v0.11.6

README.md

@@ -294,6 +294,8 @@ the same arguments can be set as volume flags, in addition to `d2d` and `d2t` fo
 
 `e2tsr` is probably always overkill, since `e2ds`/`e2dsa` would pick up any file modifications and cause `e2ts` to reindex those
 
+the rescan button in the admin panel has no effect unless the volume has `-e2ds` or higher
+
 
 ## metadata from audio files
 

bin/README.md

@@ -45,3 +45,18 @@ you could replace winfsp with [dokan](https://github.com/dokan-dev/dokany/releas
 # [`mtag/`](mtag/)
 * standalone programs which perform misc. file analysis
 * copyparty can Popen programs like these during file indexing to collect additional metadata
+
+
+# [`dbtool.py`](dbtool.py)
+upgrade utility which can show db info and help transfer data between databases, for example when a new version of copyparty recommends to wipe the DB and reindex because it now collects additional metadata during analysis, but you have some really expensive `-mtp` parsers and want to copy over the tags from the old db
+
+for that example (upgrading to v0.11.0), first move the old db aside, launch copyparty, let it rebuild the db until the point where it starts running mtp (colored messages as it adds the mtp tags), then CTRL-C and patch in the old mtp tags from the old db instead
+
+so assuming you have `-mtp` parsers to provide the tags `key` and `.bpm`:
+
+```
+~/bin/dbtool.py -ls up2k.db
+~/bin/dbtool.py -src up2k.db.v0.10.22 up2k.db -cmp
+~/bin/dbtool.py -src up2k.db.v0.10.22 up2k.db -rm-mtp-flag -copy key
+~/bin/dbtool.py -src up2k.db.v0.10.22 up2k.db -rm-mtp-flag -copy .bpm -vac
+```

bin/dbtool.py

@@ -0,0 +1,198 @@
+#!/usr/bin/env python3
+
+import os
+import sys
+import sqlite3
+import argparse
+
+DB_VER = 3
+
+
+def die(msg):
+    print("\033[31m\n" + msg + "\n\033[0m")
+    sys.exit(1)
+
+
+def read_ver(db):
+    for tab in ["ki", "kv"]:
+        try:
+            c = db.execute(r"select v from {} where k = 'sver'".format(tab))
+        except:
+            continue
+
+        rows = c.fetchall()
+        if rows:
+            return int(rows[0][0])
+
+    return "corrupt"
+
+
+def ls(db):
+    nfiles = next(db.execute("select count(w) from up"))[0]
+    ntags = next(db.execute("select count(w) from mt"))[0]
+    print(f"{nfiles} files")
+    print(f"{ntags} tags\n")
+
+    print("number of occurences for each tag,")
+    print(" 'x' = file has no tags")
+    print(" 't:mtp' = the mtp flag (file not mtp processed yet)")
+    print()
+    for k, nk in db.execute("select k, count(k) from mt group by k order by k"):
+        print(f"{nk:9} {k}")
+
+
+def compare(n1, d1, n2, d2, verbose):
+    nt = next(d1.execute("select count(w) from up"))[0]
+    n = 0
+    miss = 0
+    for w, rd, fn in d1.execute("select w, rd, fn from up"):
+        n += 1
+        if n % 25_000 == 0:
+            m = f"\033[36mchecked {n:,} of {nt:,} files in {n1} against {n2}\033[0m"
+            print(m)
+
+        q = "select w from up where substr(w,1,16) = ?"
+        hit = d2.execute(q, (w[:16],)).fetchone()
+        if not hit:
+            miss += 1
+            if verbose:
+                print(f"file in {n1} missing in {n2}: [{w}] {rd}/{fn}")
+
+    print(f" {miss} files in {n1} missing in {n2}\n")
+
+    nt = next(d1.execute("select count(w) from mt"))[0]
+    n = 0
+    miss = {}
+    nmiss = 0
+    for w, k, v in d1.execute("select * from mt"):
+        n += 1
+        if n % 100_000 == 0:
+            m = f"\033[36mchecked {n:,} of {nt:,} tags in {n1} against {n2}, so far {nmiss} missing tags\033[0m"
+            print(m)
+
+        v2 = d2.execute("select v from mt where w = ? and +k = ?", (w, k)).fetchone()
+        if v2:
+            v2 = v2[0]
+
+        # if v != v2 and v2 and k in [".bpm", "key"] and n2 == "src":
+        #    print(f"{w} [{rd}/{fn}] {k} = [{v}] / [{v2}]")
+
+        if v2 is not None:
+            if k.startswith("."):
+                try:
+                    diff = abs(float(v) - float(v2))
+                    if diff > float(v) / 0.9:
+                        v2 = None
+                    else:
+                        v2 = v
+                except:
+                    pass
+
+            if v != v2:
+                v2 = None
+
+        if v2 is None:
+            nmiss += 1
+            try:
+                miss[k] += 1
+            except:
+                miss[k] = 1
+
+            if verbose:
+                q = "select rd, fn from up where substr(w,1,16) = ?"
+                rd, fn = d1.execute(q, (w,)).fetchone()
+                print(f"missing in {n2}: [{w}] [{rd}/{fn}] {k} = {v}")
+
+    for k, v in sorted(miss.items()):
+        if v:
+            print(f"{n1} has {v:6} more {k:<6} tags than {n2}")
+
+    print(f"in total, {nmiss} missing tags in {n2}\n")
+
+
+def copy_mtp(d1, d2, tag, rm):
+    nt = next(d1.execute("select count(w) from mt where k = ?", (tag,)))[0]
+    n = 0
+    ndone = 0
+    for w, k, v in d1.execute("select * from mt where k = ?", (tag,)):
+        n += 1
+        if n % 25_000 == 0:
+            m = f"\033[36m{n:,} of {nt:,} tags checked, so far {ndone} copied\033[0m"
+            print(m)
+
+        hit = d2.execute("select v from mt where w = ? and +k = ?", (w, k)).fetchone()
+        if hit:
+            hit = hit[0]
+
+        if hit != v:
+            ndone += 1
+            if hit is not None:
+                d2.execute("delete from mt where w = ? and +k = ?", (w, k))
+
+            d2.execute("insert into mt values (?,?,?)", (w, k, v))
+            if rm:
+                d2.execute("delete from mt where w = ? and +k = 't:mtp'", (w,))
+
+    d2.commit()
+    print(f"copied {ndone} {tag} tags over")
+
+
+def main():
+    os.system("")
+    print()
+    
+    ap = argparse.ArgumentParser()
+    ap.add_argument("db", help="database to work on")
+    ap.add_argument("-src", metavar="DB", type=str, help="database to copy from")
+
+    ap2 = ap.add_argument_group("informational / read-only stuff")
+    ap2.add_argument("-v", action="store_true", help="verbose")
+    ap2.add_argument("-ls", action="store_true", help="list summary for db")
+    ap2.add_argument("-cmp", action="store_true", help="compare databases")
+
+    ap2 = ap.add_argument_group("options which modify target db")
+    ap2.add_argument("-copy", metavar="TAG", type=str, help="mtp tag to copy over")
+    ap2.add_argument(
+        "-rm-mtp-flag",
+        action="store_true",
+        help="when an mtp tag is copied over, also mark that as done, so copyparty won't run mtp on it",
+    )
+    ap2.add_argument("-vac", action="store_true", help="optimize DB")
+
+    ar = ap.parse_args()
+
+    for v in [ar.db, ar.src]:
+        if v and not os.path.exists(v):
+            die("database must exist")
+
+    db = sqlite3.connect(ar.db)
+    ds = sqlite3.connect(ar.src) if ar.src else None
+
+    for d, n in [[ds, "src"], [db, "dst"]]:
+        if not d:
+            continue
+
+        ver = read_ver(d)
+        if ver == "corrupt":
+            die("{} database appears to be corrupt, sorry")
+
+        if ver != DB_VER:
+            m = f"{n} db is version {ver}, this tool only supports version {DB_VER}, please upgrade it with copyparty first"
+            die(m)
+
+    if ar.ls:
+        ls(db)
+
+    if ar.cmp:
+        if not ds:
+            die("need src db to compare against")
+
+        compare("src", ds, "dst", db, ar.v)
+        compare("dst", db, "src", ds, ar.v)
+
+    if ar.copy:
+        copy_mtp(ds, db, ar.copy, ar.rm_mtp_flag)
+
+
+if __name__ == "__main__":
+    main()

copyparty/__main__.py

@@ -249,6 +249,10 @@ def run_argparse(argv, formatter):
     ap.add_argument("--urlform", metavar="MODE", type=str, default="print,get", help="how to handle url-forms")
     ap.add_argument("--salt", type=str, default="hunter2", help="up2k file-hash salt")
 
+    ap2 = ap.add_argument_group('admin panel options')
+    ap2.add_argument("--no-rescan", action="store_true", help="disable ?scan (volume reindexing)")
+    ap2.add_argument("--no-stack", action="store_true", help="disable ?stack (list all stacks)")
+
     ap2 = ap.add_argument_group('thumbnail options')
     ap2.add_argument("--no-thumb", action="store_true", help="disable all thumbnails")
     ap2.add_argument("--no-vthumb", action="store_true", help="disable video thumbnails")
@@ -257,7 +261,7 @@ def run_argparse(argv, formatter):
     ap2.add_argument("--th-no-jpg", action="store_true", help="disable jpg output")
     ap2.add_argument("--th-no-webp", action="store_true", help="disable webp output")
     ap2.add_argument("--th-poke", metavar="SEC", type=int, default=300, help="activity labeling cooldown")
-    ap2.add_argument("--th-clean", metavar="SEC", type=int, default=1800, help="cleanup interval")
+    ap2.add_argument("--th-clean", metavar="SEC", type=int, default=43200, help="cleanup interval")
     ap2.add_argument("--th-maxage", metavar="SEC", type=int, default=604800, help="max folder age")
 
     ap2 = ap.add_argument_group('database options')
@@ -287,6 +291,7 @@ def run_argparse(argv, formatter):
     ap2.add_argument("--log-conn", action="store_true", help="print tcp-server msgs")
     ap2.add_argument("--no-sendfile", action="store_true", help="disable sendfile")
     ap2.add_argument("--no-scandir", action="store_true", help="disable scandir")
+    ap2.add_argument("--no-fastboot", action="store_true", help="wait for up2k indexing")
     ap2.add_argument("--ihead", metavar="HEADER", action='append', help="dump incoming header")
     ap2.add_argument("--lf-url", metavar="RE", type=str, default=r"^/\.cpr/|\?th=[wj]$", help="dont log URLs matching")
     

copyparty/__version__.py

@@ -1,8 +1,8 @@
 # coding: utf-8
 
-VERSION = (0, 11, 0)
+VERSION = (0, 11, 6)
 CODENAME = "the grid"
-BUILD_DT = (2021, 5, 29)
+BUILD_DT = (2021, 6, 1)
 
 S_VERSION = ".".join(map(str, VERSION))
 S_BUILD_DT = "{0:04d}-{1:02d}-{2:02d}".format(*BUILD_DT)

copyparty/authsrv.py

@@ -14,11 +14,12 @@ from .util import IMPLICATIONS, undot, Pebkac, fsdec, fsenc, statdir, nuprint
 class VFS(object):
     """single level in the virtual fs"""
 
-    def __init__(self, realpath, vpath, uread=[], uwrite=[], flags={}):
+    def __init__(self, realpath, vpath, uread=[], uwrite=[], uadm=[], flags={}):
         self.realpath = realpath  # absolute path on host filesystem
         self.vpath = vpath  # absolute path in the virtual filesystem
         self.uread = uread  # users who can read this
         self.uwrite = uwrite  # users who can write this
+        self.uadm = uadm  # users who are regular admins
         self.flags = flags  # config switches
         self.nodes = {}  # child nodes
         self.all_vols = {vpath: self}  # flattened recursive
@@ -27,7 +28,7 @@ class VFS(object):
         return "VFS({})".format(
             ", ".join(
                 "{}={!r}".format(k, self.__dict__[k])
-                for k in "realpath vpath uread uwrite flags".split()
+                for k in "realpath vpath uread uwrite uadm flags".split()
             )
         )
 
@@ -52,6 +53,7 @@ class VFS(object):
                 "{}/{}".format(self.vpath, name).lstrip("/"),
                 self.uread,
                 self.uwrite,
+                self.uadm,
                 self.flags,
             )
             self._trk(vn)
@@ -133,7 +135,7 @@ class VFS(object):
             #
             return os.path.realpath(rp)
 
-    def ls(self, rem, uname, scandir, lstat=False):
+    def ls(self, rem, uname, scandir, incl_wo=False, lstat=False):
         """return user-readable [fsdir,real,virt] items at vpath"""
         virt_vis = {}  # nodes readable by user
         abspath = self.canonical(rem)
@@ -141,12 +143,12 @@ class VFS(object):
         real.sort()
         if not rem:
             for name, vn2 in sorted(self.nodes.items()):
-                if (
-                    uname in vn2.uread
-                    or "*" in vn2.uread
-                    or uname in vn2.uwrite
-                    or "*" in vn2.uwrite
-                ):
+                ok = uname in vn2.uread or "*" in vn2.uread
+
+                if not ok and incl_wo:
+                    ok = uname in vn2.uwrite or "*" in vn2.uwrite
+
+                if ok:
                     virt_vis[name] = vn2
 
             # no vfs nodes in the list of real inodes
@@ -160,7 +162,7 @@ class VFS(object):
         rel is a unix-style user-defined vpath (not vfs-related)
         """
 
-        fsroot, vfs_ls, vfs_virt = self.ls(rem, uname, scandir, lstat)
+        fsroot, vfs_ls, vfs_virt = self.ls(rem, uname, scandir, False, lstat)
         rfiles = [x for x in vfs_ls if not stat.S_ISDIR(x[1].st_mode)]
         rdirs = [x for x in vfs_ls if stat.S_ISDIR(x[1].st_mode)]
 
@@ -226,15 +228,19 @@ class VFS(object):
             for f in [{"vp": v, "ap": a, "st": n[1]} for v, a, n in files]:
                 yield f
 
-    def user_tree(self, uname, readable=False, writable=False):
+    def user_tree(self, uname, readable=False, writable=False, admin=False):
         ret = []
         opt1 = readable and (uname in self.uread or "*" in self.uread)
         opt2 = writable and (uname in self.uwrite or "*" in self.uwrite)
-        if opt1 or opt2:
-            ret.append(self.vpath)
+        if admin:
+            if opt1 and opt2:
+                ret.append(self.vpath)
+        else:
+            if opt1 or opt2:
+                ret.append(self.vpath)
 
         for _, vn in sorted(self.nodes.items()):
-            ret.extend(vn.user_tree(uname, readable, writable))
+            ret.extend(vn.user_tree(uname, readable, writable, admin))
 
         return ret
 
@@ -269,7 +275,7 @@ class AuthSrv(object):
 
         yield prev, True
 
-    def _parse_config_file(self, fd, user, mread, mwrite, mflags, mount):
+    def _parse_config_file(self, fd, user, mread, mwrite, madm, mflags, mount):
         vol_src = None
         vol_dst = None
         self.line_ctr = 0
@@ -301,6 +307,7 @@ class AuthSrv(object):
                 mount[vol_dst] = vol_src
                 mread[vol_dst] = []
                 mwrite[vol_dst] = []
+                madm[vol_dst] = []
                 mflags[vol_dst] = {}
                 continue
 
@@ -311,10 +318,15 @@ class AuthSrv(object):
                 uname = "*"
 
             self._read_vol_str(
-                lvl, uname, mread[vol_dst], mwrite[vol_dst], mflags[vol_dst]
+                lvl,
+                uname,
+                mread[vol_dst],
+                mwrite[vol_dst],
+                madm[vol_dst],
+                mflags[vol_dst],
             )
 
-    def _read_vol_str(self, lvl, uname, mr, mw, mf):
+    def _read_vol_str(self, lvl, uname, mr, mw, ma, mf):
         if lvl == "c":
             cval = True
             if "=" in uname:
@@ -332,6 +344,9 @@ class AuthSrv(object):
         if lvl in "wa":
             mw.append(uname)
 
+        if lvl == "a":
+            ma.append(uname)
+
     def _read_volflag(self, flags, name, value, is_list):
         if name not in ["mtp"]:
             flags[name] = value
@@ -355,6 +370,7 @@ class AuthSrv(object):
         user = {}  # username:password
         mread = {}  # mountpoint:[username]
         mwrite = {}  # mountpoint:[username]
+        madm = {}  # mountpoint:[username]
         mflags = {}  # mountpoint:[flag]
         mount = {}  # dst:src (mountpoint:realpath)
 
