1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
|
#! /usr/local/bin/python
#-*- coding: utf-8 -*-
__author__ = "Cedric Bonhomme"
__version__ = "$Revision: 0.3 $"
__date__ = "$Date: 2010/03/10 $"
__copyright__ = "Copyright (c) 2010 Cedric Bonhomme"
__license__ = "GPLv3"
IMPORT_ERROR = []
import re
try:
import pylab
except:
IMPORT_ERROR.append("pylab")
import sqlite3
import hashlib
from datetime import datetime
from string import punctuation
from collections import defaultdict
from StringIO import StringIO
try:
from oice.langdet import langdet
from oice.langdet import streams
from oice.langdet import languages
except:
IMPORT_ERROR.append("oice")
import threading
LOCKER = threading.Lock()
def detect_language(text):
"""
Detect the language of a text.
English, French or other (not detected).
"""
text = text.strip()
try:
text_stream = streams.Stream(StringIO(text))
lang = langdet.LanguageDetector.detect(text_stream)
except:
return 'other'
if lang == languages.french:
return 'french'.encode('utf-8')
elif lang == languages.english:
return 'english'.encode('utf-8')
else:
return 'other'
def remove_html_tags(data):
"""
Remove HTML tags for the search.
"""
p = re.compile(r'<[^<]*?/?>')
return p.sub('', data)
def top_words(dic_articles, n=10):
"""
Return the n most frequent words in a list.
"""
words = {}
articles_content = ""
for rss_feed_id in dic_articles.keys():
for article in dic_articles[rss_feed_id]:
articles_content += remove_html_tags(article[4].encode('utf-8'))
words_gen = (word.strip(punctuation).lower() \
for word in articles_content.split() \
if len(word) >= 5)
words = defaultdict(int)
for word in words_gen:
words[word] += 1
top_words = sorted(words.iteritems(),
key=lambda(word, count): (-count, word))[:n]
return top_words
def create_histogram(words, file_name="./var/histogram.png"):
"""
Create a histogram.
"""
length = 10
ind = pylab.arange(length) # abscissa
width = 0.35 # bars width
w = [elem[0] for elem in words]
count = [int(elem[1]) for elem in words]
max_count = max(count) # maximal weight
p = pylab.bar(ind, count, width, color='r')
pylab.ylabel("Count")
pylab.title("Most frequent words")
pylab.xticks(ind + (width / 2), range(1, len(w)+1))
pylab.xlim(-width, len(ind))
# changing the ordinate scale according to the max.
if max_count <= 100:
pylab.ylim(0, max_count + 5)
pylab.yticks(pylab.arange(0, max_count + 5, 5))
elif max_count <= 200:
pylab.ylim(0, max_count + 10)
pylab.yticks(pylab.arange(0, max_count + 10, 10))
elif max_count <= 600:
pylab.ylim(0, max_count + 25)
pylab.yticks(pylab.arange(0, max_count + 25, 25))
elif max_count <= 800:
pylab.ylim(0, max_count + 50)
pylab.yticks(pylab.arange(0, max_count + 50, 50))
pylab.savefig(file_name, dpi = 80)
pylab.close()
def compare(stringtime1, stringtime2):
"""
Compare two dates in the format 'yyyy-mm-dd hh:mm:ss'.
"""
date1, time1 = stringtime1.split(' ')
date2, time2 = stringtime2.split(' ')
year1, month1, day1 = date1.split('-')
year2, month2, day2 = date2.split('-')
hour1, minute1, second1 = time1.split(':')
hour2, minute2, second2 = time2.split(':')
datetime1 = datetime(year=int(year1), month=int(month1), day=int(day1), \
hour=int(hour1), minute=int(minute1), second=int(second1))
datetime2 = datetime(year=int(year2), month=int(month2), day=int(day2), \
hour=int(hour2), minute=int(minute2), second=int(second2))
if datetime1 < datetime2:
return -1
elif datetime1 > datetime2:
return 1
return 0
def load_feed():
"""
Load feeds and articles in a dictionary.
"""
LOCKER.acquire()
list_of_feeds = []
list_of_articles = []
try:
conn = sqlite3.connect("./var/feed.db", isolation_level = None)
c = conn.cursor()
list_of_feeds = c.execute("SELECT * FROM feeds").fetchall()
except:
pass
# articles[feed_id] = (article_id, article_date, article_title,
# article_link, article_description, article_readed,
# article_language)
# feeds[feed_id] = (nb_article, nb_article_unreaded, feed_image,
# feed_title, feed_link, feed_site_link)
articles, feeds = {}, {}
if list_of_feeds != []:
for feed in list_of_feeds:
list_of_articles = c.execute(\
"SELECT * FROM articles WHERE feed_link='" + \
feed[2] + "'").fetchall()
if list_of_articles != []:
for article in list_of_articles:
sha1_hash = hashlib.sha1()
sha1_hash.update(article[5].encode('utf-8'))
feed_id = sha1_hash.hexdigest()
sha1_hash.update(article[2].encode('utf-8'))
article_id = sha1_hash.hexdigest()
if "oice" not in IMPORT_ERROR:
if article[3] != "":
language = detect_language(remove_html_tags(article[3][:80]).encode('utf-8') + \
remove_html_tags(article[1]).encode('utf-8'))
else:
language = detect_language(remove_html_tags(article[1]).encode('utf-8'))
else:
language = "IMPORT_ERROR"
article_list = [article_id, article[0], article[1], \
article[2], article[3], article[4], language]
if feed_id not in articles:
articles[feed_id] = [article_list]
else:
articles[feed_id].append(article_list)
# sort articles by date for each feeds
for rss_feed_id in articles.keys():
articles[rss_feed_id].sort(lambda x,y: compare(y[1], x[1]))
feeds[feed_id] = (len(articles[feed_id]), \
len([article for article in articles[feed_id] \
if article[5]=="0"]), \
feed[3], feed[0], feed[2], feed[1] \
)
c.close()
LOCKER.release()
return (articles, feeds)
return (articles, feeds)
|
