v0.11.12

.vscode/launch.json

@@ -16,12 +16,9 @@
                 "-e2ts",
                 "-mtp",
                 ".bpm=f,bin/mtag/audio-bpm.py",
-                "-a",
-                "ed:wark",
-                "-v",
-                "srv::r:aed:cnodupe",
-                "-v",
-                "dist:dist:r"
+                "-aed:wark",
+                "-vsrv::r:aed:cnodupe",
+                "-vdist:dist:r"
             ]
         },
         {
@@ -43,5 +40,13 @@
                 "${file}"
             ]
         },
+        {
+            "name": "Python: Current File",
+            "type": "python",
+            "request": "launch",
+            "program": "${file}",
+            "console": "integratedTerminal",
+            "justMyCode": false
+        },
     ]
 }

.vscode/launch.py

@@ -3,14 +3,16 @@
 # launches 10x faster than mspython debugpy
 # and is stoppable with ^C
 
+import re
 import os
 import sys
+
+print(sys.executable)
+
 import shlex
-
-sys.path.insert(0, os.getcwd())
-
 import jstyleson
-from copyparty.__main__ import main as copyparty
+import subprocess as sp
+
 
 with open(".vscode/launch.json", "r", encoding="utf-8") as f:
     tj = f.read()
@@ -25,11 +27,19 @@ except:
     pass
 
 argv = [os.path.expanduser(x) if x.startswith("~") else x for x in argv]
-try:
-    copyparty(["a"] + argv)
-except SystemExit as ex:
-    if ex.code:
-        raise
+
+if re.search(" -j ?[0-9]", " ".join(argv)):
+    argv = [sys.executable, "-m", "copyparty"] + argv
+    sp.check_call(argv)
+else:
+    sys.path.insert(0, os.getcwd())
+    from copyparty.__main__ import main as copyparty
+
+    try:
+        copyparty(["a"] + argv)
+    except SystemExit as ex:
+        if ex.code:
+            raise
 
 print("\n\033[32mokke\033[0m")
 sys.exit(1)

README.md

@@ -37,6 +37,7 @@ turn your phone or raspi into a portable file server with resumable uploads/down
     * [other tricks](#other-tricks)
 * [searching](#searching)
     * [search configuration](#search-configuration)
+    * [database location](#database-location)
     * [metadata from audio files](#metadata-from-audio-files)
     * [file parser plugins](#file-parser-plugins)
     * [complete examples](#complete-examples)
@@ -68,12 +69,16 @@ you may also want these, especially on servers:
 
 ## notes
 
+general:
+* paper-printing is affected by dark/light-mode! use lightmode for color, darkmode for grayscale
+  * because no browsers currently implement the media-query to do this properly orz
+
+browser-specific:
 * iPhone/iPad: use Firefox to download files
 * Android-Chrome: increase "parallel uploads" for higher speed (android bug)
 * Android-Firefox: takes a while to select files (their fix for ☝️)
 * Desktop-Firefox: ~~may use gigabytes of RAM if your files are massive~~ *seems to be OK now*
-* paper-printing is affected by dark/light-mode! use lightmode for color, darkmode for grayscale
-  * because no browsers currently implement the media-query to do this properly orz
+* Desktop-Firefox: may stop you from deleting folders you've uploaded until you visit `about:memory` and click `Minimize memory usage`
 
 
 ## status
@@ -117,12 +122,12 @@ summary: all planned features work! now please enjoy the bloatening
 * Windows: python 3.7 and older cannot read tags with ffprobe, so use mutagen or upgrade
 * Windows: python 2.7 cannot index non-ascii filenames with `-e2d`
 * Windows: python 2.7 cannot handle filenames with mojibake
+* MacOS: `--th-ff-jpg` may fix thumbnails using macports-FFmpeg
 
 ## general bugs
 
 * all volumes must exist / be available on startup; up2k (mtp especially) gets funky otherwise
 * cannot mount something at `/d1/d2/d3` unless `d2` exists inside `d1`
-* hiding the contents at url `/d1/d2/d3` using `-v :d1/d2/d3:cd2d` has the side-effect of creating databases (for files/tags) inside folders d1 and d2, and those databases take precedence over the main db at the top of the vfs - this means all files in d2 and below will be reindexed unless you already had a vfs entry at or below d2
 * probably more, pls let me know
 
 ## not my bugs
@@ -176,6 +181,8 @@ click `[-]` and `[+]` to adjust the size, and the `[a]` toggles if the tree shou
 
 it does static images with Pillow and uses FFmpeg for video files, so you may want to `--no-thumb` or maybe just `--no-vthumb` depending on how destructive your users are
 
+images named `folder.jpg` and `folder.png` become the thumbnail of the folder they're in
+
 
 ## zip downloads
 
@@ -292,9 +299,29 @@ the same arguments can be set as volume flags, in addition to `d2d` and `d2t` fo
 * `-v ~/music::r:cd2d` disables **all** indexing, even if any `-e2*` are on
 * `-v ~/music::r:cd2t` disables all `-e2t*` (tags), does not affect `-e2d*`
 
-`e2tsr` is probably always overkill, since `e2ds`/`e2dsa` would pick up any file modifications and cause `e2ts` to reindex those
+note:
+* `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
 
-the rescan button in the admin panel has no effect unless the volume has `-e2ds` or higher
+you can choose to only index filename/path/size/last-modified (and not the hash of the file contents) by setting `--no-hash` or the volume-flag `cdhash`, this has the following consequences:
+* initial indexing is way faster, especially when the volume is on a networked disk
+* makes it impossible to [file-search](#file-search)
+* if someone uploads the same file contents, the upload will not be detected as a dupe, so it will not get symlinked or rejected
+
+if you set `--no-hash`, you can enable hashing for specific volumes using flag `cehash`
+
+
+## database location
+
+copyparty creates a subfolder named `.hist` inside each volume where it stores the database, thumbnails, and some other stuff
+
+this can instead be kept in a single place using the `--hist` argument, or the `hist=` volume flag, or a mix of both:
+* `--hist ~/.cache/copyparty -v ~/music::r:chist=-` sets `~/.cache/copyparty` as the default place to put volume info, but `~/music` gets the regular `.hist` subfolder (`-` restores default behavior)
+
+note:
+* markdown edits are always stored in a local `.hist` subdirectory
+* on windows the volflag path is cyglike, so `/c/temp` means `C:\temp` but use regular paths for `--hist`
+  * you can use cygpaths for volumes too, `-v C:\Users::r` and `-v /c/users::r` both work
 
 
 ## metadata from audio files
@@ -397,6 +424,8 @@ quick summary of more eccentric web-browsers trying to view a directory index:
   * cross-platform python client available in [./bin/](bin/)
   * [rclone](https://rclone.org/) as client can give ~5x performance, see [./docs/rclone.md](docs/rclone.md)
 
+* sharex (screenshot utility): see [./contrib/sharex.sxcu](contrib/#sharexsxcu)
+
 copyparty returns a truncated sha512sum of your PUT/POST as base64; you can generate the same checksum locally to verify uplaods:
 
     b512(){ printf "$((sha512sum||shasum -a512)|sed -E 's/ .*//;s/(..)/\\x\1/g')"|base64|head -c43;}
@@ -520,20 +549,25 @@ in the `scripts` folder:
 
 roughly sorted by priority
 
-* separate sqlite table per tag
-* audio fingerprinting
 * readme.md as epilogue
+* single sha512 across all up2k chunks? maybe
 * reduce up2k roundtrips
   * start from a chunk index and just go
   * terminate client on bad data
-* `os.copy_file_range` for up2k cloning
-* single sha512 across all up2k chunks? maybe
-* figure out the deal with pixel3a not being connectable as hotspot
-  * pixel3a having unpredictable 3sec latency in general :||||
 
 discarded ideas
 
+* separate sqlite table per tag
+  * performance fixed by skipping some indexes (`+mt.k`)
+* audio fingerprinting
+  * only makes sense if there can be a wasm client and that doesn't exist yet (except for olaf which is agpl hence counts as not existing)
+* `os.copy_file_range` for up2k cloning
+  * almost never hit this path anyways
 * up2k partials ui
+  * feels like there isn't much point
 * cache sha512 chunks on client
+  * too dangerous
 * comment field
+  * nah
 * look into android thumbnail cache file format
+  * absolutely not

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/copyparty-fuse.py

@@ -54,6 +54,12 @@ MACOS = platform.system() == "Darwin"
 info = log = dbg = None
 
 
+print("{} v{} @ {}".format(
+    platform.python_implementation(),
+    ".".join([str(x) for x in sys.version_info]),
+    sys.executable))
+
+
 try:
     from fuse import FUSE, FuseOSError, Operations
 except:

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()

contrib/README.md

@@ -9,6 +9,16 @@
 * assumes the webserver and copyparty is running on the same server/IP
 * modify `10.13.1.1` as necessary if you wish to support browsers without javascript
 
+### [`sharex.sxcu`](sharex.sxcu)
+* sharex config file to upload screenshots and grab the URL
+* `RequestURL`: full URL to the target folder
+* `pw`: password (remove the `pw` line if anon-write)
+
+however if your copyparty is behind a reverse-proxy, you may want to use [`sharex-html.sxcu`](sharex-html.sxcu) instead:
+* `RequestURL`: full URL to the target folder
+* `URL`: full URL to the root folder (with trailing slash) followed by `$regex:1|1$`
+* `pw`: password (remove `Parameters` if anon-write)
+
 ### [`explorer-nothumbs-nofoldertypes.reg`](explorer-nothumbs-nofoldertypes.reg)
 * disables thumbnails and folder-type detection in windows explorer
 * makes it way faster (especially for slow/networked locations (such as copyparty-fuse))

contrib/sharex-html.sxcu

@@ -0,0 +1,19 @@
+{
+  "Version": "13.5.0",
+  "Name": "copyparty-html",
+  "DestinationType": "ImageUploader",
+  "RequestMethod": "POST",
+  "RequestURL": "http://127.0.0.1:3923/sharex",
+  "Parameters": {
+    "pw": "wark"
+  },
+  "Body": "MultipartFormData",
+  "Arguments": {
+    "act": "bput"
+  },
+  "FileFormName": "f",
+  "RegexList": [
+    "bytes // <a href=\"/([^\"]+)\""
+  ],
+  "URL": "http://127.0.0.1:3923/$regex:1|1$"
+}

contrib/sharex.sxcu

@@ -0,0 +1,17 @@
+{
+  "Version": "13.5.0",
+  "Name": "copyparty",
+  "DestinationType": "ImageUploader",
+  "RequestMethod": "POST",
+  "RequestURL": "http://127.0.0.1:3923/sharex",
+  "Parameters": {
+    "pw": "wark",
+    "j": null
+  },
+  "Body": "MultipartFormData",
+  "Arguments": {
+    "act": "bput"
+  },
+  "FileFormName": "f",
+  "URL": "$json:files[0].url$"
+}

copyparty/__main__.py

@@ -225,6 +225,20 @@ def run_argparse(argv, formatter):
             --ciphers help = available ssl/tls ciphers,
             --ssl-ver help = available ssl/tls versions,
               default is what python considers safe, usually >= TLS1
+            
+            values for --ls:
+              "USR" is a user to browse as; * is anonymous, ** is all users
+              "VOL" is a single volume to scan, default is * (all vols)
+              "FLAG" is flags;
+                "v" in addition to realpaths, print usernames and vpaths
+                "ln" only prints symlinks leaving the volume mountpoint
+                "p" exits 1 if any such symlinks are found
+                "r" resumes startup after the listing
+            examples:
+              --ls '**'          # list all files which are possible to read
+              --ls '**,*,ln'     # check for dangerous symlinks
+              --ls '**,*,ln,p,r' # check, then start normally if safe
+            \033[0m
             """
         ),
     )
@@ -249,6 +263,9 @@ 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('appearance options')
+    ap2.add_argument("--css-browser", metavar="L", help="URL to additional CSS to include")
+
     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)")
@@ -260,8 +277,9 @@ def run_argparse(argv, formatter):
     ap2.add_argument("--th-no-crop", action="store_true", help="dynamic height; show full image")
     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-ff-jpg", action="store_true", help="force jpg for video thumbs")
     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')
@@ -271,6 +289,8 @@ def run_argparse(argv, formatter):
     ap2.add_argument("-e2t", action="store_true", help="enable metadata indexing")
     ap2.add_argument("-e2ts", action="store_true", help="enable metadata scanner, sets -e2t")
     ap2.add_argument("-e2tsr", action="store_true", help="rescan all metadata, sets -e2ts")
+    ap2.add_argument("--hist", metavar="PATH", type=str, help="where to store volume state")
+    ap2.add_argument("--no-hash", action="store_true", help="disable hashing during e2ds folder scans")
     ap2.add_argument("--no-mutagen", action="store_true", help="use ffprobe for tags instead")
     ap2.add_argument("--no-mtag-mt", action="store_true", help="disable tag-read parallelism")
     ap2.add_argument("-mtm", metavar="M=t,t,t", action="append", type=str, help="add/replace metadata mapping")
@@ -288,6 +308,7 @@ def run_argparse(argv, formatter):
     ap2.add_argument("--ssl-log", metavar="PATH", help="log master secrets")
 
     ap2 = ap.add_argument_group('debug options')
+    ap2.add_argument("--ls", metavar="U[,V[,F]]", help="scan all volumes")
     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")
@@ -363,6 +384,9 @@ def main(argv=None):
             + "  (if you crash with codec errors then that is why)"
         )
 
+    if WINDOWS and sys.version_info < (3, 6):
+        al.no_scandir = True
+
     # signal.signal(signal.SIGINT, sighandler)
 
     SvcHub(al).run()

copyparty/__version__.py

@@ -1,8 +1,8 @@
 # coding: utf-8
 
-VERSION = (0, 11, 1)
+VERSION = (0, 11, 12)
 CODENAME = "the grid"
-BUILD_DT = (2021, 5, 29)
+BUILD_DT = (2021, 6, 12)
 
 S_VERSION = ".".join(map(str, VERSION))
 S_BUILD_DT = "{0:04d}-{1:02d}-{2:02d}".format(*BUILD_DT)

copyparty/authsrv.py

@@ -5,9 +5,11 @@ import re
 import os
 import sys
 import stat
+import base64
+import hashlib
 import threading
 
-from .__init__ import PY2, WINDOWS
+from .__init__ import WINDOWS
 from .util import IMPLICATIONS, undot, Pebkac, fsdec, fsenc, statdir, nuprint
 
 
@@ -22,7 +24,15 @@ class VFS(object):
         self.uadm = uadm  # users who are regular admins
         self.flags = flags  # config switches
         self.nodes = {}  # child nodes
-        self.all_vols = {vpath: self}  # flattened recursive
+        self.histtab = None  # all realpath->histpath
+        self.dbv = None  # closest full/non-jump parent
+
+        if realpath:
+            self.histpath = os.path.join(realpath, ".hist")  # db / thumbcache
+            self.all_vols = {vpath: self}  # flattened recursive
+        else:
+            self.histpath = None
+            self.all_vols = None
 
     def __repr__(self):
         return "VFS({})".format(
@@ -32,9 +42,12 @@ class VFS(object):
             )
         )
 
