varia/etherpump
14
0
Fork 0
Code Issues 15 Pull Requests Releases Activity

Skip chunking, rely solely on trio

Browse Source
This commit is contained in:
Luke Murphy 2020-01-20 18:06:12 +01:00
parent 4facb64bf8
commit dc63cac8aa
No known key found for this signature in database
GPG Key ID: 5E2EF5A63E3718CC
1 changed files with 3 additions and 15 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
etherpump/commands/pull.py
View File

@ -15,7 +15,6 @@ from xml.etree import ElementTree as ET
import asks import asks
import html5lib import html5lib
import trio import trio
from tqdm import tqdm
from etherpump.commands.common import * # noqa from etherpump.commands.common import * # noqa
from etherpump.commands.html5tidy import html5tidy from etherpump.commands.html5tidy import html5tidy
@ -446,16 +445,6 @@ async def handle_pad(args, padid, data, info, session):
await f.write(json.dumps(meta)) await f.write(json.dumps(meta))
async def handle_pad_chunk(args, padids, data, info, session):
progress_kwargs = {}
if not istty():
progress_kwargs.update(dict(disable=True))
padids = tqdm(iterable=padids, total=len(padids), **progress_kwargs,)
for padid in padids:
await handle_pad(args, padid, data, info, session)
async def handle_pads(args): async def handle_pads(args):
session = asks.Session(connections=args.connection) session = asks.Session(connections=args.connection)
info = loadpadinfo(args.padinfo) info = loadpadinfo(args.padinfo)
@ -464,12 +453,11 @@ async def handle_pads(args):
padids = await get_padids(args, info, data, session) padids = await get_padids(args, info, data, session)
if args.skip: if args.skip:
padids = padids[args.skip : len(padids)] padids = padids[args.skip : len(padids)]
CHUNK_SIZE = math.ceil(len(padids) / 3)
async with trio.open_nursery() as nursery: async with trio.open_nursery() as nursery:
for padids in chunks(padids, CHUNK_SIZE): print("Processing {} pads now...".format(len(padids)))
_args = (args, padids, data, info, session) for padid in padids:
nursery.start_soon(handle_pad_chunk, *_args) nursery.start_soon(handle_pad, args, padid, data, info, session)
def main(args): def main(args):