]> git.rkrishnan.org Git - tahoe-lafs/zfec.git/blobdiff - zfec/setuptools-0.6c16dev3.egg/setuptools/package_index.py
zfec: rearrange files
[tahoe-lafs/zfec.git] / zfec / setuptools-0.6c16dev3.egg / setuptools / package_index.py
diff --git a/zfec/setuptools-0.6c16dev3.egg/setuptools/package_index.py b/zfec/setuptools-0.6c16dev3.egg/setuptools/package_index.py
deleted file mode 100644 (file)
index 3d82d93..0000000
+++ /dev/null
@@ -1,798 +0,0 @@
-"""PyPI and direct package downloading"""
-import sys, os.path, re, urlparse, urllib2, shutil, random, socket, cStringIO
-import httplib, urllib
-from pkg_resources import *
-from distutils import log
-from distutils.errors import DistutilsError
-try:
-    from hashlib import md5
-except ImportError:
-    from md5 import md5
-from fnmatch import translate
-EGG_FRAGMENT = re.compile(r'^egg=([-A-Za-z0-9_.]+)$')
-HREF = re.compile("""href\\s*=\\s*['"]?([^'"> ]+)""", re.I)
-# this is here to fix emacs' cruddy broken syntax highlighting
-PYPI_MD5 = re.compile(
-    '<a href="([^"#]+)">([^<]+)</a>\n\s+\\(<a (?:title="MD5 hash"\n\s+)'
-    'href="[^?]+\?:action=show_md5&amp;digest=([0-9a-f]{32})">md5</a>\\)'
-)
-URL_SCHEME = re.compile('([-+.a-z0-9]{2,}):',re.I).match
-EXTENSIONS = ".tar.gz .tar.bz2 .tar .zip .tgz".split()
-
-def is_local(url_or_fname):
-    """ Return True if url_or_fname is a "file:" url or if it is a schemaless thing (which is presumably a filename). """
-    mo = URL_SCHEME(url_or_fname)
-    return not (mo and mo.group(1).lower()!='file')
-
-def url_or_fname_to_fname(url_or_fname):
-    """ Assert that is_local(url_or_fname) then if it is a "file:" url, parse it and run url2pathname on it, else just return it. """
-    assert is_local(url_or_fname)
-
-    mo = URL_SCHEME(url_or_fname)
-    if mo:
-        return urllib2.url2pathname(urlparse.urlparse(url)[2])
-    else:
-        return url_or_fname
-
-__all__ = [
-    'PackageIndex', 'distros_for_url', 'parse_bdist_wininst',
-    'interpret_distro_name',
-]
-
-def parse_bdist_wininst(name):
-    """Return (base,pyversion) or (None,None) for possible .exe name"""
-
-    lower = name.lower()
-    base, py_ver = None, None
-
-    if lower.endswith('.exe'):
-        if lower.endswith('.win32.exe'):
-            base = name[:-10]
-        elif lower.startswith('.win32-py',-16):
-            py_ver = name[-7:-4]
-            base = name[:-16]
-
-    return base,py_ver
-
-def egg_info_for_url(url):
-    scheme, server, path, parameters, query, fragment = urlparse.urlparse(url)
-    base = urllib2.unquote(path.split('/')[-1])
-    if server=='sourceforge.net' and base=='download':    # XXX Yuck
-        base = urllib2.unquote(path.split('/')[-2])
-    if '#' in base: base, fragment = base.split('#',1)
-    return base,fragment
-
-def distros_for_url(url, metadata=None):
-    """Yield egg or source distribution objects that might be found at a URL"""
-    base, fragment = egg_info_for_url(url)
-    for dist in distros_for_location(url, base, metadata): yield dist
-    if fragment:
-        match = EGG_FRAGMENT.match(fragment)
-        if match:
-            for dist in interpret_distro_name(
-                url, match.group(1), metadata, precedence = CHECKOUT_DIST
-            ):
-                yield dist
-
-def distros_for_location(location, basename, metadata=None):
-    """Yield egg or source distribution objects based on basename"""
-    if basename.endswith('.egg.zip'):
-        basename = basename[:-4]    # strip the .zip
-    if basename.endswith('.egg') and '-' in basename:
-        # only one, unambiguous interpretation
-        return [Distribution.from_location(location, basename, metadata)]
-    if basename.endswith('.exe'):
-        win_base, py_ver = parse_bdist_wininst(basename)
-        if win_base is not None:
-            return interpret_distro_name(
-                location, win_base, metadata, py_ver, BINARY_DIST, "win32"
-            )
-    # Try source distro extensions (.zip, .tgz, etc.)
