src/tree-cutter.py

   1 #!/usr/bin/python
   2 import os
   3 import fnmatch
   4 import subprocess
   5 import amara
   6 import re
   7 import tempfile
   8 import errno
   9 import time
  10 from amara import bindery
  11 from amara.xslt import transform
  12 from Cheetah.Template import Template
  13
  14 dist = "."
  15 style = "default"
  16 style_xslt = dist+"/style/"+style+"/docbook.xsl"
  17 style_tmpl = dist+"/style/"+style+"/index.html.tmpl"
  18 outputdir = dist+"/htdocs/"
  19
  20 valid_scripts = ['.py','.pl']
  21 MAXLEVEL = 10000
  22
  23 def mkdir_p(path):
  24     try:
  25         os.makedirs(path)
  26     except OSError as exc: # Python >2.5
  27         if exc.errno == errno.EEXIST:
  28             pass
  29         else: raise
  30
  31 def generateSitemap():
  32   sitemap = []
  33   try:
  34     sfile = open('sitemap.txt')
  35     flist = sfile.read().split()
  36     sfile.close()
  37     for f in flist:
  38       sitemap.append(dict(link=f))
  39   except IOError, what_error:
  40     print 'Sitemap missing - generating one.'
  41   for dirname, dirnames, filenames in os.walk('.'):
  42     for filename in filenames:
  43       if fnmatch.fnmatch(filename, '*.xml'):
  44         xfile = os.path.join(dirname,filename)
  45         doc = bindery.parse(xfile,
  46                             prefixes={u'db': u'http://docbook.org/ns/docbook',
  47                                       u'xi': u'http://www.w3.org/2001/XInclude'})
  48         title = doc.xml_select(u'/db:article/db:info/db:title')
  49         menu  = doc.xml_select(u'/db:article/db:info/db:titleabbrev')
  50         code  = doc.xml_select(u"//xi:include[@parse='text']")
  51         exe = 0
  52         for c in code:
  53           (p, ext) = os.path.splitext(c.href)
  54           if ext in valid_scripts:
  55             exe = 1
  56
  57         if title and menu:
  58           found = 0
  59           base = os.path.splitext(xfile)[0]
  60           link = base.translate(None,'.').replace('index','')
  61           level = len(filter(None,re.split(r'(/\w*/)',link)))
  62           page = dict(title=unicode(doc.article.info.title),
  63                       menu=unicode(doc.article.info.titleabbrev),
  64                       output=os.path.join(dirname,filename.replace('xml','html')),
  65                       exe=exe,
  66                       file=xfile,
  67                       level=level)
  68           for l in sitemap:
  69             if l['link'] == link:
  70               found = 1
  71               l.update(page)
  72           if not found:
  73             print "adding "+link+" to sitemap"
  74             dd = dict(link=link)
  75             dd.update(page)
  76             sitemap.append(dd)
  77   sfile = open('sitemap.txt','w')
  78   for l in sitemap:
  79     sfile.write(l['link']+'\n')
  80   sfile.close()
  81   return sitemap
  82
  83 def expandXincludeTxt(page):
  84   doc = bindery.parse(page['file'],prefixes={u'db': u'http://docbook.org/ns/docbook',
  85                                              u'xi': u'http://www.w3.org/2001/XInclude'})
  86   if page['exe']:
  87     code  = doc.xml_select(u"//xi:include[@parse='text']")
  88     for c in code:
  89       (p, ext) = os.path.splitext(c.href)
  90       if ext in valid_scripts:
  91         exe = os.path.join(os.path.abspath(c.href))
  92         xml = subprocess.Popen([exe],stdout=subprocess.PIPE)
  93         xstr = bindery.parse(str(xml.stdout.read()))
  94         id = c.xml_index_on_parent
  95         for x in xstr.xml_children:
  96           c.xml_parent.xml_insert(id,x)
  97         c.xml_parent.xml_remove(c)
  98   return doc
  99
 100 def xsltConvert(doc):
 101 #  amara can not handle the docbook stylesheets
 102 #  xmlarticle = transform(doc,style_xslt)
 103   cwd = os.getcwd()
 104   rundir = os.path.dirname(page['file'])
 105   os.chdir(rundir)
 106   infile  = os.path.basename(tempfile.mktemp())
 107   outfile = tempfile.mktemp()
 108   tfi = open(infile,'w')
 109   tfi.write(doc.xml_encode())
 110   tfi.close()
 111   cmd = ["saxon-xslt-xinclude","-o",outfile,infile,style_xslt]
 112   retcode = subprocess.call(cmd)
 113   if retcode:
 114     print 'Error: '+' '.join(cmd)+' Returncode ['+str(retcode)+']'
 115   tfo = open(outfile,'r')
 116   result = tfo.read()
 117   tfo.close()
 118   os.remove(infile)
 119   os.remove(outfile)
 120   os.chdir(cwd)
 121   return result
 122
 123 def genMenu(page,sitemap,slevel,elevel):
 124   title = None
 125   sm = []
 126   if elevel == MAXLEVEL or elevel == 1:
 127     sm = sitemap
 128   else:
 129     idx = sitemap.index(page)
 130     while (sitemap[idx]['level'] == page['level']):
 131       idx = idx-1
 132     title = sitemap[idx]['menu']
 133     idx = idx+1
 134     while (sitemap[idx]['level'] == page['level']):
 135       sm.append(sitemap[idx])
 136       idx = idx+1
 137   oldlevel = slevel
 138
 139   html = '<ul>\n'
 140   for p in sm:
 141     if slevel > p['level'] or elevel < p['level']:
 142       continue
 143     if not title and p['link'] == '/':
 144       title = p['menu']
 145
 146     if oldlevel < p['level']:
 147       html+='<ul>\n'
 148     elif oldlevel > p['level']:
 149       if p['link'][-1] == '/':
 150         html+='</li>\n'
 151       html+='</ul>\n</li>\n'
 152     if page == p:
 153       html+='<li><a href="%s">[%s]</a>' % (p['link'],p['menu'])
 154     else:
 155       html+='<li><a href="%s">%s</a>' % (p['link'],p['menu'])
 156     if p['link'][-1] != '/' or p['link'] == '/':
 157         html+='</li>\n'
 158     oldlevel = p['level']
 159   html+='</ul>\n'
 160   return (html,title)
 161
 162 def writeToTemplate(page,doc,sitemap):
 163   (menu,menuname) = genMenu(page,sitemap,1,MAXLEVEL)
 164   (levelmenu,levelname) = genMenu(page,sitemap,page['level'],page['level'])
 165   template = Template(file=style_tmpl,
 166                       searchList=[{'menu':menu},
 167                                   {'article':doc},
 168                                   {'levelmenu':levelmenu},
 169                                   {'levelname':levelname}])
 170   outfile = outputdir+page['output']
 171   d = os.path.split(outfile)[0]
 172   if d != '':
 173     mkdir_p(d)
 174   out = open(outfile, 'w')
 175   out.write(str(template))
 176
 177 sitemap = generateSitemap()
 178 for page in sitemap:
 179   t1 = time.time()
 180   print "Page : "+page['link'],
 181   doc = expandXincludeTxt(page)
 182   pubdoc = xsltConvert(doc)
 183   writeToTemplate(page,pubdoc,sitemap)
 184 #  publishResources()
 185   t2 = time.time()
 186   print "["+str(round(t2-t1,2))+"]  done."
 187
 188