hpastetwo

New Revision

#!/usr/bin/env python

# Horribly hacked-together script to generate an HTML index page for
# Haddock-generated Haskell package docs.  Note that the docs
# directory is hard-coded, below: see the comments there.

import os
import re
import StringIO

# This is the path to the docs.  Any subdirectory which contains an
# 'html/index.html' will get an entry in the index (all the other are
# listed separately at the end).

# The index is written to 'index.html' at this path.

PATH = "/usr/local/share/doc"

class Package:

def __init__(self, name, path):
        self.name = name
        self.path = path
        self.title = self._seekTitle()

def __cmp__(self, other):
        return cmp(self.name.lower(), other.name.lower())

def index(self):
        return os.path.join(self.path, 'html', 'index.html')

titleRE = r'>.*: (.*)</TITLE'

def _seekTitle(self):
        f = open(self.index())
        d = f.read()
        f.close()
        m = re.search(self.titleRE, d)
        if not m:
            raise ValueError, "Can't extract title from %s" % self.index()
        return m.group(1)

class Builder:

def __init__(self, path):
        self.path = path
        libs = [l for l in os.listdir(self.path)
                if os.path.isdir(os.path.join(path, l))]
        self.has_html = []
        self.no_html = []
        for lib in libs:
            full = os.path.join(path, lib)
            if self.seek_html(full):
                self.has_html.append(Package(lib, full))
            else:
                self.no_html.append((lib, full))
        self.has_html.sort()
        p = os.path.join(path, 'index.html')
        o = open(p, 'w')
        o.write(self.html())
        o.close()

def html(self):
        s = StringIO.StringIO()
        s.write('<html>\n')
        s.write('<head>\n')
        s.write('<title>Local Haskell package docs (from Haddock)</title>\n')
        s.write(('<link rel="stylesheet" type="text/css" '
                 'href="http://hackage.haskell.org/packages/hackage.css"/>\n'))
        s.write('</head>\n')
        s.write('<body>\n')
        s.write('<h2>Local packages with docs</h2>\n')

split = self.splitAlpha()
        keys = split.keys()
        keys.sort()

s.write('<p class="toc">\n')
        links = ['<a href="#%s">%s</a>' % (start, start) for start in keys]
        s.write(' &bull; '.join(links))
        s.write('</p>\n')
        
        for start in keys:
            packages = split[start]
            s.write('<h3 class="category"><a name="%s">%s</a></h3>\n' % (start, start))
            s.write('  <ul class="packages">\n')
            for package in packages:
                s.write('    <li>')
                s.write('<a href="file://%s">' % package.path)
                s.write(package.name)
                s.write('</a>: %s</li>\n' % package.title)
            s.write('  </ul>\n')
            
        s.write('<hr />')
        s.write('<h2>Directories without HTML docs</h2>\n')
        s.write('<ul>\n')
        for (name, path) in self.no_html:
            s.write('  <li><a href="file://%s">%s</li>\n' % (path, name))
        s.write('</ul>\n')
        s.write('</body>\n')
        s.write('</html>\n')
        h = s.getvalue()
        return h

def splitAlpha(self):
        al = {}
        for package in self.has_html:
            start = package.name[0].upper()
            try:
                al[start].append(package)
            except KeyError:
                al[start] = [package]
        return al
            
    def seek_html(self, path):
        items = os.listdir(path)
        if 'html' not in items:
            return False
        html = os.path.join(path, 'html')
        if not os.path.isdir(html):
            return False
        return 'index.html' in os.listdir(html)
        
if __name__ == '__main__':
    builder = Builder(PATH)

author title

language channel