From 72121928017f70e210cacdf124b56cf36eb26cbf Mon Sep 17 00:00:00 2001
From: John-Mark Gurney <jmg@funkthat.com>
Date: Fri, 17 Feb 2023 17:40:15 -0800
Subject: [PATCH] use threads to read/hash multiple datablocks at a time..

as we might have a lot of work to submit, BUT it might fail early,
don't submit too much work early on, just for it to fail, so we
limit how much work we submit..
---
 ui/medashare/btv/__init__.py | 35 ++++++++++++++++++++++++++---------
 1 file changed, 26 insertions(+), 9 deletions(-)

diff --git a/ui/medashare/btv/__init__.py b/ui/medashare/btv/__init__.py
index d29ae95..f2c80dc 100644
--- a/ui/medashare/btv/__init__.py
+++ b/ui/medashare/btv/__init__.py
@@ -1,5 +1,6 @@
 
 from . import bencode
+import concurrent
 import fnmatch
 from functools import reduce
 from hashlib import sha1, sha512
@@ -247,7 +248,7 @@ def validate(torrent, basedir, with_file_hashes=None):
 
 	file_hashes = dict()
 
-	def apply_fun(data, fname, offset):
+	def apply_fun(hash, data, fname, offset):
 		if with_file_hashes is not None:
 			try:
 				hio = file_hashes[fname]
@@ -262,16 +263,32 @@ def validate(torrent, basedir, with_file_hashes=None):
 	pieces = info['pieces']
 	piecescnt = len(pieces) // 20
 	valid = [ None ] * piecescnt
-	for num, i in enumerate(pieces[x:x+20] for x in range(0, len(pieces),
-	    20)):
-		hash = sha1()
+	with concurrent.futures.ThreadPoolExecutor(max_workers=os.cpu_count()) as executor:
+		futures = {}
 
-		stor.apply_piece(num, apply_fun)
+		pending_work = [ x for x in enumerate(pieces[x:x+20] for x in range(0, len(pieces), 20)) ]
 
-		if hash.digest() == i:
-			valid[num] = True
-		else:
-			valid[num] = False
+		def submit_work(num, i):
+			hash = sha1()
+
+			fut = executor.submit(stor.apply_piece, num, lambda *args: apply_fun(hash, *args))
+
+			futures[fut] = num, hash, i
+
+		for i in range(min(len(pending_work), os.cpu_count() + 1)):
+			submit_work(*pending_work.pop(0))
+
+		for future in concurrent.futures.as_completed(futures):
+			if pending_work:
+				submit_work(*pending_work.pop(0))
+
+			future.result()
+			num, hash, i = futures[future]
+
+			if hash.digest() == i:
+				valid[num] = True
+			else:
+				valid[num] = False
 
 	if files is None:
 		filesizes = { pathlib.PurePosixPath(info['name'].decode(