fix: harden repair paths, slugify edge case, OCR meta resilience, venv python detection, and repair exit handling

paperforge/commands/repair.py

@@ -32,11 +32,15 @@ def run(args: argparse.Namespace) -> int:
     vault = getattr(args, "vault_path", None)
     paths = getattr(args, "paths", None)
     if vault is None:
-        from paperforge.config import load_vault_config, paperforge_paths, resolve_vault
+        from paperforge.config import resolve_vault
+        from paperforge.worker._utils import pipeline_paths
 
         vault = resolve_vault(cli_vault=getattr(args, "vault", None))
-        cfg = load_vault_config(vault)
-        paths = paperforge_paths(vault, cfg)
+        paths = pipeline_paths(vault)
+    elif paths is None or "config" not in paths:
+        from paperforge.worker._utils import pipeline_paths
+
+        paths = pipeline_paths(vault)
 
     run_repair = _get_run_repair()
     result = run_repair(

paperforge/plugin/main.js

@@ -1,5 +1,5 @@
 const { Plugin, Notice, ItemView, Modal, Setting, PluginSettingTab, addIcon } = require('obsidian');
-const { exec } = require('node:child_process');
+const { exec, execFile } = require('node:child_process');
 const fs = require('fs');
 const path = require('path');
 
@@ -185,6 +185,7 @@ const ACTIONS = [
         desc: 'Fix three-way state divergence, path errors, and rebuild index',
         icon: '\u21BA',  // ↺
         cmd: 'repair',
+        args: ['--fix', '--fix-paths'],
         okMsg: 'Repair complete',
     },
     {
@@ -207,6 +208,20 @@ const ACTIONS = [
     },
 ];
 
+function resolvePythonExecutable(vaultPath) {
+    const candidates = [
+        path.join(vaultPath, '.paperforge-test-venv', 'Scripts', 'python.exe'),
+        path.join(vaultPath, '.venv', 'Scripts', 'python.exe'),
+        path.join(vaultPath, 'venv', 'Scripts', 'python.exe'),
+    ];
+    for (const candidate of candidates) {
+        try {
+            if (fs.existsSync(candidate)) return candidate;
+        } catch {}
+    }
+    return 'python';
+}
+
 class PaperForgeStatusView extends ItemView {
     constructor(leaf) {
         super(leaf);
@@ -375,7 +390,8 @@ class PaperForgeStatusView extends ItemView {
             if (!quiet && !this._cachedStats) {
                 this._metricsEl.createEl('div', { cls: 'paperforge-status-loading', text: 'No index \u2014 trying CLI...' });
             }
-            exec('python -m paperforge status --json', { cwd: vp, timeout: 30000 }, (err2, stdout) => {
+            const pythonExe = resolvePythonExecutable(vp);
+            execFile(pythonExe, ['-m', 'paperforge', 'status', '--json'], { cwd: vp, timeout: 30000 }, (err2, stdout) => {
                 if (err2) {
                     if (this._cachedStats) return;
                     this._metricsEl.createEl('div', { cls: 'paperforge-status-error', text: 'Cannot reach PaperForge CLI.\nMake sure paperforge is installed and in your PATH.' });
@@ -1262,7 +1278,7 @@ class PaperForgeStatusView extends ItemView {
         this._showMessage('Processing...', 'running');
 
         // Resolve extra arguments based on action flags
-        let extraArgs = [];
+        let extraArgs = Array.isArray(a.args) ? [...a.args] : [];
 
         if (a.needsKey) {
             // Resolve zotero_key from active file frontmatter
@@ -1289,17 +1305,18 @@ class PaperForgeStatusView extends ItemView {
                 card.removeClass('running');
                 return;
             }
-            extraArgs = [key];
+            extraArgs = [...extraArgs, key];
         }
 
         if (a.needsFilter) {
             // Default to --all for the initial implementation
-            extraArgs = ['--all'];
+            extraArgs = [...extraArgs, '--all'];
         }
 
