Merge origin/master into feat/typed-models-schema-validation-24

Resolves three conflicts in scripts/export.py, all the result of PR #35
(linter cleanup) landing on master while PR #30 was open:

1. Line 41 import block — master added `# noqa: E402` to the existing
   `cursor_md_exporter` import; this branch added `ExportEntry,
   SchemaError` from `models`. Kept both: the new import also gets
   `# noqa: E402` so the linter discipline is consistent.

2. Manifest writer (local) at the `for e in exported:` loop — master
   renamed the loop variable to `entry` (to dodge a mypy
   `Assignment to variable "e" outside except: block` clash); this
   branch added `title` + `workspace` fields to each manifest row.
   Kept master's `entry` rename + this branch's wider schema.

3. Manifest writer (global) — same shape as #2, resolved identically.

Plus one secondary fix in tests/test_models.py: master removed
`continue-on-error: true` from the mypy CI step, so the stricter gate
now flags 5 `var-annotated` errors in this branch's test file (loop
variables iterating heterogeneous tuples). Fixed by hoisting each
tuple into a typed local — `bad_ids: tuple[object, ...] = (...)` —
matching the pattern this same file already uses at line 180.

Verified locally on the merged tree:
  - mypy --ignore-missing-imports --no-strict-optional . → Success
  - ruff check api/ utils/ scripts/export.py app.py → All checks passed
  - unittest discover tests → 223 / 223 OK
  - python app.py boot smoke → HTTP 200, no errors

Author: timon0305

.github/workflows/tests.yml

@@ -55,9 +55,9 @@ jobs:
   # Codebase already has type hints across most of the surface (~70+ typed
   # functions). Mypy runs in lenient mode (--ignore-missing-imports for
   # untyped third-party deps; no strict-optional) so the gate isn't a wall
-  # of false positives on first run. continue-on-error keeps findings as
-  # warnings during the surface-cleanup phase; flip to required by removing
-  # continue-on-error once the surface is clean.
+  # of false positives. The transitional `continue-on-error: true` was
+  # removed in #29 once mypy reached zero errors on this repo — type
+  # failures now block merges.
   typecheck:
     name: Typecheck (mypy)
     runs-on: ubuntu-latest
@@ -75,9 +75,8 @@ jobs:
           python -m pip install 'flask>=3.0' 'fpdf2>=2.7' 'mypy>=1.10'
 
       - name: Run mypy
-        # Transitional only (maintainer consensus): keeps CI green until `mypy` exits
-        # zero on this repo — then delete this line so type errors fail the job.
-        continue-on-error: true
+        # No `continue-on-error` — mypy now exits zero on this repo (closes #29),
+        # so type errors must fail the job from here on.
         run: mypy --ignore-missing-imports --no-strict-optional --pretty .
 
   # ── Secret scan: gitleaks ─────────────────────────────────────────────────

api/composers.py

@@ -5,6 +5,7 @@
 """
 
 import json
+import logging
 import os
 import sqlite3
 from contextlib import closing
@@ -16,6 +17,7 @@
 from models import Composer, SchemaError, Workspace, WorkspaceLocalComposer
 
 bp = Blueprint("composers", __name__)
+_logger = logging.getLogger(__name__)
 
 
 def _read_json_file(path: str):
@@ -96,8 +98,8 @@ def list_composers():
         composers.sort(key=lambda pair: to_epoch_ms(pair[0].last_updated_at), reverse=True)
         return jsonify([c for _, c in composers])
 
-    except Exception as e:
-        print(f"Failed to get composers: {e}")
+    except Exception:
+        _logger.exception("Failed to get composers")
         return jsonify({"error": "Failed to get composers"}), 500
 
 
@@ -187,6 +189,6 @@ def get_composer(composer_id):
 
         return jsonify({"error": "Composer not found"}), 404
 
-    except Exception as e:
-        print(f"Failed to get composer: {e}")
+    except Exception:
+        _logger.exception("Failed to get composer")
         return jsonify({"error": "Failed to get composer"}), 500

api/export_api.py

@@ -9,7 +9,6 @@
 import os
 import re
 import sqlite3
-import sys
 import zipfile
 from contextlib import closing
 from datetime import datetime
@@ -18,7 +17,7 @@
 from flask import Blueprint, Response, current_app, jsonify, request
 
 from utils.workspace_path import resolve_workspace_path
-from utils.path_helpers import normalize_file_path, get_workspace_folder_paths, to_epoch_ms
+from utils.path_helpers import get_workspace_folder_paths, to_epoch_ms
 from utils.text_extract import extract_text_from_bubble
 from utils.tool_parser import parse_tool_call
 from utils.exclusion_rules import build_searchable_text, is_excluded_by_rules
@@ -390,15 +389,15 @@ def export_chats():
                         status_str = f" ({tool_status})" if tool_status else ""
                         md += f"> **Tool: {tool_summary}**{status_str}\n"
                         if t.get("input"):
-                            md += f">\n> **INPUT:**\n> ```\n"
+                            md += ">\n> **INPUT:**\n> ```\n"
                             for iline in str(t["input"]).split("\n"):
                                 md += f"> {iline}\n"
-                            md += f"> ```\n"
+                            md += "> ```\n"
                         if t.get("output"):
-                            md += f">\n> **OUTPUT:**\n> ```\n"
+                            md += ">\n> **OUTPUT:**\n> ```\n"
                             for oline in str(t["output"]).split("\n"):
                                 md += f"> {oline}\n"
-                            md += f"> ```\n"
+                            md += "> ```\n"
                         md += "\n"
                     md += "---\n\n"
 