-    #
-    for ext in EXTENSIONS:
-        if basename.endswith(ext):
-            basename = basename[:-len(ext)]
-            return interpret_distro_name(location, basename, metadata)
-    return []  # no extension matched
-
-def distros_for_filename(filename, metadata=None):
-    """Yield possible egg or source distribution objects based on a filename"""
-    return distros_for_location(
-        normalize_path(filename), os.path.basename(filename), metadata
-    )
-
-
-def interpret_distro_name(location, basename, metadata,
-    py_version=None, precedence=SOURCE_DIST, platform=None
-):
-    """Generate alternative interpretations of a source distro name
-
-    Note: if `location` is a filesystem filename, you should call
-    ``pkg_resources.normalize_path()`` on it before passing it to this
-    routine!
-    """
-    # Generate alternative interpretations of a source distro name
-    # Because some packages are ambiguous as to name/versions split
-    # e.g. "adns-python-1.1.0", "egenix-mx-commercial", etc.
-    # So, we generate each possible interepretation (e.g. "adns, python-1.1.0"
-    # "adns-python, 1.1.0", and "adns-python-1.1.0, no version").  In practice,
-    # the spurious interpretations should be ignored, because in the event
-    # there's also an "adns" package, the spurious "python-1.1.0" version will
-    # compare lower than any numeric version number, and is therefore unlikely
-    # to match a request for it.  It's still a potential problem, though, and
-    # in the long run PyPI and the distutils should go for "safe" names and
-    # versions in distribution archive names (sdist and bdist).
-
-    parts = basename.split('-')
-    if not py_version:
-        for i,p in enumerate(parts[2:]):
-            if len(p)==5 and p.startswith('py2.'):
-                return # It's a bdist_dumb, not an sdist -- bail out
-
-    for p in range(1,len(parts)+1):
-        yield Distribution(
-            location, metadata, '-'.join(parts[:p]), '-'.join(parts[p:]),
-            py_version=py_version, precedence = precedence,
-            platform = platform
-        )
-
-REL = re.compile("""<([^>]*\srel\s*=\s*['"]?([^'">]+)[^>]*)>""", re.I)
-# this line is here to fix emacs' cruddy broken syntax highlighting
-
-def find_external_links(url, page):
-    """Find rel="homepage" and rel="download" links in `page`, yielding URLs"""
-
-    for match in REL.finditer(page):
-        tag, rel = match.groups()
-        rels = map(str.strip, rel.lower().split(','))
-        if 'homepage' in rels or 'download' in rels:
-            for match in HREF.finditer(tag):
-                yield urlparse.urljoin(url, htmldecode(match.group(1)))
-
-    for tag in ("<th>Home Page", "<th>Download URL"):
-        pos = page.find(tag)
-        if pos!=-1:
-            match = HREF.search(page,pos)
-            if match:
-                yield urlparse.urljoin(url, htmldecode(match.group(1)))
-
-user_agent = "Python-urllib/%s setuptools/%s" % (
-    urllib2.__version__, require('setuptools')[0].version
-)
-
-
-class PackageIndex(Environment):
-    """A distribution index that scans web pages for download URLs"""
-
-    def __init__(self, index_url="http://pypi.python.org/simple", hosts=('*',),
-        *args, **kw
-    ):
-        Environment.__init__(self,*args,**kw)
-        self.index_url = index_url + "/"[:not index_url.endswith('/')]
-        self.scanned_urls = {}
-        self.fetched_urls = {}
-        self.package_pages = {}
-        self.allows = re.compile('|'.join(map(translate,hosts))).match
-        self.to_scan = []
-
-
-
-    def process_url(self, url, retrieve=False):
-        """Evaluate a URL as a possible download, and maybe retrieve it"""
-        if url in self.scanned_urls and not retrieve:
-            return
-        self.scanned_urls[url] = True
-        if not URL_SCHEME(url):
-            self.process_filename(url)
-            return
-        else:
-            dists = list(distros_for_url(url))
-            if dists:
-                if not self.url_ok(url):
-                    return
-                self.debug("Found link: %s", url)
-
-        if dists or not retrieve or url in self.fetched_urls:
-            map(self.add, dists)
-            return  # don't need the actual page
-
-        if not self.url_ok(url):
-            self.fetched_urls[url] = True
-            return
-
-        self.info("Reading %s", url)
-        self.fetched_urls[url] = True   # prevent multiple fetch attempts
-        f = self.open_url(url, "Download error: %s -- Some packages may not be found!")
