X-Git-Url: https://jfr.im/git/z_archive/twitter.git/blobdiff_plain/907402f64301f9f87555afa5c305c40504da7145..794d6d442f9c150766568fd1626be46735829c85:/twitter/util.py diff --git a/twitter/util.py b/twitter/util.py index 4396b07..7831939 100644 --- a/twitter/util.py +++ b/twitter/util.py @@ -10,15 +10,19 @@ from __future__ import print_function import contextlib import re import sys +import textwrap import time -import urllib2 -import urlparse +import socket try: from html.entities import name2codepoint unichr = chr + import urllib.request as urllib2 + import urllib.parse as urlparse except ImportError: from htmlentitydefs import name2codepoint + import urllib2 + import urlparse def htmlentitydecode(s): return re.sub( @@ -83,18 +87,18 @@ class Fail(object): def find_links(line): """Find all links in the given line. The function returns a sprintf style format string (with %s placeholders for the links) and a list of urls.""" - l = line.replace(u"%", u"%%") + l = line.replace("%", "%%") regex = "(https?://[^ )]+)" return ( - re.sub(regex, "%s", l), + re.sub(regex, "%s", l), [m.group(1) for m in re.finditer(regex, l)]) - + def follow_redirects(link, sites= None): """Follow directs for the link as long as the redirects are on the given sites and return the resolved link.""" def follow(url): return sites == None or urlparse.urlparse(url).hostname in sites - + class RedirectHandler(urllib2.HTTPRedirectHandler): def __init__(self): self.last_url = None @@ -106,7 +110,7 @@ def follow_redirects(link, sites= None): self, req, fp, code, msg, hdrs, newurl) r.get_method = lambda : 'HEAD' return r - + if not follow(link): return link redirect_handler = RedirectHandler() @@ -114,21 +118,37 @@ def follow_redirects(link, sites= None): req = urllib2.Request(link) req.get_method = lambda : 'HEAD' try: - with contextlib.closing(opener.open(req)) as site: + with contextlib.closing(opener.open(req,timeout=1)) as site: return site.url - except (urllib2.HTTPError, urllib2.URLError): + except: return redirect_handler.last_url if redirect_handler.last_url else link def expand_line(line, sites): """Expand the links in the line for the given sites.""" - l = line.strip() - msg_format, links = find_links(l) - args = tuple(follow_redirects(l, sites) for l in links) - return msg_format % args + try: + l = line.strip() + msg_format, links = find_links(l) + args = tuple(follow_redirects(l, sites) for l in links) + line = msg_format % args + except Exception as e: + try: + err("expanding line %s failed due to %s" % (line, unicode(e))) + except: + pass + return line def parse_host_list(list_of_hosts): """Parse the comma separated list of hosts.""" p = set( m.group(1) for m in re.finditer("\s*([^,\s]+)\s*,?\s*", list_of_hosts)) return p - + + +def align_text(text, left_margin=17, max_width=160): + lines = [] + for line in text.split('\n'): + temp_lines = textwrap.wrap(line, max_width - left_margin) + temp_lines = [(' ' * left_margin + line) for line in temp_lines] + lines.append('\n'.join(temp_lines)) + ret = '\n'.join(lines) + return ret.lstrip()