         const { spawn } = require('node:child_process');
         const cmdTimeout = a.needsFilter ? 60000 : (a.needsKey ? 30000 : 600000);
-        const child = spawn('python', ['-m', 'paperforge', a.cmd, ...extraArgs], { cwd: vp, timeout: cmdTimeout });
+        const pythonExe = resolvePythonExecutable(vp);
+        const child = spawn(pythonExe, ['-m', 'paperforge', a.cmd, ...extraArgs], { cwd: vp, timeout: cmdTimeout });
         const log = [];
         const startTime = Date.now();
         const pollTimer = setInterval(() => this._fetchStats(true), 4000);
@@ -1324,8 +1341,14 @@ class PaperForgeStatusView extends ItemView {
             const elapsed = ((Date.now() - startTime) / 1000).toFixed(1);
             if (code !== 0) {
                 const last = log.slice(-3).join(' | ') || 'exit code ' + code;
-                this._showMessage('[!!] ' + last, 'error');
-                new Notice('[!!] ' + a.cmd + ' failed: ' + last, 8000);
+                if (a.cmd === 'repair' && code === 1) {
+                    this._showMessage('[WARN] ' + last, 'running');
+                    new Notice('[WARN] Repair completed with remaining issues: ' + last, 8000);
+                    this._fetchStats(true);
+                } else {
+                    this._showMessage('[!!] ' + last, 'error');
+                    new Notice('[!!] ' + a.cmd + ' failed: ' + last, 8000);
+                }
             } else if (a.needsKey || a.needsFilter) {
                 // Context actions: copy JSON output to clipboard
                 const output = log.join('\n');

paperforge/plugin/styles.css

@@ -1162,7 +1162,7 @@
 }
 
 .paperforge-paper-year::before {
-    content: " \u00B7  ";
+    content: " · ";
 }
 
 .paperforge-paper-actions {

paperforge/worker/_utils.py

@@ -124,7 +124,8 @@ def yaml_list(key: str, values) -> list[str]:
 
 def slugify_filename(text: str) -> str:
     cleaned = re.sub('[<>:"/\\\\|?*]+', "", text).strip()
-    return cleaned[:120] or "untitled"
+    cleaned = cleaned[:120].rstrip(" .")
+    return cleaned or "untitled"
 
 
 def _extract_year(value: str) -> str:

paperforge/worker/ocr.py

@@ -65,6 +65,13 @@ def ensure_ocr_meta(vault: Path, row: dict) -> dict:
     return meta
 
 
+def _read_meta_or_empty(meta_path: Path) -> dict:
+    try:
+        return read_json(meta_path) if meta_path.exists() else {}
+    except Exception:
+        return {}
+
+
 def validate_ocr_meta(paths: dict[str, Path], meta: dict) -> tuple[str, str]:
     status = str(meta.get("ocr_status", "pending") or "pending").strip().lower()
     if status != "done":
@@ -161,7 +168,7 @@ def sync_ocr_queue(paths: dict[str, Path], target_rows: list[dict]) -> list[dict
         if not target.get("has_pdf"):
             continue
         meta_path = paths["ocr"] / key / "meta.json"
-        meta = read_json(meta_path) if meta_path.exists() else {}
+        meta = _read_meta_or_empty(meta_path)
         status = str(meta.get("ocr_status", "pending") or "pending").strip().lower()
         if status in {"done", "blocked"}:
             continue
@@ -181,7 +188,7 @@ def sync_ocr_queue(paths: dict[str, Path], target_rows: list[dict]) -> list[dict
         if not row.get("has_pdf"):
             continue
         meta_path = paths["ocr"] / key / "meta.json"
-        meta = read_json(meta_path) if meta_path.exists() else {}
+        meta = _read_meta_or_empty(meta_path)
         status = str(meta.get("ocr_status", "pending") or "pending").strip().lower()
         if status in {"done", "blocked"}:
             continue

paperforge/worker/sync.py

@@ -618,8 +618,8 @@ def has_deep_reading_content(text: str) -> bool:
 def library_record_markdown(row: dict) -> str:
     lines = [
         "---",
-        f"zotero_key: {row.get('zotero_key', '')}",
-        f"domain: {row.get('domain', '')}",
+        f"zotero_key: {yaml_quote(row.get('zotero_key', ''))}",
+        f"domain: {yaml_quote(row.get('domain', ''))}",
         f"title: {yaml_quote(row.get('title', ''))}",
         f"year: {row.get('year', '')}",
         f"doi: {yaml_quote(row.get('doi', ''))}",

tests/test_e2e_pipeline.py

@@ -18,7 +18,7 @@ import pytest
 
 from paperforge.config import paperforge_paths
 from paperforge.worker._utils import scan_library_records
-from paperforge.worker.sync import run_index_refresh, run_selection_sync
+from paperforge.worker.sync import run_index_refresh, run_selection_sync, update_frontmatter_field
 
 
 class TestSelectionSyncProducesLibraryRecords:
@@ -164,9 +164,16 @@ class TestIndexRefreshProducesFormalNotes:
 class TestOcrQueueStates:
     """E2E: OCR queue and state transitions are correct after pipeline run."""
 