-    def _trk(self, vol):
-        self.all_vols[vol.vpath] = vol
-        return vol
+    def get_all_vols(self, outdict):
+        if self.realpath:
+            outdict[self.vpath] = self
+
+        for v in self.nodes.values():
+            v.get_all_vols(outdict)
 
     def add(self, src, dst):
         """get existing, or add new path to the vfs"""
@@ -46,19 +59,19 @@ class VFS(object):
             name, dst = dst.split("/", 1)
             if name in self.nodes:
                 # exists; do not manipulate permissions
-                return self._trk(self.nodes[name].add(src, dst))
+                return self.nodes[name].add(src, dst)
 
             vn = VFS(
-                "{}/{}".format(self.realpath, name),
+                os.path.join(self.realpath, name) if self.realpath else None,
                 "{}/{}".format(self.vpath, name).lstrip("/"),
                 self.uread,
                 self.uwrite,
                 self.uadm,
-                self.flags,
+                self._copy_flags(name),
             )
-            self._trk(vn)
+            vn.dbv = self.dbv or self
             self.nodes[name] = vn
-            return self._trk(vn.add(src, dst))
+            return vn.add(src, dst)
 
         if dst in self.nodes:
             # leaf exists; return as-is
@@ -67,8 +80,26 @@ class VFS(object):
         # leaf does not exist; create and keep permissions blank
         vp = "{}/{}".format(self.vpath, dst).lstrip("/")
         vn = VFS(src, vp)
+        vn.dbv = self.dbv or self
         self.nodes[dst] = vn
-        return self._trk(vn)
+        return vn
+
+    def _copy_flags(self, name):
+        flags = {k: v for k, v in self.flags.items()}
+        hist = flags.get("hist")
+        if hist and hist != "-":
+            flags["hist"] = "{}/{}".format(hist.rstrip("/"), name)
+
+        return flags
+
+    def bubble_flags(self):
+        if self.dbv:
+            for k, v in self.dbv.flags.items():
+                if k not in ["hist"]:
+                    self.flags[k] = v
+
+        for v in self.nodes.values():
+            v.bubble_flags()
 
     def _find(self, vpath):
         """return [vfs,remainder]"""
@@ -96,6 +127,7 @@ class VFS(object):
         ]
 
     def get(self, vpath, uname, will_read, will_write):
+        # type: (str, str, bool, bool) -> tuple[VFS, str]
         """returns [vfsnode,fs_remainder] if user has the requested permissions"""
         vn, rem = self._find(vpath)
 
@@ -107,6 +139,15 @@ class VFS(object):
 
         return vn, rem
 
+    def get_dbv(self, vrem):
+        dbv = self.dbv
+        if not dbv:
+            return self, vrem
+
+        vrem = [self.vpath[len(dbv.vpath) + 1 :], vrem]
+        vrem = "/".join([x for x in vrem if x])
+        return dbv, vrem
+
     def canonical(self, rem):
         """returns the canonical path (fully-resolved absolute fs path)"""
         rp = self.realpath
@@ -135,7 +176,8 @@ 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):
+        # type: (str, str, bool, bool, bool) -> tuple[str, str, dict[str, VFS]]
         """return user-readable [fsdir,real,virt] items at vpath"""
         virt_vis = {}  # nodes readable by user
         abspath = self.canonical(rem)
@@ -143,12 +185,12 @@ 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
@@ -156,13 +198,21 @@ class VFS(object):
 
         return [abspath, real, virt_vis]
 
-    def walk(self, rel, rem, uname, dots, scandir, lstat=False):
+    def walk(self, rel, rem, seen, uname, dots, scandir, lstat):
         """
         recursively yields from ./rem;
         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, incl_wo=False, lstat=lstat
+        )
+
+        if seen and not fsroot.startswith(seen[-1]) and fsroot in seen:
+            print("bailing from symlink loop,\n  {}\n  {}".format(seen[-1], fsroot))
+            return
+
+        seen = seen[:] + [fsroot]
         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)]
 
@@ -177,7 +227,7 @@ class VFS(object):
 
             wrel = (rel + "/" + rdir).lstrip("/")
             wrem = (rem + "/" + rdir).lstrip("/")
-            for x in self.walk(wrel, wrem, uname, scandir, lstat):
+            for x in self.walk(wrel, wrem, seen, uname, dots, scandir, lstat):
                 yield x
 
         for n, vfs in sorted(vfs_virt.items()):
@@ -185,14 +235,16 @@ class VFS(object):
                 continue
 
             wrel = (rel + "/" + n).lstrip("/")
-            for x in vfs.walk(wrel, "", uname, scandir, lstat):
+            for x in vfs.walk(wrel, "", seen, uname, dots, scandir, lstat):
                 yield x
 
     def zipgen(self, vrem, flt, uname, dots, scandir):
         if flt:
             flt = {k: True for k in flt}
 
-        for vpath, apath, files, rd, vd in self.walk("", vrem, uname, dots, scandir):
+        for vpath, apath, files, rd, vd in self.walk(
+            "", vrem, [], uname, dots, scandir, False
+        ):
             if flt:
                 files = [x for x in files if x[0] in flt]
 
@@ -228,21 +280,19 @@ 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, 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 admin:
-            if opt1 and opt2:
-                ret.append(self.vpath)
-        else:
-            if opt1 or opt2:
-                ret.append(self.vpath)
+    def user_tree(self, uname, readable, writable, admin):
+        is_readable = False
+        if uname in self.uread or "*" in self.uread:
+            readable.append(self.vpath)
+            is_readable = True
+
+        if uname in self.uwrite or "*" in self.uwrite:
+            writable.append(self.vpath)
+            if is_readable:
+                admin.append(self.vpath)
 
         for _, vn in sorted(self.nodes.items()):
-            ret.extend(vn.user_tree(uname, readable, writable, admin))
-
-        return ret
+            vn.user_tree(uname, readable, writable, admin)
 
 
 class AuthSrv(object):
@@ -263,7 +313,8 @@ class AuthSrv(object):
         self.reload()
 
     def log(self, msg, c=0):
-        self.log_func("auth", msg, c)
+        if self.log_func:
+            self.log_func("auth", msg, c)
 
     def laggy_iter(self, iterable):
         """returns [value,isFinalValue]"""
@@ -388,6 +439,9 @@ class AuthSrv(object):
                     raise Exception("invalid -v argument: [{}]".format(v_str))
 
                 src, dst, perms = m.groups()
+                if WINDOWS and src.startswith("/"):
+                    src = "{}:\\{}".format(src[1], src[3:])
+
                 # print("\n".join([src, dst, perms]))
                 src = fsdec(os.path.abspath(fsenc(src)))
                 dst = dst.strip("/")
@@ -420,7 +474,7 @@ class AuthSrv(object):
             vfs = VFS(os.path.abspath("."), "", ["*"], ["*"])
         elif "" not in mount:
             # there's volumes but no root; make root inaccessible
-            vfs = VFS(os.path.abspath("."), "")
+            vfs = VFS(None, "")
             vfs.flags["d2d"] = True
 
         maxdepth = 0
@@ -441,6 +495,10 @@ class AuthSrv(object):
             v.uwrite = mwrite[dst]
             v.uadm = madm[dst]
             v.flags = mflags[dst]
+            v.dbv = None
+
+        vfs.all_vols = {}
+        vfs.get_all_vols(vfs.all_vols)
 
         missing_users = {}
         for d in [mread, mwrite]:
@@ -457,6 +515,69 @@ class AuthSrv(object):
             )
             raise Exception("invalid config")
 
+        promote = []
+        demote = []
+        for vol in vfs.all_vols.values():
+            hid = hashlib.sha512(fsenc(vol.realpath)).digest()
+            hid = base64.b32encode(hid).decode("ascii").lower()
+            vflag = vol.flags.get("hist")
+            if vflag == "-":
+                pass
+            elif vflag:
+                if WINDOWS and vflag.startswith("/"):
+                    vflag = "{}:\\{}".format(vflag[1], vflag[3:])
+                vol.histpath = vflag
+            elif self.args.hist:
+                for nch in range(len(hid)):
+                    hpath = os.path.join(self.args.hist, hid[: nch + 1])
+                    try:
+                        os.makedirs(hpath)
+                    except:
+                        pass
+
+                    powner = os.path.join(hpath, "owner.txt")
+                    try:
+                        with open(powner, "rb") as f:
+                            owner = f.read().rstrip()
+                    except:
+                        owner = None
+
+                    me = fsenc(vol.realpath).rstrip()
+                    if owner not in [None, me]:
+                        continue
+
+                    if owner is None:
+                        with open(powner, "wb") as f:
+                            f.write(me)
+
+                    vol.histpath = hpath
+                    break
+
+            vol.histpath = os.path.realpath(vol.histpath)
+            if vol.dbv:
+                if os.path.exists(os.path.join(vol.histpath, "up2k.db")):
+                    promote.append(vol)
+                    vol.dbv = None
+                else:
+                    demote.append(vol)
+
+        # discard jump-vols
+        for v in demote:
+            vfs.all_vols.pop(v.vpath)
+
+        if promote:
+            msg = [
+                "\n  the following jump-volumes were generated to assist the vfs.\n  As they contain a database (probably from v0.11.11 or older),\n  they are promoted to full volumes:"
+            ]
+            for vol in promote:
+                msg.append(
+                    "  /{}  ({})  ({})".format(vol.vpath, vol.realpath, vol.histpath)
+                )
+
+            self.log("\n\n".join(msg) + "\n", c=3)
+
+        vfs.histtab = {v.realpath: v.histpath for v in vfs.all_vols.values()}
+
         all_mte = {}
         errors = False
         for vol in vfs.all_vols.values():
@@ -466,6 +587,10 @@ class AuthSrv(object):
             if self.args.e2d or "e2ds" in vol.flags:
                 vol.flags["e2d"] = True
 
+            if self.args.no_hash:
+                if "ehash" not in vol.flags:
+                    vol.flags["dhash"] = True
+
             for k in ["e2t", "e2ts", "e2tsr"]:
                 if getattr(self.args, k):
                     vol.flags[k] = True
@@ -543,6 +668,8 @@ class AuthSrv(object):
         if errors:
             sys.exit(1)
 
+        vfs.bubble_flags()
+
         try:
             v, _ = vfs.get("/", "*", False, True)
             if self.warn_anonwrite and os.getcwd() == v.realpath:
@@ -559,3 +686,90 @@ class AuthSrv(object):
 
         # import pprint
         # pprint.pprint({"usr": user, "rd": mread, "wr": mwrite, "mnt": mount})
+
+    def dbg_ls(self):
+        users = self.args.ls
+        vols = "*"
+        flags = []
+
+        try:
+            users, vols = users.split(",", 1)
+        except:
+            pass
+
+        try:
+            vols, flags = vols.split(",", 1)
+            flags = flags.split(",")
+        except:
+            pass
+
+        if users == "**":
+            users = list(self.user.keys()) + ["*"]
+        else:
+            users = [users]
+
+        for u in users:
+            if u not in self.user and u != "*":
+                raise Exception("user not found: " + u)
+
+        if vols == "*":
+            vols = ["/" + x for x in self.vfs.all_vols.keys()]
+        else:
+            vols = [vols]
+
+        for v in vols:
+            if not v.startswith("/"):
+                raise Exception("volumes must start with /")
+
+            if v[1:] not in self.vfs.all_vols:
+                raise Exception("volume not found: " + v)
+
+        self.log({"users": users, "vols": vols, "flags": flags})
+        for k, v in self.vfs.all_vols.items():
+            self.log("/{}: read({}) write({})".format(k, v.uread, v.uwrite))
+
+        flag_v = "v" in flags
+        flag_ln = "ln" in flags
+        flag_p = "p" in flags
+        flag_r = "r" in flags
+
+        n_bads = 0
+        for v in vols:
+            v = v[1:]
+            vtop = "/{}/".format(v) if v else "/"
+            for u in users:
+                self.log("checking /{} as {}".format(v, u))
+                try:
+                    vn, _ = self.vfs.get(v, u, True, False)
+                except:
+                    continue
+
+                atop = vn.realpath
+                g = vn.walk("", "", [], u, True, not self.args.no_scandir, False)
+                for vpath, apath, files, _, _ in g:
+                    fnames = [n[0] for n in files]
+                    vpaths = [vpath + "/" + n for n in fnames] if vpath else fnames
+                    vpaths = [vtop + x for x in vpaths]
+                    apaths = [os.path.join(apath, n) for n in fnames]
+                    files = [[vpath + "/", apath + os.sep]] + list(zip(vpaths, apaths))
+
+                    if flag_ln:
+                        files = [x for x in files if not x[1].startswith(atop + os.sep)]
+                        n_bads += len(files)
+
+                    if flag_v:
+                        msg = [
+                            '# user "{}", vpath "{}"\n{}'.format(u, vp, ap)
+                            for vp, ap in files
+                        ]
+                    else:
+                        msg = [x[1] for x in files]
+
+                    if msg:
+                        nuprint("\n".join(msg))
+
+                if n_bads and flag_p:
+                    raise Exception("found symlink leaving volume, and strict is set")
+
+        if not flag_r:
+            sys.exit(0)

copyparty/broker_mp.py

@@ -44,7 +44,9 @@ class BrokerMp(object):
             proc.clients = {}
             proc.workload = 0
 
-            thr = threading.Thread(target=self.collector, args=(proc,))
+            thr = threading.Thread(
+                target=self.collector, args=(proc,), name="mp-collector"
+            )
             thr.daemon = True
             thr.start()
 
@@ -52,14 +54,19 @@ class BrokerMp(object):
             proc.start()
 
         if not self.args.q:
-            thr = threading.Thread(target=self.debug_load_balancer)
+            thr = threading.Thread(
+                target=self.debug_load_balancer, name="mp-dbg-loadbalancer"
+            )
             thr.daemon = True
             thr.start()
 
     def shutdown(self):
         self.log("broker", "shutting down")
-        for proc in self.procs:
-            thr = threading.Thread(target=proc.q_pend.put([0, "shutdown", []]))
+        for n, proc in enumerate(self.procs):
+            thr = threading.Thread(
+                target=proc.q_pend.put([0, "shutdown", []]),
+                name="mp-shutdown-{}-{}".format(n, len(self.procs)),
+            )
             thr.start()
 
         with self.mutex:

copyparty/broker_mpw.py

@@ -1,5 +1,6 @@
 # coding: utf-8
 from __future__ import print_function, unicode_literals
+from copyparty.authsrv import AuthSrv
 
 import sys
 import time
@@ -27,20 +28,23 @@ class MpWorker(object):
         self.retpend = {}
         self.retpend_mutex = threading.Lock()
         self.mutex = threading.Lock()
-        self.workload_thr_active = False
+        self.workload_thr_alive = False
 
         # we inherited signal_handler from parent,
         # replace it with something harmless
         if not FAKE_MP:
             signal.signal(signal.SIGINT, self.signal_handler)
 
+        # starting to look like a good idea
+        self.asrv = AuthSrv(args, None, False)
+
         # instantiate all services here (TODO: inheritance?)
-        self.httpsrv = HttpSrv(self)
+        self.httpsrv = HttpSrv(self, True)
         self.httpsrv.disconnect_func = self.httpdrop
 
