hgweb: parse WSGI request into a data structure

Currently, our WSGI applications (hgweb_mod and hgwebdir_mod) process
the raw WSGI request instance themselves. This means they have to
talk in terms of system strings. And they need to know details
about what's in the WSGI request. And in the case of hgweb_mod, it
is doing some very funky things with URL parsing to impact
dispatching. The code is difficult to read and maintain.

This commit introduces parsing of the WSGI request into a higher-level
and easier-to-reason-about data structure.

To prove it works, we hook it up to hgweb_mod and use it for populating
the relative URL on the request instance.

We hold off on using it in more places because the logic in hgweb_mod
is crazy and I don't want to involve those changes with review of
the parsing code.

The URL construction code has variations that use the HTTP: Host header
(the canonical WSGI way of reconstructing the URL) and with the use
of SERVER_NAME. We need to differentiate because hgweb is currently
using SERVER_NAME for URL construction.

Differential Revision: https://phab.mercurial-scm.org/D2734

mercurial/hgweb/hgweb_mod.py

@@ -1,472 +1,466 b''
 # hgweb/hgweb_mod.py - Web interface for a repository.
 #
 # Copyright 21 May 2005 - (c) 2005 Jake Edge <[e-mail]>
 # Copyright 2005-2007 Matt Mackall <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import contextlib
 import os
 
 from .common import (
     ErrorResponse,
     HTTP_BAD_REQUEST,
     HTTP_NOT_FOUND,
     HTTP_NOT_MODIFIED,
     HTTP_OK,
     HTTP_SERVER_ERROR,
     caching,
     cspvalues,
     permhooks,
 )
 
 from .. import (
     encoding,
     error,
     formatter,
     hg,
     hook,
     profiling,
     pycompat,
     repoview,
     templatefilters,
     templater,
     ui as uimod,
     util,
     wireprotoserver,
 )
 
 from . import (
     request as requestmod,
     webcommands,
     webutil,
     wsgicgi,
 )
 
 archivespecs = util.sortdict((
     ('zip', ('application/zip', 'zip', '.zip', None)),
     ('gz', ('application/x-gzip', 'tgz', '.tar.gz', None)),
     ('bz2', ('application/x-bzip2', 'tbz2', '.tar.bz2', None)),
 ))
 
 def getstyle(req, configfn, templatepath):
     fromreq = req.form.get('style', [None])[0]
     styles = (
         fromreq,
         configfn('web', 'style'),
         'paper',
     )
     return styles, templater.stylemap(styles, templatepath)
 
 def makebreadcrumb(url, prefix=''):
     '''Return a 'URL breadcrumb' list
 
     A 'URL breadcrumb' is a list of URL-name pairs,
     corresponding to each of the path items on a URL.
     This can be used to create path navigation entries.
     '''
     if url.endswith('/'):
         url = url[:-1]
     if prefix:
         url = '/' + prefix + url
     relpath = url
     if relpath.startswith('/'):
         relpath = relpath[1:]
 
     breadcrumb = []
     urlel = url
     pathitems = [''] + relpath.split('/')
     for pathel in reversed(pathitems):
         if not pathel or not urlel:
             break
         breadcrumb.append({'url': urlel, 'name': pathel})
         urlel = os.path.dirname(urlel)
     return reversed(breadcrumb)
 
 class requestcontext(object):
     """Holds state/context for an individual request.
 
     Servers can be multi-threaded. Holding state on the WSGI application
     is prone to race conditions. Instances of this class exist to hold
     mutable and race-free state for requests.
     """
     def __init__(self, app, repo):
         self.repo = repo
         self.reponame = app.reponame
 
         self.archivespecs = archivespecs
 
         self.maxchanges = self.configint('web', 'maxchanges')
         self.stripecount = self.configint('web', 'stripes')
         self.maxshortchanges = self.configint('web', 'maxshortchanges')
         self.maxfiles = self.configint('web', 'maxfiles')
         self.allowpull = self.configbool('web', 'allow-pull')
 
         # we use untrusted=False to prevent a repo owner from using
         # web.templates in .hg/hgrc to get access to any file readable
         # by the user running the CGI script
         self.templatepath = self.config('web', 'templates', untrusted=False)
 
         # This object is more expensive to build than simple config values.
         # It is shared across requests. The app will replace the object
         # if it is updated. Since this is a reference and nothing should
         # modify the underlying object, it should be constant for the lifetime
         # of the request.
         self.websubtable = app.websubtable
 
         self.csp, self.nonce = cspvalues(self.repo.ui)
 
