FreshRSS/rssguard
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

update translation scraper

Browse Source 
¨
This commit is contained in:
Martin Rotter 2022-04-26 08:01:42 +02:00
parent fe02722662
commit 2eb016c895
2 changed files with 7 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
resources/desktop/com.github.rssguard.appdata.xml
View File

@ -26,7 +26,7 @@
<url type="donation">https://github.com/sponsors/martinrotter</url>
<content_rating type="oars-1.1" />
<releases>
<release version="4.2.1" date="2022-04-22"/>
<release version="4.2.1" date="2022-04-26"/>
</releases>
<content_rating type="oars-1.0">
<content_attribute id="violence-cartoon">none</content_attribute>

18
resources/scripts/scrapers/translate-feed.py
View File

@ -1,9 +1,9 @@
# Translates entries of RSS 2.0 feed into different locale.
# Translates entries of RSS 2.0 (or ATOM) feed into different locale.
#
# Requires Python 3.10+.
#
# Make sure to have all dependencies installed:
# pip3 install googletrans
# pip3 install asyncio (if using parallel version of the script)
# pip3 install hyper (for HTTP/2 support, much faster than default)
# pip3 install googletrans-py lxml bs4 httpx httpcore asyncio --upgrade
#
# You must provide raw RSS 2.0 (or ATOM) UTF-8 feed XML data as input, for example with curl:
# curl 'https://phys.org/rss-feed/' | python ./translate-feed.py "en" "pt_BR" "true"
@ -11,14 +11,10 @@
# You must provide three command line arguments:
# translate-feed.py [FROM-LANGUAGE] [TO-LANGUAGE] [RUN-PARALLEL] [FEED-ENCODING (optional)]
import json
import re
import io
import sys
import time
import html
import requests
import distutils.util
import setuptools._distutils.util
import xml.etree.ElementTree as ET
import itertools as IT
from googletrans import Translator
@ -26,7 +22,7 @@ from bs4 import BeautifulSoup
lang_from = sys.argv[1]
lang_to = sys.argv[2]
parallel = bool(distutils.util.strtobool(sys.argv[3]))
parallel = bool(setuptools._distutils.util.strtobool(sys.argv[3]))
if (len(sys.argv) >= 5):
src_enc = sys.argv[4]
@ -40,8 +36,6 @@ if parallel:
sys.stdin.reconfigure(encoding = src_enc)
rss_data = sys.stdin.read()
#print(rss_data)
try:
rss_document = ET.fromstring(rss_data)
except ET.ParseError as err: