fix: batch generate JSON datetime serialization, unique PDF names, approved-only ZIP/exports

main.py

@@ -544,7 +544,11 @@ async def generate_cvs_for_requirement(req_id: str, request: Request):
 @app.get("/api/generated-cvs")
 async def list_generated_cvs(req_id: str = None, candidate_id: str = None, batch_id: str = None):
     """List batch-generated CV exports (plus any legacy generated_cvs rows)."""
-    where_parts = ["bi.generated_cv_path IS NOT NULL", "bi.generated_cv_path <> ''"]
+    where_parts = [
+        "bi.generated_cv_path IS NOT NULL",
+        "bi.generated_cv_path <> ''",
+        "bi.status = 'approved'",
+    ]
     params = []
     if candidate_id:
         where_parts.append("bi.candidate_id = %s")
@@ -901,7 +905,11 @@ async def list_batches():
                    COUNT(*) FILTER (WHERE status = 'approved') AS approved_items,
                    COUNT(*) FILTER (WHERE status = 'proposed') AS proposed_items,
                    COUNT(*) FILTER (WHERE status = 'removed') AS removed_items,
-                   COUNT(*) FILTER (WHERE generated_cv_path IS NOT NULL AND generated_cv_path <> '') AS generated_items
+                   COUNT(*) FILTER (
+                       WHERE generated_cv_path IS NOT NULL
+                         AND generated_cv_path <> ''
+                         AND status = 'approved'
+                   ) AS generated_items
             FROM batch_items
             GROUP BY batch_id
         ) stats ON stats.batch_id = b.id
@@ -1070,11 +1078,41 @@ async def analyze_batch(batch_id: str):
 async def update_batch_item(batch_id: str, item_id: str, request: Request):
     """Update a batch item (approve, remove, edit realignment)."""
     body = await request.json()
-    row = db.execute(
-        """UPDATE batch_items SET status = %s, realigned_cv_data = %s, updated_at = NOW()
-           WHERE id = %s AND batch_id = %s RETURNING *""",
-        (body.get("status", "proposed"), Json(body.get("realigned_cv_data")), item_id, batch_id)
-    )
+    status = body.get("status", "proposed")
+    realigned = body.get("realigned_cv_data")
+
+    # If removing a candidate, clear any stale generated PDF path so exports/ZIPs stay clean.
+    if status == "removed":
+        existing = db.query(
+            "SELECT generated_cv_path FROM batch_items WHERE id = %s AND batch_id = %s",
+            (item_id, batch_id),
+            fetch="one",
+        )
+        if existing and existing.get("generated_cv_path"):
+            rel = str(existing["generated_cv_path"]).lstrip("/")
+            if rel.startswith("static/"):
+                file_path = static_dir / rel[len("static/"):]
+            else:
+                file_path = static_dir / "generated" / Path(rel).name
+            try:
+                if file_path.exists() and file_path.is_file():
+                    # Only delete files that belong to this item id tag when possible.
+                    if str(item_id)[:8] in file_path.name:
+                        file_path.unlink()
+            except Exception:
+                pass
+        row = db.execute(
+            """UPDATE batch_items
+               SET status = %s, realigned_cv_data = %s, generated_cv_path = NULL, updated_at = NOW()
+               WHERE id = %s AND batch_id = %s RETURNING *""",
+            (status, Json(realigned), item_id, batch_id),
+        )
+    else:
+        row = db.execute(
+            """UPDATE batch_items SET status = %s, realigned_cv_data = %s, updated_at = NOW()
+               WHERE id = %s AND batch_id = %s RETURNING *""",
+            (status, Json(realigned), item_id, batch_id),
+        )
     if not row:
         raise HTTPException(404, "Batch item not found")
     return dict(row)
@@ -1183,6 +1221,25 @@ async def update_batch_positions(batch_id: str, request: Request):
 async def generate_batch_cvs(batch_id: str):
     """Generate CVs for all approved items in the batch using the selected template."""
     import httpx
+    from datetime import date as _date, datetime as _datetime
+    from decimal import Decimal
+    from uuid import UUID
+
+    def _jsonable(value):
+        """Recursively convert DB row values into JSON-safe types."""
+        if isinstance(value, dict):
+            return {k: _jsonable(v) for k, v in value.items()}
+        if isinstance(value, (list, tuple)):
+            return [_jsonable(v) for v in value]
+        if isinstance(value, (_datetime, _date)):
+            return value.isoformat()
+        if isinstance(value, UUID):
+            return str(value)
+        if isinstance(value, Decimal):
+            return float(value)
+        if isinstance(value, (bytes, bytearray)):
+            return value.decode("utf-8", errors="replace")
+        return value
 
