Compute stats summary

CITATION.cff

@@ -29,5 +29,5 @@ keywords:
   - bot
   - software
 license: MIT
-version: 0.5.3
-date-released: 2026-07-06
+version: 0.5.4.dev1
+date-released: 2026-07-08

README.md

@@ -3,7 +3,7 @@
 [![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.19468976.svg)](https://doi.org/10.5281/zenodo.19468976)
 [![SWH](https://archive.softwareheritage.org/badge/swh:1:dir:9f3d474c040158cc675e8db135767e0d2d3fba7b/)](https://archive.softwareheritage.org/swh:1:dir:9f3d474c040158cc675e8db135767e0d2d3fba7b;origin=https://github.com/SoftwareUnderstanding/sw-metadata-bot;visit=swh:1:snp:03da984544982fdf5aa1f556012b917250f1e4c3;anchor=swh:1:rev:fc64150edf2e35f045cf3858d81b00ca6e7f6e68)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
-[![Version](https://img.shields.io/badge/version-0.5.3-blue.svg)](https://github.com/SoftwareUnderstanding/sw-metadata-bot/releases)
+[![Version](https://img.shields.io/badge/version-0.5.4.dev1-blue.svg)](https://github.com/SoftwareUnderstanding/sw-metadata-bot/releases)
 [![Python 3.11--3.12](https://img.shields.io/badge/python-3.11--3.12-blue.svg)](https://www.python.org/downloads/)
 [![CI (build)](https://github.com/SoftwareUnderstanding/sw-metadata-bot/actions/workflows/ci.yml/badge.svg)](https://github.com/SoftwareUnderstanding/sw-metadata-bot/actions/workflows/ci.yml)
 [![Code style: ruff](https://img.shields.io/badge/code%20style-ruff-000000.svg)](https://github.com/astral-sh/ruff)

codemeta.json

@@ -56,7 +56,7 @@
   ],
   "codeRepository": "https://github.com/SoftwareUnderstanding/sw-metadata-bot",
   "dateCreated": "2025-12-15",
-  "dateModified": "2026-07-06",
+  "dateModified": "2026-07-08",
   "datePublished": "2026-01-16",
   "description": "A repository to keep the code of the RSMetaCheck bot for pushing issues with existing repository metadata",
   "downloadUrl": "https://github.com/SoftwareUnderstanding/sw-metadata-bot/releases",
@@ -83,8 +83,8 @@
     "python-dotenv>=1.0.0",
     "requests>=2.32.5"
   ],
-  "version": "0.5.3",
-  "softwareVersion": "0.5.3",
+  "version": "0.5.4.dev1",
+  "softwareVersion": "0.5.4.dev1",
   "maintainer": {
     "@id": "https://orcid.org/0000-0001-5226-3089",
     "givenName": "Tom",

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "sw-metadata-bot"
-version = "0.5.3"
+version = "0.5.4.dev1"
 description = "Metadata quality bot for software repositories, leveraging metacheck for analysis and GitHub/GitLab APIs for issue management."
 readme = "README.md"
 requires-python = ">=3.11,<3.13"

src/sw_metadata_bot/constants.py

@@ -134,6 +134,8 @@
         FILENAME_CONFIG_SNAPSHOT,
     }
 )
+
+FILENAME_REPORT_SUMMARY = "report_summary.json"
 """Files to exclude when cleaning up metacheck output directories."""
 
 

src/sw_metadata_bot/main.py

@@ -4,6 +4,7 @@
 
 from .pipeline import run_analysis_command
 from .publish import publish_command, simulate_publish_command
+from .report_summary import summarize_report_command
 from .verify_tokens import verify_tokens_command
 
 
@@ -22,6 +23,7 @@ def cli():
 cli.add_command(run_analysis_command, name="run-analysis")
 cli.add_command(publish_command, name="publish")
 cli.add_command(simulate_publish_command, name="simulate-publish")
+cli.add_command(summarize_report_command, name="report-summary")
 
 
 def main():