         # on winxp and some other platforms,
         # use thr.join() to block all signals
-        thr = threading.Thread(target=self.main)
+        thr = threading.Thread(target=self.main, name="mpw-main")
         thr.daemon = True
         thr.start()
         thr.join()
@@ -75,13 +79,15 @@ class MpWorker(object):
 
                 if self.args.log_conn:
                     self.log("%s %s" % addr, "|%sC-qpop" % ("-" * 4,), c="1;30")
-                
+
                 self.httpsrv.accept(sck, addr)
 
                 with self.mutex:
-                    if not self.workload_thr_active:
+                    if not self.workload_thr_alive:
                         self.workload_thr_alive = True
-                        thr = threading.Thread(target=self.thr_workload)
+                        thr = threading.Thread(
+                            target=self.thr_workload, name="mpw-workload"
+                        )
                         thr.daemon = True
                         thr.start()
 

copyparty/broker_thr.py

@@ -3,6 +3,7 @@ from __future__ import print_function, unicode_literals
 
 import threading
 
+from .authsrv import AuthSrv
 from .httpsrv import HttpSrv
 from .broker_util import ExceptionalQueue, try_exec
 
@@ -14,6 +15,7 @@ class BrokerThr(object):
         self.hub = hub
         self.log = hub.log
         self.args = hub.args
+        self.asrv = hub.asrv
 
         self.mutex = threading.Lock()
 

copyparty/httpcli.py

@@ -16,6 +16,7 @@ import calendar
 
 from .__init__ import E, PY2, WINDOWS, ANYWIN
 from .util import *  # noqa  # pylint: disable=unused-wildcard-import
+from .authsrv import AuthSrv
 from .szip import StreamZip
 from .star import StreamTar
 
@@ -35,12 +36,13 @@ class HttpCli(object):
     def __init__(self, conn):
         self.t0 = time.time()
         self.conn = conn
-        self.s = conn.s
-        self.sr = conn.sr
+        self.s = conn.s  # type: socket
+        self.sr = conn.sr  # type: Unrecv
         self.ip = conn.addr[0]
-        self.addr = conn.addr
+        self.addr = conn.addr  # type: tuple[str, int]
         self.args = conn.args
-        self.auth = conn.auth
+        self.is_mp = conn.is_mp
+        self.asrv = conn.asrv  # type: AuthSrv
         self.ico = conn.ico
         self.thumbcli = conn.thumbcli
         self.log_func = conn.log_func
@@ -152,11 +154,9 @@ class HttpCli(object):
         self.vpath = unquotep(vpath)
 
         pwd = uparam.get("pw")
-        self.uname = self.auth.iuser.get(pwd, "*")
-        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)
+        self.uname = self.asrv.iuser.get(pwd, "*")
+        self.rvol, self.wvol, self.avol = [[], [], []]
+        self.asrv.vfs.user_tree(self.uname, self.rvol, self.wvol, self.avol)
 
         ua = self.headers.get("user-agent", "")
         self.is_rclone = ua.startswith("rclone/")
@@ -258,7 +258,14 @@ class HttpCli(object):
         return "?" + "&".join(r)
 
     def redirect(
-        self, vpath, suf="", msg="aight", flavor="go to", click=True, use302=False
+        self,
+        vpath,
+        suf="",
+        msg="aight",
+        flavor="go to",
+        click=True,
+        status=200,
+        use302=False,
     ):
         html = self.j2(
             "msg",
@@ -273,7 +280,7 @@ class HttpCli(object):
             h = {"Location": "/" + vpath, "Cache-Control": "no-cache"}
             self.reply(html, status=302, headers=h)
         else:
-            self.reply(html)
+            self.reply(html, status=status)
 
     def handle_get(self):
         if self.do_log:
@@ -314,9 +321,7 @@ class HttpCli(object):
                     self.redirect(vpath, flavor="redirecting to", use302=True)
                     return True
 
-        self.readable, self.writable = self.conn.auth.vfs.can_access(
-            self.vpath, self.uname
-        )
+        self.readable, self.writable = self.asrv.vfs.can_access(self.vpath, self.uname)
         if not self.readable and not self.writable:
             if self.vpath:
                 self.log("inaccessible: [{}]".format(self.vpath))
@@ -433,7 +438,7 @@ class HttpCli(object):
 
     def dump_to_file(self):
         reader, remains = self.get_body_reader()
-        vfs, rem = self.conn.auth.vfs.get(self.vpath, self.uname, False, True)
+        vfs, rem = self.asrv.vfs.get(self.vpath, self.uname, False, True)
         fdir = os.path.join(vfs.realpath, rem)
 
         addr = self.ip.replace(":", ".")
@@ -443,8 +448,10 @@ class HttpCli(object):
         with open(fsenc(path), "wb", 512 * 1024) as f:
             post_sz, _, sha_b64 = hashcopy(self.conn, reader, f)
 
+        vfs, vrem = vfs.get_dbv(rem)
+
         self.conn.hsrv.broker.put(
-            False, "up2k.hash_file", vfs.realpath, vfs.flags, rem, fn
+            False, "up2k.hash_file", vfs.realpath, vfs.flags, vrem, fn
         )
 
         return post_sz, sha_b64, remains, path
@@ -500,7 +507,7 @@ class HttpCli(object):
         if v is None:
             raise Pebkac(422, "need zip or tar keyword")
 
-        vn, rem = self.auth.vfs.get(self.vpath, self.uname, True, False)
+        vn, rem = self.asrv.vfs.get(self.vpath, self.uname, True, False)
         items = self.parser.require("files", 1024 * 1024)
         if not items:
             raise Pebkac(422, "need files list")
@@ -508,6 +515,7 @@ class HttpCli(object):
         items = items.replace("\r", "").split("\n")
         items = [unquotep(x) for x in items if items]
 
+        self.parser.drop()
         return self.tx_zip(k, v, vn, rem, items, self.args.ed)
 
     def handle_post_json(self):
@@ -549,13 +557,14 @@ class HttpCli(object):
             self.vpath = "/".join([self.vpath, sub]).strip("/")
             body["name"] = name
 
-        vfs, rem = self.conn.auth.vfs.get(self.vpath, self.uname, False, True)
+        vfs, rem = self.asrv.vfs.get(self.vpath, self.uname, False, True)
+        dbv, vrem = vfs.get_dbv(rem)
 
-        body["vtop"] = vfs.vpath
-        body["ptop"] = vfs.realpath
-        body["prel"] = rem
+        body["vtop"] = dbv.vpath
+        body["ptop"] = dbv.realpath
+        body["prel"] = vrem
         body["addr"] = self.ip
-        body["vcfg"] = vfs.flags
+        body["vcfg"] = dbv.flags
 
         if sub:
             try:
@@ -577,8 +586,14 @@ class HttpCli(object):
 
     def handle_search(self, body):
         vols = []
+        seen = {}
         for vtop in self.rvol:
-            vfs, _ = self.conn.auth.vfs.get(vtop, self.uname, True, False)
+            vfs, _ = self.asrv.vfs.get(vtop, self.uname, True, False)
+            vfs = vfs.dbv or vfs
+            if vfs in seen:
+                continue
+
+            seen[vfs] = True
             vols.append([vfs.vpath, vfs.realpath, vfs.flags])
 
         idx = self.conn.get_u2idx()
@@ -600,8 +615,9 @@ 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()
@@ -633,8 +649,8 @@ class HttpCli(object):
         except KeyError:
             raise Pebkac(400, "need hash and wark headers for binary POST")
 
-        vfs, _ = self.conn.auth.vfs.get(self.vpath, self.uname, False, True)
-        ptop = vfs.realpath
+        vfs, _ = self.asrv.vfs.get(self.vpath, self.uname, False, True)
+        ptop = (vfs.dbv or vfs).realpath
 
         x = self.conn.hsrv.broker.put(True, "up2k.handle_chunk", ptop, wark, chash)
         response = x.get()
@@ -706,7 +722,7 @@ class HttpCli(object):
         pwd = self.parser.require("cppwd", 64)
         self.parser.drop()
 
-        if pwd in self.auth.iuser:
+        if pwd in self.asrv.iuser:
             msg = "login ok"
             dt = datetime.utcfromtimestamp(time.time() + 60 * 60 * 24 * 365)
             exp = dt.strftime("%a, %d %b %Y %H:%M:%S GMT")
@@ -725,7 +741,7 @@ class HttpCli(object):
         self.parser.drop()
 
         nullwrite = self.args.nw
-        vfs, rem = self.conn.auth.vfs.get(self.vpath, self.uname, False, True)
+        vfs, rem = self.asrv.vfs.get(self.vpath, self.uname, False, True)
         self._assert_safe_rem(rem)
 
         sanitized = sanitize_fn(new_dir)
@@ -754,7 +770,7 @@ class HttpCli(object):
         self.parser.drop()
 
         nullwrite = self.args.nw
-        vfs, rem = self.conn.auth.vfs.get(self.vpath, self.uname, False, True)
+        vfs, rem = self.asrv.vfs.get(self.vpath, self.uname, False, True)
         self._assert_safe_rem(rem)
 
         if not new_file.endswith(".md"):
@@ -778,7 +794,7 @@ class HttpCli(object):
 
     def handle_plain_upload(self):
         nullwrite = self.args.nw
-        vfs, rem = self.conn.auth.vfs.get(self.vpath, self.uname, False, True)
+        vfs, rem = self.asrv.vfs.get(self.vpath, self.uname, False, True)
         self._assert_safe_rem(rem)
 
         files = []
@@ -815,8 +831,14 @@ class HttpCli(object):
                             raise Pebkac(400, "empty files in post")
 
                         files.append([sz, sha512_hex, p_file, fname])
+                        dbv, vrem = vfs.get_dbv(rem)
                         self.conn.hsrv.broker.put(
-                            False, "up2k.hash_file", vfs.realpath, vfs.flags, rem, fname
+                            False,
+                            "up2k.hash_file",
+                            dbv.realpath,
+                            dbv.flags,
+                            vrem,
+                            fname,
                         )
                         self.conn.nbyte += sz
 
@@ -846,18 +868,36 @@ class HttpCli(object):
         status = "OK"
         if errmsg:
             self.log(errmsg)
-            errmsg = "ERROR: " + errmsg
             status = "ERROR"
 
         msg = "{} // {} bytes // {:.3f} MiB/s\n".format(status, sz_total, spd)
+        jmsg = {"status": status, "sz": sz_total, "mbps": round(spd, 3), "files": []}
+
+        if errmsg:
+            msg += errmsg + "\n"
+            jmsg["error"] = errmsg
+            errmsg = "ERROR: " + errmsg
 
         for sz, sha512, ofn, lfn in files:
-            vpath = self.vpath + "/" + lfn
+            vpath = (self.vpath + "/" if self.vpath else "") + lfn
             msg += 'sha512: {} // {} bytes // <a href="/{}">{}</a>\n'.format(
                 sha512[:56], sz, quotep(vpath), html_escape(ofn, crlf=True)
             )
             # truncated SHA-512 prevents length extension attacks;
             # using SHA-512/224, optionally SHA-512/256 = :64
+            jpart = {
+                "url": "{}://{}/{}".format(
+                    "https" if self.tls else "http",
+                    self.headers.get("host", "copyparty"),
+                    vpath,
+                ),
+                "sha512": sha512[:56],
+                "sz": sz,
+                "fn": lfn,
+                "fn_orig": ofn,
+                "path": vpath,
+            }
+            jmsg["files"].append(jpart)
 
         vspd = self._spd(sz_total, False)
         self.log("{} {}".format(vspd, msg))
@@ -869,7 +909,22 @@ class HttpCli(object):
                 ft = "{}\n{}\n{}\n".format(ft, msg.rstrip(), errmsg)
                 f.write(ft.encode("utf-8"))
 
-        self.redirect(self.vpath, msg=msg, flavor="return to", click=False)
+        status = 400 if errmsg else 200
+        if "j" in self.uparam:
+            jtxt = json.dumps(jmsg, indent=2, sort_keys=True).encode("utf-8", "replace")
+            self.reply(jtxt, mime="application/json", status=status)
+        else:
+            self.redirect(
+                self.vpath,
+                msg=msg,
+                flavor="return to",
+                click=False,
+                status=status,
+            )
+
+        if errmsg:
+            return False
+
         self.parser.drop()
         return True
 
@@ -880,7 +935,7 @@ class HttpCli(object):
             raise Pebkac(400, "could not read lastmod from request")
 
         nullwrite = self.args.nw
-        vfs, rem = self.conn.auth.vfs.get(self.vpath, self.uname, False, True)
+        vfs, rem = self.asrv.vfs.get(self.vpath, self.uname, False, True)
         self._assert_safe_rem(rem)
 
         # TODO:
@@ -973,6 +1028,8 @@ class HttpCli(object):
         cli_lastmod = self.headers.get("if-modified-since")
         if cli_lastmod:
             try:
+                # some browser append "; length=573"
+                cli_lastmod = cli_lastmod.split(";")[0].strip()
                 cli_dt = time.strptime(cli_lastmod, HTTP_TS_FMT)
                 cli_ts = calendar.timegm(cli_dt)
                 return file_lastmod, int(file_ts) > int(cli_ts)
@@ -1142,7 +1199,8 @@ class HttpCli(object):
             if use_sendfile:
                 remains = sendfile_kern(lower, upper, f, self.s)
             else:
-                remains = sendfile_py(lower, upper, f, self.s)
+                actor = self.conn if self.is_mp else None
+                remains = sendfile_py(lower, upper, f, self.s, actor)
 
         if remains > 0:
             logmsg += " \033[31m" + unicode(upper - remains) + "\033[0m"
@@ -1310,21 +1368,30 @@ 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]
+        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())
+            x = self.conn.hsrv.broker.put(True, "up2k.get_state")
+            vs = json.loads(x.get())
+            vstate = {("/" + k).rstrip("/") + "/": v for k, v in vs["volstate"].items()}
+        else:
+            vstate = {}
+            vs = {"scanning": None, "hashq": None, "tagq": None, "mtpq": None}
 
         html = self.j2(
             "splash",
             this=self,
             rvol=rvol,
             wvol=wvol,
-            avol=self.avol,
+            avol=avol,
             vstate=vstate,
+            scanning=vs["scanning"],
+            hashq=vs["hashq"],
+            tagq=vs["tagq"],
+            mtpq=vs["mtpq"],
             url_suf=suf,
         )
         self.reply(html.encode("utf-8"), headers=NO_STORE)
@@ -1337,9 +1404,10 @@ class HttpCli(object):
         if self.args.no_rescan:
             raise Pebkac(403, "disabled by argv")
 
-        vn, _ = self.auth.vfs.get(self.vpath, self.uname, True, True)
+        vn, _ = self.asrv.vfs.get(self.vpath, self.uname, True, True)
+
+        args = [self.asrv.vfs.all_vols, [vn.vpath]]
 
-        args = [self.auth.vfs.all_vols, [vn.vpath]]
         x = self.conn.hsrv.broker.put(True, "up2k.rescan", *args)
         x = x.get()
         if not x:
@@ -1355,15 +1423,31 @@ class HttpCli(object):
         if self.args.no_stack:
             raise Pebkac(403, "disabled by argv")
 