+    def _set_analyze_true(self, test_vault: Path) -> None:
+        paths = paperforge_paths(test_vault)
+        record_path = paths["library_records"] / "骨科" / "TSTONE001.md"
+        text = record_path.read_text(encoding="utf-8")
+        record_path.write_text(update_frontmatter_field(text, "analyze", True), encoding="utf-8")
+
     def test_scan_library_records_returns_tstone001(self, test_vault: Path) -> None:
         """Verify TSTONE001 appears in library records scan with analyze=true."""
         run_selection_sync(test_vault)
+        self._set_analyze_true(test_vault)
         records = scan_library_records(test_vault)
         tstone = [r for r in records if r["zotero_key"] == "TSTONE001"]
         assert len(tstone) >= 1, "TSTONE001 should be in library records scan"
@@ -176,6 +183,7 @@ class TestOcrQueueStates:
     def test_tstone001_ocr_status_done(self, test_vault: Path) -> None:
         """Verify TSTONE001 has OCR status 'done' (fixture has meta.json with ocr_status: done)."""
         run_selection_sync(test_vault)
+        self._set_analyze_true(test_vault)
         records = scan_library_records(test_vault)
         tstone = [r for r in records if r["zotero_key"] == "TSTONE001"]
         assert len(tstone) >= 1
@@ -210,6 +218,7 @@ class TestOcrQueueStates:
             )
 
         # Stage 4: OCR state is consistent
+        self._set_analyze_true(test_vault)
         records = scan_library_records(test_vault)
         tstone = [r for r in records if r["zotero_key"] == "TSTONE001"]
         assert len(tstone) >= 1

tests/test_ocr_state_machine.py

@@ -11,6 +11,7 @@ from pathlib import Path
 from unittest.mock import MagicMock, patch
 
 import pytest
+import requests
 
 
 def _make_vault(tmp_path: Path) -> tuple[Path, Path, Path, Path]:

tests/test_repair.py

@@ -2,8 +2,10 @@
 
 from __future__ import annotations
 
+import argparse
 import json
 import re
+from unittest.mock import patch
 
 from paperforge.worker.repair import run_repair
 from paperforge.worker.sync import pipeline_paths
@@ -438,3 +440,43 @@ class TestRunRepairReturnStructure:
         assert "formal_note_ocr_status" in item
         assert "meta_ocr_status" in item
         assert "reason" in item
+
+
+class TestRepairCommandPaths:
+    def test_repair_command_uses_pipeline_paths_with_config(self, tmp_path):
+        vault = _make_vault(tmp_path)
+        captured = {}
+
+        def _fake_run_repair(vault_arg, paths_arg, verbose=False, fix=False, fix_paths=False):
+            captured["vault"] = vault_arg
+            captured["paths"] = paths_arg
+            return {"scanned": 0, "divergent": [], "fixed": 0, "errors": [], "path_errors": {"total": 0}, "rebuilt": 0}
+
+        from paperforge.commands import repair as repair_cmd
+
+        args = argparse.Namespace(vault_path=None, paths=None, vault=str(vault), verbose=False, fix=False, fix_paths=False)
+        with patch.object(repair_cmd, "_get_run_repair", return_value=_fake_run_repair):
+            code = repair_cmd.run(args)
+
+        assert code == 0
+        assert captured["vault"] == vault
+        assert "config" in captured["paths"]
+
+    def test_repair_command_upgrades_legacy_paths_arg(self, tmp_path):
+        vault = _make_vault(tmp_path)
+        captured = {}
+
+        def _fake_run_repair(vault_arg, paths_arg, verbose=False, fix=False, fix_paths=False):
+            captured["paths"] = paths_arg
+            return {"scanned": 0, "divergent": [], "fixed": 0, "errors": [], "path_errors": {"total": 0}, "rebuilt": 0}
+
+        from paperforge.commands import repair as repair_cmd
+        from paperforge.config import paperforge_paths
+
+        legacy_paths = paperforge_paths(vault)
+        args = argparse.Namespace(vault_path=vault, paths=legacy_paths, vault=str(vault), verbose=False, fix=False, fix_paths=False)
+        with patch.object(repair_cmd, "_get_run_repair", return_value=_fake_run_repair):
+            code = repair_cmd.run(args)
+
+        assert code == 0
+        assert "config" in captured["paths"]

tests/test_utils_slugify.py

@@ -53,6 +53,16 @@ class TestSlugifyFilename:
     def test_leading_trailing_spaces(self) -> None:
         assert slugify_filename("  hello  ") == "hello"
 
+    def test_truncation_does_not_leave_trailing_space(self) -> None:
+        title = (
+            "Dynamics and Crosstalk between Gut Microbiota, Metabolome, and Fecal "
+            "Calprotectin in Very Preterm Infants: Insights into Feeding Intolerance"
+        )
+        assert slugify_filename(title) == (
+            "Dynamics and Crosstalk between Gut Microbiota, Metabolome, and Fecal "
+            "Calprotectin in Very Preterm Infants Insights into"
+        )
+
 
 class TestExtractYear:
     """_extract_year(value: str) -> str"""