Work in progress. Functional programming is making me mad. Mad I say.

cygwinapi_check.py

from contextlib import closing
#from collections import defaultdict
import re
import shelve

from urllib2 import Request, urlopen, URLError

from BeautifulSoup import BeautifulSoup

apiurls = { 'std-unix': 'http://cygwin.com/cygwin-api/compatibility.html',
            'std-bsd': 'http://cygwin.com/cygwin-api/std-bsd.html',
            'std-gnu': 'http://cygwin.com/cygwin-api/std-gnu.html',
            'std-sunos': 'http://cygwin.com/cygwin-api/std-solaris.html',
            'std-deprec': 'http://cygwin.com/cygwin-api/std-deprec.html',
            # std-notimpl must be present. If unavailable or gone,
            # set to None, the download routine will skip it.
            'std-notimpl': 'http://cygwin.com/cygwin-api/std-notimpl.html',
          }

cache = shelve.open('.cygwinapicheck.cache')

def webapipages(apiurls={}):
    for api, url in apiurls.iteritems():

        if url is None: break

        print "Downloading Cygwin API Reference set:", api
        req = Request(url)
        with closing(urlopen(req)) as apipage:
            try:
                html = apipage.read()
                yield (api, html)
            except URLError as e:
                def err():
                    if hasattr(e, 'reason'):
                        return ' '.join("Reason: ", e.reason)
                    elif hasattr(e, 'code'):
                        return ' '.join("Errcode: ", e.code)

                print ("Unable to download %s from: %s! (%s)" %
                        (api, url, err()))
                print ("%s has been skipped." % api)

def syscalls(html):

    sc_pattern = re.compile(r'(\S+)(?:\\t)?(\S+)?')

    dtree = BeautifulSoup(html)
    apidata = dtree.find('pre', attrs={'class': 'screen'}).text
    apititle = dtree.find('h2', attrs={'class': 'title',
                                               'style': 'clear: both'},
                                  ).text

    if not len(apititle): apititle = u'(no title)'

    if apidata:
        print "API: %s" % (apititle)
        _lines = (l.trim() for l in apidata.split('\n'))
        _rlines = (sc_pattern.match(l) for l in _lines)
        _syscalls = (sc.groups() for sc in _rlines if sc)

        return _syscalls

    else:
        print "No API data found!"

if __name__ == '__main__':
    for api, html in