-        if f is None: return
-        self.fetched_urls[f.url] = True
-        if 'html' not in f.headers.get('content-type', '').lower():
-            f.close()   # not html, we can't process it
-            return
-
-        base = f.url     # handle redirects
-        page = f.read()
-        f.close()
-        if url.startswith(self.index_url) and getattr(f,'code',None)!=404:
-            page = self.process_index(url, page)
-        for match in HREF.finditer(page):
-            link = urlparse.urljoin(base, htmldecode(match.group(1)))
-            self.process_url(link)
-
-    def process_filename(self, fn, nested=False):
-        # process filenames or directories
-        if not os.path.exists(fn):
-            self.warn("Not found: %s", fn)
-            return
-
-        if os.path.isdir(fn) and not nested:
-            path = os.path.realpath(fn)
-            for item in os.listdir(path):
-                self.process_filename(os.path.join(path,item), True)
-
-        dists = distros_for_filename(fn)
-        if dists:
-            self.debug("Found: %s", fn)
-            map(self.add, dists)
-
-    def url_ok(self, url, fatal=False):
-        s = URL_SCHEME(url)
-        if (s and s.group(1).lower()=='file') or self.allows(urlparse.urlparse(url)[1]):
-            return True
-        msg = "\nLink to % s ***BLOCKED*** by --allow-hosts\n"
-        if fatal:
-            raise DistutilsError(msg % url)
-        else:
-            self.warn(msg, url)
-
-    def scan_egg_links(self, search_path):
-        for item in search_path:
-            if os.path.isdir(item):
-                for entry in os.listdir(item):
-                    if entry.endswith('.egg-link'):
-                        self.scan_egg_link(item, entry)
-
-    def scan_egg_link(self, path, entry):
-        lines = filter(None, map(str.strip, file(os.path.join(path, entry))))
-        if len(lines)==2:
-            for dist in find_distributions(os.path.join(path, lines[0])):
-                dist.location = os.path.join(path, *lines)
-                dist.precedence = SOURCE_DIST
-                self.add(dist)
-
-    def process_index(self,url,page):
-        """Process the contents of a PyPI page"""
-        def scan(link):
-            # Process a URL to see if it's for a package page
-            if link.startswith(self.index_url):
-                parts = map(
-                    urllib2.unquote, link[len(self.index_url):].split('/')
-                )
-                if len(parts)==2 and '#' not in parts[1]:
-                    # it's a package page, sanitize and index it
-                    pkg = safe_name(parts[0])
-                    ver = safe_version(parts[1])
-                    self.package_pages.setdefault(pkg.lower(),{})[link] = True
-                    return to_filename(pkg), to_filename(ver)
-            return None, None
-
-        # process an index page into the package-page index
-        for match in HREF.finditer(page):
-            scan( urlparse.urljoin(url, htmldecode(match.group(1))) )
-
-        pkg, ver = scan(url)   # ensure this page is in the page index
-        if pkg:
-            # process individual package page
-            for new_url in find_external_links(url, page):
-                # Process the found URL
-                base, frag = egg_info_for_url(new_url)
-                if base.endswith('.py') and not frag:
-                    if ver:
-                        new_url+='#egg=%s-%s' % (pkg,ver)
-                    else:
-                        self.need_version_info(url)
-                self.scan_url(new_url)
-
-            return PYPI_MD5.sub(
-                lambda m: '<a href="%s#md5=%s">%s</a>' % m.group(1,3,2), page
-            )
-        else:
-            return ""   # no sense double-scanning non-package pages
-
-
-
-    def need_version_info(self, url):
-        self.scan_all(
-            "Page at %s links to .py file(s) without version info; an index "
-            "scan is required.", url
-        )
-
-    def scan_all(self, msg=None, *args):
-        if self.index_url not in self.fetched_urls:
-            if msg: self.warn(msg,*args)
-            self.info(
-                "Scanning index of all packages (this may take a while)"
-            )
-        self.scan_url(self.index_url)
-
-    def find_packages(self, requirement):
-        self.scan_url(self.index_url + requirement.unsafe_name+'/')
-
-        if not self.package_pages.get(requirement.key):
-            # Fall back to safe version of the name
-            self.scan_url(self.index_url + requirement.project_name+'/')
-
-        if not self.package_pages.get(requirement.key):
-            # We couldn't find the target package, so search the index page too
-            self.not_found_in_index(requirement)
-
-        for url in list(self.package_pages.get(requirement.key,())):
-            # scan each page that might be related to the desired package
-            self.scan_url(url)
-
-    def obtain(self, requirement, installer=None):
-        self.prescan(); self.find_packages(requirement)
-        for dist in self[requirement.key]:
-            if dist in requirement:
-                return dist
-            self.debug("%s does not match %s", requirement, dist)
-        return super(PackageIndex, self).obtain(requirement,installer)
-
-
-
-
-
-    def check_md5(self, cs, info, filename, tfp):
-        if re.match('md5=[0-9a-f]{32}$', info):
-            self.debug("Validating md5 checksum for %s", filename)
-            if cs.hexdigest()!=info[4:]:
-                tfp.close()
-                os.unlink(filename)
-                raise DistutilsError(
-                    "MD5 validation failed for "+os.path.basename(filename)+
-                    "; possible download problem?"
-                )
-
-    def add_find_links(self, urls):
-        """Add `urls` to the list that will be prescanned for searches"""
-        for url in urls:
-            if (
-                self.to_scan is None        # if we have already "gone online"
-                or not URL_SCHEME(url)      # or it's a local file/directory
-                or url.startswith('file:')
-                or list(distros_for_url(url))   # or a direct package link
-            ):
-                # then go ahead and process it now
-                self.scan_url(url)
-            else:
-                # otherwise, defer retrieval till later
-                self.to_scan.append(url)
-
-    def prescan(self):
-        """Scan urls scheduled for prescanning (e.g. --find-links)"""
-        if self.to_scan:
-            map(self.scan_url, self.to_scan)
-        self.to_scan = None     # from now on, go ahead and process immediately
-
-    def not_found_in_index(self, requirement):
-        if self[requirement.key]:   # we've seen at least one distro
-            meth, msg = self.info, "Couldn't retrieve index page for %r"
-        else:   # no distros seen for this name, might be misspelled
-            meth, msg = (self.warn,
-                "Couldn't find index page for %r (maybe misspelled?)")
-        meth(msg, requirement.unsafe_name)
-        self.scan_all()
-
-    def download(self, spec, tmpdir):
-        """Locate and/or download `spec` to `tmpdir`, returning a local path
-
-        `spec` may be a ``Requirement`` object, or a string containing a URL,
-        an existing local filename, or a project/version requirement spec
-        (i.e. the string form of a ``Requirement`` object).  If it is the URL
-        of a .py file with an unambiguous ``#egg=name-version`` tag (i.e., one
-        that escapes ``-`` as ``_`` throughout), a trivial ``setup.py`` is
-        automatically created alongside the downloaded file.
-
-        If `spec` is a ``Requirement`` object or a string containing a
-        project/version requirement spec, this method returns the location of
-        a matching distribution (possibly after downloading it to `tmpdir`).
-        If `spec` is a locally existing file or directory name, it is simply
-        returned unchanged.  If `spec` is a URL, it is downloaded to a subpath
-        of `tmpdir`, and the local filename is returned.  Various errors may be
-        raised if a problem occurs during downloading.
