feat: Support special github scenarios via configuration (#52)

Define behaviour for stripping PR information from github descriptions.
Extract PR information as a footer, and extract common github keywords
as footers.

closes #50

changelog_gen/config.py

@@ -27,10 +27,6 @@
 
 FOOTER_PARSERS = [
     r"(Refs)(: )(#?[\w-]+)",
-    # TODO(edgy): Parse github behind a github config
-    # https://github.com/NRWLDev/changelog-gen/issues/50
-    r"(closes)( )(#[\w-]+)",
-    r"(fixes)( )(#[\w-]+)",
     r"(Authors)(: )(.*)",
 ]
 
@@ -51,6 +47,15 @@ class PostProcessConfig:
     auth_env: str | None = None
 
 
+@dataclasses.dataclass
+class GithubConfig:
+    """Github specific helpers."""
+
+    strip_pr_from_description: bool = False
+    extract_pr_from_description: bool = False
+    extract_common_footers: bool = False
+
+
 STRICT_VALIDATOR = re.compile(
     r"^(?P<major>0|[1-9]\d*)\.(?P<minor>0|[1-9]\d*)\.(?P<patch>0|[1-9]\d*)(?:-(?P<prerelease>(?:0|[1-9]\d*|\d*[a-zA-Z-][0-9a-zA-Z-]*)(?:\.(?:0|[1-9]\d*|\d*[a-zA-Z-][0-9a-zA-Z-]*))*))?$",
 )
@@ -88,6 +93,9 @@ class Config:
     # Version bumping
     files: dict = dataclasses.field(default_factory=dict)
 
+    # Github helpers
+    github: GithubConfig | None = None
+
     # Changelog configuration
     date_format: str | None = None
     version_string: str = "v{new_version}"
@@ -186,6 +194,7 @@ def _process_pyproject(pyproject: Path) -> dict:
         data["tool"]["changelog_gen"]["footer_parsers"] = footer_parsers
         data["tool"]["changelog_gen"]["commit_types"] = commit_types
         data["tool"]["changelog_gen"]["type_headers"] = type_headers
+
         return data["tool"]["changelog_gen"]
 
 
@@ -227,6 +236,10 @@ def read(path: str = "pyproject.toml", **kwargs) -> Config:
             msg = f"Failed to create post_process: {e!s}"
             raise RuntimeError(msg) from e
 
+    if "github" in cfg:
+        github = GithubConfig(**cfg["github"])
+        cfg["github"] = github
+
     try:
         return Config(**cfg)
     except TypeError as e:

changelog_gen/extractor.py

@@ -102,8 +102,12 @@ def extract(self: t.Self) -> list[Change]:  # noqa: C901, PLR0912, PLR0915
                 prm = re.search(r"\(#\d+\)$", description)
                 if prm is not None:
                     # Strip githubs additional link information from description.
-                    description = re.sub(r" \(#\d+\)$", "", description)
-                    footers["pr"] = Footer("PR", ": ", prm.group()[1:-1])
+                    if self.context.config.github and self.context.config.github.strip_pr_from_description:
+                        description = re.sub(r" \(#\d+\)$", "", description)
+
+                    if self.context.config.github and self.context.config.github.extract_pr_from_description:
+                        footers["pr"] = Footer("PR", ": ", prm.group()[1:-1])
+
                 details = m[5] or ""
 
                 # Handle missing refs in commit message, skip link generation in writer
@@ -118,6 +122,20 @@ def extract(self: t.Self) -> list[Change]:  # noqa: C901, PLR0912, PLR0915
                 if breaking:
                     self.context.info("  Breaking change detected:\n    %s: %s", commit_type, description)
 
+                footer_parsers = self.context.config.footer_parsers
+                if self.context.config.github and self.context.config.github.extract_common_footers:
+                    footer_parsers.extend([
+                        r"(close)( )(#[\w-]+)",
+                        r"(closes)( )(#[\w-]+)",
+                        r"(closed)( )(#[\w-]+)",
+                        r"(fix)( )(#[\w-]+)",
+                        r"(fixes)( )(#[\w-]+)",
+                        r"(fixed)( )(#[\w-]+)",
+                        r"(resolve)( )(#[\w-]+)",
+                        r"(resolves)( )(#[\w-]+)",
+                        r"(resolved)( )(#[\w-]+)",
+                    ])
+
                 for line in details.split("\n"):
                     for parser in self.context.config.footer_parsers:
                         m = re.match(parser, line, re.IGNORECASE)

docs/configuration.md

@@ -418,6 +418,52 @@ pre_l = ["dev", "rc"]
   Enforce strict rules based on SemVer 2.0.0 and error if non conforming parser or
   serialisers are configured.
 
+
+## Github
+
+### `strip_pr_from_description`
+  _**[optional]**_<br />
+  **default**: False
+
+  Strip the `(#\d+)` from the end of github PR commit descriptions.
+
+  Example:
+
+```toml
+[tool.changelog_gen.github]
+strip_pr_from_description = true
+```
+
+### `extract_pr_from_description`
+  _**[optional]**_<br />
+  **default**: False
+
+  Extract the `(#\d+)` from the end of github PR commit descriptions, and track
+  it as a footer for later extraction and link generation. Creates a `PR` footer entry.
+
+  Example:
+
+```toml
+[tool.changelog_gen.github]
+extract_pr_from_description = true
+```
+
+### `extract_common_footers`
+  _**[optional]**_<br />
+  **default**: False
+
+  Extract supported keyword footers from github commits, `closes #1` etc.
+
+  Supported footers can be found
+  [here](https://docs.github.com/en/issues/tracking-your-work-with-issues/linking-a-pull-request-to-an-issue).
+
+  Example:
+
+```toml
+[tool.changelog_gen.github]
+extract_common_footers = true
+```
+
 ## Post processing
 
 ### `post_process`

pyproject.toml

@@ -71,6 +71,11 @@ allowed_branches = [
 ]
 date_format = "- %Y-%m-%d"
 
+[tool.changelog_gen.github]
+strip_pr_from_description = true
+extract_pr_from_description = true
+extract_common_footers = true
+
 [[tool.changelog_gen.extractors]]
 footer = ["closes", "fixes", "Refs"]
 pattern = '#(?P<issue_ref>\d+)'

tests/cli/test_config.py

@@ -35,8 +35,6 @@ def test_config_displayed(cli_runner):
 serialisers = ['{major}.{minor}.{patch}']
 footer_parsers = [
     '(Refs)(: )(#?[\w-]+)',
-    '(closes)( )(#[\w-]+)',
-    '(fixes)( )(#[\w-]+)',
     '(Authors)(: )(.*)',
 ]
 extractors = []

tests/test_config.py

@@ -179,6 +179,25 @@ def test_read_picks_up_type_headers(self, config_factory):
             "test": "Miscellaneous",
         }
 
+    def test_read_picks_up_github_config(self, config_factory):
+        config_factory(
+            """
+[tool.changelog_gen]
+current_version = "0.0.0"
+[tool.changelog_gen.github]
+strip_pr_from_description = true
+extract_pr_from_description = true
+extract_common_footers = true
+""",
+        )
+
+        c = config.read()
+        assert c.github == config.GithubConfig(
+            strip_pr_from_description=True,
+            extract_pr_from_description=True,
+            extract_common_footers=True,
+        )
+
 
 class TestPostProcessConfig:
     def test_read_picks_up_no_post_process_config(self, config_factory):