Page de statistiques

tools/stats/words_histogram.py

@@ -0,0 +1,59 @@
+#!/usr/bin/env python3
+
+import sys
+import json
+import os
+
+CURRENT_DIR = os.path.dirname(os.path.abspath(__file__))
+PARENT_DIR = os.path.abspath(os.path.join(CURRENT_DIR, os.pardir))
+if CURRENT_DIR not in sys.path:
+    sys.path.append(CURRENT_DIR)
+if PARENT_DIR not in sys.path:
+    sys.path.append(PARENT_DIR)
+
+from common import load_articles, write_result  # noqa: E402
+
+
+def main():
+    try:
+        payload = json.load(sys.stdin)
+    except Exception as exc:  # noqa: BLE001
+        print(f"Failed to read JSON: {exc}", file=sys.stderr)
+        sys.exit(1)
+
+    content_dir = payload.get("contentDir") or "content"
+    output_path = payload.get("outputPath")
+    public_path = payload.get("publicPath")
+
+    articles = load_articles(content_dir)
+    values = [a.get("wordCount") or 0 for a in articles if a.get("wordCount")]
+
+    try:
+        from render_stats_charts import render_words_histogram, setup_rcparams
+    except ImportError as exc:  # noqa: BLE001
+        print(f"Failed to import renderer: {exc}", file=sys.stderr)
+        sys.exit(1)
+
+    setup_rcparams()
+    render_words_histogram(
+        {
+            "values": values,
+            "title": "Distribution des longueurs d'article",
+            "bins": 20,
+        },
+        output_path,
+    )
+
+    write_result(
+        {
+            "image": public_path,
+            "meta": {
+                "articles": len(values),
+            },
+        }
+    )
+
+
+if __name__ == "__main__":
+    main()
+