-        """
-        if not isinstance(spec,Requirement):
-            scheme = URL_SCHEME(spec)
-            if scheme:
-                # It's a url, download it to tmpdir
-                found = self._download_url(scheme.group(1), spec, tmpdir)
-                base, fragment = egg_info_for_url(spec)
-                if base.endswith('.py'):
-                    found = self.gen_setup(found,fragment,tmpdir)
-                return found
-            elif os.path.exists(spec):
-                # Existing file or directory, just return it
-                return spec
-            else:
-                try:
-                    spec = Requirement.parse(spec)
-                except ValueError:
-                    raise DistutilsError(
-                        "Not a URL, existing file, or requirement spec: %r" %
-                        (spec,)
-                    )
-        return getattr(self.fetch_distribution(spec, tmpdir),'location',None)
-
-
-    def fetch_distribution(self,
-        requirement, tmpdir, force_scan=False, source=False, develop_ok=False,
-        local_index=None, 
-    ):
-        """Obtain a distribution suitable for fulfilling `requirement`
-
-        `requirement` must be a ``pkg_resources.Requirement`` instance.
-        If necessary, or if the `force_scan` flag is set, the requirement is
-        searched for in the (online) package index as well as the locally
-        installed packages.  If a distribution matching `requirement` is found,
-        the returned distribution's ``location`` is the value you would have
-        gotten from calling the ``download()`` method with the matching
-        distribution's URL or filename.  If no matching distribution is found,
-        ``None`` is returned.
-
-        If the `source` flag is set, only source distributions and source
-        checkout links will be considered.  Unless the `develop_ok` flag is
-        set, development and system eggs (i.e., those using the ``.egg-info``
-        format) will be ignored.
-        """
-        # process a Requirement
-        self.info("Searching for %s", requirement)
-        skipped = {}
-        dist = None
-
-        def find(env, req):
-            # Find a matching distribution; may be called more than once
-
-            # first try to find a local dist
-            for allow_remote in (False, True):
-                # then try to find a platform-dependent dist
-                for allow_platform_independent in (False, True):
-                    for dist in env[req.key]:
-                        if dist.precedence==DEVELOP_DIST and not develop_ok:
-                            if dist not in skipped:
-                                self.warn("Skipping development or system egg: %s",dist)
-                                skipped[dist] = 1
-                            continue
-
-                        if ((is_local(dist.location) or allow_remote) and
-                            (dist in req) and
-                            ((allow_platform_independent or dist.platform is not None) and
-                             (dist.precedence<=SOURCE_DIST or not source))):
-                            return dist
-
-        if force_scan:
-            self.prescan()
-            self.find_packages(requirement)
-            dist = find(self, requirement)
-            
-        if local_index is not None:
-            dist = dist or find(local_index, requirement)
-
-        if dist is None and self.to_scan is not None:
-            self.prescan()
-            dist = find(self, requirement)
-
-        if dist is None and not force_scan:
-            self.find_packages(requirement)
-            dist = find(self, requirement)
-
-        if dist is None:
-            self.warn(
-                "No local packages or download links found for %s%s",
-                (source and "a source distribution of " or ""),
-                requirement,
-            )
-        else:
-            self.info("Best match: %s", dist)
-            return dist.clone(location=self.download(dist.location, tmpdir))
-
-
-    def fetch(self, requirement, tmpdir, force_scan=False, source=False):
-        """Obtain a file suitable for fulfilling `requirement`
-
-        DEPRECATED; use the ``fetch_distribution()`` method now instead.  For
-        backward compatibility, this routine is identical but returns the
-        ``location`` of the downloaded distribution instead of a distribution
-        object.
-        """
-        dist = self.fetch_distribution(requirement,tmpdir,force_scan,source)
-        if dist is not None:
-            return dist.location
-        return None
-
-
-    def gen_setup(self, filename, fragment, tmpdir):
-        match = EGG_FRAGMENT.match(fragment)
-        dists = match and [d for d in
-            interpret_distro_name(filename, match.group(1), None) if d.version
-        ] or []
-
-        if len(dists)==1:   # unambiguous ``#egg`` fragment
-            basename = os.path.basename(filename)
-
-            # Make sure the file has been downloaded to the temp dir.