-        ret = []
+        threads = {}
         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))
+            name = "{} ({:x})".format(names.get(tid), tid)
+            threads[name] = stack
+
+        rret = []
+        bret = []
+        for name, stack in sorted(threads.items()):
+            ret = ["\n\n# {}".format(name)]
+            pad = None
             for fn, lno, name, line in traceback.extract_stack(stack):
+                fn = os.sep.join(fn.split(os.sep)[-3:])
                 ret.append('File: "{}", line {}, in {}'.format(fn, lno, name))
                 if line:
                     ret.append("  " + str(line.strip()))
+                    if "self.not_empty.wait()" in line:
+                        pad = " " * 4
 
+            if pad:
+                bret += [ret[0]] + [pad + x for x in ret[1:]]
+            else:
+                rret += ret
+
+        ret = rret + bret
         ret = ("<pre>" + "\n".join(ret)).encode("utf-8")
         self.reply(ret)
 
@@ -1395,8 +1479,10 @@ class HttpCli(object):
             ret["k" + quotep(excl)] = sub
 
         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)
+            vn, rem = self.asrv.vfs.get(top, self.uname, True, False)
+            fsroot, vfs_ls, vfs_virt = vn.ls(
+                rem, self.uname, not self.args.no_scandir, incl_wo=True
+            )
         except:
             vfs_ls = []
             vfs_virt = {}
@@ -1434,35 +1520,51 @@ class HttpCli(object):
 
                 vpnodes.append([quotep(vpath) + "/", html_escape(node, crlf=True)])
 
-        vn, rem = self.auth.vfs.get(
+        vn, rem = self.asrv.vfs.get(
             self.vpath, self.uname, self.readable, self.writable
         )
         abspath = vn.canonical(rem)
+        dbv, vrem = vn.get_dbv(rem)
 
         try:
             st = os.stat(fsenc(abspath))
         except:
             raise Pebkac(404)
 
-        if self.readable and not stat.S_ISDIR(st.st_mode):
+        if self.readable:
             if rem.startswith(".hist/up2k."):
                 raise Pebkac(403)
 
+            is_dir = stat.S_ISDIR(st.st_mode)
             th_fmt = self.uparam.get("th")
             if th_fmt is not None:
+                if is_dir:
+                    for fn in ["folder.png", "folder.jpg"]:
+                        fp = os.path.join(abspath, fn)
+                        if os.path.exists(fp):
+                            vrem = "{}/{}".format(vrem.rstrip("/"), fn)
+                            is_dir = False
+                            break
+
+                    if is_dir:
+                        return self.tx_ico("a.folder")
+
                 thp = None
                 if self.thumbcli:
-                    thp = self.thumbcli.get(vn.realpath, rem, int(st.st_mtime), th_fmt)
+                    thp = self.thumbcli.get(
+                        dbv.realpath, vrem, int(st.st_mtime), th_fmt
+                    )
 
                 if thp:
                     return self.tx_file(thp)
 
                 return self.tx_ico(rem)
 
-            if abspath.endswith(".md") and "raw" not in self.uparam:
-                return self.tx_md(abspath)
+            if not is_dir:
+                if abspath.endswith(".md") and "raw" not in self.uparam:
+                    return self.tx_md(abspath)
 
-            return self.tx_file(abspath)
+                return self.tx_file(abspath)
 
         srv_info = []
 
@@ -1561,7 +1663,9 @@ 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, incl_wo=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())
@@ -1593,7 +1697,7 @@ class HttpCli(object):
         icur = None
         if "e2t" in vn.flags:
             idx = self.conn.get_u2idx()
-            icur = idx.get_cur(vn.realpath)
+            icur = idx.get_cur(dbv.realpath)
 
         dirs = []
         files = []
@@ -1661,6 +1765,9 @@ class HttpCli(object):
             rd = f["rd"]
             del f["rd"]
             if icur:
+                if vn != dbv:
+                    _, rd = vn.get_dbv(rd)
+
                 q = "select w from up where rd = ? and fn = ?"
                 try:
                     r = icur.execute(q, (rd, fn)).fetchone()
@@ -1701,9 +1808,13 @@ class HttpCli(object):
         j2a["files"] = dirs + files
         j2a["logues"] = logues
         j2a["taglist"] = taglist
+
         if "mte" in vn.flags:
             j2a["tag_order"] = json.dumps(vn.flags["mte"].split(","))
 
+        if self.args.css_browser:
+            j2a["css"] = self.args.css_browser
+
         html = self.j2(tpl, **j2a)
         self.reply(html.encode("utf-8", "replace"), headers=NO_STORE)
         return True

copyparty/httpconn.py

@@ -34,7 +34,8 @@ class HttpConn(object):
         self.hsrv = hsrv
 
         self.args = hsrv.args
-        self.auth = hsrv.auth
+        self.asrv = hsrv.asrv
+        self.is_mp = hsrv.is_mp
         self.cert_path = hsrv.cert_path
 
         enth = HAVE_PIL and not self.args.no_thumb
@@ -70,7 +71,7 @@ class HttpConn(object):
 
     def get_u2idx(self):
         if not self.u2idx:
-            self.u2idx = U2idx(self.args, self.log_func)
+            self.u2idx = U2idx(self)
 
         return self.u2idx
 
@@ -174,6 +175,11 @@ class HttpConn(object):
             self.sr = Unrecv(self.s)
 
         while True:
+            if self.is_mp:
+                self.workload += 50
+                if self.workload >= 2 ** 31:
+                    self.workload = 100
+
             cli = HttpCli(self)
             if not cli.run():
                 return

copyparty/httpsrv.py

@@ -25,8 +25,8 @@ except ImportError:
     sys.exit(1)
 
 from .__init__ import E, MACOS
-from .httpconn import HttpConn
 from .authsrv import AuthSrv
+from .httpconn import HttpConn
 
 
 class HttpSrv(object):
@@ -35,10 +35,12 @@ class HttpSrv(object):
     relying on MpSrv for performance (HttpSrv is just plain threads)
     """
 
-    def __init__(self, broker):
+    def __init__(self, broker, is_mp=False):
         self.broker = broker
+        self.is_mp = is_mp
         self.args = broker.args
         self.log = broker.log
+        self.asrv = broker.asrv
 
         self.disconnect_func = None
         self.mutex = threading.Lock()
@@ -46,7 +48,6 @@ class HttpSrv(object):
         self.clients = {}
         self.workload = 0
         self.workload_thr_alive = False
-        self.auth = AuthSrv(self.args, self.log)
 
         env = jinja2.Environment()
         env.loader = jinja2.FileSystemLoader(os.path.join(E.mod, "web"))
@@ -66,7 +67,11 @@ class HttpSrv(object):
         if self.args.log_conn:
             self.log("%s %s" % addr, "|%sC-cthr" % ("-" * 5,), c="1;30")
 
-        thr = threading.Thread(target=self.thr_client, args=(sck, addr))
+        thr = threading.Thread(
+            target=self.thr_client,
+            args=(sck, addr),
+            name="httpsrv-{}-{}".format(addr[0].split(".", 2)[-1][-6:], addr[1]),
+        )
         thr.daemon = True
         thr.start()
 
@@ -84,13 +89,16 @@ class HttpSrv(object):
         cli = HttpConn(sck, addr, self)
         with self.mutex:
             self.clients[cli] = 0
-            self.workload += 50
 
-            if not self.workload_thr_alive:
-                self.workload_thr_alive = True
-                thr = threading.Thread(target=self.thr_workload)
-                thr.daemon = True
-                thr.start()
+            if self.is_mp:
+                self.workload += 50
+                if not self.workload_thr_alive:
+                    self.workload_thr_alive = True
+                    thr = threading.Thread(
+                        target=self.thr_workload, name="httpsrv-workload"
+                    )
+                    thr.daemon = True
+                    thr.start()
 
         try:
             if self.args.log_conn:
@@ -99,6 +107,7 @@ class HttpSrv(object):
             cli.run()
 
         finally:
+            sck = cli.s
             if self.args.log_conn:
                 self.log("%s %s" % addr, "|%sC-cdone" % ("-" * 7,), c="1;30")
 

copyparty/ico.py

@@ -1,3 +1,6 @@
+# coding: utf-8
+from __future__ import print_function, unicode_literals
+
 import hashlib
 import colorsys
 

copyparty/mtag.py

@@ -1,7 +1,6 @@
 # coding: utf-8
 from __future__ import print_function, unicode_literals
 
-import re
 import os
 import sys
 import json

copyparty/star.py

@@ -1,3 +1,6 @@
+# coding: utf-8
+from __future__ import print_function, unicode_literals
+
 import os
 import tarfile
 import threading
@@ -42,7 +45,7 @@ class StreamTar(object):
         fmt = tarfile.GNU_FORMAT
         self.tar = tarfile.open(fileobj=self.qfile, mode="w|", format=fmt)
 
-        w = threading.Thread(target=self._gen)
+        w = threading.Thread(target=self._gen, name="star-gen")
         w.daemon = True
         w.start()
 

copyparty/sutil.py

@@ -1,3 +1,6 @@
+# coding: utf-8
+from __future__ import print_function, unicode_literals
+
 import os
 import time
 import tempfile

copyparty/svchub.py

@@ -37,12 +37,13 @@ class SvcHub(object):
 
         self.log = self._log_disabled if args.q else self._log_enabled
 
-        # jank goes here
-        auth = AuthSrv(self.args, self.log, False)
-
         # initiate all services to manage
+        self.asrv = AuthSrv(self.args, self.log, False)
+        if args.ls:
+            self.asrv.dbg_ls()
+
         self.tcpsrv = TcpSrv(self)
-        self.up2k = Up2k(self, auth.vfs.all_vols)
+        self.up2k = Up2k(self)
 
         self.thumbsrv = None
         if not args.no_thumb:
@@ -52,7 +53,7 @@ class SvcHub(object):
                     msg = "setting --th-no-webp because either libwebp is not available or your Pillow is too old"
                     self.log("thumb", msg, c=3)
 
-                self.thumbsrv = ThumbSrv(self, auth.vfs.all_vols)
+                self.thumbsrv = ThumbSrv(self)
             else:
                 msg = "need Pillow to create thumbnails; for example:\n{}{} -m pip install --user Pillow\n"
                 self.log(
@@ -69,7 +70,7 @@ class SvcHub(object):
         self.broker = Broker(self)
 
     def run(self):
-        thr = threading.Thread(target=self.tcpsrv.run)
+        thr = threading.Thread(target=self.tcpsrv.run, name="svchub-main")
         thr.daemon = True
         thr.start()
 
@@ -93,9 +94,11 @@ class SvcHub(object):
                         break
 
                     if n == 3:
-                        print("waiting for thumbsrv...")
+                        print("waiting for thumbsrv (10sec)...")
 
-            print("nailed it")
+            print("nailed it", end="")
+        finally:
+            print("\033[0m")
 
     def _log_disabled(self, src, msg, c=0):
         pass

copyparty/szip.py

@@ -1,3 +1,6 @@
+# coding: utf-8
+from __future__ import print_function, unicode_literals
+
 import os
 import time
 import zlib

copyparty/th_cli.py

@@ -1,5 +1,7 @@
+# coding: utf-8
+from __future__ import print_function, unicode_literals
+
 import os
-import time
 
 from .util import Cooldown
 from .th_srv import thumb_path, THUMBABLE, FMT_FF
@@ -9,6 +11,7 @@ class ThumbCli(object):
     def __init__(self, broker):
         self.broker = broker
         self.args = broker.args
+        self.asrv = broker.asrv
 
         # cache on both sides for less broker spam
         self.cooldown = Cooldown(self.args.th_poke)
@@ -18,16 +21,19 @@ class ThumbCli(object):
         if ext not in THUMBABLE:
             return None
 
-        if self.args.no_vthumb and ext in FMT_FF:
+        is_vid = ext in FMT_FF
+        if is_vid and self.args.no_vthumb:
             return None
 
         if fmt == "j" and self.args.th_no_jpg:
             fmt = "w"
 
-        if fmt == "w" and self.args.th_no_webp:
-            fmt = "j"
+        if fmt == "w":
+            if self.args.th_no_webp or (is_vid and self.args.th_ff_jpg):
+                fmt = "j"
 
-        tpath = thumb_path(ptop, rem, mtime, fmt)
+        histpath = self.asrv.vfs.histtab[ptop]
+        tpath = thumb_path(histpath, rem, mtime, fmt)
         ret = None
         try:
             st = os.stat(tpath)

copyparty/th_srv.py

@@ -1,5 +1,7 @@
+# coding: utf-8
+from __future__ import print_function, unicode_literals
+
 import os
-import sys
 import time
 import shutil
 import base64
@@ -8,7 +10,7 @@ import threading
 import subprocess as sp
 
 from .__init__ import PY2
-from .util import fsenc, mchkcmd, Queue, Cooldown, BytesIO
+from .util import fsenc, runcmd, Queue, Cooldown, BytesIO, min_ex
 from .mtag import HAVE_FFMPEG, HAVE_FFPROBE, ffprobe
 
 
@@ -51,7 +53,7 @@ except:
 # https://pillow.readthedocs.io/en/stable/handbook/image-file-formats.html
 # ffmpeg -formats
 FMT_PIL = "bmp dib gif icns ico jpg jpeg jp2 jpx pcx png pbm pgm ppm pnm sgi tga tif tiff webp xbm dds xpm"
-FMT_FF = "av1 asf avi flv m4v mkv mjpeg mjpg mpg mpeg mpg2 mpeg2 mov 3gp mp4 ts mpegts nut ogv ogm rm vob webm wmv"
+FMT_FF = "av1 asf avi flv m4v mkv mjpeg mjpg mpg mpeg mpg2 mpeg2 h264 avc h265 hevc mov 3gp mp4 ts mpegts nut ogv ogm rm vob webm wmv"
 
 if HAVE_HEIF:
     FMT_PIL += " heif heifs heic heics"
@@ -71,7 +73,7 @@ if HAVE_FFMPEG and HAVE_FFPROBE:
     THUMBABLE.update(FMT_FF)
 
 
-def thumb_path(ptop, rem, mtime, fmt):
+def thumb_path(histpath, rem, mtime, fmt):
     # base16 = 16 = 256
     # b64-lc = 38 = 1444
     # base64 = 64 = 4096
@@ -92,16 +94,15 @@ def thumb_path(ptop, rem, mtime, fmt):
     h = hashlib.sha512(fsenc(fn)).digest()[:24]
     fn = base64.urlsafe_b64encode(h).decode("ascii")[:24]
 
-    return "{}/.hist/th/{}/{}.{:x}.{}".format(
-        ptop, rd, fn, int(mtime), "webp" if fmt == "w" else "jpg"
+    return "{}/th/{}/{}.{:x}.{}".format(
+        histpath, rd, fn, int(mtime), "webp" if fmt == "w" else "jpg"
     )
 
 
 class ThumbSrv(object):
-    def __init__(self, hub, vols):
+    def __init__(self, hub):
         self.hub = hub
-        self.vols = [v.realpath for v in vols.values()]
-
+        self.asrv = hub.asrv
         self.args = hub.args
         self.log_func = hub.log
 
@@ -114,8 +115,10 @@ class ThumbSrv(object):
         self.stopping = False
         self.nthr = os.cpu_count() if hasattr(os, "cpu_count") else 4
         self.q = Queue(self.nthr * 4)
-        for _ in range(self.nthr):
-            t = threading.Thread(target=self.worker)
+        for n in range(self.nthr):
+            t = threading.Thread(
+                target=self.worker, name="thumb-{}-{}".format(n, self.nthr)
+            )
             t.daemon = True
             t.start()
 
@@ -129,9 +132,9 @@ class ThumbSrv(object):
 
             msg = "cannot create video thumbnails because some of the required programs are not available: "
             msg += ", ".join(missing)
-            self.log(msg, c=1)
+            self.log(msg, c=3)
 
-        t = threading.Thread(target=self.cleaner)
+        t = threading.Thread(target=self.cleaner, name="thumb-cleaner")
         t.daemon = True
         t.start()
 
@@ -148,7 +151,8 @@ class ThumbSrv(object):
             return not self.nthr
 
     def get(self, ptop, rem, mtime, fmt):
-        tpath = thumb_path(ptop, rem, mtime, fmt)
+        histpath = self.asrv.vfs.histtab[ptop]
+        tpath = thumb_path(histpath, rem, mtime, fmt)
         abspath = os.path.join(ptop, rem)
         cond = threading.Condition()
         with self.mutex:
@@ -206,9 +210,9 @@ class ThumbSrv(object):
             if fun:
                 try:
                     fun(abspath, tpath)
-                except Exception as ex:
-                    msg = "{} failed on {}\n  {!r}"
-                    self.log(msg.format(fun.__name__, abspath, ex), 3)
+                except:
+                    msg = "{} failed on {}\n{}"
+                    self.log(msg.format(fun.__name__, abspath, min_ex()), 3)
                     with open(tpath, "wb") as _:
                         pass
 
@@ -240,8 +244,8 @@ class ThumbSrv(object):
             except:
                 im.thumbnail(self.res)
 
-            if im.mode not in ("RGB", "L"):
-                im = im.convert("RGB")
+            fmts = ["RGB", "L"]
+            args = {"quality": 40}
 
             if tpath.endswith(".webp"):
                 # quality 80 = pillow-default
@@ -249,15 +253,27 @@ class ThumbSrv(object):
                 # method 0 = pillow-default, fast
                 # method 4 = ffmpeg-default
                 # method 6 = max, slow
-                im.save(tpath, quality=40, method=6)
+                fmts += ["RGBA", "LA"]
+                args["method"] = 6
             else:
-                im.save(tpath, quality=40)  # default=75
+                pass  # default q = 75
+
+            if im.mode not in fmts:
+                print("conv {}".format(im.mode))
+                im = im.convert("RGB")
+
+            im.save(tpath, quality=40, method=6)
 
     def conv_ffmpeg(self, abspath, tpath):
         ret, _ = ffprobe(abspath)
 
-        dur = ret[".dur"][1] if ".dur" in ret else 4
-        seek = "{:.0f}".format(dur / 3)
+        ext = abspath.rsplit(".")[-1]
+        if ext in ["h264", "h265"]:
+            seek = []
+        else:
+            dur = ret[".dur"][1] if ".dur" in ret else 4
+            seek = "{:.0f}".format(dur / 3)
+            seek = [b"-ss", seek.encode("utf-8")]
 
         scale = "scale={0}:{1}:force_original_aspect_ratio="
         if self.args.th_no_crop:
@@ -266,19 +282,20 @@ class ThumbSrv(object):
             scale += "increase,crop={0}:{1},setsar=1:1"
 
         scale = scale.format(*list(self.res)).encode("utf-8")
+        # fmt: off
         cmd = [
             b"ffmpeg",
             b"-nostdin",
-            b"-hide_banner",
-            b"-ss",
-            seek,
-            b"-i",
-            fsenc(abspath),
-            b"-vf",
-            scale,
-            b"-vframes",
-            b"1",
+            b"-v", b"error",
+            b"-hide_banner"
         ]
+        cmd += seek
+        cmd += [
+            b"-i", fsenc(abspath),
+            b"-vf", scale,
+            b"-vframes", b"1",
+        ]
+        # fmt: on
 
         if tpath.endswith(".jpg"):
             cmd += [
@@ -295,7 +312,11 @@ class ThumbSrv(object):
 
         cmd += [fsenc(tpath)]
 
-        mchkcmd(cmd)
+        ret, sout, serr = runcmd(*cmd)
+        if ret != 0:
+            msg = ["ff: {}".format(x) for x in serr.split("\n")]
+            self.log("FFmpeg failed:\n" + "\n".join(msg), c="1;30")
+            raise sp.CalledProcessError(ret, (cmd[0], b"...", cmd[-1]))
 
     def poke(self, tdir):
         if not self.poke_cd.poke(tdir):
@@ -314,26 +335,29 @@ class ThumbSrv(object):
         interval = self.args.th_clean
         while True:
             time.sleep(interval)
-            for vol in self.vols:
-                vol += "/.hist/th"
-                self.log("cln {}/".format(vol))
-                self.clean(vol)
+            for vol, histpath in self.asrv.vfs.histtab.items():
+                if histpath.startswith(vol):
+                    self.log("\033[Jcln {}/\033[A".format(histpath))
+                else:
+                    self.log("\033[Jcln {} ({})/\033[A".format(histpath, vol))
 
-            self.log("cln ok")
+                self.clean(histpath)
 
-    def clean(self, vol):
-        # self.log("cln {}".format(vol))
+            self.log("\033[Jcln ok")
+
+    def clean(self, histpath):
+        # self.log("cln {}".format(histpath))
         maxage = self.args.th_maxage
         now = time.time()
         prev_b64 = None
         prev_fp = None
         try:
-            ents = os.listdir(vol)
+            ents = os.listdir(histpath)
         except:
             return
 
         for f in sorted(ents):
-            fp = os.path.join(vol, f)
+            fp = os.path.join(histpath, f)
             cmp = fp.lower().replace("\\", "/")
 
             # "top" or b64 prefix/full (a folder)

copyparty/u2idx.py

@@ -7,7 +7,7 @@ import time
 import threading
 from datetime import datetime
 
-from .util import u8safe, s3dec, html_escape, Pebkac
+from .util import s3dec, Pebkac, min_ex
 from .up2k import up2k_wark_from_hashlist
 
 
@@ -19,10 +19,11 @@ except:
 
 
 class U2idx(object):
-    def __init__(self, args, log_func):
-        self.args = args
-        self.log_func = log_func
-        self.timeout = args.srch_time
+    def __init__(self, conn):
+        self.log_func = conn.log_func
+        self.asrv = conn.asrv
+        self.args = conn.args
+        self.timeout = self.args.srch_time
 
         if not HAVE_SQLITE3:
             self.log("could not load sqlite3; searchign wqill be disabled")
@@ -47,57 +48,143 @@ 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]
-        except Exception as ex:
-            raise Pebkac(500, repr(ex))
+            return self.run_query(vols, uq, uv)[0]
+        except:
+            raise Pebkac(500, min_ex())
 
     def get_cur(self, ptop):
         cur = self.cur.get(ptop)
         if cur:
             return cur
 
-        cur = _open(ptop)
-        if not cur:
+        histpath = self.asrv.vfs.histtab[ptop]
+        db_path = os.path.join(histpath, "up2k.db")
+        if not os.path.exists(db_path):
             return None
 
+        cur = sqlite3.connect(db_path).cursor()
         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
@@ -108,39 +195,19 @@ class U2idx(object):
                 self.active_id,
                 done_flag,
             ),
+            name="u2idx-terminator",
         )
         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 +230,7 @@ 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:
@@ -178,6 +245,7 @@ class U2idx(object):
                 hit["tags"] = tags
 
             ret.extend(sret)
+            # print("[{}] {}".format(ptop, sret))
 
         done_flag.append(True)
         self.active_id = None
@@ -198,84 +266,3 @@ class U2idx(object):
 
         if identifier == self.active_id:
             self.active_cur.connection.interrupt()
-
-
-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

@@ -48,8 +48,9 @@ class Up2k(object):
         * ~/.config flatfiles for active jobs
     """
 
-    def __init__(self, hub, all_vols):
+    def __init__(self, hub):
         self.hub = hub
+        self.asrv = hub.asrv
         self.args = hub.args
         self.log_func = hub.log
 
@@ -60,6 +61,8 @@ class Up2k(object):
         self.mutex = threading.Lock()
         self.hashq = Queue()
         self.tagq = Queue()
+        self.n_hashq = 0
+        self.n_tagq = 0
         self.volstate = {}
         self.registry = {}
         self.entags = {}
@@ -83,7 +86,7 @@ class Up2k(object):
         if ANYWIN:
             # usually fails to set lastmod too quickly
             self.lastmod_q = Queue()
-            thr = threading.Thread(target=self._lastmodder)
+            thr = threading.Thread(target=self._lastmodder, name="up2k-lastmod")
             thr.daemon = True
             thr.start()
 
@@ -94,45 +97,73 @@ class Up2k(object):
             self.log("could not initialize sqlite3, will use in-memory registry only")
 
         if self.args.no_fastboot:
-            self.deferred_init(all_vols)
+            self.deferred_init()
         else:
-            t = threading.Thread(target=self.deferred_init, args=(all_vols,))
+            t = threading.Thread(
+                target=self.deferred_init,
+                name="up2k-deferred-init",
+            )
             t.daemon = True
             t.start()
 
-    def deferred_init(self, all_vols):
+    def deferred_init(self):
+        all_vols = self.asrv.vfs.all_vols
         have_e2d = self.init_indexes(all_vols)
 
         if have_e2d:
-            thr = threading.Thread(target=self._snapshot)
+            thr = threading.Thread(target=self._snapshot, name="up2k-snapshot")
             thr.daemon = True
             thr.start()
 
-            thr = threading.Thread(target=self._hasher)
+            thr = threading.Thread(target=self._hasher, name="up2k-hasher")
             thr.daemon = True
             thr.start()
 
             if self.mtag:
-                thr = threading.Thread(target=self._tagger)
+                thr = threading.Thread(target=self._tagger, name="up2k-tagger")
                 thr.daemon = True
                 thr.start()
 
-                thr = threading.Thread(target=self._run_all_mtp)
+                thr = threading.Thread(target=self._run_all_mtp, name="up2k-mtp-init")
                 thr.daemon = True
                 thr.start()
 
     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 get_state(self):
+        mtpq = 0
+        q = "select count(w) from mt where k = 't:mtp'"
+        got_lock = self.mutex.acquire(timeout=0.5)
+        if got_lock:
+            for cur in self.cur.values():
+                try:
+                    mtpq += cur.execute(q).fetchone()[0]
+                except:
+                    pass
+            self.mutex.release()
+        else:
+            mtpq = "?"
+
+        ret = {
+            "volstate": self.volstate,
+            "scanning": hasattr(self, "pp"),
+            "hashq": self.n_hashq,
+            "tagq": self.n_tagq,
+            "mtpq": mtpq,
+        }
+        return json.dumps(ret, 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 = threading.Thread(
+            target=self.init_indexes,
+            args=args,
+            name="up2k-rescan-{}".format(scan_vols[0]),
+        )
         t.daemon = True
         t.start()
         return None
@@ -178,20 +209,27 @@ class Up2k(object):
                 self.log(msg, c=3)
 
         live_vols = []
-        for vol in vols:
-            try:
-                os.listdir(vol.realpath)
-                if not self.register_vpath(vol.realpath, vol.flags):
-                    raise Exception()
+        with self.mutex:
+            # only need to protect register_vpath but all in one go feels right
+            for vol in vols:
+                try:
+                    os.listdir(vol.realpath)
+                except:
+                    self.volstate[vol.vpath] = "OFFLINE (cannot access folder)"
+                    self.log("cannot access " + vol.realpath, c=1)
+                    continue
 
-                if vol.vpath in scan_vols or not scan_vols:
-                    live_vols.append(vol)
+                if scan_vols and vol.vpath not in scan_vols:
+                    continue
+
+                if not self.register_vpath(vol.realpath, vol.flags):
+                    # self.log("db not enable for {}".format(m, vol.realpath))
+                    continue
+
+                live_vols.append(vol)
 
                 if vol.vpath not in self.volstate:
-                    self.volstate[vol.vpath] = "OFFLINE (not initialized)"
-            except:
-                # self.log("db not enabled for {}".format(m, vol.realpath))
-                pass
+                    self.volstate[vol.vpath] = "OFFLINE (pending initialization)"
 
         vols = live_vols
         need_vac = {}
@@ -268,7 +306,7 @@ class Up2k(object):
         if self.mtag:
             m = "online (running mtp)"
             if scan_vols:
-                thr = threading.Thread(target=self._run_all_mtp)
+                thr = threading.Thread(target=self._run_all_mtp, name="up2k-mtp-scan")
                 thr.daemon = True
         else:
             del self.pp
@@ -283,9 +321,13 @@ class Up2k(object):
         return have_e2d
 
     def register_vpath(self, ptop, flags):
-        db_path = os.path.join(ptop, ".hist", "up2k.db")
+        histpath = self.asrv.vfs.histtab[ptop]
+        db_path = os.path.join(histpath, "up2k.db")
         if ptop in self.registry:
-            return [self.cur[ptop], db_path]
+            try:
+                return [self.cur[ptop], db_path]
+            except:
+                return None
 
         _, flags = self._expr_idx_filter(flags)
 
@@ -300,7 +342,7 @@ class Up2k(object):
             self.log(" ".join(sorted(a)) + "\033[0m")
 
         reg = {}
-        path = os.path.join(ptop, ".hist", "up2k.snap")
+        path = os.path.join(histpath, "up2k.snap")
         if "e2d" in flags and os.path.exists(path):
             with gzip.GzipFile(path, "rb") as f:
                 j = f.read().decode("utf-8")
@@ -324,7 +366,7 @@ class Up2k(object):
             return None
 
         try:
-            os.mkdir(os.path.join(ptop, ".hist"))
+            os.makedirs(histpath)
         except:
             pass
 
@@ -341,6 +383,7 @@ class Up2k(object):
     def _build_file_index(self, vol, all_vols):
         do_vac = False
         top = vol.realpath
+        nohash = "dhash" in vol.flags
         with self.mutex:
             cur, _ = self.register_vpath(top, vol.flags)
 
@@ -355,7 +398,7 @@ class Up2k(object):
             if WINDOWS:
                 excl = [x.replace("/", "\\") for x in excl]
 