src/sw_metadata_bot/report_summary.py

@@ -0,0 +1,131 @@
+"""Create a summary from existing report for dashboard"""
+
+import json
+from collections import Counter
+from pathlib import Path
+
+import click
+
+from . import constants
+from .reporting import RunReport, load_report
+
+
+@click.command()
+@click.option(
+    "--analysis-root",
+    type=click.Path(exists=True, dir_okay=True, path_type=Path),
+    default=None,
+    help="Previous run_report.json used for incremental issue handling.",
+)
+def summarize_report_command(analysis_root: Path):
+    """Summarize report.json file and save the summary file in the analysis root folder."""
+    summary = summarize_analysis_folder(analysis_root)
+
+    save_summary_report(analysis_root, summary)
+
+
+def summarize_report(report: RunReport) -> dict[str, object]:
+    """Build a compact summary for plotting and comparison."""
+    repository_count = len(report.records)
+    toolmetadata = report.get_tool_metadata()
+
+    pitfalls_by_id: Counter[str] = Counter()
+    warnings_by_id: Counter[str] = Counter()
+    total_pitfalls = 0
+    total_warnings = 0
+    issues_created = 0
+    actions: Counter[str] = Counter()
+    reason_codes: Counter[str] = Counter()
+    reason_codes_by_action: dict[str, Counter[str]] = {}
+
+    for record in report.records:
+        for pitfall_id in record.pitfalls_ids:
+            pitfalls_by_id[pitfall_id] += 1
+        for warning_id in record.warnings_ids:
+            warnings_by_id[warning_id] += 1
+
+        total_pitfalls += int(record.pitfalls_count or 0)
+        total_warnings += int(record.warnings_count or 0)
+
+        if record.action == constants.ACTION_CREATED:
+            issues_created += 1
+
+        if record.action is not None:
+            actions[record.action] += 1
+            reason_codes_by_action.setdefault(record.action, Counter())
+
+        if record.reason_code is not None:
+            reason_codes[record.reason_code] += 1
+            if record.action is not None:
+                reason_codes_by_action[record.action][record.reason_code] += 1
+
+    reason_code_percentages_by_action: dict[str, dict[str, float]] = {}
+    for action_name, counter in reason_codes_by_action.items():
+        total_for_action = sum(counter.values())
+        if total_for_action == 0:
+            reason_code_percentages_by_action[action_name] = {}
+            continue
+        reason_code_percentages_by_action[action_name] = {
+            reason_code: (count / total_for_action) * 100.0
+            for reason_code, count in sorted(counter.items())
+        }
+
+    return {
+        "tool_metadata": toolmetadata.to_dict(),
+        "run_metadata": report.run_metadata,
+        "repository_count": repository_count,
+        "pitfalls_by_id": dict(sorted(pitfalls_by_id.items())),
+        "warnings_by_id": dict(sorted(warnings_by_id.items())),
+        "total_pitfalls": total_pitfalls,
+        "total_warnings": total_warnings,
+        "issues_created": issues_created,
+        "actions": dict(sorted(actions.items())),
+        "reason_codes": dict(sorted(reason_codes.items())),
+        "reason_codes_by_action": {
+            action_name: dict(sorted(counter.items()))
+            for action_name, counter in sorted(reason_codes_by_action.items())
+        },
+        "reason_code_percentages_by_action": {
+            action_name: dict(sorted(percentages.items()))
+            for action_name, percentages in sorted(
+                reason_code_percentages_by_action.items()
+            )
+        },
+        "pitfalls_per_repository": (
+            total_pitfalls / repository_count if repository_count else 0.0
+        ),
+        "warnings_per_repository": (
+            total_warnings / repository_count if repository_count else 0.0
+        ),
+        "issues_created_per_repository": (
+            issues_created / repository_count if repository_count else 0.0
+        ),
+    }
+
+
+def summarize_analysis_folder(analysis_folder: Path | str) -> dict[str, object]:
+    """Load run_report.json from an analysis directory and summarize it."""
+    analysis_path = Path(analysis_folder)
+    report_path = analysis_path / constants.FILENAME_RUN_REPORT
+    if analysis_path.is_file():
+        report_path = analysis_path
+
+    return summarize_report(load_report(report_path))
+
+
+def save_summary_report(analysis_folder: Path, summary_data: dict):
+    """Save summary of the analysis"""
+    if summary_data:
+        summary_path = analysis_folder / constants.FILENAME_REPORT_SUMMARY
+        with open(summary_path, "w", encoding="utf-8") as f:
+            json.dump(summary_data, f, indent=2)
+    else:
+        print("Error: No summary info")
+
+
+if __name__ == "__main__":
+    analysis_root = Path("./outputs/ossr/20260702/")
+
+    summary = summarize_analysis_folder(analysis_root)
+
+    save_summary_report(analysis_root, summary)