Toot-Mastodon-CLI-TUI-clien.../toot/utils.py

91 lines
2.1 KiB
Python
Raw Normal View History

2017-04-24 16:25:34 +02:00
# -*- coding: utf-8 -*-
import os
2017-04-24 16:25:34 +02:00
import re
2017-12-29 14:26:40 +01:00
import socket
2018-01-21 16:39:40 +01:00
import unicodedata
import warnings
2017-04-24 16:25:34 +02:00
from bs4 import BeautifulSoup
from toot.exceptions import ConsoleError
2017-04-24 16:25:34 +02:00
2019-01-24 11:18:28 +01:00
def str_bool(b):
"""Convert boolean to string, in the way expected by the API."""
return "true" if b else "false"
2017-04-24 16:25:34 +02:00
def get_text(html):
"""Converts html to text, strips all tags."""
# Ignore warnings made by BeautifulSoup, if passed something that looks like
# a file (e.g. a dot which matches current dict), it will warn that the file
# should be opened instead of passing a filename.
with warnings.catch_warnings():
warnings.simplefilter("ignore")
text = BeautifulSoup(html.replace(''', "'"), "html.parser").get_text()
2018-01-21 16:39:40 +01:00
return unicodedata.normalize('NFKC', text)
2017-04-24 16:25:34 +02:00
def parse_html(html):
"""Attempt to convert html to plain text while keeping line breaks.
Returns a list of paragraphs, each being a list of lines.
"""
paragraphs = re.split("</?p[^>]*>", html)
# Convert <br>s to line breaks and remove empty paragraphs
paragraphs = [re.split("<br */?>", p) for p in paragraphs if p]
# Convert each line in each paragraph to plain text:
return [[get_text(l) for l in p] for p in paragraphs]
def format_content(content):
"""Given a Status contents in HTML, converts it into lines of plain text.
Returns a generator yielding lines of content.
"""
paragraphs = parse_html(content)
first = True
for paragraph in paragraphs:
if not first:
yield ""
for line in paragraph:
yield line
first = False
2017-12-29 14:26:40 +01:00
def domain_exists(name):
try:
socket.gethostbyname(name)
return True
except OSError:
return False
def assert_domain_exists(domain):
if not domain_exists(domain):
raise ConsoleError("Domain {} not found".format(domain))
EOF_KEY = "Ctrl-Z" if os.name == 'nt' else "Ctrl-D"
def multiline_input():
"""Lets user input multiple lines of text, terminated by EOF."""
lines = []
while True:
try:
lines.append(input())
except EOFError:
break
return "\n".join(lines).strip()