@@ -378,17 +394,22 @@ class AuthSrv(object):
                 mount[dst] = src
                 mread[dst] = []
                 mwrite[dst] = []
+                madm[dst] = []
                 mflags[dst] = {}
 
                 perms = perms.split(":")
                 for (lvl, uname) in [[x[0], x[1:]] for x in perms]:
-                    self._read_vol_str(lvl, uname, mread[dst], mwrite[dst], mflags[dst])
+                    self._read_vol_str(
+                        lvl, uname, mread[dst], mwrite[dst], madm[dst], mflags[dst]
+                    )
 
         if self.args.c:
             for cfg_fn in self.args.c:
                 with open(cfg_fn, "rb") as f:
                     try:
-                        self._parse_config_file(f, user, mread, mwrite, mflags, mount)
+                        self._parse_config_file(
+                            f, user, mread, mwrite, madm, mflags, mount
+                        )
                     except:
                         m = "\n\033[1;31m\nerror in config file {} on line {}:\n\033[0m"
                         print(m.format(cfg_fn, self.line_ctr))
@@ -410,12 +431,15 @@ class AuthSrv(object):
 
             if dst == "":
                 # rootfs was mapped; fully replaces the default CWD vfs
-                vfs = VFS(mount[dst], dst, mread[dst], mwrite[dst], mflags[dst])
+                vfs = VFS(
+                    mount[dst], dst, mread[dst], mwrite[dst], madm[dst], mflags[dst]
+                )
                 continue
 
             v = vfs.add(mount[dst], dst)
             v.uread = mread[dst]
             v.uwrite = mwrite[dst]
+            v.uadm = madm[dst]
             v.flags = mflags[dst]
 
         missing_users = {}

copyparty/httpcli.py

@@ -10,6 +10,7 @@ import json
 import string
 import socket
 import ctypes
+import traceback
 from datetime import datetime
 import calendar
 
@@ -155,6 +156,7 @@ class HttpCli(object):
         if self.uname:
             self.rvol = self.auth.vfs.user_tree(self.uname, readable=True)
             self.wvol = self.auth.vfs.user_tree(self.uname, writable=True)
+            self.avol = self.auth.vfs.user_tree(self.uname, True, True, True)
 
         ua = self.headers.get("user-agent", "")
         self.is_rclone = ua.startswith("rclone/")
@@ -326,6 +328,12 @@ class HttpCli(object):
             self.vpath = None
             return self.tx_mounts()
 
+        if "scan" in self.uparam:
+            return self.scanvol()
+
+        if "stack" in self.uparam:
+            return self.tx_stack()
+
         return self.tx_browser()
 
     def handle_options(self):
@@ -592,8 +600,9 @@ class HttpCli(object):
             taglist = {}
         else:
             # search by query params
-            self.log("qj: " + repr(body))
-            hits, taglist = idx.search(vols, body)
+            q = body["q"]
+            self.log("qj: " + q)
+            hits, taglist = idx.search(vols, q)
             msg = len(hits)
 
         idx.p_end = time.time()
@@ -1302,12 +1311,66 @@ class HttpCli(object):
 
     def tx_mounts(self):
         suf = self.urlq(rm=["h"])
-        rvol = [x + "/" if x else x for x in self.rvol]
-        wvol = [x + "/" if x else x for x in self.wvol]
-        html = self.j2("splash", this=self, rvol=rvol, wvol=wvol, url_suf=suf)
+        rvol, wvol, avol = [
+            [("/" + x).rstrip("/") + "/" for x in y]
+            for y in [self.rvol, self.wvol, self.avol]
+        ]
+
+        vstate = {}
+        if self.avol and not self.args.no_rescan:
+            x = self.conn.hsrv.broker.put(True, "up2k.get_volstate")
+            vstate = json.loads(x.get())
+            vstate = {("/" + k).rstrip("/") + "/": v for k, v in vstate.items()}
+
+        html = self.j2(
+            "splash",
+            this=self,
+            rvol=rvol,
+            wvol=wvol,
+            avol=avol,
+            vstate=vstate,
+            url_suf=suf,
+        )
         self.reply(html.encode("utf-8"), headers=NO_STORE)
         return True
 
+    def scanvol(self):
+        if not self.readable or not self.writable:
+            raise Pebkac(403, "not admin")
+
+        if self.args.no_rescan:
+            raise Pebkac(403, "disabled by argv")
+
+        vn, _ = self.auth.vfs.get(self.vpath, self.uname, True, True)
+
+        args = [self.auth.vfs.all_vols, [vn.vpath]]
+        x = self.conn.hsrv.broker.put(True, "up2k.rescan", *args)
+        x = x.get()
+        if not x:
+            self.redirect("", "?h")
+            return ""
+
+        raise Pebkac(500, x)
+
+    def tx_stack(self):
+        if not self.readable or not self.writable:
+            raise Pebkac(403, "not admin")
+
+        if self.args.no_stack:
+            raise Pebkac(403, "disabled by argv")
+
+        ret = []
+        names = dict([(t.ident, t.name) for t in threading.enumerate()])
+        for tid, stack in sys._current_frames().items():
+            ret.append("\n\n# {} ({:x})".format(names.get(tid), tid))
+            for fn, lno, name, line in traceback.extract_stack(stack):
+                ret.append('File: "{}", line {}, in {}'.format(fn, lno, name))
+                if line:
+                    ret.append("  " + str(line.strip()))
+
+        ret = ("<pre>" + "\n".join(ret)).encode("utf-8")
+        self.reply(ret)
+
     def tx_tree(self):
         top = self.uparam["tree"] or ""
         dst = self.vpath
@@ -1337,7 +1400,9 @@ class HttpCli(object):
 
         try:
             vn, rem = self.auth.vfs.get(top, self.uname, True, False)
-            fsroot, vfs_ls, vfs_virt = vn.ls(rem, self.uname, not self.args.no_scandir)
+            fsroot, vfs_ls, vfs_virt = vn.ls(
+                rem, self.uname, not self.args.no_scandir, True
+            )
         except:
             vfs_ls = []
             vfs_virt = {}
@@ -1502,7 +1567,9 @@ class HttpCli(object):
             if v is not None:
                 return self.tx_zip(k, v, vn, rem, [], self.args.ed)
 
-        fsroot, vfs_ls, vfs_virt = vn.ls(rem, self.uname, not self.args.no_scandir)
+        fsroot, vfs_ls, vfs_virt = vn.ls(
+            rem, self.uname, not self.args.no_scandir, True
+        )
         stats = {k: v for k, v in vfs_ls}
         vfs_ls = [x[0] for x in vfs_ls]
         vfs_ls.extend(vfs_virt.keys())

copyparty/th_srv.py

@@ -316,10 +316,10 @@ class ThumbSrv(object):
             time.sleep(interval)
             for vol in self.vols:
                 vol += "/.hist/th"
-                self.log("cln {}/".format(vol))
+                self.log("\033[Jcln {}/\033[A".format(vol))
                 self.clean(vol)
 
-            self.log("cln ok")
+            self.log("\033[Jcln ok")
 
     def clean(self, vol):
         # self.log("cln {}".format(vol))

copyparty/u2idx.py

@@ -47,11 +47,11 @@ class U2idx(object):
         fhash = body["hash"]
         wark = up2k_wark_from_hashlist(self.args.salt, fsize, fhash)
 
-        uq = "substr(w,1,16) = ? and w = ?"
+        uq = "where substr(w,1,16) = ? and w = ?"
         uv = [wark[:16], wark]
 
         try:
-            return self.run_query(vols, uq, uv, {})[0]
+            return self.run_query(vols, uq, uv)[0]
         except Exception as ex:
             raise Pebkac(500, repr(ex))
 
@@ -67,37 +67,121 @@ class U2idx(object):
         self.cur[ptop] = cur
         return cur
 
-    def search(self, vols, body):
+    def search(self, vols, uq):
         """search by query params"""
         if not HAVE_SQLITE3:
             return []
 