api/logs.py

@@ -4,6 +4,7 @@
 """
 
 import json
+import logging
 import os
 import re
 import sqlite3
@@ -16,6 +17,7 @@
 from utils.path_helpers import to_epoch_ms
 
 bp = Blueprint("logs", __name__)
+_logger = logging.getLogger(__name__)
 
 
 def _extract_chat_id_from_bubble_key(key: str) -> str | None:
@@ -69,8 +71,8 @@ def get_logs():
                         "type": "chat",
                         "messageCount": len(bubbles),
                     })
-            except Exception as e:
-                print(f"Error reading global storage: {e}")
+            except Exception:
+                _logger.exception("Error reading global storage")
 
         # Per-workspace (legacy)
         try:
@@ -133,9 +135,9 @@ def get_logs():
         except Exception:
             pass
 
-        logs.sort(key=lambda l: l.get("timestamp") or 0, reverse=True)
+        logs.sort(key=lambda log: log.get("timestamp") or 0, reverse=True)
         return jsonify({"logs": logs})
 
-    except Exception as e:
-        print(f"Failed to get logs: {e}")
+    except Exception:
+        _logger.exception("Failed to get logs")
         return jsonify({"error": "Failed to get logs", "logs": []}), 500

api/search.py

@@ -4,6 +4,7 @@
 """
 
 import json
+import logging
 import os
 import re
 import sqlite3
@@ -15,12 +16,13 @@
 
 from utils.exclusion_rules import build_searchable_text, is_excluded_by_rules
 from utils.workspace_path import resolve_workspace_path, get_cli_chats_path
-from utils.path_helpers import normalize_file_path, get_workspace_folder_paths, to_epoch_ms
+from utils.path_helpers import to_epoch_ms
 from utils.text_extract import extract_text_from_bubble
 from utils.cli_chat_reader import list_cli_projects, traverse_blobs, messages_to_bubbles
 from models import Bubble, Composer, SchemaError
 
 bp = Blueprint("search", __name__)
+_logger = logging.getLogger(__name__)
 
 
 def _json_dump_safe(value) -> str:
@@ -52,7 +54,7 @@ def _build_exclusion_searchable(
     metadata_parts: list[str] | None = None,
 ) -> str:
     """Build broad searchable text so exclusion rules cover visible output."""
-    combined = []
+    combined: list[str] = []
     if content_parts:
         combined.extend(p for p in content_parts if p)
     if metadata_parts:
@@ -243,7 +245,7 @@ def search():
                                 # Derive title from first bubble
                                 for text in bubble_texts:
                                     if text:
-                                        first_lines = [l for l in text.split("\n") if l.strip()]
+                                        first_lines = [ln for ln in text.split("\n") if ln.strip()]
                                         if first_lines:
                                             title = first_lines[0][:100]
                                         break
@@ -262,8 +264,8 @@ def search():
                     except Exception:
                         pass
 
-            except Exception as e:
-                print(f"Error searching global storage: {e}")
+            except Exception:
+                _logger.exception("Error searching global storage")
             finally:
                 if conn is not None:
                     conn.close()
@@ -447,8 +449,8 @@ def search():
                                 "type": "cli_agent",
                                 "source": "cli",
                             })
-            except Exception as e:
-                print(f"Error searching CLI sessions: {e}")
+            except Exception:
+                _logger.exception("Error searching CLI sessions")
 
         # Sort by timestamp descending
         def _ts(r):
@@ -463,6 +465,6 @@ def _ts(r):
 
         return jsonify({"results": results})
 
-    except Exception as e:
-        print(f"Search failed: {e}")
+    except Exception:
+        _logger.exception("Search failed")
         return jsonify({"error": "Search failed", "results": []}), 500

api/workspaces.py

@@ -8,6 +8,7 @@
 from __future__ import annotations
 
 import json
+import logging
 import os
 import re
 import sqlite3
@@ -32,10 +33,12 @@
     to_epoch_ms,
 )
 from utils.text_extract import extract_text_from_bubble, format_tool_action
+from utils.tool_parser import parse_tool_call as _parse_tool_call
 from utils.exclusion_rules import build_searchable_text, is_excluded_by_rules
 from models import Bubble, Composer, SchemaError, Workspace
 
 bp = Blueprint("workspaces", __name__)
+_logger = logging.getLogger(__name__)
 
 
 def _get_workspace_display_name(workspace_path: str, workspace_id: str) -> str:
@@ -674,7 +677,6 @@ def list_workspaces():
             primary = group[0]
             all_ws_ids = [e["name"] for e in group]
 
