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 publish(src,target):
  38     cmd = ["rsync","-a",src,target]
  39     retcode = subprocess.call(cmd)
  40     if retcode:
  41         print 'Error: '+' '.join(cmd)+' Returncode ['+str(retcode)+']'
  42
  43 def generateSitemap():
  44   sitemap = []
  45   try:
  46     sfile = open('sitemap.txt')
  47     flist = sfile.read().split()
  48     sfile.close()
  49     for f in flist:
  50       sitemap.append(dict(link=f))
  51   except IOError, what_error:
  52     print 'Sitemap missing - generating one.'
  53   for dirname, dirnames, filenames in os.walk('.'):
  54     for filename in filenames:
  55       if fnmatch.fnmatch(filename, '*.xml'):
  56         xfile = os.path.join(dirname,filename)
  57         doc = bindery.parse(xfile,
  58                             prefixes={u'db': u'http://docbook.org/ns/docbook',
  59                                       u'xi': u'http://www.w3.org/2001/XInclude'})
  60         title = doc.xml_select(u'/db:article/db:info/db:title')
  61         menu  = doc.xml_select(u'/db:article/db:info/db:titleabbrev')
  62         code  = doc.xml_select(u"//xi:include[@parse='text']")
  63         exe = 0
  64         for c in code:
  65           (p, ext) = os.path.splitext(c.href)
  66           if ext in valid_scripts:
  67             exe = 1
  68
  69         if title and menu:
  70           found = 0
  71           base = xfile.split('.')[1]
  72           link = base.replace('index','')
  73           level = len(filter(None,re.split(r'(/\w*/)',link)))
  74           page = dict(title=unicode(doc.article.info.title),
  75                       menu=unicode(doc.article.info.titleabbrev),
  76                       output=os.path.join(dirname,
  77                                           filename.replace('xml','html')),
  78                       exe=exe,
  79                       file=xfile,
  80                       level=level)
  81           for l in sitemap:
  82             if l['link'] == link:
  83               found = 1
  84               l.update(page)
  85           if not found:
  86             print "adding "+link+" to sitemap"
  87             dd = dict(link=link)
  88             dd.update(page)
  89             sitemap.append(dd)
  90   sfile = open('sitemap.txt','w')
  91   for l in sitemap:
  92     sfile.write(l['link']+'\n')
  93   sfile.close()
  94   return sitemap
  95
  96 def expandXincludeTxt(page):
  97   doc = bindery.parse(page['file'],
  98                       prefixes={u'db': u'http://docbook.org/ns/docbook',
  99                                 u'xi': u'http://www.w3.org/2001/XInclude'})
 100   if page['exe']:
 101     code  = doc.xml_select(u"//xi:include[@parse='text']")
 102     for c in code:
 103       (p, ext) = os.path.splitext(c.href)
 104       if ext in valid_scripts:
 105         exe = os.path.join(os.path.abspath(c.href))
 106         xml = subprocess.Popen([exe],stdout=subprocess.PIPE)
 107         xstr = bindery.parse(str(xml.stdout.read()))
 108         id = c.xml_index_on_parent
 109         for x in xstr.xml_children:
 110           c.xml_parent.xml_insert(id,x)
 111         c.xml_parent.xml_remove(c)
 112   return doc
 113
 114 def xsltConvert(doc):
 115 #  amara can not handle the docbook stylesheets
 116 #  xmlarticle = transform(doc,style_xslt)
 117   cwd = os.getcwd()
 118   rundir = os.path.dirname(page['file'])
 119   os.chdir(rundir)
 120   infile  = os.path.basename(tempfile.mktemp())
 121   outfile = tempfile.mktemp()
 122   tfi = open(infile,'w')
 123   tfi.write(doc.xml_encode())
 124   tfi.close()
 125 #  cmd = ["saxon-xslt-xinclude","-o",outfile,infile,style_xslt]
 126   cmd = ["xsltproc","--xinclude","--output",outfile,style_xslt,infile]
 127   retcode = subprocess.call(cmd)
 128   if retcode:
 129     print 'Error: '+' '.join(cmd)+' Returncode ['+str(retcode)+']'
 130   tfo = open(outfile,'r')
 131   result = tfo.read()
 132   tfo.close()
 133   os.remove(infile)
 134   os.remove(outfile)
 135   os.chdir(cwd)
 136   return result
 137
 138 def genMenu(page,sitemap,slevel,elevel):
 139   title = None
 140   sm = []
 141   if elevel == MAXLEVEL or elevel == 1:
 142     sm = sitemap
 143   else:
 144     idx = sitemap.index(page)
 145     while (sitemap[idx]['level'] == page['level']):
 146       idx = idx-1
 147     title = sitemap[idx]['menu']
 148     idx = idx+1
 149     while (idx < len(sitemap) and sitemap[idx]['level'] == page['level']):
 150       sm.append(sitemap[idx])
 151       idx = idx+1
 152   oldlevel = slevel
 153   html = '<ul>\n'
 154   for p in sm:
 155     if slevel > p['level'] or elevel < p['level']:
 156       continue
 157     if not title and p['link'] == '/':
 158       title = p['menu']
 159
 160     if oldlevel < p['level']:
 161       html+='<ul>\n'
 162     elif oldlevel > p['level']:
 163       if p['link'][-1] == '/':
 164         html+='</li>\n'
 165       html+='</ul>\n</li>\n'
 166     if page == p:
 167       html+='<li><a href="%s">[%s]</a>' % (p['link'],p['menu'])
 168     else:
 169       html+='<li><a href="%s">%s</a>' % (p['link'],p['menu'])
 170     if p['link'][-1] != '/' or p['link'] == '/':
 171         html+='</li>\n'
 172     oldlevel = p['level']
 173   html+='</ul>\n'
 174   return (html,title)
 175
 176 def writeToTemplate(page,doc,sitemap):
 177   (menu,menuname) = genMenu(page,sitemap,1,MAXLEVEL)
 178   (levelmenu,levelname) = genMenu(page,sitemap,page['level'],page['level'])
 179   template = Template(file=style_tmpl,
 180                       searchList=[{'title':page['title']},
 181                                   {'menu':menu},
 182                                   {'article':doc},
 183                                   {'levelmenu':levelmenu},
 184                                   {'levelname':levelname}])
 185   outfile = args.output+page['output']
 186   out = open('result', 'w')
 187   out.write(str(template))
 188   out.close()
 189   publish('result',outfile)
 190   os.remove('result')
 191
 192 sitemap = generateSitemap()
 193 for page in sitemap:
 194   t1 = time.time()
 195   print "Page : %-30s %30s" % (page['link'],
 196                       time.ctime(os.stat(page['file']).st_mtime)),
 197   doc = expandXincludeTxt(page)
 198   pubdoc = xsltConvert(doc)
 199   writeToTemplate(page,pubdoc,sitemap)
 200   publish(args.style+"css", args.output)
 201   publish(args.style+"images",args.output)
 202   t2 = time.time()
 203   print "[%5.2f s]" % (round(t2-t1,2))