-        qobj = {}
-        _conv_sz(qobj, body, "sz_min", "up.sz >= ?")
-        _conv_sz(qobj, body, "sz_max", "up.sz <= ?")
-        _conv_dt(qobj, body, "dt_min", "up.mt >= ?")
-        _conv_dt(qobj, body, "dt_max", "up.mt <= ?")
-        for seg, dk in [["path", "up.rd"], ["name", "up.fn"]]:
-            if seg in body:
-                _conv_txt(qobj, body, seg, dk)
+        q = ""
+        va = []
+        joins = ""
+        is_key = True
+        is_size = False
+        is_date = False
+        kw_key = ["(", ")", "and ", "or ", "not "]
+        kw_val = ["==", "=", "!=", ">", ">=", "<", "<=", "like "]
+        ptn_mt = re.compile(r"^\.?[a-z]+$")
+        mt_ctr = 0
+        mt_keycmp = "substr(up.w,1,16)"
+        mt_keycmp2 = None
 
-        uq, uv = _sqlize(qobj)
+        while True:
+            uq = uq.strip()
+            if not uq:
+                break
 
-        qobj = {}
-        if "tags" in body:
-            _conv_txt(qobj, body, "tags", "mt.v")
+            ok = False
+            for kw in kw_key + kw_val:
+                if uq.startswith(kw):
+                    is_key = kw in kw_key
+                    uq = uq[len(kw) :]
+                    ok = True
+                    q += kw
+                    break
 
-        if "adv" in body:
-            _conv_adv(qobj, body, "adv")
+            if ok:
+                continue
+
+            v, uq = (uq + " ").split(" ", 1)
+            if is_key:
+                is_key = False
+
+                if v == "size":
+                    v = "up.sz"
+                    is_size = True
+
+                elif v == "date":
+                    v = "up.mt"
+                    is_date = True
+
+                elif v == "path":
+                    v = "up.rd"
+
+                elif v == "name":
+                    v = "up.fn"
+
+                elif v == "tags" or ptn_mt.match(v):
+                    mt_ctr += 1
+                    mt_keycmp2 = "mt{}.w".format(mt_ctr)
+                    joins += "inner join mt mt{} on {} = {} ".format(
+                        mt_ctr, mt_keycmp, mt_keycmp2
+                    )
+                    mt_keycmp = mt_keycmp2
+                    if v == "tags":
+                        v = "mt{0}.v".format(mt_ctr)
+                    else:
+                        v = "+mt{0}.k = '{1}' and mt{0}.v".format(mt_ctr, v)
+
+                else:
+                    raise Pebkac(400, "invalid key [" + v + "]")
+
+                q += v + " "
+                continue
+
+            head = ""
+            tail = ""
+
+            if is_date:
+                is_date = False
+                v = v.upper().rstrip("Z").replace(",", " ").replace("T", " ")
+                while "  " in v:
+                    v = v.replace("  ", " ")
+
+                for fmt in [
+                    "%Y-%m-%d %H:%M:%S",
+                    "%Y-%m-%d %H:%M",
+                    "%Y-%m-%d %H",
+                    "%Y-%m-%d",
+                ]:
+                    try:
+                        v = datetime.strptime(v, fmt).timestamp()
+                        break
+                    except:
+                        pass
+
+            elif is_size:
+                is_size = False
+                v = int(float(v) * 1024 * 1024)
+
+            else:
+                if v.startswith("*"):
+                    head = "'%'||"
+                    v = v[1:]
+
+                if v.endswith("*"):
+                    tail = "||'%'"
+                    v = v[:-1]
+
+            q += " {}?{} ".format(head, tail)
+            va.append(v)
+            is_key = True
 
         try:
-            return self.run_query(vols, uq, uv, qobj)
+            return self.run_query(vols, joins + "where " + q, va)
         except Exception as ex:
             raise Pebkac(500, repr(ex))
 
-    def run_query(self, vols, uq, uv, targs):
-        self.log("qs: {} {} ,  {}".format(uq, repr(uv), repr(targs)))
-
+    def run_query(self, vols, uq, uv):
         done_flag = []
         self.active_id = "{:.6f}_{}".format(
             time.time(), threading.current_thread().ident
@@ -112,35 +196,14 @@ class U2idx(object):
         thr.daemon = True
         thr.start()
 
-        if not targs:
-            if not uq:
-                q = "select * from up"
-                v = ()
-            else:
-                q = "select * from up where " + uq
-                v = tuple(uv)
+        if not uq or not uv:
+            q = "select * from up"
+            v = ()
         else:
-            q = "select up.* from up"
-            keycmp = "substr(up.w,1,16)"
-            where = []
-            v = []
-            ctr = 0
-            for tq, tv in sorted(targs.items()):
-                ctr += 1
-                tq = tq.split("\n")[0]
-                keycmp2 = "mt{}.w".format(ctr)
-                q += " inner join mt mt{} on {} = {}".format(ctr, keycmp, keycmp2)
-                keycmp = keycmp2
-                where.append(tq.replace("mt.", keycmp[:-1]))
-                v.append(tv)
+            q = "select up.* from up " + uq
+            v = tuple(uv)
 
-            if uq:
-                where.append(uq)
-                v.extend(uv)
-
-            q += " where " + (" and ".join(where))
-
-        # self.log("q2: {} {}".format(q, repr(v)))
+        self.log("qs: {!r} {!r}".format(q, v))
 
         ret = []
         lim = 1000
@@ -163,7 +226,7 @@ class U2idx(object):
                 if rd.startswith("//") or fn.startswith("//"):
                     rd, fn = s3dec(rd, fn)
 
-                rp = "/".join([vtop, rd, fn])
+                rp = "/".join([x for x in [vtop, rd, fn] if x])
                 sret.append({"ts": int(ts), "sz": sz, "rp": rp, "w": w[:16]})
 
             for hit in sret:
@@ -204,78 +267,3 @@ def _open(ptop):
     db_path = os.path.join(ptop, ".hist", "up2k.db")
     if os.path.exists(db_path):
         return sqlite3.connect(db_path).cursor()
-
-
-def _conv_sz(q, body, k, sql):
-    if k in body:
-        q[sql] = int(float(body[k]) * 1024 * 1024)
-
-
-def _conv_dt(q, body, k, sql):
-    if k not in body:
-        return
-
-    v = body[k].upper().rstrip("Z").replace(",", " ").replace("T", " ")
-    while "  " in v:
-        v = v.replace("  ", " ")
-
-    for fmt in ["%Y-%m-%d %H:%M:%S", "%Y-%m-%d %H:%M", "%Y-%m-%d %H", "%Y-%m-%d"]:
-        try:
-            ts = datetime.strptime(v, fmt).timestamp()
-            break
-        except:
-            ts = None
-
-    if ts:
-        q[sql] = ts
-
-
-def _conv_txt(q, body, k, sql):
-    for v in body[k].split(" "):
-        inv = ""
-        if v.startswith("-"):
-            inv = "not"
-            v = v[1:]
-
-        if not v:
-            continue
-
-        head = "'%'||"
-        if v.startswith("^"):
-            head = ""
-            v = v[1:]
-
-        tail = "||'%'"
-        if v.endswith("$"):
-            tail = ""
-            v = v[:-1]
-
-        qk = "{} {} like {}?{}".format(sql, inv, head, tail)
-        q[qk + "\n" + v] = u8safe(v)
-
-
-def _conv_adv(q, body, k):
-    ptn = re.compile(r"^(\.?[a-z]+) *(==?|!=|<=?|>=?) *(.*)$")
-
-    parts = body[k].split(" ")
-    parts = [x.strip() for x in parts if x.strip()]
-
-    for part in parts:
-        m = ptn.match(part)
-        if not m:
-            p = html_escape(part)
-            raise Pebkac(400, "invalid argument [" + p + "]")
-
-        k, op, v = m.groups()
-        qk = "mt.k = '{}' and mt.v {} ?".format(k, op)
-        q[qk + "\n" + v] = u8safe(v)
-
-
-def _sqlize(qobj):
-    keys = []
-    values = []
-    for k, v in sorted(qobj.items()):
-        keys.append(k.split("\n")[0])
-        values.append(v)
-
-    return " and ".join(keys), values

copyparty/up2k.py

@@ -52,7 +52,6 @@ class Up2k(object):
         self.hub = hub
         self.args = hub.args
         self.log_func = hub.log
-        self.all_vols = all_vols
 
