🐛 Route analyse/ logging by environment instead of a stderr handler at import

src/sphinx_codelinks/analyse/analyse.py

@@ -1,7 +1,6 @@
 from collections.abc import Generator
 from dataclasses import dataclass
 import json
-import logging
 from pathlib import Path
 from typing import Any, TypedDict
 
@@ -26,14 +25,14 @@
     OneLineCommentStyle,
     SourceAnalyseConfig,
 )
+from sphinx_codelinks.logger import get_logger
 
-# initialize logger
-logger = logging.getLogger(__name__)
-logger.setLevel(logging.INFO)
-# log to the console
-console = logging.StreamHandler()
-console.setLevel(logging.INFO)
-logger.addHandler(console)
+logger = get_logger(__name__)
+
+
+def _count(n: int, noun: str) -> str:
+    """Format ``n noun`` with a naive (append-s) plural for progress summaries."""
+    return f"{n} {noun}" if n == 1 else f"{n} {noun}s"
 
 
 class AnalyseWarningType(TypedDict):
@@ -57,7 +56,10 @@ class SourceAnalyse:
     def __init__(
         self,
         analyse_config: SourceAnalyseConfig,
+        *,
+        name: str = "",
     ) -> None:
+        self.name = name
         self.analyse_config = analyse_config
         self.src_files: list[SourceFile] = []
         self.src_comments: list[SourceComment] = []
@@ -110,9 +112,6 @@ def create_src_objects(self) -> None:
             self.src_files.append(src_file)
             self.src_comments.extend(src_comments)
 
-        logger.info(f"Source files loaded: {len(self.src_files)}")
-        logger.info(f"Source comments extracted: {len(self.src_comments)}")
-
     def extract_marker(
         self,
         text: str,
@@ -353,13 +352,6 @@ def extract_marked_content(self) -> None:
                 if marked_rst:
                     self.marked_rst.append(marked_rst)
 
-        if self.analyse_config.get_need_id_refs:
-            logger.info(f"Need-id-refs extracted: {len(self.need_id_refs)}")
-        if self.analyse_config.get_oneline_needs:
-            logger.info(f"Oneline needs extracted: {len(self.oneline_needs)}")
-        if self.analyse_config.get_rst:
-            logger.info(f"Marked rst extracted: {len(self.marked_rst)}")
-
     def merge_marked_content(self) -> None:
         self.all_marked_content.extend(self.need_id_refs)
         self.oneline_needs.sort(key=lambda x: x.source_map["start"]["row"])
@@ -378,9 +370,23 @@ def dump_marked_content(self, outdir: Path) -> None:
         ]
         with output_path.open("w") as f:
             json.dump(to_dump, f)
-        logger.info(f"Marked content dumped to {output_path}")
 
     def run(self) -> None:
         self.create_src_objects()
         self.extract_marked_content()
         self.merge_marked_content()
+        self._log_summary()
+
+    def _log_summary(self) -> None:
+        """Emit a per-project marker (default-visible) plus a -v breakdown."""
+        label = f"codelinks [{self.name}]" if self.name else "codelinks"
+        logger.info(
+            f"{label}: {_count(len(self.src_files), 'file')}, "
+            f"{_count(len(self.all_marked_content), 'marker')}"
+        )
+        logger.debug(
+            f"{label}: {_count(len(self.src_comments), 'comment')}, "
+            f"{_count(len(self.oneline_needs), 'oneline need')}, "
+            f"{_count(len(self.need_id_refs), 'id-ref')}, "
+            f"{_count(len(self.marked_rst), 'marked-rst block')}"
+        )

src/sphinx_codelinks/analyse/oneline_parser.py

@@ -1,17 +1,8 @@
 from dataclasses import dataclass
 from enum import Enum
-import logging
 
 from sphinx_codelinks.config import ESCAPE, UNIX_NEWLINE, OneLineCommentStyle
 
-# initialize logger
-logger = logging.getLogger(__name__)
-logger.setLevel(logging.INFO)
-# log to the console
-console = logging.StreamHandler()
-console.setLevel(logging.INFO)
-logger.addHandler(console)
-
 
 class WarningSubTypeEnum(str, Enum):
     """Enum for warning sub types."""

src/sphinx_codelinks/analyse/projects.py

@@ -1,5 +1,4 @@
 import json
-import logging
 from pathlib import Path
 from typing import cast
 
@@ -9,14 +8,9 @@
     SourceAnalyse,
 )
 from sphinx_codelinks.config import CodeLinksConfig, CodeLinksProjectConfigType
+from sphinx_codelinks.logger import get_logger
 
-# initialize logger
-logger = logging.getLogger(__name__)
-logger.setLevel(logging.INFO)
-# log to the console
-console = logging.StreamHandler()
-console.setLevel(logging.INFO)
-logger.addHandler(console)
+logger = get_logger(__name__)
 
 
 class AnalyseProjects:
@@ -32,7 +26,7 @@ def __init__(self, codelink_config: CodeLinksConfig) -> None:
 
     def run(self) -> None:
         for project, config in self.projects_configs.items():
-            src_analyse = SourceAnalyse(config["analyse_config"])
+            src_analyse = SourceAnalyse(config["analyse_config"], name=project)
             src_analyse.run()
             self.projects_analyse[project] = src_analyse
 
@@ -46,7 +40,7 @@ def dump_markers(self) -> None:
         }
         with output_path.open("w") as f:
             json.dump(to_dump, f)
-        logger.info(f"Marked content dumped to {output_path}")
+        logger.debug(f"codelinks: marked content dumped to {output_path}")
 
