v4: folder picker for Work drive + smart content-sniff scan (any .txt, any filename)

app.py

@@ -8,6 +8,8 @@ APP_DIR = r"C:\cookievault"
 DB_PATH = os.path.join(APP_DIR, "cookies.db")
 WATCH_DIR = os.path.join(APP_DIR, "incoming")
 SESSION_DIR = os.path.join(APP_DIR, "sessions")
+WORK_DRIVE = r"D:\\"          # the "Work" removable drive
+SCAN_ROOTS = [WATCH_DIR, WORK_DRIVE]
 os.makedirs(WATCH_DIR, exist_ok=True)
 os.makedirs(SESSION_DIR, exist_ok=True)
 
@@ -49,11 +51,28 @@ def parse_netscape(text):
         out.append((domain, flag, path, secure, expiry, name, value))
     return out
 
+def _safe_expiry(val):
+    try:
+        iv = int(float(val))
+        return iv if iv > 0 else 0
+    except (TypeError, ValueError):
+        return 0
+
 def import_file(path):
     fname = os.path.basename(path)
     with open(path, "r", encoding="utf-8", errors="replace") as f:
         text = f.read()
     cookies = parse_netscape(text)
+    # guard: every cookie must have sane types (name+value non-empty, expiry numeric-ish)
+    clean = []
+    for ck in cookies:
+        if not ck[5] or not ck[6]:
+            continue
+        if re.match(r"^[0-9]+$", ck[4].strip() or "0") is None:
+            # header junk line that survived parsing (e.g. split words) — skip
+            continue
+        clean.append(ck)
+    cookies = clean
     c = db()
     before = c.execute("SELECT COUNT(*) FROM cookies").fetchone()[0]
     for ck in cookies:
@@ -98,7 +117,10 @@ td.val{color:#9ece6a;max-width:240px;overflow:hidden;text-overflow:ellipsis;whit
     </div>
     <p><button type=submit>IMPORT</button> <span class=muted>also watches C:\cookievault\incoming &mdash; dump files there and hit rescan</span></p>
   </form>
-  <form method=post action=/scan style=display:inline><button class="btn gray" type=submit>RESCAN incoming folder</button></form>
+  <form method=post action=/scan style=display:inline><button class="btn gray" type=submit>RESCAN incoming</button></form>
+  <a class=btn href="/browse">&#128193; PICK FOLDER ON WORK DRIVE</a>
+  <form method=post action=/scan_work style=display:inline><button type=submit>&#128269; SMART-SCAN ALL OF D:\</button></form>
+  <span class=muted>smart scan: any .txt containing Netscape cookies gets imported, whatever its name</span>
   {% if msg %}<p class=ok>{{msg|safe}}</p>{% endif %}
 </div>
 <div class=box>
@@ -124,16 +146,48 @@ drop.addEventListener('drop',ev=>{m.files=ev.dataTransfer.files;document.getElem
 m.addEventListener('change',()=>document.getElementById('f').submit());
 </script></body></html>"""
 