         # config
         self.salt = self.args.salt
@@ -61,12 +60,14 @@ class Up2k(object):
         self.mutex = threading.Lock()
         self.hashq = Queue()
         self.tagq = Queue()
+        self.volstate = {}
         self.registry = {}
         self.entags = {}
         self.flags = {}
         self.cur = {}
         self.mtag = None
         self.pending_tags = None
+        self.mtp_parsers = {}
 
         self.mem_cur = None
         self.sqlite_ver = None
@@ -92,7 +93,15 @@ class Up2k(object):
         if not HAVE_SQLITE3:
             self.log("could not initialize sqlite3, will use in-memory registry only")
 
-        have_e2d = self.init_indexes()
+        if self.args.no_fastboot:
+            self.deferred_init(all_vols)
+        else:
+            t = threading.Thread(target=self.deferred_init, args=(all_vols,))
+            t.daemon = True
+            t.start()
+
+    def deferred_init(self, all_vols):
+        have_e2d = self.init_indexes(all_vols)
 
         if have_e2d:
             thr = threading.Thread(target=self._snapshot)
@@ -115,6 +124,19 @@ class Up2k(object):
     def log(self, msg, c=0):
         self.log_func("up2k", msg + "\033[K", c)
 
+    def get_volstate(self):
+        return json.dumps(self.volstate, indent=4)
+
+    def rescan(self, all_vols, scan_vols):
+        if hasattr(self, "pp"):
+            return "cannot initiate; scan is already in progress"
+
+        args = (all_vols, scan_vols)
+        t = threading.Thread(target=self.init_indexes, args=args)
+        t.daemon = True
+        t.start()
+        return None
+
     def _vis_job_progress(self, job):
         perc = 100 - (len(job["need"]) * 100.0 / len(job["hash"]))
         path = os.path.join(job["ptop"], job["prel"], job["name"])
@@ -137,9 +159,9 @@ class Up2k(object):
 
         return True, ret
 
-    def init_indexes(self):
+    def init_indexes(self, all_vols, scan_vols=[]):
         self.pp = ProgressPrinter()
-        vols = self.all_vols.values()
+        vols = all_vols.values()
         t0 = time.time()
         have_e2d = False
 
@@ -159,24 +181,35 @@ class Up2k(object):
         for vol in vols:
             try:
                 os.listdir(vol.realpath)
-                live_vols.append(vol)
             except:
+                self.volstate[vol.vpath] = "OFFLINE (cannot access folder)"
                 self.log("cannot access " + vol.realpath, c=1)
+                continue
+
+            if not self.register_vpath(vol.realpath, vol.flags):
+                # self.log("db not enabled for {}".format(m, vol.realpath))
+                continue
+
+            if vol.vpath in scan_vols or not scan_vols:
+                live_vols.append(vol)
+
+            if vol.vpath not in self.volstate:
+                self.volstate[vol.vpath] = "OFFLINE (pending initialization)"
 
         vols = live_vols
+        need_vac = {}
 
         need_mtag = False
         for vol in vols:
             if "e2t" in vol.flags:
                 need_mtag = True
 
-        if need_mtag:
+        if need_mtag and not self.mtag:
             self.mtag = MTag(self.log_func, self.args)
             if not self.mtag.usable:
                 self.mtag = None
 
-        # e2ds(a) volumes first,
-        # also covers tags where e2ts is set
+        # e2ds(a) volumes first
         for vol in vols:
             en = {}
             if "mte" in vol.flags:
@@ -188,26 +221,45 @@ class Up2k(object):
                 have_e2d = True
 
             if "e2ds" in vol.flags:
-                r = self._build_file_index(vol, vols)
-                if not r:
-                    needed_mutagen = True
+                self.volstate[vol.vpath] = "busy (hashing files)"
+                _, vac = self._build_file_index(vol, list(all_vols.values()))
+                if vac:
+                    need_vac[vol] = True
+
+            if "e2ts" not in vol.flags:
+                m = "online, idle"
+            else:
+                m = "online (tags pending)"
+
+            self.volstate[vol.vpath] = m
 
         # open the rest + do any e2ts(a)
         needed_mutagen = False
         for vol in vols:
-            r = self.register_vpath(vol.realpath, vol.flags)
-            if not r or "e2ts" not in vol.flags:
+            if "e2ts" not in vol.flags:
                 continue
 
-            cur, db_path, sz0 = r
-            n_add, n_rm, success = self._build_tags_index(vol.realpath)
+            m = "online (reading tags)"
+            self.volstate[vol.vpath] = m
+            self.log("{} [{}]".format(m, vol.realpath))
+
+            nadd, nrm, success = self._build_tags_index(vol)
             if not success:
                 needed_mutagen = True
 
-            if n_add or n_rm:
-                self.vac(cur, db_path, n_add, n_rm, sz0)
+            if nadd or nrm:
+                need_vac[vol] = True
+
+            self.volstate[vol.vpath] = "online (mtp soon)"
+
+        for vol in need_vac:
+            cur, _ = self.register_vpath(vol.realpath, vol.flags)
+            with self.mutex:
+                cur.connection.commit()
+                cur.execute("vacuum")
 
         self.pp.end = True
+
         msg = "{} volumes in {:.2f} sec"
         self.log(msg.format(len(vols), time.time() - t0))
 
@@ -215,110 +267,104 @@ class Up2k(object):
             msg = "could not read tags because no backends are available (mutagen or ffprobe)"
             self.log(msg, c=1)
 
+        thr = None
+        if self.mtag:
+            m = "online (running mtp)"
+            if scan_vols:
+                thr = threading.Thread(target=self._run_all_mtp)
+                thr.daemon = True
+        else:
+            del self.pp
+            m = "online, idle"
+
+        for vol in vols:
+            self.volstate[vol.vpath] = m
+
+        if thr:
+            thr.start()
+
         return have_e2d
 
     def register_vpath(self, ptop, flags):
-        with self.mutex:
-            if ptop in self.registry:
-                return None
+        db_path = os.path.join(ptop, ".hist", "up2k.db")
+        if ptop in self.registry:
+            return [self.cur[ptop], db_path]
 
-            _, flags = self._expr_idx_filter(flags)
+        _, flags = self._expr_idx_filter(flags)
 
-            ft = "\033[0;32m{}{:.0}"
-            ff = "\033[0;35m{}{:.0}"
-            fv = "\033[0;36m{}:\033[1;30m{}"
-            a = [
-                (ft if v is True else ff if v is False else fv).format(k, str(v))
-                for k, v in flags.items()
-            ]
-            if a:
-                self.log(" ".join(sorted(a)) + "\033[0m")
+        ft = "\033[0;32m{}{:.0}"
+        ff = "\033[0;35m{}{:.0}"
+        fv = "\033[0;36m{}:\033[1;30m{}"
+        a = [
+            (ft if v is True else ff if v is False else fv).format(k, str(v))
+            for k, v in flags.items()
+        ]
+        if a:
+            self.log(" ".join(sorted(a)) + "\033[0m")
 
-            reg = {}
-            path = os.path.join(ptop, ".hist", "up2k.snap")
-            if "e2d" in flags and os.path.exists(path):
-                with gzip.GzipFile(path, "rb") as f:
-                    j = f.read().decode("utf-8")
+        reg = {}
+        path = os.path.join(ptop, ".hist", "up2k.snap")
+        if "e2d" in flags and os.path.exists(path):
+            with gzip.GzipFile(path, "rb") as f:
+                j = f.read().decode("utf-8")
 
-                reg2 = json.loads(j)
-                for k, job in reg2.items():
-                    path = os.path.join(job["ptop"], job["prel"], job["name"])
-                    if os.path.exists(fsenc(path)):
-                        reg[k] = job
-                        job["poke"] = time.time()
-                    else:
-                        self.log("ign deleted file in snap: [{}]".format(path))
+            reg2 = json.loads(j)
+            for k, job in reg2.items():
+                path = os.path.join(job["ptop"], job["prel"], job["name"])
+                if os.path.exists(fsenc(path)):
+                    reg[k] = job
+                    job["poke"] = time.time()
+                else:
+                    self.log("ign deleted file in snap: [{}]".format(path))
 