-            if os.path.dirname(filename) != tmpdir:
-                dst = os.path.join(tmpdir, basename)
-                from setuptools.command.easy_install import samefile
-                if not samefile(filename, dst):
-                    shutil.copy2(filename, dst)
-                    filename=dst
-
-            file = open(os.path.join(tmpdir, 'setup.py'), 'w')
-            file.write(
-                "from setuptools import setup\n"
-                "setup(name=%r, version=%r, py_modules=[%r])\n"
-                % (
-                    dists[0].project_name, dists[0].version,
-                    os.path.splitext(basename)[0]
-                )
-            )
-            file.close()
-            return filename
-
-        elif match:
-            raise DistutilsError(
-                "Can't unambiguously interpret project/version identifier %r; "
-                "any dashes in the name or version should be escaped using "
-                "underscores. %r" % (fragment,dists)
-            )
-        else:
-            raise DistutilsError(
-                "Can't process plain .py files without an '#egg=name-version'"
-                " suffix to enable automatic setup script generation."
-            )
-
-    dl_blocksize = 8192
-    def _download_to(self, url, filename):
-        self.info("Downloading %s", url)
-        # Download the file
-        fp, tfp, info = None, None, None
-        try:
-            if '#' in url:
-                url, info = url.split('#', 1)
-            fp = self.open_url(url)
-            if isinstance(fp, urllib2.HTTPError):
-                raise DistutilsError(
-                    "Can't download %s: %s %s" % (url, fp.code,fp.msg)
-                )
-            cs = md5()
-            headers = fp.info()
-            blocknum = 0
-            bs = self.dl_blocksize
-            size = -1
-            if "content-length" in headers:
-                size = int(headers["Content-Length"])
-                self.reporthook(url, filename, blocknum, bs, size)
-            tfp = open(filename,'wb')
-            while True:
-                block = fp.read(bs)
-                if block:
-                    cs.update(block)
-                    tfp.write(block)
-                    blocknum += 1
-                    self.reporthook(url, filename, blocknum, bs, size)
-                else:
-                    break
-            if info: self.check_md5(cs, info, filename, tfp)
-            return headers
-        finally:
-            if fp: fp.close()
-            if tfp: tfp.close()
-
-    def reporthook(self, url, filename, blocknum, blksize, size):
-        pass    # no-op
-
-
-    def open_url(self, url, warning=None):
-        if url.startswith('file:'): return local_open(url)
-        try:
-            return open_with_auth(url)
-        except urllib2.HTTPError, v:
-            return v
-        except urllib2.URLError, v:
-            reason = v.reason
-        except httplib.HTTPException, v: 
-            reason = "%s: %s" % (v.__doc__ or v.__class__.__name__, v)
-        if warning:
-            self.warn(warning, reason)
-        else:
-            raise DistutilsError("Download error for %s: %s" % (url, reason))
-
-    def _download_url(self, scheme, url, tmpdir):
-        # Determine download filename
-        #
-        name, fragment = egg_info_for_url(url)
-        if name:
-            while '..' in name:
-                name = name.replace('..','.').replace('\\','_')
-        else:
-            name = "__downloaded__"    # default if URL has no path contents
-
-        if name.endswith('.egg.zip'):
-            name = name[:-4]    # strip the extra .zip before download
-
-        filename = os.path.join(tmpdir,name)
-
-        # Download the file
-        #
-        if scheme=='svn' or scheme.startswith('svn+'):
-            return self._download_svn(url, filename)
-        elif scheme=='file':
-            return urllib2.url2pathname(urlparse.urlparse(url)[2])
-        else:
-            self.url_ok(url, True)   # raises error if not allowed
-            return self._attempt_download(url, filename)
-
-
-    def scan_url(self, url):
-        self.process_url(url, True)
-
-
-    def _attempt_download(self, url, filename):
-        headers = self._download_to(url, filename)
-        if 'html' in headers.get('content-type','').lower():
-            return self._download_html(url, headers, filename)
-        else:
-            return filename
-
-    def _download_html(self, url, headers, filename):
-        file = open(filename)
-        for line in file:
-            if line.strip():
-                # Check for a subversion index page
-                if re.search(r'<title>([^- ]+ - )?Revision \d+:', line):
-                    # it's a subversion index page:
-                    file.close()
-                    os.unlink(filename)
-                    return self._download_svn(url, filename)
-                break   # not an index page
-        file.close()
-        os.unlink(filename)
-        raise DistutilsError("Unexpected HTML page found at "+url)
-
-    def _download_svn(self, url, filename):
-        url = url.split('#',1)[0]   # remove any fragment for svn's sake
-        self.info("Doing subversion checkout from %s to %s", url, filename)
-        os.system("svn checkout -q %s %s" % (url, filename))
-        return filename
-
-    def debug(self, msg, *args):
-        log.debug(msg, *args)
-
-    def info(self, msg, *args):
-        log.info(msg, *args)
-
-    def warn(self, msg, *args):
-        log.warn(msg, *args)
-
-# This pattern matches a character entity reference (a decimal numeric
-# references, a hexadecimal numeric reference, or a named reference).