-            n_add = self._build_dir(dbw, top, set(excl), top)
+            n_add = self._build_dir(dbw, top, set(excl), top, nohash)
             n_rm = self._drop_lost(dbw[0], top)
             if dbw[1]:
                 self.log("commit {} new files".format(dbw[1]))
@@ -363,23 +406,25 @@ class Up2k(object):
 
             return True, n_add or n_rm or do_vac
 
-    def _build_dir(self, dbw, top, excl, cdir):
+    def _build_dir(self, dbw, top, excl, cdir, nohash):
         self.pp.msg = "a{} {}".format(self.pp.n, cdir)
-        histdir = os.path.join(top, ".hist")
+        histpath = self.asrv.vfs.histtab[top]
         ret = 0
-        for iname, inf in statdir(self.log, not self.args.no_scandir, False, cdir):
+        g = statdir(self.log, not self.args.no_scandir, False, cdir)
+        for iname, inf in sorted(g):
             abspath = os.path.join(cdir, iname)
             lmod = int(inf.st_mtime)
+            sz = inf.st_size
             if stat.S_ISDIR(inf.st_mode):
-                if abspath in excl or abspath == histdir:
+                if abspath in excl or abspath == histpath:
                     continue
                 # self.log(" dir: {}".format(abspath))
-                ret += self._build_dir(dbw, top, excl, abspath)
+                ret += self._build_dir(dbw, top, excl, abspath, nohash)
             else:
                 # self.log("file: {}".format(abspath))
                 rp = abspath[len(top) :].replace("\\", "/").strip("/")
                 rd, fn = rp.rsplit("/", 1) if "/" in rp else ["", rp]
-                sql = "select * from up where rd = ? and fn = ?"
+                sql = "select w, mt, sz from up where rd = ? and fn = ?"
                 try:
                     c = dbw[0].execute(sql, (rd, fn))
                 except:
@@ -388,18 +433,18 @@ class Up2k(object):
                 in_db = list(c.fetchall())
                 if in_db:
                     self.pp.n -= 1
-                    _, dts, dsz, _, _ = in_db[0]
+                    dw, dts, dsz = in_db[0]
                     if len(in_db) > 1:
                         m = "WARN: multiple entries: [{}] => [{}] |{}|\n{}"
                         rep_db = "\n".join([repr(x) for x in in_db])
                         self.log(m.format(top, rp, len(in_db), rep_db))
                         dts = -1
 
-                    if dts == lmod and dsz == inf.st_size:
+                    if dts == lmod and dsz == sz and (nohash or dw[0] != "#"):
                         continue
 
                     m = "reindex [{}] => [{}] ({}/{}) ({}/{})".format(
-                        top, rp, dts, lmod, dsz, inf.st_size
+                        top, rp, dts, lmod, dsz, sz
                     )
                     self.log(m)
                     self.db_rm(dbw[0], rd, fn)
@@ -408,17 +453,22 @@ class Up2k(object):
                     in_db = None
 
                 self.pp.msg = "a{} {}".format(self.pp.n, abspath)
-                if inf.st_size > 1024 * 1024:
-                    self.log("file: {}".format(abspath))
 
-                try:
-                    hashes = self._hashlist_from_file(abspath)
-                except Exception as ex:
-                    self.log("hash: {} @ [{}]".format(repr(ex), abspath))
-                    continue
+                if nohash:
+                    wark = up2k_wark_from_metadata(self.salt, sz, lmod, rd, fn)
+                else:
+                    if sz > 1024 * 1024:
+                        self.log("file: {}".format(abspath))
 
-                wark = up2k_wark_from_hashlist(self.salt, inf.st_size, hashes)
-                self.db_add(dbw[0], wark, rd, fn, lmod, inf.st_size)
+                    try:
+                        hashes = self._hashlist_from_file(abspath)
+                    except Exception as ex:
+                        self.log("hash: {} @ [{}]".format(repr(ex), abspath))
+                        continue
+
+                    wark = up2k_wark_from_hashlist(self.salt, sz, hashes)
+
+                self.db_add(dbw[0], wark, rd, fn, lmod, sz)
                 dbw[1] += 1
                 ret += 1
                 td = time.time() - dbw[2]
@@ -549,9 +599,10 @@ class Up2k(object):
                     last_write = time.time()
                     n_buf = 0
 
-            self._stop_mpool(mpool)
-            with self.mutex:
-                n_add += len(self._flush_mpool(c3))
+            if mpool:
+                self._stop_mpool(mpool)
+                with self.mutex:
+                    n_add += len(self._flush_mpool(c3))
 
             conn.commit()
             c3.close()
@@ -585,7 +636,8 @@ class Up2k(object):
 
         del self.pp
         for k in list(self.volstate.keys()):
-            self.volstate[k] = "online, idle"
+            if "OFFLINE" not in self.volstate[k]:
+                self.volstate[k] = "online, idle"
 
     def _run_one_mtp(self, ptop):
         entags = self.entags[ptop]
@@ -748,7 +800,9 @@ class Up2k(object):
 
         mpool = Queue(nw)
         for _ in range(nw):
-            thr = threading.Thread(target=self._tag_thr, args=(mpool,))
+            thr = threading.Thread(
+                target=self._tag_thr, args=(mpool,), name="up2k-mpool"
+            )
             thr.daemon = True
             thr.start()
 
@@ -909,7 +963,7 @@ class Up2k(object):
     def _create_v3(self, cur):
         """
         collision in 2^(n/2) files where n = bits (6 bits/ch)
-          10*6/2 = 2^30 =       1'073'741'824, 24.1mb idx
+          10*6/2 = 2^30 =       1'073'741'824, 24.1mb idx  1<<(3*10)
           12*6/2 = 2^36 =      68'719'476'736, 24.8mb idx
           16*6/2 = 2^48 = 281'474'976'710'656, 26.1mb idx
         """
@@ -957,9 +1011,10 @@ class Up2k(object):
         return self._orz(db_path)
 
     def handle_json(self, cj):
-        if not self.register_vpath(cj["ptop"], cj["vcfg"]):
-            if cj["ptop"] not in self.registry:
-                raise Pebkac(410, "location unavailable")
+        with self.mutex:
+            if not self.register_vpath(cj["ptop"], cj["vcfg"]):
+                if cj["ptop"] not in self.registry:
+                    raise Pebkac(410, "location unavailable")
 
         cj["name"] = sanitize_fn(cj["name"], bad=[".prologue.html", ".epilogue.html"])
         cj["poke"] = time.time()
@@ -1200,6 +1255,7 @@ class Up2k(object):
 
         if "e2t" in self.flags[ptop]:
             self.tagq.put([ptop, wark, rd, fn])
+            self.n_tagq += 1
 
         return True
 
@@ -1240,12 +1296,15 @@ 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
@@ -1322,11 +1381,12 @@ class Up2k(object):
                 for k, reg in self.registry.items():
                     self._snap_reg(prev, k, reg, discard_interval)
 
-    def _snap_reg(self, prev, k, reg, discard_interval):
+    def _snap_reg(self, prev, ptop, reg, discard_interval):
         now = time.time()
+        histpath = self.asrv.vfs.histtab[ptop]
         rm = [x for x in reg.values() if now - x["poke"] > discard_interval]
         if rm:
-            m = "dropping {} abandoned uploads in {}".format(len(rm), k)
+            m = "dropping {} abandoned uploads in {}".format(len(rm), ptop)
             vis = [self._vis_job_progress(x) for x in rm]
             self.log("\n".join([m] + vis))
             for job in rm:
@@ -1344,21 +1404,21 @@ class Up2k(object):
                 except:
                     pass
 
-        path = os.path.join(k, ".hist", "up2k.snap")
+        path = os.path.join(histpath, "up2k.snap")
         if not reg:
-            if k not in prev or prev[k] is not None:
-                prev[k] = None
+            if ptop not in prev or prev[ptop] is not None:
+                prev[ptop] = None
                 if os.path.exists(fsenc(path)):
                     os.unlink(fsenc(path))
             return
 
         newest = max(x["poke"] for _, x in reg.items()) if reg else 0
         etag = [len(reg), newest]
-        if etag == prev.get(k, None):
+        if etag == prev.get(ptop, None):
             return
 
         try:
-            os.mkdir(os.path.join(k, ".hist"))
+            os.makedirs(histpath)
         except:
             pass
 
@@ -1370,14 +1430,21 @@ class Up2k(object):
         atomic_move(path2, path)
 
         self.log("snap: {} |{}|".format(path, len(reg.keys())))
-        prev[k] = etag
+        prev[ptop] = etag
 
     def _tagger(self):
+        with self.mutex:
+            self.n_tagq += 1
+
         while True:
+            with self.mutex:
+                self.n_tagq -= 1
+
             ptop, wark, rd, fn = self.tagq.get()
             if "e2t" not in self.flags[ptop]:
                 continue
 
+            # self.log("\n  " + repr([ptop, rd, fn]))
             abspath = os.path.join(ptop, rd, fn)
             tags = self.mtag.get(abspath)
             ntags1 = len(tags)
@@ -1403,8 +1470,16 @@ class Up2k(object):
             self.log("tagged {} ({}+{})".format(abspath, ntags1, len(tags) - ntags1))
 
     def _hasher(self):
+        with self.mutex:
+            self.n_hashq += 1
+
         while True:
+            with self.mutex:
+                self.n_hashq -= 1
+            # self.log("hashq {}".format(self.n_hashq))
+
             ptop, rd, fn = self.hashq.get()
+            # self.log("hashq {} pop {}/{}/{}".format(self.n_hashq, ptop, rd, fn))
             if "e2d" not in self.flags[ptop]:
                 continue
 
@@ -1417,8 +1492,11 @@ class Up2k(object):
                 self.idx_wark(ptop, wark, rd, fn, inf.st_mtime, inf.st_size)
 
     def hash_file(self, ptop, flags, rd, fn):
-        self.register_vpath(ptop, flags)
-        self.hashq.put([ptop, rd, fn])
+        with self.mutex:
+            self.register_vpath(ptop, flags)
+            self.hashq.put([ptop, rd, fn])
+            self.n_hashq += 1
+        # self.log("hashq {} push {}/{}/{}".format(self.n_hashq, ptop, rd, fn))
 
 
 def up2k_chunksize(filesize):
@@ -1440,9 +1518,12 @@ def up2k_wark_from_hashlist(salt, filesize, hashes):
     ident.extend(hashes)
     ident = "\n".join(ident)
 
-    hasher = hashlib.sha512()
-    hasher.update(ident.encode("utf-8"))
-    digest = hasher.digest()[:32]
+    wark = hashlib.sha512(ident.encode("utf-8")).digest()
+    wark = base64.urlsafe_b64encode(wark)
+    return wark.decode("ascii")[:43]
 
-    wark = base64.urlsafe_b64encode(digest)
-    return wark.decode("utf-8").rstrip("=")
+
+def up2k_wark_from_metadata(salt, sz, lastmod, rd, fn):
+    ret = fsenc("{}\n{}\n{}\n{}\n{}".format(salt, lastmod, sz, rd, fn))
+    ret = base64.urlsafe_b64encode(hashlib.sha512(ret).digest())
+    return "#{}".format(ret[:42].decode("ascii"))

copyparty/util.py

@@ -193,7 +193,7 @@ class ProgressPrinter(threading.Thread):
     """
 
     def __init__(self):
-        threading.Thread.__init__(self)
+        threading.Thread.__init__(self, name="pp")
         self.daemon = True
         self.msg = None
         self.end = False
@@ -208,6 +208,8 @@ class ProgressPrinter(threading.Thread):
 
             msg = self.msg
             uprint(" {}\033[K\r".format(msg))
+            if PY2:
+                sys.stdout.flush()
 
         print("\033[K", end="")
         sys.stdout.flush()  # necessary on win10 even w/ stderr btw
@@ -252,6 +254,17 @@ def trace(*args, **kwargs):
     nuprint(msg)
 
 
+def min_ex():
+    et, ev, tb = sys.exc_info()
+    tb = traceback.extract_tb(tb, 2)
+    ex = [
+        "{} @ {} <{}>: {}".format(fp.split(os.sep)[-1], ln, fun, txt)
+        for fp, ln, fun, txt in tb
+    ]
+    ex.append("{}: {}".format(et.__name__, ev))
+    return "\n".join(ex)
+
+
 @contextlib.contextmanager
 def ren_open(fname, *args, **kwargs):
     fdir = kwargs.pop("fdir", None)
@@ -262,6 +275,11 @@ def ren_open(fname, *args, **kwargs):
             yield {"orz": [f, fname]}
             return
 
+    if suffix:
+        ext = fname.split(".")[-1]
+        if len(ext) < 7:
+            suffix += "." + ext
+
     orig_name = fname
     bname = fname
     ext = ""
@@ -561,8 +579,10 @@ def read_header(sr):
             else:
                 continue
 
-        sr.unrecv(ret[ofs + 4 :])
-        return ret[:ofs].decode("utf-8", "surrogateescape").split("\r\n")
+        if len(ret) > ofs + 4:
+            sr.unrecv(ret[ofs + 4 :])
+
+        return ret[:ofs].decode("utf-8", "surrogateescape").lstrip("\r\n").split("\r\n")
 
 
 def humansize(sz, terse=False):
@@ -847,13 +867,14 @@ def yieldfile(fn):
 
 
 def hashcopy(actor, fin, fout):
-    u32_lim = int((2 ** 31) * 0.9)
+    is_mp = actor.is_mp
     hashobj = hashlib.sha512()
     tlen = 0
     for buf in fin:
-        actor.workload += 1
-        if actor.workload > u32_lim:
-            actor.workload = 100  # prevent overflow
+        if is_mp:
+            actor.workload += 1
+            if actor.workload > 2 ** 31:
+                actor.workload = 100
 
         tlen += len(buf)
         hashobj.update(buf)
@@ -865,12 +886,17 @@ def hashcopy(actor, fin, fout):
     return tlen, hashobj.hexdigest(), digest_b64
 
 
-def sendfile_py(lower, upper, f, s):
+def sendfile_py(lower, upper, f, s, actor=None):
     remains = upper - lower
     f.seek(lower)
     while remains > 0:
+        if actor:
+            actor.workload += 1
+            if actor.workload > 2 ** 31:
+                actor.workload = 100
+
         # time.sleep(0.01)
-        buf = f.read(min(4096, remains))
+        buf = f.read(min(1024 * 32, remains))
         if not buf:
             return remains
 
@@ -972,8 +998,8 @@ def guess_mime(url, fallback="application/octet-stream"):
 def runcmd(*argv):
     p = sp.Popen(argv, stdout=sp.PIPE, stderr=sp.PIPE)
     stdout, stderr = p.communicate()
-    stdout = stdout.decode("utf-8")
-    stderr = stderr.decode("utf-8")
+    stdout = stdout.decode("utf-8", "replace")
+    stderr = stderr.decode("utf-8", "replace")
     return [p.returncode, stdout, stderr]
 
 

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;
@@ -633,7 +644,6 @@ input[type="checkbox"]:checked+label {
 }
 #treeul a+a {
 	width: calc(100% - 2em);
-	background: #333;
 	line-height: 1em;
 }
 #treeul a+a:hover {
@@ -787,6 +797,13 @@ html.light #ghead {
 	padding: .2em .3em;
 	display: block;
 }
+#ggrid span.dir:before {
+	content: '📂';
+	line-height: 0;
+	font-size: 2em;
+	display: inline-block;
+	margin: -.7em 0 -.5em -.3em;
+}
 #ggrid a:hover {
 	background: #444;
 	border-color: #555;
@@ -899,6 +916,7 @@ html.light #files {
 }
 html.light #files thead th {
 	background: #eee;
+	border-radius: 0;
 }
 html.light #files tr td {
 	border-top: 1px solid #ddd;

copyparty/web/browser.html

@@ -8,6 +8,9 @@
     <meta name="viewport" content="width=device-width, initial-scale=0.8">
     <link rel="stylesheet" type="text/css" media="screen" href="/.cpr/browser.css{{ ts }}">
     <link rel="stylesheet" type="text/css" media="screen" href="/.cpr/upload.css{{ ts }}">
+    {%- if css %}
+    <link rel="stylesheet" type="text/css" media="screen" href="{{ css }}{{ ts }}">
+    {%- endif %}
 </head>
 