+BROWSE_PAGE = r"""<!doctype html><html><head><title>Pick a folder — COOKIE VAULT</title><style>
+body{background:#0d0f14;color:#d7dae0;font-family:Consolas,monospace;margin:0}
+header{background:#141821;padding:18px 24px;border-bottom:1px solid #232a36}
+h1{margin:0 0 4px;font-size:20px;color:#7aa2f7;letter-spacing:2px}
+.crumb{color:#565f89;font-size:13px}
+.crumb a{color:#7aa2f7;text-decoration:none}
+main{max-width:800px;margin:24px auto;padding:0 16px}
+ul{list-style:none;padding:0}
+li{border-bottom:1px solid #1a2029}
+li a{display:block;padding:10px 8px;color:#e0af68;text-decoration:none;font-size:15px}
+li a:hover{background:#141821}
+.btnrow{margin-top:18px}
+button,.btn{background:#7aa2f7;color:#0d0f14;border:0;padding:9px 18px;border-radius:6px;font-weight:bold;cursor:pointer;font-family:inherit;font-size:14px;text-decoration:none;display:inline-block}
+.btn.green{background:#9ece6a}
+.muted{color:#565f89;font-size:13px}
+</style></head><body>
+<header><h1>&#128193; Pick a folder on WORK (D:\)</h1>
+<div class=crumb>D:\ {% if rel != '.' %}&raquo; {{rel}}{% endif %}</div></header>
+<main>
+{% if parent is not none %}<div style=margin-bottom:10px><a class=btn href="/browse?path={{parent|urlencode}}">&#8592; UP</a></div>{% endif %}
+<ul>
+{% for e in entries %}<li><a href="/browse?path={{(rel ~ '/' ~ e) if rel != '.' else e|urlencode}}">&#128193; {{e}}/</a></li>{% endfor %}
+{% if not entries %}<li class=muted style=padding:10px>no subfolders here</li>{% endif %}
+</ul>
+<div class=btnrow>
+  <form method=post action=/scan_folder><input type=hidden name=path value="{{rel}}">
+  <button class=green>SMART-SCAN THIS FOLDER &#128269;</button></form>
+  <a class=btn href="/" style=margin-left:8px>Cancel</a>
+</div>
+<p class=muted>Scans this folder + all subfolders. Any .txt that contains Netscape cookies (any filename) gets imported.</p>
+</main></body></html>"""
+
 @app.route("/", methods=["GET"])
 def index():
     q = request.args.get("q", "").strip()
     c = db()
     if q:
         like = f"%{q}%"
-        domains = c.execute("""SELECT domain, COUNT(*), MAX(CAST(expiry AS INTEGER)) FROM cookies
+        domains = c.execute("""SELECT domain, COUNT(*), MAX(CASE WHEN expiry GLOB '[0-9]*' THEN CAST(expiry AS INTEGER) ELSE 0 END) FROM cookies
                                WHERE domain LIKE ? GROUP BY domain ORDER BY domain LIMIT 500""", (like,)).fetchall()
     else:
-        domains = c.execute("""SELECT domain, COUNT(*), MAX(CAST(expiry AS INTEGER)) FROM cookies
+        domains = c.execute("""SELECT domain, COUNT(*), MAX(CASE WHEN expiry GLOB '[0-9]*' THEN CAST(expiry AS INTEGER) ELSE 0 END) FROM cookies
                                GROUP BY domain ORDER BY domain LIMIT 500""").fetchall()
     stats = (c.execute("SELECT COUNT(*) FROM cookies").fetchone()[0],
              c.execute("SELECT COUNT(DISTINCT domain) FROM cookies").fetchone()[0],
@@ -182,16 +236,98 @@ def do_import():
 