-            db_path = os.path.join(workspace_path, primary["name"], "state.vscdb")
             try:
                 mtime = max(
                     os.path.getmtime(os.path.join(workspace_path, e["name"], "state.vscdb"))
@@ -769,14 +771,14 @@ def list_workspaces():
                     ),
                     "source": "cli",
                 })
-        except Exception as e:
-            print(f"Failed to load CLI projects: {e}")
+        except Exception:
+            _logger.exception("Failed to load CLI projects")
 
         projects.sort(key=lambda p: p["lastModified"], reverse=True)
         return jsonify(projects)
 
-    except Exception as e:
-        print(f"Failed to get workspaces: {e}")
+    except Exception:
+        _logger.exception("Failed to get workspaces")
         return jsonify({"error": "Failed to get workspaces"}), 500
 
 
@@ -850,16 +852,15 @@ def get_workspace(workspace_id):
             "lastModified": datetime.fromtimestamp(mtime, tz=timezone.utc).isoformat(),
         })
 
-    except Exception as e:
-        print(f"Failed to get workspace: {e}")
+    except Exception:
+        _logger.exception("Failed to get workspace")
         return jsonify({"error": "Failed to get workspace"}), 500
 
 
 # ---------------------------------------------------------------------------
 # GET /api/workspaces/<id>/tabs
 # ---------------------------------------------------------------------------
 
-from utils.tool_parser import parse_tool_call as _parse_tool_call
 
 
 def _get_cli_workspace_tabs(workspace_id: str):
@@ -883,8 +884,8 @@ def _get_cli_workspace_tabs(workspace_id: str):
 
             try:
                 messages = traverse_blobs(session["db_path"])
-            except Exception as e:
-                print(f"CLI: could not read session {session_id}: {e}")
+            except Exception:  # noqa: BLE001 — best-effort per-session skip; one corrupted session must not 500 the endpoint, and the failure mode is logged with exc_info so the concrete type is preserved.
+                _logger.warning("CLI: could not read session %s", session_id, exc_info=True)
                 continue
 
             bubbles = messages_to_bubbles(messages, created_ms)
@@ -896,7 +897,7 @@ def _get_cli_workspace_tabs(workspace_id: str):
             if not title or title.startswith("New Agent"):
                 for b in bubbles:
                     if b["type"] == "user" and b.get("text"):
-                        first_lines = [l for l in b["text"].split("\n") if l.strip()]
+                        first_lines = [ln for ln in b["text"].split("\n") if ln.strip()]
                         if first_lines:
                             title = first_lines[0][:100]
                             if len(title) == 100:
@@ -948,8 +949,8 @@ def _get_cli_workspace_tabs(workspace_id: str):
         tabs.sort(key=lambda t: t.get("timestamp") or 0, reverse=True)
         return jsonify({"tabs": tabs})
 
-    except Exception as e:
-        print(f"Failed to get CLI workspace tabs: {e}")
+    except Exception:
+        _logger.exception("Failed to get CLI workspace tabs")
         return jsonify({"error": "Failed to get CLI workspace tabs"}), 500
 
 
@@ -1275,7 +1276,7 @@ def get_workspace_tabs(workspace_id):
                     if not cd.get("name") and bubbles:
                         first_msg = bubbles[0].get("text", "")
                         if first_msg:
-                            first_lines = [l for l in first_msg.split("\n") if l.strip()]
+                            first_lines = [ln for ln in first_msg.split("\n") if ln.strip()]
                             if first_lines:
                                 title = first_lines[0][:100]
                                 if len(title) == 100:
@@ -1418,14 +1419,14 @@ def get_workspace_tabs(workspace_id):
 
                     response["tabs"].append(tab)
 
-                except Exception as e:
-                    print(f"Error parsing composer data for {composer_id}: {e}")
+                except Exception:  # noqa: BLE001 — best-effort per-composer skip in a read-many loop; one malformed row must not 500 the tabs endpoint, and exc_info captures the concrete type for debugging.
+                    _logger.warning("Error parsing composer data for %s", composer_id, exc_info=True)
 
             # Sort tabs by timestamp descending (newest first)
             response["tabs"].sort(key=lambda t: t.get("timestamp") or 0, reverse=True)
 
             return jsonify(response)
 
-    except Exception as e:
-        print(f"Failed to get workspace tabs: {e}")
+    except Exception:
+        _logger.exception("Failed to get workspace tabs")
         return jsonify({"error": "Failed to get workspace tabs"}), 500

app.py

@@ -90,7 +90,6 @@ def favicon():
 
 if __name__ == "__main__":
     import argparse
-    import sys
 
     parser = argparse.ArgumentParser(description="Cursor Chat Browser (Python)")
     parser.add_argument("--port", type=int, default=3000)