 <body>

copyparty/web/browser.js

@@ -38,7 +38,7 @@ var have_webp = null;
 	img.onerror = function () {
 		have_webp = false;
 	};
-	img.src = "data:image/webp;base64,UklGRiIAAABXRUJQVlA4IBYAAAAwAQCdASoBAAEADsD+JaQAA3AAAAAA";
+	img.src = "data:image/webp;base64,UklGRhoAAABXRUJQVlA4TA0AAAAvAAAAEAcQERGIiP4HAA==";
 })();
 
 
@@ -48,7 +48,6 @@ function MPlayer() {
 	this.au = null;
 	this.au_native = null;
 	this.au_ogvjs = null;
-	this.cover_url = '';
 	this.tracks = {};
 	this.order = [];
 
@@ -803,7 +802,10 @@ var thegrid = (function () {
 			r.sz = v;
 			swrite('gridsz', r.sz);
 		}
-		document.documentElement.style.setProperty('--grid-sz', r.sz + 'em');
+		try {
+			document.documentElement.style.setProperty('--grid-sz', r.sz + 'em');
+		}
+		catch (ex) { }
 	}
 	setsz();
 
@@ -820,12 +822,25 @@ var thegrid = (function () {
 		this.setAttribute('class', tr.getAttribute('class'));
 	}
 
+	function bgopen(e) {
+		ev(e);
+		var url = this.getAttribute('href');
+		window.open(url, '_blank');
+	}
+
 	r.loadsel = function () {
-		var ths = QSA('#ggrid>a');
+		var ths = QSA('#ggrid>a'),
+			have_sel = !!QS('#files tr.sel');
+
 		for (var a = 0, aa = ths.length; a < aa; a++) {
-			ths[a].onclick = r.sel ? seltgl : null;
+			ths[a].onclick = r.sel ? seltgl : have_sel ? bgopen : 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() {
@@ -836,20 +851,21 @@ 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] == '/',
+				ac = isdir ? ' class="dir"' : '',
 				ihref = href;
 
-			if (isdir) {
-				ihref = '/.cpr/ico/folder'
-			}
-			else if (r.thumbs) {
+			if (r.thumbs) {
 				ihref += (ihref.indexOf('?') === -1 ? '?' : '&') + 'th=' + (have_webp ? 'w' : 'j');
 			}
+			else if (isdir) {
+				ihref = '/.cpr/ico/folder';
+			}
 			else {
 				var ar = href.split('?')[0].split('.');
 				if (ar.length > 1)
@@ -870,7 +886,7 @@ var thegrid = (function () {
 			}
 
 			html.push('<a href="' + href + '" ref="' + ref + '"><img src="' +
-				ihref + '" /><span>' + ao.innerHTML + '</span></a>');
+				ihref + '" /><span' + ac + '>' + ao.innerHTML + '</span></a>');
 		}
 		lfiles.style.display = 'none';
 		gfiles.style.display = 'block';
@@ -960,7 +976,7 @@ document.onkeydown = function (e) {
 	if (k == 'KeyT')
 		return ebi('thumbs').click();
 
-	if (window['thegrid'] && thegrid.en) {
+	if (thegrid.en) {
 		if (k == 'KeyS')
 			return ebi('gridsel').click();
 
@@ -1026,6 +1042,7 @@ 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 +1067,83 @@ 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() {
@@ -1381,7 +1448,7 @@ var treectl = (function () {
 		if (hpush)
 			get_tree('.', xhr.top);
 
-		enspin('#files');
+		enspin(thegrid.en ? '#gfiles' : '#files');
 	}
 
 	function treegrow(e) {
@@ -1461,6 +1528,7 @@ var treectl = (function () {
 
 		apply_perms(res.perms);
 		despin('#files');
+		despin('#gfiles');
 
 		ebi('pro').innerHTML = res.logues ? res.logues[0] || "" : "";
 		ebi('epi').innerHTML = res.logues ? res.logues[1] || "" : "";
@@ -1995,8 +2063,7 @@ var msel = (function () {
 	}
 	function selui() {
 		clmod(ebi('wtoggle'), 'sel', getsel().length);
-		if (window['thegrid'])
-			thegrid.loadsel();
+		thegrid.loadsel();
 	}
 	function seltgl(e) {
 		ev(e);

copyparty/web/splash.css

@@ -26,10 +26,23 @@ a {
 	border-radius: .2em;
 	padding: .2em .8em;
 }
-td, th {
+table {
+	border-collapse: collapse;
+}
+.vols td,
+.vols th {
 	padding: .3em .6em;
 	text-align: left;
 }
+.num {
+	border-right: 1px solid #bbb;
+}
+.num td {
+	padding: .1em .7em .1em 0;
+}
+.num td:first-child {
+	text-align: right;
+}
 .btns {
 	margin: 1em 0;
 }
@@ -57,4 +70,7 @@ html.dark input {
 	border-radius: .5em;
 	padding: .5em .7em;
 	margin: 0 .5em 0 0;
+}
+html.dark .num {
+	border-color: #777;
 }

copyparty/web/splash.html

@@ -15,18 +15,27 @@
 
         {%- 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>/{{ mp }}</td><td><a href="/{{ mp }}?scan">rescan</a></td><td>{{ vstate[mp] }}</td></tr>
-                {%- endif %}
-                {% endfor %}
-            </tbody>
-        </table>
+        <table><tr><td> <!-- hehehe -->
+            <table class="num">
+                <tr><td>scanning</td><td>{{ scanning }}</td></tr>
+                <tr><td>hash-q</td><td>{{ hashq }}</td></tr>
+                <tr><td>tag-q</td><td>{{ tagq }}</td></tr>
+                <tr><td>mtp-q</td><td>{{ mtpq }}</td></tr>
+            </table>
+        </td><td>
+            <table class="vols">
+                <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>
+        </td></tr></table>
         <div class="btns">
-            <a href="/{{ avol[0] }}?stack">dump stack</a>
+            <a href="{{ avol[0] }}?stack">dump stack</a>
         </div>
         {%- endif %}
 
@@ -34,7 +43,7 @@
         <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 %}
@@ -43,7 +52,7 @@
         <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 %}

copyparty/web/up2k.js

@@ -17,6 +17,7 @@ function goto_up2k() {
 // chrome requires https to use crypto.subtle,
 // usually it's undefined but some chromes throw on invoke
 var up2k = null;
+var sha_js = window.WebAssembly ? 'hw' : 'ac';  // ff53,c57,sa11
 try {
     var cf = crypto.subtle || crypto.webkitSubtle;
     cf.digest('SHA-512', new Uint8Array(1)).then(
@@ -430,13 +431,15 @@ function up2k_init(subtle) {
     // upload ui hidden by default, clicking the header shows it
     function init_deps() {
         if (!subtle && !window.asmCrypto) {
-            showmodal('<h1>loading sha512.js</h1><h2>since ' + shame + '</h2><h4>thanks chrome</h4>');
-            import_js('/.cpr/deps/sha512.js', unmodal);
+            var fn = 'sha512.' + sha_js + '.js';
+            showmodal('<h1>loading ' + fn + '</h1><h2>since ' + shame + '</h2><h4>thanks chrome</h4>');
+            import_js('/.cpr/deps/' + fn, unmodal);
 
             if (is_https)
                 ebi('u2foot').innerHTML = shame + ' so <em>this</em> uploader will do like 500kB/s at best';
             else
-                ebi('u2foot').innerHTML = 'seems like ' + shame + ' so do that if you want more performance';
+                ebi('u2foot').innerHTML = 'seems like ' + shame + ' so do that if you want more performance <span style="color:#' +
+                    (sha_js == 'ac' ? 'c84">(expecting 20' : '8a5">(but dont worry too much, expect 100') + ' MiB/s)</span>';
         }
     }
 
@@ -886,6 +889,10 @@ function up2k_init(subtle) {
         return base64;
     }
 
+    function hex2u8(txt) {
+        return new Uint8Array(txt.match(/.{2}/g).map(function (b) { return parseInt(b, 16); }));
+    }
+
     function get_chunksize(filesize) {
         var chunksize = 1024 * 1024,
             stepsize = 512 * 1024;
@@ -987,10 +994,18 @@ function up2k_init(subtle) {
             if (subtle)
                 subtle.digest('SHA-512', buf).then(hash_done);
             else setTimeout(function () {
-                var hasher = new asmCrypto.Sha512();
-                hasher.process(new Uint8Array(buf));
-                hasher.finish();
-                hash_done(hasher.result);
+                var u8buf = new Uint8Array(buf);
+                if (sha_js == 'hw') {
+                    hashwasm.sha512(u8buf).then(function (v) {
+                        hash_done(hex2u8(v))
+                    });
+                }
+                else {
+                    var hasher = new asmCrypto.Sha512();
+                    hasher.process(u8buf);
+                    hasher.finish();
+                    hash_done(hasher.result);
+                }
             }, 1);
         };
 

copyparty/web/upload.css

@@ -238,6 +238,10 @@
 	color: #fff;
 	font-style: italic;
 }
+#u2foot span {
+	color: #999;
+	font-size: .9em;
+}
 #u2footfoot {
 	margin-bottom: -1em;
 }

docs/README.md

@@ -0,0 +1,11 @@
+## [`minimal-up2k.html`](minimal-up2k.html)
+* save as `.epilogue.html` inside a folder to [simplify the ui](https://user-images.githubusercontent.com/241032/118311195-dd6ca380-b4ef-11eb-86f3-75a3ff2e1332.png)
+
+## [`browser.css`](browser.css)
+* example for `--css-browser`
+
+## [`rclone.md`](rclone.md)
+* notes on using rclone as a fuse client/server
+
+## [`example.conf`](example.conf)
+* example config file for `-c` which never really happened

docs/browser.css

@@ -0,0 +1,29 @@
+html {
+    background: url('/wp/wallhaven-mdjrqy.jpg') center / cover no-repeat fixed;
+}
+#files th {
+    background: rgba(32, 32, 32, 0.9) !important;
+}
+#ops,
+#treeul,
+#files td {
+    background: rgba(32, 32, 32, 0.3) !important;
+}
+
+
+html.light {
+    background: url('/wp/wallhaven-dpxl6l.png') center / cover no-repeat fixed;
+}
+html.light #files th {
+    background: rgba(255, 255, 255, 0.9) !important;
+}
+html.light #ops,
+html.light #treeul,
+html.light #files td {
+    background: rgba(248, 248, 248, 0.8) !important;
+}
+
+
+#files * {
+    background: transparent !important;
+}

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
 
@@ -146,6 +153,9 @@ dbg.asyncStore.pendingBreakpoints = {}
 # fix firefox phantom breakpoints
 about:config >> devtools.debugger.prefs-schema-version = -1
 
+# determine server version
+git reset --hard origin/HEAD && git log --format=format:"%H %ai %d" --decorate=full > /dev/shm/revs && cat /dev/shm/revs | while read -r rev extra; do (git reset --hard $rev >/dev/null 2>/dev/null && dsz=$(cat copyparty/web/{util,browser,up2k}.js 2>/dev/null | diff -wNarU0 - <(cat /mnt/Users/ed/Downloads/ref/{util,browser,up2k}.js) | wc -c) && printf '%s %6s %s\n' "$rev" $dsz "$extra") </dev/null; done                
+
 
 ##
 ## http 206

scripts/deps-docker/Dockerfile

@@ -9,6 +9,12 @@ ENV     ver_asmcrypto=5b994303a9d3e27e0915f72a10b6c2c51535a4dc \
         ver_zopfli=1.0.3
 
 
+# TODO
+# sha512.hw.js https://github.com/Daninet/hash-wasm
+# sha512.kc.js https://github.com/chm-diederichs/sha3-wasm
+# awk '/HMAC state/{o=1}  /var HEAP/{o=0}  /function hmac_reset/{o=1}  /return \{/{o=0}  /var __extends =/{o=1}  /var Hash =/{o=0}  /hmac_|pbkdf2_/{next}  o{next}  {gsub(/IllegalStateError/,"Exception")}  {sub(/^ +/,"");sub(/^\/\/ .*/,"");sub(/;$/," ;")}  1' <sha512.ac.js.orig >sha512.ac.js; for fn in sha512.ac.js.orig sha512.ac.js; do wc -c <$fn; wc -c <$fn.gz ; for n in {1..9}; do printf '%8d %d bz\n' $(bzip2 -c$n <$fn | wc -c) $n; done; done
+
+
 # download;
 # the scp url is latin from https://fonts.googleapis.com/css2?family=Source+Code+Pro&display=swap
 RUN     mkdir -p /z/dist/no-pk \

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/make-sfx.sh

@@ -32,6 +32,10 @@ gtar=$(command -v gtar || command -v gnutar) || true
 	[ -e /opt/local/bin/bzip2 ] &&
 		bzip2() { /opt/local/bin/bzip2 "$@"; }
 }
+
+gawk=$(command -v gawk || command -v gnuawk || command -v awk)
+awk() { $gawk "$@"; }
+
 pybin=$(command -v python3 || command -v python) || {
 	echo need python
 	exit 1
@@ -194,17 +198,46 @@ tmv "$f"
 
 # up2k goes from 28k to 22k laff
 echo entabbening
-find | grep -E '\.(js|css|html)$' | while IFS= read -r f; do
+find | grep -E '\.css$' | while IFS= read -r f; do
+	awk '{
+		sub(/^[ \t]+/,"");
+		sub(/[ \t]+$/,"");
+		$0=gensub(/^([a-z-]+) *: *(.*[^ ]) *;$/,"\\1:\\2;","1");
+		sub(/ +\{$/,"{");
+		gsub(/, /,",")
+	}
+	!/\}$/ {printf "%s",$0;next}
+	1
+	' <$f | gsed 's/;\}$/}/' >t
+	tmv "$f"
+done
+find | grep -E '\.(js|html)$' | while IFS= read -r f; do
 	unexpand -t 4 --first-only <"$f" >t
 	tmv "$f"
 done
 
+
+gzres() {
+command -v pigz &&
+	pk='pigz -11 -J 34 -I 100' ||
+	pk='gzip'
+
+echo "$pk"
+find | grep -E '\.(js|css)$' | while IFS= read -r f; do
+	echo -n .
+	$pk "$f"
+done
+echo
+}
+gzres
+
+
 echo gen tarlist
 for d in copyparty dep-j2; do find $d -type f; done |
 sed -r 's/(.*)\.(.*)/\2 \1/' | LC_ALL=C sort |
 sed -r 's/([^ ]*) (.*)/\2.\1/' | grep -vE '/list1?$' > list1
 