-                m = "loaded snap {} |{}|".format(path, len(reg.keys()))
-                m = [m] + self._vis_reg_progress(reg)
-                self.log("\n".join(m))
-
-            self.flags[ptop] = flags
-            self.registry[ptop] = reg
-            if not HAVE_SQLITE3 or "e2d" not in flags or "d2d" in flags:
-                return None
-
-            try:
-                os.mkdir(os.path.join(ptop, ".hist"))
-            except:
-                pass
-
-            db_path = os.path.join(ptop, ".hist", "up2k.db")
-            if ptop in self.cur:
-                return None
-
-            try:
-                sz0 = 0
-                if os.path.exists(db_path):
-                    sz0 = os.path.getsize(db_path) // 1024
-
-                cur = self._open_db(db_path)
-                self.cur[ptop] = cur
-                return [cur, db_path, sz0]
-            except:
-                msg = "cannot use database at [{}]:\n{}"
-                self.log(msg.format(ptop, traceback.format_exc()))
+            m = "loaded snap {} |{}|".format(path, len(reg.keys()))
+            m = [m] + self._vis_reg_progress(reg)
+            self.log("\n".join(m))
 
+        self.flags[ptop] = flags
+        self.registry[ptop] = reg
+        if not HAVE_SQLITE3 or "e2d" not in flags or "d2d" in flags:
             return None
 
+        try:
+            os.mkdir(os.path.join(ptop, ".hist"))
+        except:
+            pass
+
+        try:
+            cur = self._open_db(db_path)
+            self.cur[ptop] = cur
+            return [cur, db_path]
+        except:
+            msg = "cannot use database at [{}]:\n{}"
+            self.log(msg.format(ptop, traceback.format_exc()))
+
+        return None
+
     def _build_file_index(self, vol, all_vols):
         do_vac = False
         top = vol.realpath
-        reg = self.register_vpath(top, vol.flags)
-        if not reg:
-            return
+        with self.mutex:
+            cur, _ = self.register_vpath(top, vol.flags)
 
-        _, db_path, sz0 = reg
-        dbw = [reg[0], 0, time.time()]
-        self.pp.n = next(dbw[0].execute("select count(w) from up"))[0]
+            dbw = [cur, 0, time.time()]
+            self.pp.n = next(dbw[0].execute("select count(w) from up"))[0]
 
-        excl = [
-            vol.realpath + "/" + d.vpath[len(vol.vpath) :].lstrip("/")
-            for d in all_vols
-            if d != vol and (d.vpath.startswith(vol.vpath + "/") or not vol.vpath)
-        ]
-        n_add = self._build_dir(dbw, top, set(excl), top)
-        n_rm = self._drop_lost(dbw[0], top)
-        if dbw[1]:
-            self.log("commit {} new files".format(dbw[1]))
-            dbw[0].connection.commit()
+            excl = [
+                vol.realpath + "/" + d.vpath[len(vol.vpath) :].lstrip("/")
+                for d in all_vols
+                if d != vol and (d.vpath.startswith(vol.vpath + "/") or not vol.vpath)
+            ]
+            if WINDOWS:
+                excl = [x.replace("/", "\\") for x in excl]
 
-        n_add, n_rm, success = self._build_tags_index(vol.realpath)
+            n_add = self._build_dir(dbw, top, set(excl), top)
+            n_rm = self._drop_lost(dbw[0], top)
+            if dbw[1]:
+                self.log("commit {} new files".format(dbw[1]))
+                dbw[0].connection.commit()
 
-        dbw[0].connection.commit()
-        if n_add or n_rm or do_vac:
-            self.vac(dbw[0], db_path, n_add, n_rm, sz0)
-
-        return success
-
-    def vac(self, cur, db_path, n_add, n_rm, sz0):
-        sz1 = os.path.getsize(db_path) // 1024
-        cur.execute("vacuum")
-        sz2 = os.path.getsize(db_path) // 1024
-        msg = "{} new, {} del, {} kB vacced, {} kB gain, {} kB now".format(
-            n_add, n_rm, sz1 - sz2, sz2 - sz0, sz2
-        )
-        self.log(msg)
+            return True, n_add or n_rm or do_vac
 
     def _build_dir(self, dbw, top, excl, cdir):
         self.pp.msg = "a{} {}".format(self.pp.n, cdir)
@@ -413,45 +459,53 @@ class Up2k(object):
 
         return len(rm)
 
-    def _build_tags_index(self, ptop):
-        entags = self.entags[ptop]
-        flags = self.flags[ptop]
-        cur = self.cur[ptop]
+    def _build_tags_index(self, vol):
+        ptop = vol.realpath
+        with self.mutex:
+            _, db_path = self.register_vpath(ptop, vol.flags)
+            entags = self.entags[ptop]
+            flags = self.flags[ptop]
+            cur = self.cur[ptop]
+
         n_add = 0
         n_rm = 0
         n_buf = 0
         last_write = time.time()
 
         if "e2tsr" in flags:
-            n_rm = cur.execute("select count(w) from mt").fetchone()[0]
-            if n_rm:
-                self.log("discarding {} media tags for a full rescan".format(n_rm))
-                cur.execute("delete from mt")
-            else:
-                self.log("volume has e2tsr but there are no media tags to discard")
+            with self.mutex:
+                n_rm = cur.execute("select count(w) from mt").fetchone()[0]
+                if n_rm:
+                    self.log("discarding {} media tags for a full rescan".format(n_rm))
+                    cur.execute("delete from mt")
 
         # integrity: drop tags for tracks that were deleted
         if "e2t" in flags:
-            drops = []
-            c2 = cur.connection.cursor()
-            up_q = "select w from up where substr(w,1,16) = ?"
-            for (w,) in cur.execute("select w from mt"):
-                if not c2.execute(up_q, (w,)).fetchone():
-                    drops.append(w[:16])
-            c2.close()
+            with self.mutex:
+                drops = []
+                c2 = cur.connection.cursor()
+                up_q = "select w from up where substr(w,1,16) = ?"
+                for (w,) in cur.execute("select w from mt"):
+                    if not c2.execute(up_q, (w,)).fetchone():
+                        drops.append(w[:16])
+                c2.close()
 
-            if drops:
-                msg = "discarding media tags for {} deleted files"
-                self.log(msg.format(len(drops)))
-                n_rm += len(drops)
-                for w in drops:
-                    cur.execute("delete from mt where w = ?", (w,))
+                if drops:
+                    msg = "discarding media tags for {} deleted files"
+                    self.log(msg.format(len(drops)))
+                    n_rm += len(drops)
+                    for w in drops:
+                        cur.execute("delete from mt where w = ?", (w,))
 
         # bail if a volume flag disables indexing
         if "d2t" in flags or "d2d" in flags:
             return n_add, n_rm, True
 
         # add tags for new files
+        gcur = cur
+        with self.mutex:
+            gcur.connection.commit()
+
         if "e2ts" in flags:
             if not self.mtag:
                 return n_add, n_rm, False
@@ -460,8 +514,10 @@ class Up2k(object):
             if self.mtag.prefer_mt and not self.args.no_mtag_mt:
                 mpool = self._start_mpool()
 
-            c2 = cur.connection.cursor()
-            c3 = cur.connection.cursor()
+            conn = sqlite3.connect(db_path, timeout=15)
+            cur = conn.cursor()
+            c2 = conn.cursor()
+            c3 = conn.cursor()
             n_left = cur.execute("select count(w) from up").fetchone()[0]
             for w, rd, fn in cur.execute("select w, rd, fn from up"):
                 n_left -= 1
@@ -483,7 +539,8 @@ class Up2k(object):
                     n_tags = self._tag_file(c3, *args)
                 else:
                     mpool.put(["mtag"] + args)
-                    n_tags = len(self._flush_mpool(c3))
+                    with self.mutex:
+                        n_tags = len(self._flush_mpool(c3))
 
                 n_add += n_tags
                 n_buf += n_tags
@@ -495,26 +552,32 @@ class Up2k(object):
                     last_write = time.time()
                     n_buf = 0
 
-            self._stop_mpool(mpool, c3)
+            self._stop_mpool(mpool)
+            with self.mutex:
+                n_add += len(self._flush_mpool(c3))
 