 @app.route("/scan", methods=["POST"])
 def scan():
-    msg, errs = [], []
+    return _scan_dirs(SCAN_ROOTS)
-    for fn in os.listdir(WATCH_DIR):
+
-        if fn.lower().endswith((".txt", ".json")):
+@app.route("/scan_work", methods=["POST"])
+def scan_work():
+    """Deep-scan the Work drive: every .txt file that LOOKS like Netscape cookie format."""
+    return _smart_scan(WORK_DRIVE)
+
+@app.route("/browse", methods=["GET"])
+def browse():
+    """List subfolders of a path on the Work drive so the user can pick one."""
+    sub = request.args.get("path", "").strip()
+    base = os.path.abspath(WORK_DRIVE)
+    target = os.path.abspath(os.path.join(base, sub.lstrip("/\\"))) if sub else base
+    if not target.startswith(base) or not os.path.isdir(target):
+        return redirect(url_for("index", msg="<span class=warn>invalid folder</span>"))
+    entries = []
     try:
-                n, new, total = import_file(os.path.join(WATCH_DIR, fn))
+        for name in sorted(os.listdir(target)):
+            full = os.path.join(target, name)
+            if os.path.isdir(full) and name not in ("$RECYCLE.BIN", "System Volume Information"):
+                entries.append(name)
+    except Exception as e:
+        return redirect(url_for("index", msg=f"<span class=warn>{e}</span>"))
+    rel = os.path.relpath(target, base)
+    parent = os.path.dirname(rel) if rel != "." else None
+    return render_template_string(BROWSE_PAGE, entries=entries, rel=rel, parent=parent)
+
+@app.route("/scan_folder", methods=["POST"])
+def scan_folder():
+    sub = request.form.get("path", "").strip()
+    base = os.path.abspath(WORK_DRIVE)
+    target = os.path.abspath(os.path.join(base, sub.lstrip("/\\"))) if sub else base
+    if not target.startswith(base) or not os.path.isdir(target):
+        return redirect(url_for("index", msg="<span class=warn>invalid folder</span>"))
+    return _smart_scan(target)
+
+def _smart_scan(root):
+    """Smart scan: walk root, sniff every .txt for Netscape format regardless of filename."""
+    found = []
+    for rootpath, dirs, files in os.walk(root):
+        dirs[:] = [d for d in dirs if d not in ("$RECYCLE.BIN", "System Volume Information", "node_modules")]
+        for fn in files:
+            if fn.lower().endswith(".txt"):
+                found.append(os.path.join(rootpath, fn))
+    if not found:
+        return redirect(url_for("index", msg="no .txt files found in that folder"))
+    msg, errs = [], []
+    new_total, imported_files = 0, 0
+    for p in found:
+        try:
+            looks_like = False
+            with open(p, "r", encoding="utf-8", errors="replace") as f:
+                for i, line in enumerate(f):
+                    line = line.strip()
+                    if not line or line.startswith("#"):
+                        continue
+                    parts = line.split("\t") if "\t" in line else line.split()
+                    if len(parts) >= 7 and parts[4].isdigit():
+                        looks_like = True
+                    if looks_like or i > 30:
+                        break
+            if not looks_like:
+                continue
+            n, new, total = import_file(p)
+            imported_files += 1
+            new_total += new
+            try:
+                shown = os.path.relpath(p, root)
+            except ValueError:
+                shown = p
+            msg.append(f"{shown}: {new} new")
+        except Exception as e:
+            errs.append(f"{p}: {e}")
+    if not imported_files and not errs:
+        return redirect(url_for("index", msg=f"scanned {len(found)} .txt files — none were Netscape cookie format"))
+    msg += ["<span class=warn>%s</span>" % e for e in errs]
+    return redirect(url_for("index", msg=f"<b>Scan of {root}:</b> {imported_files} cookie files of {len(found)} txt, {new_total} new cookies &middot; " + " &middot; ".join(msg[:25])))
+
+def _scan_dirs(dirs):
+    msg, errs = [], []
+    for d in dirs:
+        if not os.path.isdir(d):
+            continue
+        for fn in os.listdir(d):
+            if fn.lower().endswith((".txt", ".json")) and not fn.startswith("_pasted"):
+                try:
+                    n, new, total = import_file(os.path.join(d, fn))
                     msg.append(f"{fn}: {new} new")
                 except Exception as e:
                     errs.append(f"{fn}: {e}")
     msg += ["<span class=warn>%s</span>" % e for e in errs]
-    return redirect(url_for("index", msg=" &middot; ".join(msg) or "nothing new in incoming"))
+    return redirect(url_for("index", msg=" &middot; ".join(msg) or "nothing new found"))
 
 def _launch_login(domain, cookies):
     """Runs in background thread: selenium Chrome with cookies injected."""
@@ -215,18 +351,11 @@ def _launch_login(domain, cookies):
     time.sleep(1)
     for ck in cookies:
         c = {"name": ck[1], "value": ck[2]}
-        dom = ck[0]
+        c["domain"] = ck[0]
-        if not dom.startswith("."):
-            c["domain"] = dom
-        else:
-            c["domain"] = dom
         c["path"] = ck[3] or "/"
-        try:
+        exp = _safe_expiry(ck[4])
-            exp = int(ck[4])
         if exp > 0:
             c["expiry"] = exp
-        except Exception:
-            pass
         c["secure"] = str(ck[5]).upper() == "TRUE"
         try:
             driver.add_cookie(c)
@@ -271,8 +400,8 @@ def export():
         rows = c.execute("SELECT domain,name,value,path,expiry,secure FROM cookies").fetchall()
     c.close()
     jar = [{"domain": r[0], "name": r[1], "value": r[2], "path": r[3],
-            "expirationDate": int(r[4]) if r[4] and int(r[4]) > 0 else None,
+            "expirationDate": _safe_expiry(r[4]) or None,
-            "secure": r[5].upper() == "TRUE", "httpOnly": False} for r in rows]
+            "secure": str(r[5]).upper() == "TRUE", "httpOnly": False} for r in rows]
     return Response(json.dumps(jar, indent=2), mimetype="application/json",
                     headers={"Content-Disposition": "attachment; filename=cookies.json"})
 
@@ -287,7 +416,7 @@ def export_ns():
         rows = c.execute("SELECT domain,flag,path,secure,expiry,name,value FROM cookies").fetchall()
     c.close()
     lines = ["# Netscape HTTP Cookie File", "# Exported by COOKIE VAULT"]
-    lines += ["\t".join(str(x) for x in r) for r in rows]
+    lines += ["\t".join(str(x) for x in r) for r in rows if _safe_expiry(r[4]) >= 0]
     return Response("\n".join(lines) + "\n", mimetype="text/plain",
                     headers={"Content-Disposition": "attachment; filename=cookies.txt"})