-(grep -vE 'gz$' list1; grep -E 'gz$' list1) >list
+(grep -vE '\.(gz|br)$' list1; grep -E '\.(gz|br)$' list1) >list || true
 
 echo creating tar
 args=(--owner=1000 --group=1000)

scripts/profile.py

@@ -0,0 +1,34 @@
+#!/usr/bin/env python3
+
+import sys
+
+sys.path.insert(0, ".")
+cmd = sys.argv[1]
+
+if cmd == "cpp":
+    from copyparty.__main__ import main
+
+    argv = ["__main__", "-v", "srv::r", "-v", "../../yt:yt:r"]
+    main(argv=argv)
+
+elif cmd == "test":
+    from unittest import main
+
+    argv = ["__main__", "discover", "-s", "tests"]
+    main(module=None, argv=argv)
+
+else:
+    raise Exception()
+
+# import dis; print(dis.dis(main))
+
+
+# macos:
+#   option1) python3.9 -m pip install --user -U vmprof==0.4.9
+#   option2) python3.9 -m pip install --user -U https://github.com/vmprof/vmprof-python/archive/refs/heads/master.zip
+#
+# python -m vmprof -o prof --lines ./scripts/profile.py test
+
+# linux: ~/.local/bin/vmprofshow prof tree | grep -vF '[1m  0.'
+# macos: ~/Library/Python/3.9/bin/vmprofshow prof tree | grep -vF '[1m  0.'
+#   win: %appdata%\..\Roaming\Python\Python39\Scripts\vmprofshow.exe prof tree

scripts/run-tests.sh

@@ -0,0 +1,15 @@
+#!/bin/bash
+set -ex
+
+pids=()
+for py in python{2,3}; do
+    nice $py -m unittest discover -s tests >/dev/null &
+    pids+=($!)
+done
+
+python3 scripts/test/smoketest.py &
+pids+=($!)
+
+for pid in ${pids[@]}; do
+    wait $pid
+done

scripts/sfx.sh

@@ -47,7 +47,7 @@ grep -E '/(python|pypy)[0-9\.-]*$' >$dir/pys || true
 	printf '\033[1;30mlooking for jinja2 in [%s]\033[0m\n' "$_py" >&2
 	$_py -c 'import jinja2' 2>/dev/null || continue
 	printf '%s\n' "$_py"
-	mv $dir/{,x.}jinja2
+	mv $dir/{,x.}dep-j2
 	break
 done)"
 

scripts/test/smoketest.py

@@ -0,0 +1,209 @@
+import os
+import sys
+import time
+import shlex
+import shutil
+import signal
+import tempfile
+import requests
+import threading
+import subprocess as sp
+
+
+CPP = []
+
+
+class Cpp(object):
+    def __init__(self, args):
+        args = [sys.executable, "-m", "copyparty"] + args
+        print(" ".join([shlex.quote(x) for x in args]))
+
+        self.ls_pre = set(list(os.listdir()))
+        self.p = sp.Popen(args)
+        # , stdout=sp.PIPE, stderr=sp.PIPE)
+
+        self.t = threading.Thread(target=self._run)
+        self.t.daemon = True
+        self.t.start()
+
+    def _run(self):
+        self.so, self.se = self.p.communicate()
+
+    def stop(self, wait):
+        if wait:
+            os.kill(self.p.pid, signal.SIGINT)
+            self.t.join(timeout=2)
+        else:
+            self.p.kill()  # macos py3.8
+
+    def clean(self):
+        t = os.listdir()
+        for f in t:
+            if f not in self.ls_pre and f.startswith("up."):
+                os.unlink(f)
+
+    def await_idle(self, ub, timeout):
+        req = ["scanning</td><td>False", "hash-q</td><td>0", "tag-q</td><td>0"]
+        lim = int(timeout * 10)
+        u = ub + "?h"
+        for n in range(lim):
+            try:
+                time.sleep(0.1)
+                r = requests.get(u, timeout=0.1)
+                for x in req:
+                    if x not in r.text:
+                        print("ST: {}/{} miss {}".format(n, lim, x))
+                        raise Exception()
+                print("ST: idle")
+                return
+            except:
+                pass
+
+
+def tc1():
+    ub = "http://127.0.0.1:4321/"
+    td = os.path.join("srv", "smoketest")
+    try:
+        shutil.rmtree(td)
+    except:
+        if os.path.exists(td):
+            raise
+
+    for _ in range(10):
+        try:
+            os.mkdir(td)
+        except:
+            time.sleep(0.1)  # win10
+
+    assert os.path.exists(td)
+
+    vidp = os.path.join(tempfile.gettempdir(), "smoketest.h264")
+    if not os.path.exists(vidp):
+        cmd = "ffmpeg -f lavfi -i testsrc=48x32:3 -t 1 -c:v libx264 -tune animation -preset veryslow -crf 69"
+        sp.check_call(cmd.split(" ") + [vidp])
+
+    with open(vidp, "rb") as f:
+        ovid = f.read()
+
+    args = [
+        "-p4321",
+        "-e2dsa",
+        "-e2tsr",
+        "--no-mutagen",
+        "--th-ff-jpg",
+        "--hist",
+        os.path.join(td, "dbm"),
+    ]
+    pdirs = []
+    hpaths = {}
+
+    for d1 in ["r", "w", "a"]:
+        pdirs.append("{}/{}".format(td, d1))
+        pdirs.append("{}/{}/j".format(td, d1))
+        for d2 in ["r", "w", "a"]:
+            d = os.path.join(td, d1, "j", d2)
+            pdirs.append(d)
+            os.makedirs(d)
+
+    pdirs = [x.replace("\\", "/") for x in pdirs]
+    udirs = [x.split("/", 2)[2] for x in pdirs]
+    perms = [x.rstrip("j/")[-1] for x in pdirs]
+    for pd, ud, p in zip(pdirs, udirs, perms):
+        if ud[-1] == "j":
+            continue
+
+        hp = None
+        if pd.endswith("st/a"):
+            hp = hpaths[ud] = os.path.join(td, "db1")
+        elif pd[:-1].endswith("a/j/"):
+            hpaths[ud] = os.path.join(td, "dbm")
+            hp = None
+        else:
+            hp = "-"
+            hpaths[ud] = os.path.join(pd, ".hist")
+
+        arg = "{}:{}:{}".format(pd, ud, p, hp)
+        if hp:
+            arg += ":chist=" + hp
+
+        args += ["-v", arg]
+
+    # return
+    cpp = Cpp(args)
+    CPP.append(cpp)
+    cpp.await_idle(ub, 3)
+
+    for d in udirs:
+        vid = ovid + "\n{}".format(d).encode("utf-8")
+        try:
+            requests.post(ub + d, data={"act": "bput"}, files={"f": ("a.h264", vid)})
+        except:
+            pass
+
+    cpp.clean()
+
+    # GET permission
+    for d, p in zip(udirs, perms):
+        u = "{}{}/a.h264".format(ub, d)
+        r = requests.get(u)
+        ok = bool(r)
+        if ok != (p in ["a"]):
+            raise Exception("get {} with perm {} at {}".format(ok, p, u))
+
+    # stat filesystem
+    for d, p in zip(pdirs, perms):
+        u = "{}/a.h264".format(d)
+        ok = os.path.exists(u)
+        if ok != (p in ["a", "w"]):
+            raise Exception("stat {} with perm {} at {}".format(ok, p, u))
+
+    # GET thumbnail, vreify contents
+    for d, p in zip(udirs, perms):
+        u = "{}{}/a.h264?th=j".format(ub, d)
+        r = requests.get(u)
+        ok = bool(r and r.content[:3] == b"\xff\xd8\xff")
+        if ok != (p in ["a"]):
+            raise Exception("thumb {} with perm {} at {}".format(ok, p, u))
+
+    # check tags
+    cpp.await_idle(ub, 5)
+    for d, p in zip(udirs, perms):
+        u = "{}{}?ls".format(ub, d)
+        r = requests.get(u)
+        j = r.json() if r else False
+        tag = None
+        if j:
+            for f in j["files"]:
+                tag = tag or f["tags"].get("res")
+
+        r_ok = bool(j)
+        w_ok = bool(r_ok and j.get("files"))
+
+        if not r_ok or w_ok != (p in ["a"]):
+            raise Exception("ls {} with perm {} at {}".format(ok, p, u))
+
+        if (tag and p != "a") or (not tag and p == "a"):
+            raise Exception("tag {} with perm {} at {}".format(tag, p, u))
+
+        if tag is not None and tag != "48x32":
+            raise Exception("tag [{}] at {}".format(tag, u))
+
+    cpp.stop(True)
+
+
+def run(tc):
+    try:
+        tc()
+    finally:
+        try:
+            CPP[0].stop(False)
+        except:
+            pass
+
+
+def main():
+    run(tc1)
+
+
+if __name__ == "__main__":
+    main()

tests/test_httpcli.py

@@ -8,13 +8,13 @@ import time
 import shutil
 import pprint
 import tarfile
+import tempfile
 import unittest
-
 from argparse import Namespace
-from copyparty.authsrv import AuthSrv
-from copyparty.httpcli import HttpCli
 
 from tests import util as tu
+from copyparty.authsrv import AuthSrv
+from copyparty.httpcli import HttpCli
 
 
 def hdr(query):
@@ -32,21 +32,27 @@ class Cfg(Namespace):
             no_zip=False,
             no_scandir=False,
             no_sendfile=True,
+            no_rescan=True,
+            ihead=False,
             nih=True,
             mtp=[],
             mte="a",
+            hist=None,
+            no_hash=False,
             **{k: False for k in "e2d e2ds e2dsa e2t e2ts e2tsr".split()}
         )
 
 
 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)
 
@@ -95,7 +101,7 @@ class TestHttpCli(unittest.TestCase):
             pprint.pprint(vcfg)
 
             self.args = Cfg(v=vcfg, a=["o:o", "x:x"])
-            self.auth = AuthSrv(self.args, self.log)
+            self.asrv = AuthSrv(self.args, self.log)
             vfiles = [x for x in allfiles if x.startswith(top)]
             for fp in vfiles:
                 rok, wok = self.can_rw(fp)
@@ -184,12 +190,12 @@ class TestHttpCli(unittest.TestCase):
     def put(self, url):
         buf = "PUT /{0} HTTP/1.1\r\nCookie: cppwd=o\r\nConnection: close\r\nContent-Length: {1}\r\n\r\nok {0}\n"
         buf = buf.format(url, len(url) + 4).encode("utf-8")
-        conn = tu.VHttpConn(self.args, self.auth, self.log, buf)
+        conn = tu.VHttpConn(self.args, self.asrv, self.log, buf)
         HttpCli(conn).run()
         return conn.s._reply.decode("utf-8").split("\r\n\r\n", 1)
 
     def curl(self, url, binary=False):
-        conn = tu.VHttpConn(self.args, self.auth, self.log, hdr(url))
+        conn = tu.VHttpConn(self.args, self.asrv, self.log, hdr(url))
         HttpCli(conn).run()
         if binary:
             h, b = conn.s._reply.split(b"\r\n\r\n", 1)

tests/test_vfs.py

@@ -7,24 +7,30 @@ import json
 import shutil
 import tempfile
 import unittest
-
 from textwrap import dedent
 from argparse import Namespace
-from copyparty.authsrv import AuthSrv
-from copyparty import util
 
 from tests import util as tu
+from copyparty.authsrv import AuthSrv, VFS
+from copyparty import util
 
 
 class Cfg(Namespace):
     def __init__(self, a=[], v=[], c=None):
         ex = {k: False for k in "e2d e2ds e2dsa e2t e2ts e2tsr".split()}
-        ex["mtp"] = []
-        ex["mte"] = "a"
+        ex2 = {"mtp": [], "mte": "a", "hist": None, "no_hash": False}
+        ex.update(ex2)
         super(Cfg, self).__init__(a=a, v=v, c=c, **ex)
 
 
 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__))
 
@@ -41,6 +47,7 @@ class TestVFS(unittest.TestCase):
         self.assertEqual(util.undot(query), response)
 
     def ls(self, vfs, vpath, uname):
+        # type: (VFS, str, str) -> tuple[str, str, str]
         """helper for resolving and listing a folder"""
         vn, rem = vfs.get(vpath, uname, True, False)
         r1 = vn.ls(rem, uname, False)
@@ -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)
 
@@ -111,13 +113,13 @@ class TestVFS(unittest.TestCase):
         n = vfs.nodes["a"]
         self.assertEqual(len(vfs.nodes), 1)
         self.assertEqual(n.vpath, "a")
-        self.assertEqual(n.realpath, td + "/a")
+        self.assertEqual(n.realpath, os.path.join(td, "a"))
         self.assertEqual(n.uread, ["*", "k"])
         self.assertEqual(n.uwrite, ["k"])
         n = n.nodes["ac"]
         self.assertEqual(len(vfs.nodes), 1)
         self.assertEqual(n.vpath, "a/ac")
-        self.assertEqual(n.realpath, td + "/a/ac")
+        self.assertEqual(n.realpath, os.path.join(td, "a", "ac"))
         self.assertEqual(n.uread, ["*", "k"])
         self.assertEqual(n.uwrite, ["k"])
         n = n.nodes["acb"]
@@ -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(
@@ -249,7 +251,7 @@ class TestVFS(unittest.TestCase):
         n = au.vfs
         # root was not defined, so PWD with no access to anyone
         self.assertEqual(n.vpath, "")
-        self.assertEqual(n.realpath, td)
+        self.assertEqual(n.realpath, None)
         self.assertEqual(n.uread, [])
         self.assertEqual(n.uwrite, [])
         self.assertEqual(len(n.nodes), 1)
@@ -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

@@ -1,16 +1,36 @@
 import os
+import sys
 import time
+import shutil
 import jinja2
 import tempfile
+import platform
 import subprocess as sp
 
-from copyparty.util import Unrecv
 
+WINDOWS = platform.system() == "Windows"
+ANYWIN = WINDOWS or sys.platform in ["msys"]
+MACOS = platform.system() == "Darwin"
 
 J2_ENV = jinja2.Environment(loader=jinja2.BaseLoader)
 J2_FILES = J2_ENV.from_string("{{ files|join('\n') }}")
 
 
+def nah(*a, **ka):
+    return False
+
+
+if MACOS:
+    import posixpath
+
+    posixpath.islink = nah
+    os.path.islink = nah
+    # 25% faster; until any tests do symlink stuff
+
+
+from copyparty.util import Unrecv
+
+
 def runcmd(*argv):
     p = sp.Popen(argv, stdout=sp.PIPE, stderr=sp.PIPE)
     stdout, stderr = p.communicate()
@@ -28,18 +48,25 @@ 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://131072")
         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 +77,7 @@ def get_ramdisk():
     try:
         os.mkdir(ret)
     finally:
-        return ret
+        return subdir(ret)
 
 
 class NullBroker(object):
@@ -83,15 +110,19 @@ class VHttpSrv(object):
 
 
 class VHttpConn(object):
-    def __init__(self, args, auth, log, buf):
+    def __init__(self, args, asrv, log, buf):
         self.s = VSock(buf)
         self.sr = Unrecv(self.s)
         self.addr = ("127.0.0.1", "42069")
         self.args = args
-        self.auth = auth
+        self.asrv = asrv
+        self.is_mp = False
         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()