     batch = db.query("SELECT * FROM cv_batches WHERE id = %s", (batch_id,), fetch='one')
     if not batch:
@@ -1218,6 +1275,7 @@ async def generate_batch_cvs(batch_id: str):
         cand["experience"] = [dict(e) for e in db.query("SELECT * FROM experience WHERE candidate_id = %s", (candidate_id,))]
         cand["education"] = [dict(e) for e in db.query("SELECT * FROM education WHERE candidate_id = %s", (candidate_id,))]
         cand["certifications"] = [dict(c) for c in db.query("SELECT * FROM certifications WHERE candidate_id = %s", (candidate_id,))]
+        cand = _jsonable(cand)
         
         # Use realigned data if available, else original
         realigned = item.get("realigned_cv_data")
@@ -1231,11 +1289,13 @@ async def generate_batch_cvs(batch_id: str):
             
             if resp.status_code == 200:
                 # Save the PDF
-                import tempfile
                 pdf_dir = os.path.join(str(static_dir), "generated")
                 os.makedirs(pdf_dir, exist_ok=True)
                 safe_name = f"{item['first_name']}_{item['last_name']}".replace(" ", "_")
-                filename = f"{batch['name'].replace(' ', '_')}_{safe_name}_{candidate_id[:8]}.pdf"
+                safe_batch = str(batch['name']).replace(" ", "_").replace("/", "-")
+                safe_pos = str(item.get("position_title") or "role").replace(" ", "_").replace("/", "-")[:40]
+                item_tag = str(item["id"])[:8]
+                filename = f"{safe_batch}_{safe_pos}_{safe_name}_{item_tag}.pdf"
                 filepath = os.path.join(pdf_dir, filename)
                 with open(filepath, "wb") as f:
                     f.write(resp.content)
@@ -1245,16 +1305,24 @@ async def generate_batch_cvs(batch_id: str):
                     "UPDATE batch_items SET generated_cv_path = %s, updated_at = NOW() WHERE id = %s",
                     (f"/static/generated/{filename}", item["id"])
                 )
-                results.append({"candidate": f"{item['first_name']} {item['last_name']}", "status": "ok", "file": filename})
+                results.append({
+                    "candidate": f"{item['first_name']} {item['last_name']}",
+                    "position_title": item.get("position_title"),
+                    "status": "ok",
+                    "file": filename,
+                    "path": f"/static/generated/{filename}",
+                })
             else:
                 results.append({"candidate": f"{item['first_name']} {item['last_name']}", "status": "error", "error": resp.text[:200]})
         except Exception as e:
             results.append({"candidate": f"{item['first_name']} {item['last_name']}", "status": "error", "error": str(e)})
     
-    # Update batch status to exported
-    db.execute("UPDATE cv_batches SET status = 'exported', updated_at = NOW() WHERE id = %s", (batch_id,))
+    ok_count = len([r for r in results if r["status"] == "ok"])
+    # Only mark exported if at least one CV succeeded
+    if ok_count:
+        db.execute("UPDATE cv_batches SET status = 'exported', updated_at = NOW() WHERE id = %s", (batch_id,))
     
-    return {"success": True, "generated": len([r for r in results if r["status"] == "ok"]), "results": results}
+    return {"success": True, "generated": ok_count, "results": results}
 
 
 @app.get("/api/batches/{batch_id}/download-zip")
@@ -1273,6 +1341,7 @@ async def download_batch_zip(batch_id: str):
         FROM batch_items bi
         JOIN candidates c ON c.id = bi.candidate_id
         WHERE bi.batch_id = %s
+          AND bi.status = 'approved'
           AND bi.generated_cv_path IS NOT NULL
           AND bi.generated_cv_path <> ''
         ORDER BY bi.position_title, c.last_name, c.first_name