Mercurial > hg > index.fcgi > www > www-1

annotate myrss/myrss_app.py @ 45:1d492a6d4237

find changesets by author, revision, files, or words in the commit message
add index.fcgi
author paulo
date Tue, 05 Feb 2013 02:28:40 -0700
parents df6a1a347584
children aca02ce71274
rev   line source
paulo@39 1 import os
paulo@40 2 import sys
paulo@39 3 import re
paulo@40 4 import urllib2
paulo@40 5 import threading
paulo@40 6 import Queue
paulo@41 7 import datetime
paulo@41 8 import time
paulo@42 9 import logging
paulo@42 10 logging.basicConfig(level=logging.INFO)
paulo@39 11
paulo@39 12 import html
paulo@39 13 import xml.etree.ElementTree
paulo@39 14
paulo@39 15
paulo@41 16 FEEDS_FILE = "FEEDS"
paulo@41 17 CACHE_HTML_FILE = "__cache__.html"
paulo@41 18
paulo@44 19 CACHE_LIFE = 1200 # [seconds]
paulo@39 20 MAX_ITEMS = 30
paulo@39 21 MAX_LINK_Z = 4
paulo@40 22 MAX_THREADS = 20
paulo@39 23
paulo@39 24
paulo@39 25 _PARSE_ROOT_TAG_RE = re.compile(r"(\{(.+)\})?(.+)")
paulo@39 26
paulo@39 27 def _parse_root_tag(root_tag):
paulo@39 28 re_match = _PARSE_ROOT_TAG_RE.match(root_tag)
paulo@39 29
paulo@39 30 if re_match is None:
paulo@39 31 return (None, None)
paulo@39 32 else:
paulo@39 33 return re_match.group(2, 3)
paulo@39 34
paulo@39 35
paulo@39 36 def _go_rss(elementTree):
paulo@39 37 title = elementTree.find("channel/title").text.strip()
paulo@39 38 link = elementTree.find("channel/link").text
paulo@39 39
paulo@39 40 items = []
paulo@39 41
paulo@39 42 for i in elementTree.findall("channel/item")[:MAX_ITEMS]:
paulo@39 43 it_title = i.find("title").text.strip()
paulo@39 44 it_link = i.find("link").text
paulo@39 45
paulo@39 46 items.append((it_title, it_link))
paulo@39 47
paulo@39 48 return (title, link, items)
paulo@39 49
paulo@39 50
paulo@39 51 def _go_atom(elementTree):
paulo@39 52 ns = "http://www.w3.org/2005/Atom"
paulo@39 53
paulo@39 54 title = elementTree.find("{%s}title" % ns).text.strip()
paulo@39 55 link = ''
paulo@39 56
paulo@39 57 for i in elementTree.findall("{%s}link" % ns):
paulo@39 58 if i.get("type") == "text/html" and i.get("rel") == "alternate":
paulo@39 59 link = i.get("href")
paulo@39 60 break
paulo@39 61
paulo@39 62 items = []
paulo@39 63
paulo@39 64 for i in elementTree.findall("{%s}entry" % ns)[:MAX_ITEMS]:
paulo@39 65 it_title = i.find("{%s}title" % ns).text.strip()
paulo@39 66 it_link = ''
paulo@39 67
paulo@39 68 for j in i.findall("{%s}link" % ns):
paulo@39 69 if j.get("type") == "text/html" and j.get("rel") == "alternate":
paulo@39 70 it_link = j.get("href")
paulo@39 71 break
paulo@39 72
paulo@39 73 items.append((it_title, it_link))
paulo@39 74
paulo@39 75 return (title, link, items)
paulo@39 76
paulo@39 77
paulo@41 78 def _to_html(dtnow, docstruct):
paulo@41 79 datetime_str = dtnow.strftime("%Y-%m-%d %H:%M %Z")
paulo@41 80 page_title = "myrss -- %s" % datetime_str
paulo@41 81
paulo@42 82 root = html.HTML("html")
paulo@39 83
paulo@39 84 header = root.header
paulo@41 85 header.title(page_title)
paulo@39 86 header.link(rel="stylesheet", type="text/css", href="index.css")
paulo@39 87
paulo@41 88 body = root.body
paulo@41 89 body.h1(page_title)
paulo@41 90
paulo@39 91 link_z = 0
paulo@39 92
paulo@39 93 for feed in docstruct:
paulo@40 94 if feed is None:
paulo@40 95 continue
paulo@40 96
paulo@39 97 (title, link, items) = feed
paulo@39 98
paulo@41 99 body.h2.a(title, href=link, klass="z%d" % (link_z % MAX_LINK_Z))
paulo@39 100 link_z += 1
paulo@41 101 p = body.p
paulo@39 102
paulo@39 103 for (i, (it_title, it_link)) in enumerate(items):
paulo@39 104 if i > 0:
paulo@39 105 p += " - "
paulo@39 106
paulo@39 107 p.a(it_title, href=it_link, klass="z%d" % (link_z % MAX_LINK_Z))
paulo@39 108 link_z += 1
paulo@39 109
paulo@39 110 return unicode(root).encode("utf-8")
