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,filename.replace('xml','html')),
  71                       exe=exe,
  72                       file=xfile,
  73                       level=level)
  74           for l in sitemap:
  75             if l['link'] == link:
  76               found = 1
  77               l.update(page)
  78           if not found:
  79             print "adding "+link+" to sitemap"
  80             dd = dict(link=link)
  81             dd.update(page)
  82             sitemap.append(dd)
  83   sfile = open('sitemap.txt','w')
  84   for l in sitemap:
  85     sfile.write(l['link']+'\n')
  86   sfile.close()
  87   return sitemap
  88
  89 def expandXincludeTxt(page):
  90   doc = bindery.parse(page['file'],prefixes={u'db': u'http://docbook.org/ns/docbook',
  91                                              u'xi': u'http://www.w3.org/2001/XInclude'})
  92   if page['exe']:
  93     code  = doc.xml_select(u"//xi:include[@parse='text']")
  94     for c in code:
  95       (p, ext) = os.path.splitext(c.href)
  96       if ext in valid_scripts:
  97         exe = os.path.join(os.path.abspath(c.href))
  98         xml = subprocess.Popen([exe],stdout=subprocess.PIPE)
  99         xstr = bindery.parse(str(xml.stdout.read()))
 100         id = c.xml_index_on_parent
 101         for x in xstr.xml_children:
 102           c.xml_parent.xml_insert(id,x)
 103         c.xml_parent.xml_remove(c)
 104   return doc
 105
 106 def xsltConvert(doc):
 107 #  amara can not handle the docbook stylesheets
 108 #  xmlarticle = transform(doc,style_xslt)
 109   cwd = os.getcwd()
 110   rundir = os.path.dirname(page['file'])
 111   os.chdir(rundir)
 112   infile  = os.path.basename(tempfile.mktemp())
 113   outfile = tempfile.mktemp()
 114   tfi = open(infile,'w')
 115   tfi.write(doc.xml_encode())
 116   tfi.close()
 117 #  cmd = ["saxon-xslt-xinclude","-o",outfile,infile,style_xslt]
 118   cmd = ["xsltproc","--xinclude","--output",outfile,style_xslt,infile]
 119   retcode = subprocess.call(cmd)
 120   if retcode:
 121     print 'Error: '+' '.join(cmd)+' Returncode ['+str(retcode)+']'
 122   tfo = open(outfile,'r')
 123   result = tfo.read()
 124   tfo.close()
 125   os.remove(infile)
 126   os.remove(outfile)
 127   os.chdir(cwd)
 128   return result
 129
 130 def genMenu(page,sitemap,slevel,elevel):
 131   title = None
 132   sm = []
 133   if elevel == MAXLEVEL or elevel == 1:
 134     sm = sitemap
 135   else:
 136     idx = sitemap.index(page)
 137     while (sitemap[idx]['level'] == page['level']):
 138       idx = idx-1
 139     title = sitemap[idx]['menu']
 140     idx = idx+1
 141     while (idx < len(sitemap) and sitemap[idx]['level'] == page['level']):
 142       sm.append(sitemap[idx])
 143       idx = idx+1
 144   oldlevel = slevel
 145   html = '<ul>\n'
 146   for p in sm:
 147     if slevel > p['level'] or elevel < p['level']:
 148       continue
 149     if not title and p['link'] == '/':
 150       title = p['menu']
 151
 152     if oldlevel < p['level']:
 153       html+='<ul>\n'
 154     elif oldlevel > p['level']:
 155       if p['link'][-1] == '/':
 156         html+='</li>\n'
 157       html+='</ul>\n</li>\n'
 158     if page == p:
 159       html+='<li><a href="%s">[%s]</a>' % (p['link'],p['menu'])
 160     else:
 161       html+='<li><a href="%s">%s</a>' % (p['link'],p['menu'])
 162     if p['link'][-1] != '/' or p['link'] == '/':
 163         html+='</li>\n'
 164     oldlevel = p['level']
 165   html+='</ul>\n'
 166   return (html,title)
 167
 168 def writeToTemplate(page,doc,sitemap):
 169   (menu,menuname) = genMenu(page,sitemap,1,MAXLEVEL)
 170   (levelmenu,levelname) = genMenu(page,sitemap,page['level'],page['level'])
 171   template = Template(file=style_tmpl,
 172                       searchList=[{'menu':menu},
 173                                   {'article':doc},
 174                                   {'levelmenu':levelmenu},
 175                                   {'levelname':levelname}])
 176   outfile = outputdir+page['output']
 177   d = os.path.split(outfile)[0]
 178   if d != '':
 179     mkdir_p(d)
 180   out = open(outfile, 'w')
 181   out.write(str(template))
 182
 183 sitemap = generateSitemap()
 184 for page in sitemap:
 185   t1 = time.time()
 186   print "Page : "+page['link'],
 187   doc = expandXincludeTxt(page)
 188   pubdoc = xsltConvert(doc)
 189   writeToTemplate(page,pubdoc,sitemap)
 190 #  publishResources()
 191   t2 = time.time()
 192   print "["+str(round(t2-t1,2))+"]  done."