github/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-21 20:46:36 -05:00
Code Issues Projects Releases Packages Wiki Activity

Randomize archive order before populating search tree

Browse source 
This doesn't result in an elegant, perfectly balanced search tree,
but it's absolutely good enough. This commit completely mitigates
the worst-case scenario where the archive file is sorted.

Signed-off-by: Jody Bruchon <jody@jodybruchon.com>
This commit is contained in:
Jody Bruchon 2020-09-17 21:45:40 -04:00
parent 1d74d8d9f6
commit fda63a4e87
1 changed files with 12 additions and 35 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

47
youtube_dlc/YoutubeDL.py
View file

@ -122,17 +122,14 @@ def __init__(self, line):
# Tree insertion
def at_insert(self, line):
# print("at_insert: ", line)
cur = self
while True:
# print("comparing ", line, cur.line)
if cur.line:
if line < cur.line:
if cur.left is None:
cur.left = ArchiveTree(line)
return
else:
# print("LEFT")
cur = cur.left
continue
elif line > cur.line:
@ -140,7 +137,6 @@ def at_insert(self, line):
cur.right = ArchiveTree(line)
return
else:
# print("RIGHT")
cur = cur.right
continue
else:
@ -426,43 +422,24 @@ def preload_download_archive(self):
if ioe.errno != errno.ENOENT:
raise
lmax = len(lines)
if lmax >= 4:
if lmax > 10:
# Populate binary search tree by splitting the archive list in half
# and then adding from the outside edges inward
# This mitigates the worst case where the archive has been sorted
ptrLL = 0
ptrLR = lmax // 2
ptrRL = ptrLR + 1
ptrRR = lmax - 1
inserted = 0
while True:
# print("ptrs: %d %d %d %d" % (ptrLL, ptrLR, ptrRL, ptrRR))
if ptrLR > ptrLL:
self.archive.at_insert(lines[ptrLR])
inserted += 1
ptrLR -= 1;
if ptrRL < ptrRR:
self.archive.at_insert(lines[ptrRL])
inserted += 1
ptrRL += 1;
if ptrLL < ptrLR:
self.archive.at_insert(lines[ptrLL])
inserted += 1
ptrLL += 1;
if ptrRR > ptrRL:
self.archive.at_insert(lines[ptrRR])
inserted += 1
ptrRR -= 1;
if ptrLL == ptrLR and ptrRL == ptrRR:
print("inserted: %d, lmax: %d" % (inserted, lmax))
pos = 0
while pos < lmax:
if lmax - pos <= 2:
break
elif lmax > 0:
# Skip multi-line logic for a single line
for idx in lines:
self.archive.at_insert(idx)
else:
target = random.randrange(pos + 1, lmax - 1)
temp = lines[pos]
lines[pos] = lines[target]
lines[target] = lines[pos]
pos += 1
elif lmax < 1:
# No lines were loaded
return False
for x in lines:
self.archive.at_insert(x)
return True
def check_deprecated(param, option, suggestion):