+            conn.commit()
             c3.close()
             c2.close()
+            cur.close()
+            conn.close()
+
+        with self.mutex:
+            gcur.connection.commit()
 
         return n_add, n_rm, True
 
     def _flush_mpool(self, wcur):
-        with self.mutex:
-            ret = []
-            for x in self.pending_tags:
-                self._tag_file(wcur, *x)
-                ret.append(x[1])
+        ret = []
+        for x in self.pending_tags:
+            self._tag_file(wcur, *x)
+            ret.append(x[1])
 
-            self.pending_tags = []
-            return ret
+        self.pending_tags = []
+        return ret
 
     def _run_all_mtp(self):
         t0 = time.time()
-        self.mtp_parsers = {}
         for ptop, flags in self.flags.items():
             if "mtp" in flags:
                 self._run_one_mtp(ptop)
@@ -523,10 +586,12 @@ class Up2k(object):
         msg = "mtp finished in {:.2f} sec ({})"
         self.log(msg.format(td, s2hms(td, True)))
 
-    def _run_one_mtp(self, ptop):
-        db_path = os.path.join(ptop, ".hist", "up2k.db")
-        sz0 = os.path.getsize(db_path) // 1024
+        del self.pp
+        for k in list(self.volstate.keys()):
+            if "OFFLINE" not in self.volstate[k]:
+                self.volstate[k] = "online, idle"
 
+    def _run_one_mtp(self, ptop):
         entags = self.entags[ptop]
 
         parsers = {}
@@ -585,9 +650,8 @@ class Up2k(object):
                     jobs.append([parsers, None, w, abspath])
                     in_progress[w] = True
 
-            done = self._flush_mpool(wcur)
-
             with self.mutex:
+                done = self._flush_mpool(wcur)
                 for w in done:
                     to_delete[w] = True
                     in_progress.pop(w)
@@ -628,15 +692,16 @@ class Up2k(object):
             with self.mutex:
                 cur.connection.commit()
 
-        done = self._stop_mpool(mpool, wcur)
+        self._stop_mpool(mpool)
         with self.mutex:
+            done = self._flush_mpool(wcur)
             for w in done:
                 q = "delete from mt where w = ? and k = 't:mtp'"
                 cur.execute(q, (w,))
 
             cur.connection.commit()
             if n_done:
-                self.vac(cur, db_path, n_done, 0, sz0)
+                cur.execute("vacuum")
 
             wcur.close()
             cur.close()
@@ -693,7 +758,7 @@ class Up2k(object):
 
         return mpool
 
-    def _stop_mpool(self, mpool, wcur):
+    def _stop_mpool(self, mpool):
         if not mpool:
             return
 
@@ -701,8 +766,6 @@ class Up2k(object):
             mpool.put(None)
 
         mpool.join()
-        done = self._flush_mpool(wcur)
-        return done
 
     def _tag_thr(self, q):
         while True:
@@ -1181,12 +1244,15 @@ class Up2k(object):
         return wark
 
     def _hashlist_from_file(self, path):
+        pp = self.pp if hasattr(self, "pp") else None
         fsz = os.path.getsize(fsenc(path))
         csz = up2k_chunksize(fsz)
         ret = []
         with open(fsenc(path), "rb", 512 * 1024) as f:
             while fsz > 0:
-                self.pp.msg = "{} MB, {}".format(int(fsz / 1024 / 1024), path)
+                if pp:
+                    pp.msg = "{} MB, {}".format(int(fsz / 1024 / 1024), path)
+
                 hashobj = hashlib.sha512()
                 rem = min(csz, fsz)
                 fsz -= rem

copyparty/web/browser.css

@@ -529,6 +529,17 @@ input[type="checkbox"]:checked+label {
 	height: 1em;
 	margin: .2em 0 -1em 1.6em;
 }
+#tq_raw {
+	width: calc(100% - 2em);
+	margin: .3em 0 0 1.4em;
+}
+#tq_raw td+td {
+	width: 100%;
+}
+#op_search #q_raw {
+	width: 100%;
+	display: block;
+}
 #files td div span {
 	color: #fff;
 	padding: 0 .4em;

copyparty/web/browser.js

@@ -826,6 +826,11 @@ var thegrid = (function () {
 			ths[a].onclick = r.sel ? seltgl : null;
 			ths[a].setAttribute('class', ebi(ths[a].getAttribute('ref')).parentNode.parentNode.getAttribute('class'));
 		}
+		var uns = QS('#ggrid a[ref="unsearch"]');
+		if (uns)
+			uns.onclick = function () {
+				ebi('unsearch').click();
+			};
 	}
 
 	function loadgrid() {
@@ -836,9 +841,9 @@ var thegrid = (function () {
 			return r.loadsel();
 
 		var html = [];
-		var tr = lfiles.tBodies[0].rows;
-		for (var a = 0; a < tr.length; a++) {
-			var ao = tr[a].cells[1].firstChild,
+		var files = QSA('#files>tbody>tr>td:nth-child(2) a[id]');
+		for (var a = 0, aa = files.length; a < aa; a++) {
+			var ao = files[a],
 				href = esc(ao.getAttribute('href')),
 				ref = ao.getAttribute('id'),
 				isdir = href.split('?')[0].slice(-1)[0] == '/',
@@ -1026,6 +1031,7 @@ document.onkeydown = function (e) {
 	for (var a = 0; a < trs.length; a += 2) {
 		html.push('<table>' + (trs[a].concat(trs[a + 1])).join('\n') + '</table>');
 	}
+	html.push('<table id="tq_raw"><tr><td>raw</td><td><input id="q_raw" type="text" name="q" /></td></tr></table>');
 	ebi('srch_form').innerHTML = html.join('\n');
 
 	var o = QSA('#op_search input');
@@ -1050,33 +1056,83 @@ document.onkeydown = function (e) {
 			var chk = ebi(id.slice(0, -1) + 'c');
 			chk.checked = ((v + '').length > 0);
 		}
+
+		if (id != "q_raw")
+			encode_query();
+
 		clearTimeout(search_timeout);
 		if (Date.now() - search_in_progress > 30 * 1000)
 			search_timeout = setTimeout(do_search, 200);
 	}
 
+	function encode_query() {
+		var q = '';
+		for (var a = 0; a < sconf.length; a++) {
+			for (var b = 1; b < sconf[a].length; b++) {
+				var k = sconf[a][b][0],
+					chk = 'srch_' + k + 'c',
+					tvs = ebi('srch_' + k + 'v').value.split(/ /g);
+
+				if (!ebi(chk).checked)
+					continue;
+
+				for (var c = 0; c < tvs.length; c++) {
+					var tv = tvs[c];
+					if (!tv.length)
+						break;
+
+					q += ' and ';
+
+					if (k == 'adv') {
+						q += tv.replace(/ /g, " and ").replace(/([=!><]=?)/, " $1 ");
+						continue;
+					}
+
+					if (k.length == 3) {
+						q += k.replace(/sz/, 'size').replace(/dt/, 'date').replace(/l$/, ' >= ').replace(/u$/, ' <= ') + tv;
+						continue;
+					}
+
+					if (k == 'path' || k == 'name' || k == 'tags') {
+						var not = ' ';
+						if (tv.slice(0, 1) == '-') {
+							tv = tv.slice(1);
+							not = ' not ';
+						}
+
+						if (tv.slice(0, 1) == '^') {
+							tv = tv.slice(1);
+						}
+						else {
+							tv = '*' + tv;
+						}
+
+						if (tv.slice(-1) == '$') {
+							tv = tv.slice(0, -1);
+						}
+						else {
+							tv += '*';
+						}
+
+						q += k + not + 'like ' + tv;
+					}
+				}
+			}
+		}
+		ebi('q_raw').value = q.slice(5);
+	}
+
 	function do_search() {
 		search_in_progress = Date.now();
 		srch_msg(false, "searching...");
 		clearTimeout(search_timeout);
 
