将oschina导出的html转化为 rss文件

convertToRss.py

#coding:utf8
#将oschina导出的html 转化为rss格式
import re, sys, codecs, types
import htmlentitydefs
import urlparse
import HTMLParser
import urllib

rssheader = '''<?xml version="1.0" encoding="utf-8" ?>
<?xml-stylesheet type="text/xsl" title="XSL Formatting" href="/rss.xsl" media="all" ?>
<rss version="2.0">

%s
</rss>
'''

channel = '''
<channel>
    <title>liyonghelpme</title>
    <description>liyonghelpme</description>
    <language>zh-cn</language>
    <ttl>5</ttl>
    <copyright><![CDATA[Copyright &copy; liyonghelpme]]></copyright> 
    <pubDate>2015/2/4 23:13:49</pubDate>

%s
</channel>
'''

item = '''
<item>
<title><![CDATA[%s]]></title>
<author>liyonghelpme</author>
<pubDate>%s</pubDate>
<description>
<![CDATA[
%s
]]>
</description>
</item>
'''
 
saveRss = open('rss.xml', 'w')

pat = re.compile("<div class='content'>(.*?)</div>", re.MULTILINE|re.DOTALL)

class HTML2RSS(HTMLParser.HTMLParser):
    def __init__(self, out=None, baseurl=''):
        HTMLParser.HTMLParser.__init__(self)
        self.outrss = []
        
        self.htmlData = ''
        
        self.state = 0 #not read blog
        self.title = ''
        self.time = 0
        self.content = ''
        self.divNest = 0
        self.conStartPos = -1
        self.conEndPos = -1
        
    def feed(self, data):
        #data = data.replace("</' + 'script>", "</ignore>")
        self.htmlData = data
        HTMLParser.HTMLParser.feed(self, data)
    def handle_starttag(self, tag, attrs, startPos = -1):
        print 'start tag', tag, attrs
        if tag == 'div' and len(attrs) == 1:
            if len(attrs[0]) == 2:
                firstAttr = attrs[0]
                if attrs[0][0] == 'class' and attrs[0][1] == 'blog' and self.state == 0:
                    self.state = 1#begin blog
                elif firstAttr[0] == 'class' and firstAttr[1] == 'date' and self.state == 3:
                    self.state = 4#begin read date
                elif firstAttr[0] == 'class' and firstAttr[1] == 'content' and self.state == 5:
                    self.state = 6
                    #self.content = ''
                    self.content = self.get_starttag_text()
                    self.conStartPos = startPos
                    self.divNest = 0

        if tag == 'div' and self.state == 6:
            self.divNest += 1
                    
        if tag == 'a' and len(attrs) == 2 and self.state == 1:
            if attrs[0][0] == 'name' and attrs[0][1].find('blog') != -1:
                self.state = 2 #read title
            
    def handle_endtag(self, tag, endPos=-1):
        print 'end tag', tag
        if self.state == 2:
            self.state = 3#read time
        if self.state == 4:
            self.state = 5 #read content
        if self.state == 6 and tag == 'div':
            self.divNest -= 1

            if self.divNest == 0:
                self.conEndPos = endPos
                #self.content = self.get_starttag_text()
                print self.conStartPos
                print self.conEndPos
                self.content = self.htmlData[self.conStartPos:self.conEndPos]

                res = pat.findall(self.content)
                print 'read blog '
                self.state = 0
                #saveRss.write('read blog '+'\n\n')
                #saveRss.write(self.title+'\n')
                #saveRss.write(self.time+'\n')
                #saveRss.write(self.content+'\n\n')
                print self.content
                rss = [self.title, self.time, res[0]]
                self.outrss.append(rss)
             

    def handle_data(self, data):
        print 'some data', data
        if self.state == 2:
            self.title = data
        if self.state == 4:
            self.time = data
        if self.state == 6:
            pass

            


parser = HTML2RSS()
f = open('blogs_20150204.html').read()

parser.feed(f)

#aBlog = re.compile("<div class='blog'>.*</div>")
saveRss.close()


#allContent = []

#print 'allconent '
#for r in res:
#    print r

#print 'content len', len(res)
#print len(parser.outrss)

items = []
mpa = dict.fromkeys(range(32))
a = ''.join(chr(n) for n in range(32))

for r in parser.outrss:
    r[1] = r[1].replace('时间：', '').replace('-0', '/').replace('-', '/')
    print 'r2'
    print r[2]
    # r[2] = r[2].translate(mpa)
    r[2] = re.sub("[%s]" % (a), '', r[2])
    print r[2]

    items.append(item % (r[0], r[1], r[2]))
    #break

it = str.join('', items)
ch = channel % it



full = rssheader % (ch)
f = open('out.xml', 'w')
f.write(full)
f.close()