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