    """Convert marked objects extracted by anaylse CLI to needextend in RST"""
+    """Convert marked objects extracted by analyse CLI to needextend in RST"""
     errors = []
     needextend_texts: list[str] = []
     intersted_objs = [

src/sphinx_codelinks/source_discover/config.py

@@ -1,11 +1,16 @@
-from dataclasses import MISSING, dataclass, field, fields
+from dataclasses import dataclass, field, fields
 from enum import Enum
 from pathlib import Path
 from typing import Any, Required, TypedDict, cast
 
 from jsonschema import ValidationError, validate
 
-COMMENT_FILETYPE = {"cpp": ["c", "cpp", "h", "hpp"], "python": ["py"]}
+# TODO: COMMENT_FILETYPE is probably not a good name, as C# uses the same comment style as C++, but it's grammatically different.
+# Therefore, C# requires a different parser of tree-sitter to analyse its AST
+COMMENT_FILETYPE = {
+    "cpp": ["c", "ci", "cpp", "cc", "cxx", "h", "hpp", "hxx", "hh", "ihl"],
+    "python": ["py"],
+}
 
 
 class CommentType(str, Enum):
@@ -73,7 +78,7 @@ def field_names(cls) -> set[str]:
     @classmethod
     def get_schema(cls, name: str) -> dict[str, Any] | None:  # type: ignore[explicit-any]
         _field = next(_field for _field in fields(cls) if _field.name is name)
-        if _field.metadata is not MISSING and "schema" in _field.metadata:
+        if _field.metadata and "schema" in _field.metadata:
             return cast(dict[str, Any], _field.metadata["schema"])  # type: ignore[explicit-any]
         return None
 

tests/test_source_discover.py

@@ -3,6 +3,7 @@
 import pytest
 
 from sphinx_codelinks.source_discover.config import (
+    COMMENT_FILETYPE,
     SourceDiscoverConfig,
     SourceDiscoverConfigType,
 )
@@ -145,3 +146,30 @@ def test_source_discover(
     assert len(source_discover.source_paths) == num_files
     if suffix:
         assert all(path.suffix == ".cpp" for path in source_discover.source_paths)
+
+
+@pytest.fixture(scope="function")
+def create_source_files(tmp_path: Path) -> Path:
+    for file_types in COMMENT_FILETYPE.values():
+        for ext in file_types:
+            (tmp_path / f"file.{ext}").touch()
+    return tmp_path
+
+
+@pytest.mark.parametrize(
+    ("comment_type", "nums_files"),
+    [
+        ("cpp", len(COMMENT_FILETYPE["cpp"])),
+        ("python", len(COMMENT_FILETYPE["python"])),
+    ],
+)
+def test_comment_filetype(
+    comment_type: str, nums_files: int, create_source_files: Path
+) -> None:
+    src_dir = create_source_files
+
+    config = SourceDiscoverConfig(
+        src_dir=src_dir, comment_type=comment_type, gitignore=False
+    )
+    source_discover = SourceDiscover(config)
+    assert len(source_discover.source_paths) == nums_files