-entity_sub = re.compile(r'&(#(\d+|x[\da-fA-F]+)|[\w.:-]+);?').sub
-
-def uchr(c):
-    if not isinstance(c, int):
-        return c
-    if c>255: return unichr(c)
-    return chr(c)
-
-def decode_entity(match):
-    what = match.group(1)
-    if what.startswith('#x'):
-        what = int(what[2:], 16)
-    elif what.startswith('#'):
-        what = int(what[1:])
-    else:
-        from htmlentitydefs import name2codepoint
-        what = name2codepoint.get(what, match.group(0))
-    return uchr(what)
-
-def htmldecode(text):
-    """Decode HTML entities in the given text."""
-    return entity_sub(decode_entity, text)
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-def open_with_auth(url):
-    """Open a urllib2 request, handling HTTP authentication"""
-
-    scheme, netloc, path, params, query, frag = urlparse.urlparse(url)
-
-    if scheme in ('http', 'https'):
-        auth, host = urllib.splituser(netloc)
-    else:
-        auth = None
-
-    if auth:
-        auth = "Basic " + urllib2.unquote(auth).encode('base64').strip()
-        new_url = urlparse.urlunparse((scheme,host,path,params,query,frag))
-        request = urllib2.Request(new_url)
-        request.add_header("Authorization", auth)
-    else:
-        request = urllib2.Request(url)
-
-    request.add_header('User-Agent', user_agent)
-    fp = urllib2.urlopen(request)
-
-    if auth:
-        # Put authentication info back into request URL if same host,
-        # so that links found on the page will work
-        s2, h2, path2, param2, query2, frag2 = urlparse.urlparse(fp.url)
-        if s2==scheme and h2==host:
-            fp.url = urlparse.urlunparse((s2,netloc,path2,param2,query2,frag2))
-
-    return fp
-
-
-
-
-
-
-
-
-
-
-
-
-def fix_sf_url(url):
-    return url      # backward compatibility
-
-def local_open(url):
-    """Read a local path, with special support for directories"""
-    scheme, server, path, param, query, frag = urlparse.urlparse(url)
-    filename = urllib2.url2pathname(path)
-    if os.path.isfile(filename):
-        return urllib2.urlopen(url)
-    elif path.endswith('/') and os.path.isdir(filename):
-        files = []
-        for f in os.listdir(filename):
-            if f=='index.html':
-                body = open(os.path.join(filename,f),'rb').read()
-                break
-            elif os.path.isdir(os.path.join(filename,f)):
-                f+='/'
-            files.append("<a href=%r>%s</a>" % (f,f))
-        else:
-            body = ("<html><head><title>%s</title>" % url) + \
-                "</head><body>%s</body></html>" % '\n'.join(files)
-        status, message = 200, "OK"
-    else:
-        status, message, body = 404, "Path not found", "Not found"
-
-    return urllib2.HTTPError(url, status, message,
-            {'content-type':'text/html'}, cStringIO.StringIO(body))
-
-
-
-
-
-
-
-
-
-
-
-
-
-# this line is a kludge to keep the trailing blank lines for pje's editor