treecutter/page.py

   1 #!/usr/bin/python
   2 import os
   3 import subprocess
   4 import tempfile
   5 import re
   6 from amara import bindery
   7 from amara.xslt import transform
   8 from Cheetah.Template import Template
   9 from pkg_resources import resource_filename, resource_listdir
  10 from time import time
  11 import treecutter.constants as const
  12 from treecutter.tools import mkdir_p
  13
  14 class Page():
  15     """Class representing a version of a webpage"""
  16     def __init__(self,link,page):
  17         self._link = link
  18         self._file = page[1]
  19         self._lang = page[0]
  20         self._doc = None
  21         self._resources = []
  22         self._title = None
  23         self._menu = None
  24         self._rendered_article = None
  25
  26     def language(self):
  27         return self._lang
  28
  29     def resources(self):
  30         return set(self._resources)
  31
  32     def menu(self):
  33         return self._menu
  34
  35     def set_article(self,art):
  36         self._rendered_article = art
  37
  38     def prepare(self):
  39         self._doc = bindery.parse(self._file, prefixes=const.PREFIXES)
  40         if self._doc.xml_select(u'/db:article/db:info/db:title'):
  41             self._title = unicode(self._doc.article.info.title)
  42         if self._doc.xml_select(u'/db:article/db:info/db:titleabbrev'):
  43             self._menu = unicode(self._doc.article.info.titleabbrev)
  44
  45         dirname = os.path.dirname(self._file)
  46         cwd = os.getcwd()
  47         code  = self._doc.xml_select(u"//xi:include[@parse='text']")
  48         if code:
  49             for c in code:
  50                 (p, ext) = os.path.splitext(c.href)
  51                 if ext in const.valid_scripts:
  52                     exe = []
  53                     script = os.path.join(os.path.abspath(dirname)+'/'+c.href)
  54                     if os.path.isfile(script):
  55                         exe.append(script)
  56                     else:
  57                         if c.href in resource_listdir('xinclude', ''):
  58                             script = resource_filename('xinclude', c.href)
  59                             exe.append(script)
  60                         else:
  61                             print "Script "+c.href+" in "+self._file+" missing"
  62                     if c.xml_select(u"//xi:include[@accept-language]"):
  63                         alang = c.xml_attributes[None, "accept-language"]
  64                         exe.append("lang="+alang)
  65                     if c.xml_select(u"//xi:include[@xpointer]"):
  66                         exe.append("xptr="+c.xpointer)
  67                     print "  executing %15s" % (c.href),
  68                     ts = time()
  69                     os.chdir(dirname)
  70                     xml = subprocess.Popen(exe,stdout=subprocess.PIPE)
  71                     os.chdir(cwd)
  72                     xmlblock = str(xml.stdout.read())
  73                     te = time()
  74                     print " [%5.2f s]  (%s)" % (round(te-ts,2),c.xpointer)
  75                     xstr = bindery.parse(xmlblock)
  76                     idp = c.xml_index_on_parent
  77                     for x in xstr.xml_children:
  78                         c.xml_parent.xml_insert(idp,x)
  79                         c.xml_parent.xml_remove(c)
  80
  81         for r in self._doc.xml_select(u"//db:link[@xl:href]"):
  82             rf = os.path.join(dirname,r.href)
  83             if os.path.isfile(rf):
  84                 self._resources.append(rf)
  85         for i in self._doc.xml_select(u"//db:imagedata[@fileref]"):
  86             im = os.path.join(dirname,i.fileref)
  87             if os.path.isfile(im):
  88                 self._resources.append(im)
  89         for i in self._doc.xml_select(u"//html:form[@action]"):
  90             pyscript = re.split('\.py',i.action,1)[0]+'.py'
  91             im = os.path.join(dirname,pyscript)
  92             if os.path.isfile(im):
  93                 self._resources.append(im)
  94
  95     def render(self, style):
  96         #  amara can not handle the docbook stylesheets
  97         #  xmlarticle = transform(doc,style_xslt)
  98         cwd = os.getcwd()
  99         dirname = os.path.dirname(self._file)
 100         os.chdir(dirname)
 101         infile  = os.path.basename(tempfile.mktemp())
 102         outfile = tempfile.mktemp()
 103         tfi = open(infile,'w')
 104         tfi.write(self._doc.xml_encode(omit_xml_declaration=True))
 105         tfi.close()
 106 #  cmd = ["saxon-xslt-xinclude","-o",outfile,infile,style_xslt]
 107         cmd = ["xsltproc","--xinclude","--output",outfile,style+"docbook.xsl",infile]
 108         retcode = subprocess.call(cmd)
 109         if retcode:
 110             print 'Error: '+' '.join(cmd)+' Returncode ['+str(retcode)+']'
 111         tfo = open(outfile,'r')
 112         self._rendered_article = tfo.read()
 113         tfo.close()
 114         os.remove(infile)
 115         os.remove(outfile)
 116         os.chdir(cwd)
 117
 118     def template(self,sitemap,style,tdir):
 119         htmlmenu =  sitemap.gen_menu(self._lang,None,"menu")
 120         levelmenu = sitemap.gen_menu(self._lang,self,"tree")
 121         langmenu = sitemap.lang_menu(self._lang,self._link)
 122         template = Template(file=style+'index.'+self._lang+'.html.tmpl',
 123                             searchList=[{'title':self._title},
 124                                         {'menu':htmlmenu},
 125                                         {'article':self._rendered_article},
 126                                         {'levelmenu':levelmenu},
 127                                         {'langmenu':langmenu}])
 128         outfile = tdir+'html'.join(self._file.rsplit('xml',1))
 129         mkdir_p(os.path.dirname(outfile))
 130         out = open(outfile, 'w')
 131         out.write(str(template))
 132         out.close()