paulo@39 111
paulo@39 112
paulo@40 113 def _process_url(url):
paulo@40 114 ret = None
paulo@40 115
paulo@40 116 try:
paulo@42 117 logging.info("processing %s" % url)
paulo@41 118 feed = urllib2.urlopen(urllib2.Request(url, headers={"User-Agent": ''}))
paulo@40 119 except urllib2.HTTPError as e:
paulo@42 120 logging.info("(%s) %s" % (url, e))
paulo@40 121 return ret
paulo@40 122
paulo@40 123 elementTree = xml.etree.ElementTree.parse(feed)
paulo@40 124 root = elementTree.getroot()
paulo@40 125
paulo@40 126 parsed_root_tag = _parse_root_tag(root.tag)
paulo@40 127
paulo@40 128 if parsed_root_tag == (None, "rss"):
paulo@40 129 version = float(root.get("version", 0.0))
paulo@40 130 if version >= 2.0:
paulo@40 131 ret = _go_rss(elementTree)
paulo@40 132 else:
paulo@40 133 raise NotImplementedError("Unsupported rss version")
paulo@40 134 elif parsed_root_tag == ("http://www.w3.org/2005/Atom", "feed"):
paulo@40 135 ret = _go_atom(elementTree)
paulo@40 136 else:
paulo@40 137 raise NotImplementedError("Unknown root tag")
paulo@40 138
paulo@40 139 return ret
paulo@40 140
paulo@40 141
paulo@40 142 class WorkerThread(threading.Thread):
paulo@40 143 def __init__(self, *args, **kwargs):
paulo@40 144 self._input_queue = kwargs.pop("input_queue")
paulo@40 145 self._output_queue = kwargs.pop("output_queue")
paulo@40 146 threading.Thread.__init__(self, *args, **kwargs)
paulo@40 147 self.daemon = True
paulo@40 148
paulo@40 149 def run(self):
paulo@40 150 while True:
paulo@40 151 (idx, url) = self._input_queue.get()
paulo@40 152 docfeed = None
paulo@40 153 try:
paulo@40 154 docfeed = _process_url(url)
paulo@40 155 except Exception as e:
paulo@42 156 logging.info("(%s) exception: %s" % (url, e))
paulo@40 157 self._output_queue.put((idx, docfeed))
paulo@40 158
paulo@40 159
paulo@44 160 def main(input_queue, output_queue, lock):
paulo@41 161 ret = ''
paulo@41 162
paulo@44 163 with lock:
paulo@44 164 epoch_now = time.time()
paulo@44 165 dtnow = datetime.datetime.fromtimestamp(epoch_now)
paulo@41 166
paulo@44 167 if os.path.exists(CACHE_HTML_FILE) and (epoch_now - os.stat(CACHE_HTML_FILE).st_mtime) < float(CACHE_LIFE):
paulo@44 168 with open(CACHE_HTML_FILE) as cache_html_file:
paulo@44 169 ret = cache_html_file.read()
paulo@41 170
paulo@44 171 else:
paulo@44 172 with open(FEEDS_FILE) as feeds_file:
paulo@44 173 feedlines = feeds_file.readlines()
paulo@41 174
paulo@44 175 docstruct = [None]*len(feedlines)
paulo@44 176 num_input = 0
paulo@44 177 for (i, l) in enumerate(feedlines):
paulo@44 178 if l[0] != '#':
paulo@44 179 l = l.strip()
paulo@44 180 input_queue.put((i, l))
paulo@44 181 num_input += 1
paulo@41 182
paulo@44 183 for _ in range(num_input):
paulo@44 184 (idx, docfeed) = output_queue.get()
paulo@44 185 docstruct[idx] = docfeed
paulo@41 186
paulo@44 187 ret = _to_html(dtnow, docstruct)
paulo@41 188
paulo@44 189 with open(CACHE_HTML_FILE, 'w') as cache_html_file:
paulo@44 190 cache_html_file.write(ret)
paulo@41 191
paulo@41 192 return ret
paulo@41 193
paulo@41 194
paulo@42 195 class MyRssApp:
paulo@42 196 def __init__(self):
paulo@42 197 self._iq = Queue.Queue(MAX_THREADS)
paulo@42 198 self._oq = Queue.Queue(MAX_THREADS)
paulo@44 199 self._main_lock = threading.Lock()
paulo@39 200
paulo@42 201 for _ in range(MAX_THREADS):
paulo@42 202 WorkerThread(input_queue=self._iq, output_queue=self._oq).start()
paulo@42 203
paulo@42 204 def __call__(self, environ, start_response):
paulo@44 205 response_body = main(self._iq, self._oq, self._main_lock)
paulo@42 206 response_headers = [
paulo@42 207 ("Content-Type", "text/html"),
paulo@42 208 ("Content-Length", str(len(response_body))),
paulo@42 209 ]
paulo@42 210 start_response("200 OK", response_headers)
paulo@42 211
paulo@42 212 return [response_body]