     @classmethod
     def load_warnings(cls, warnings_dir: Path) -> list[AnalyseWarning] | None:

src/sphinx_codelinks/analyse/utils.py

@@ -1,6 +1,5 @@
 from collections.abc import ByteString, Callable
 import configparser
-import logging
 from pathlib import Path
 from typing import TypedDict
 from urllib.request import pathname2url
@@ -10,6 +9,7 @@
 from tree_sitter import Node as TreeSitterNode
 
 from sphinx_codelinks.config import UNIX_NEWLINE, CommentCategory
+from sphinx_codelinks.logger import get_logger
 from sphinx_codelinks.source_discover.config import CommentType
 
 # Language-specific node types for scope detection
@@ -31,13 +31,7 @@
     },
 }
 
-# initialize logger
-logger = logging.getLogger(__name__)
-logger.setLevel(logging.INFO)
-# log to the console
-console = logging.StreamHandler()
-console.setLevel(logging.INFO)
-logger.addHandler(console)
+logger = get_logger(__name__)
 
 GIT_HOST_URL_TEMPLATE = {
     "github": "https://github.com/{owner}/{repo}/blob/{rev}/{path}#L{lineno}",
@@ -270,15 +264,23 @@ def locate_git_root(src_dir: Path) -> Path | None:
     for parent in parents:
         if (parent / ".git").exists() and (parent / ".git").is_dir():
             return parent
-    logger.warning(f"git root is not found in the parent of {src_dir}")
+    logger.warning(
+        f"git root is not found in the parent of {src_dir}",
+        subtype="git_root",
+        location=str(src_dir),
+    )
     return None
 
 
 def get_remote_url(git_root: Path, remote_name: str = "origin") -> str | None:
     """Get remote url from .git/config."""
     config_path = git_root / ".git" / "config"
     if not config_path.exists():
-        logging.warning(f"{config_path} does not exist")
+        logger.warning(
+            f"{config_path} does not exist",
+            subtype="git_config",
+            location=str(config_path),
+        )
         return None
 
     config = configparser.ConfigParser(allow_no_value=True, strict=False)
@@ -287,24 +289,37 @@ def get_remote_url(git_root: Path, remote_name: str = "origin") -> str | None:
     if section in config and "url" in config[section]:
         url: str = config[section]["url"]
         return url
-    logger.warning(f"remote-url is not found in {config_path}")
+    logger.warning(
+        f"remote-url is not found in {config_path}",
+        subtype="git_remote",
+        location=str(config_path),
+    )
     return None
 
 
 def get_current_rev(git_root: Path) -> str | None:
     """Get current commit rev from .git/HEAD."""
     head_path = git_root / ".git" / "HEAD"
     if not head_path.exists():
-        logging.warning(f"{head_path} does not exist")
+        logger.warning(
+            f"{head_path} does not exist",
+            subtype="git_head",
+            location=str(head_path),
+        )
         return None
     head_content = head_path.read_text().strip()
     if not head_content.startswith("ref: "):
-        logging.warning(f"Expect starting with 'ref: ' in {head_path}")
-        return None
+        # Detached HEAD (e.g. CI checkouts): .git/HEAD holds the commit SHA
+        # directly, which is exactly the rev we want.
+        return head_content
 
     ref_path = git_root / ".git" / head_content.split(":", 1)[1].strip()
     if not ref_path.exists():
-        logging.warning(f"{ref_path} does not exist")
+        logger.warning(
+            f"{ref_path} does not exist",
+            subtype="git_ref",
+            location=str(ref_path),
+        )
         return None
     return ref_path.read_text().strip()
 
@@ -315,7 +330,10 @@ def form_https_url(
     parsed_url = parse(git_url)
     template = GIT_HOST_URL_TEMPLATE.get(parsed_url.platform)
     if not template:
-        logging.warning(f"Unsupported Git host: {parsed_url.platform}")
+        logger.warning(
+            f"Unsupported Git host: {parsed_url.platform}",
+            subtype="git_host",
+        )
         return git_url
     https_url = template.format(
         owner=parsed_url.owner,

src/sphinx_codelinks/cmd.py

@@ -14,7 +14,7 @@
     CodeLinksProjectConfigType,
     generate_project_configs,
 )
-from sphinx_codelinks.logger import logger
+from sphinx_codelinks.logger import configure_cli, logger
 from sphinx_codelinks.needextend_write import MarkedObjType, convert_marked_content
 from sphinx_codelinks.source_discover.config import (
     CommentType,
@@ -88,9 +88,12 @@ def analyse(  # noqa: PLR0912   # for CLI, so it needs the branches
             exists=True,
         ),
     ] = None,
+    verbose: OptVerbose = False,
+    quiet: OptQuiet = False,
 ) -> None:
     """Analyse marked content in source code."""
     # @CLI command to analyse source code and extract traceability markers, IMPL_CLI_ANALYZE, impl, [FE_CLI_ANALYZE]
+    configure_cli(verbose, quiet)
 
     data: CodeLinksConfigType = load_config_from_toml(config)
 
@@ -291,7 +294,7 @@ def write_rst(  # noqa: PLR0913  # for CLI, so it takes as many as it requires
     quiet: OptQuiet = False,
 ) -> None:
     """Generate needextend.rst from the extracted obj in JSON."""
-    logger.configure(verbose, quiet)
+    configure_cli(verbose, quiet)
     try:
         with jsonpath.open("r") as f:
             marked_content = json.load(f)