-		var params = {},
-			o = QSA('#op_search input[type="text"]');
-
-		for (var a = 0; a < o.length; a++) {
-			var chk = ebi(o[a].getAttribute('id').slice(0, -1) + 'c');
-			if (!chk.checked)
-				continue;
-
-			params[o[a].getAttribute('name')] = o[a].value;
-		}
-		// ebi('srch_q').textContent = JSON.stringify(params, null, 4);
 		var xhr = new XMLHttpRequest();
 		xhr.open('POST', '/?srch', true);
 		xhr.setRequestHeader('Content-Type', 'text/plain');
 		xhr.onreadystatechange = xhr_search_results;
 		xhr.ts = Date.now();
-		xhr.send(JSON.stringify(params));
+		xhr.send(JSON.stringify({ "q": ebi('q_raw').value }));
 	}
 
 	function xhr_search_results() {

copyparty/web/splash.css

@@ -26,6 +26,13 @@ a {
 	border-radius: .2em;
 	padding: .2em .8em;
 }
+td, th {
+	padding: .3em .6em;
+	text-align: left;
+}
+.btns {
+	margin: 1em 0;
+}
 
 
 html.dark,

copyparty/web/splash.html

@@ -13,11 +13,28 @@
     <div id="wrap">
         <p>hello {{ this.uname }}</p>
 
+        {%- if avol %}
+        <h1>admin panel:</h1>
+        <table>
+            <thead><tr><th>vol</th><th>action</th><th>status</th></tr></thead>
+            <tbody>
+                {% for mp in avol %}
+                {%- if mp in vstate and vstate[mp] %}
+                <tr><td><a href="{{ mp }}{{ url_suf }}">{{ mp }}</a></td><td><a href="{{ mp }}?scan">rescan</a></td><td>{{ vstate[mp] }}</td></tr>
+                {%- endif %}
+                {% endfor %}
+            </tbody>
+        </table>
+        <div class="btns">
+            <a href="{{ avol[0] }}?stack">dump stack</a>
+        </div>
+        {%- endif %}
+
         {%- if rvol %}
         <h1>you can browse these:</h1>
         <ul>
             {% for mp in rvol %}
-            <li><a href="/{{ mp }}{{ url_suf }}">/{{ mp }}</a></li>
+            <li><a href="{{ mp }}{{ url_suf }}">{{ mp }}</a></li>
             {% endfor %}
         </ul>
         {%- endif %}
@@ -26,7 +43,7 @@
         <h1>you can upload to:</h1>
         <ul>
             {% for mp in wvol %}
-            <li><a href="/{{ mp }}{{ url_suf }}">/{{ mp }}</a></li>
+            <li><a href="{{ mp }}{{ url_suf }}">{{ mp }}</a></li>
             {% endfor %}
         </ul>
         {%- endif %}

docs/notes.sh

@@ -80,6 +80,13 @@ command -v gdate && date() { gdate "$@"; }; while true; do t=$(date +%s.%N); (ti
 var t=[]; var b=document.location.href.split('#')[0].slice(0, -1); document.querySelectorAll('#u2tab .prog a').forEach((x) => {t.push(b+encodeURI(x.getAttribute("href")))}); console.log(t.join("\n"));
 
 
+##
+## bash oneliners
+
+# get the size and video-id of all youtube vids in folder, assuming filename ends with -id.ext, and create a copyparty search query
+find -maxdepth 1 -printf '%s %p\n' | sort -n | awk '!/-([0-9a-zA-Z_-]{11})\.(mkv|mp4|webm)$/{next} {sub(/\.[^\.]+$/,"");n=length($0);v=substr($0,n-10);print $1, v}' | tee /dev/stderr | awk 'BEGIN {p="("} {printf("%s name like *-%s.* ",p,$2);p="or"} END {print ")\n"}' | cat >&2
+
+
 ##
 ## sqlite3 stuff
 

scripts/install-githooks.sh

@@ -0,0 +1,12 @@
+#!/bin/bash
+set -ex
+
+[ -e setup.py ] || ..
+[ -e setup.py ] || {
+    echo u wot
+    exit 1
+}
+
+cd .git/hooks
+rm -f pre-commit
+ln -s ../../scripts/run-tests.sh pre-commit

scripts/run-tests.sh

@@ -0,0 +1,12 @@
+#!/bin/bash
+set -ex
+
+pids=()
+for py in python{2,3}; do
+    $py -m unittest discover -s tests >/dev/null &
+    pids+=($!)
+done
+
+for pid in ${pids[@]}; do
+    wait $pid
+done

tests/test_httpcli.py

@@ -8,6 +8,7 @@ import time
 import shutil
 import pprint
 import tarfile
+import tempfile
 import unittest
 
 from argparse import Namespace
@@ -32,6 +33,8 @@ class Cfg(Namespace):
             no_zip=False,
             no_scandir=False,
             no_sendfile=True,
+            no_rescan=True,
+            ihead=False,
             nih=True,
             mtp=[],
             mte="a",
@@ -40,13 +43,15 @@ class Cfg(Namespace):
 
 
 class TestHttpCli(unittest.TestCase):
-    def test(self):
-        td = os.path.join(tu.get_ramdisk(), "vfs")
-        try:
-            shutil.rmtree(td)
-        except OSError:
-            pass
+    def setUp(self):
+        self.td = tu.get_ramdisk()
 
+    def tearDown(self):
+        os.chdir(tempfile.gettempdir())
+        shutil.rmtree(self.td)
+
+    def test(self):
+        td = os.path.join(self.td, "vfs")
         os.mkdir(td)
         os.chdir(td)
 

tests/test_vfs.py

@@ -25,6 +25,13 @@ class Cfg(Namespace):
 
 
 class TestVFS(unittest.TestCase):
+    def setUp(self):
+        self.td = tu.get_ramdisk()
+
+    def tearDown(self):
+        os.chdir(tempfile.gettempdir())
+        shutil.rmtree(self.td)
+
     def dump(self, vfs):
         print(json.dumps(vfs, indent=4, sort_keys=True, default=lambda o: o.__dict__))
 
@@ -55,12 +62,7 @@ class TestVFS(unittest.TestCase):
         pass
 
     def test(self):
-        td = os.path.join(tu.get_ramdisk(), "vfs")
-        try:
-            shutil.rmtree(td)
-        except OSError:
-            pass
-
+        td = os.path.join(self.td, "vfs")
         os.mkdir(td)
         os.chdir(td)
 
@@ -227,7 +229,7 @@ class TestVFS(unittest.TestCase):
         self.assertEqual(list(v1), list(v2))
 
         # config file parser
-        cfg_path = os.path.join(tu.get_ramdisk(), "test.cfg")
+        cfg_path = os.path.join(self.td, "test.cfg")
         with open(cfg_path, "wb") as f:
             f.write(
                 dedent(
@@ -260,6 +262,4 @@ class TestVFS(unittest.TestCase):
         self.assertEqual(n.uwrite, ["asd"])
         self.assertEqual(len(n.nodes), 0)
 
-        os.chdir(tempfile.gettempdir())
-        shutil.rmtree(td)
         os.unlink(cfg_path)

tests/util.py

@@ -1,5 +1,6 @@
 import os
 import time
+import shutil
 import jinja2
 import tempfile
 import subprocess as sp
@@ -28,18 +29,25 @@ def chkcmd(*argv):
 
 
 def get_ramdisk():
+    def subdir(top):
+        ret = os.path.join(top, "cptd-{}".format(os.getpid()))
+        shutil.rmtree(ret, True)
+        os.mkdir(ret)
+        return ret
+
     for vol in ["/dev/shm", "/Volumes/cptd"]:  # nosec (singleton test)
         if os.path.exists(vol):
-            return vol
+            return subdir(vol)
 
     if os.path.exists("/Volumes"):
-        devname, _ = chkcmd("hdiutil", "attach", "-nomount", "ram://32768")
+        # hdiutil eject /Volumes/cptd/
+        devname, _ = chkcmd("hdiutil", "attach", "-nomount", "ram://65536")
         devname = devname.strip()
         print("devname: [{}]".format(devname))
         for _ in range(10):
             try:
                 _, _ = chkcmd("diskutil", "eraseVolume", "HFS+", "cptd", devname)
-                return "/Volumes/cptd"
+                return subdir("/Volumes/cptd")
             except Exception as ex:
                 print(repr(ex))
                 time.sleep(0.25)
@@ -50,7 +58,7 @@ def get_ramdisk():
     try:
         os.mkdir(ret)
     finally:
-        return ret
+        return subdir(ret)
 
 
 class NullBroker(object):
@@ -91,7 +99,10 @@ class VHttpConn(object):
         self.auth = auth
         self.log_func = log
         self.log_src = "a"
+        self.lf_url = None
         self.hsrv = VHttpSrv()
         self.nbyte = 0
         self.workload = 0
+        self.ico = None
+        self.thumbcli = None
         self.t0 = time.time()