10 from lxml.builder import ElementMaker
12 from treecutter import constants as const
13 from treecutter.trie import Trie
14 from treecutter.link import Link
15 from treecutter.tools import ssh_cmd, publish, mkdir_p
18 """Class keeping the internal site structure"""
19 def __init__(self,args):
20 self._output = args.output
21 self._style = args.style
22 self._subdir = args.subdir
23 self._file = 'sitemap.txt'
25 self._sitelang = set()
26 self._isocode = etree.parse('/usr/share/xml/iso-codes/iso_639_3.xml')
28 self._tmptarget = tempfile.mkdtemp()+'/'
30 # The sitemap uses a trie structure to keep track of links
31 # A link represents the path to the document and the link
32 # representing the text on the site.
33 # A link can have several pages in different languages.
34 def add_link(self, link):
35 tokens = filter(None,re.split(r'(^/[\w\.:-]*$|^/[\w\.:-]*/|[\w\.:-]*/)',link,flags=re.UNICODE))
36 self._tree.add(tokens,Link(link))
39 f = codecs.open(self._file,'w','utf-8')
40 s = '\n'.join(link.link() for link in self._tree)
46 f = codecs.open(self._file, 'r', 'utf-8')
47 sml = f.read().split()
51 except IOError, what_error:
52 print 'INFO: Could not read sitemap.txt - one will be created'
54 # Create a set of the current tree for comparison with the
57 return set(link.link() for link in self._tree)
59 # Main driver in the application processing the documents
60 # in the collected sitemap
63 print "Prepareing the input"
64 for link in self._tree:
67 print "Prepare [%5.2f s]" % (round(t2-t1,2))
68 for link in self._tree:
69 self._sitelang = self._sitelang.union(set(link.languages()))
70 for tran in self._sitelang:
72 self._tranlang[tran] = gettext.translation('iso_639_3',
75 print "Language [%5.2f s]" % (round(t3-t2,2))
77 transform['xhtml5'] = etree.XSLT(etree.parse(self._style+"docbook.xhtml5.xsl"))
78 for link in self._tree:
79 link.render(transform)
81 print "Render [%5.2f s]" % (round(t4-t3,2))
82 for link in self._tree:
83 link.template(self, self._style, self._tmptarget,self._subdir)
85 print "Template [%5.2f s]" % (round(t5-t4,2))
88 # Collect all files used by the documents
89 for link in self._tree:
90 res = res.union(link.resources())
92 outfile = self._tmptarget+f
93 mkdir_p(os.path.dirname(outfile))
94 shutil.copyfile(f,outfile)
95 print "Resources[%5.2f s]" % (round(t6-t5,2))
96 # TODO: Improve the sitemap, it is a page that is generated from
97 # the ground up and added a bit adhoc.
98 sitmaplink = Link('/sitemap')
99 for l in self._sitelang:
100 sitmaplink.add_page((l,'/sitemap.'+l+'.xml'))
101 for l in self._sitelang:
102 sitmaplink.page(l).set_article(self.gen_menu(l,None,"tree sitemap"))
103 sitmaplink.page(l).template(self,self._style,self._tmptarget,self._subdir)
105 print "Sitemap [%5.2f s]" % (round(t7-t6,2))
110 def gen_menu(self,lang,page,cssclass):
111 return self._tree.menu(lang,page,cssclass,self._subdir)
113 def lang_menu(self,lang,link):
114 html = ElementMaker()
116 for l in link.languages():
117 isoxml = u"//iso_639_3_entry[@*='"+l+"']"
118 ln = self._isocode.xpath(isoxml)[0].get('name')
120 ln = self._tranlang[lang].gettext(ln)
121 p = unicode(link.link())
125 li = html.li(html.a(ln.decode('utf-8'),
126 href=self._subdir+p,hreflang=l))
128 return etree.tostring(menu,encoding='UTF-8',pretty_print=False)
131 ssh_cmd(self._output,"mkdir -p")
132 publish(self._tmptarget, self._output)
133 for res in ["stylesheets","images","js","fonts","favicon.ico"]:
134 if (os.path.exists(self._style+res)):
135 publish(self._style+res, self._output)
136 ssh_cmd(self._output,"chmod a+rx")