     # Trust the settings from the .hg/hgrc files by default.
     def config(self, section, name, default=uimod._unset, untrusted=True):
         return self.repo.ui.config(section, name, default,
                                    untrusted=untrusted)
 
     def configbool(self, section, name, default=uimod._unset, untrusted=True):
         return self.repo.ui.configbool(section, name, default,
                                        untrusted=untrusted)
 
     def configint(self, section, name, default=uimod._unset, untrusted=True):
         return self.repo.ui.configint(section, name, default,
                                       untrusted=untrusted)
 
     def configlist(self, section, name, default=uimod._unset, untrusted=True):
         return self.repo.ui.configlist(section, name, default,
                                        untrusted=untrusted)
 
     def archivelist(self, nodeid):
         allowed = self.configlist('web', 'allow_archive')
         for typ, spec in self.archivespecs.iteritems():
             if typ in allowed or self.configbool('web', 'allow%s' % typ):
                 yield {'type': typ, 'extension': spec[2], 'node': nodeid}
 
     def templater(self, wsgireq):
         # determine scheme, port and server name
         # this is needed to create absolute urls
 
         proto = wsgireq.env.get('wsgi.url_scheme')
         if proto == 'https':
             proto = 'https'
             default_port = '443'
         else:
             proto = 'http'
             default_port = '80'
 
         port = wsgireq.env[r'SERVER_PORT']
         port = port != default_port and (r':' + port) or r''
         urlbase = r'%s://%s%s' % (proto, wsgireq.env[r'SERVER_NAME'], port)
         logourl = self.config('web', 'logourl')
         logoimg = self.config('web', 'logoimg')
         staticurl = (self.config('web', 'staticurl')
                      or pycompat.sysbytes(wsgireq.url) + 'static/')
         if not staticurl.endswith('/'):
             staticurl += '/'
 
         # some functions for the templater
 
         def motd(**map):
             yield self.config('web', 'motd')
 
         # figure out which style to use
 
         vars = {}
         styles, (style, mapfile) = getstyle(wsgireq, self.config,
                                             self.templatepath)
         if style == styles[0]:
             vars['style'] = style
 
         sessionvars = webutil.sessionvars(vars, '?')
 
         if not self.reponame:
             self.reponame = (self.config('web', 'name', '')
                              or wsgireq.env.get('REPO_NAME')
                              or wsgireq.url.strip(r'/') or self.repo.root)
 
         def websubfilter(text):
             return templatefilters.websub(text, self.websubtable)
 
         # create the templater
         # TODO: export all keywords: defaults = templatekw.keywords.copy()
         defaults = {
             'url': pycompat.sysbytes(wsgireq.url),
             'logourl': logourl,
             'logoimg': logoimg,
             'staticurl': staticurl,
             'urlbase': urlbase,
             'repo': self.reponame,
             'encoding': encoding.encoding,
             'motd': motd,
             'sessionvars': sessionvars,
             'pathdef': makebreadcrumb(pycompat.sysbytes(wsgireq.url)),
             'style': style,
             'nonce': self.nonce,
         }
         tres = formatter.templateresources(self.repo.ui, self.repo)
         tmpl = templater.templater.frommapfile(mapfile,
                                                filters={'websub': websubfilter},
                                                defaults=defaults,
                                                resources=tres)
         return tmpl
 
 
 class hgweb(object):
     """HTTP server for individual repositories.
 
     Instances of this class serve HTTP responses for a particular
     repository.
 
     Instances are typically used as WSGI applications.
 
     Some servers are multi-threaded. On these servers, there may
     be multiple active threads inside __call__.
     """
     def __init__(self, repo, name=None, baseui=None):
         if isinstance(repo, str):
             if baseui:
                 u = baseui.copy()
             else:
                 u = uimod.ui.load()
             r = hg.repository(u, repo)
         else:
             # we trust caller to give us a private copy
             r = repo
 
         r.ui.setconfig('ui', 'report_untrusted', 'off', 'hgweb')
         r.baseui.setconfig('ui', 'report_untrusted', 'off', 'hgweb')
         r.ui.setconfig('ui', 'nontty', 'true', 'hgweb')
         r.baseui.setconfig('ui', 'nontty', 'true', 'hgweb')
         # resolve file patterns relative to repo root
         r.ui.setconfig('ui', 'forcecwd', r.root, 'hgweb')
         r.baseui.setconfig('ui', 'forcecwd', r.root, 'hgweb')
         # displaying bundling progress bar while serving feel wrong and may
         # break some wsgi implementation.
         r.ui.setconfig('progress', 'disable', 'true', 'hgweb')
         r.baseui.setconfig('progress', 'disable', 'true', 'hgweb')
         self._repos = [hg.cachedlocalrepo(self._webifyrepo(r))]
         self._lastrepo = self._repos[0]
         hook.redirect(True)
         self.reponame = name
 
     def _webifyrepo(self, repo):
         repo = getwebview(repo)
         self.websubtable = webutil.getwebsubs(repo)
         return repo
 
     @contextlib.contextmanager
     def _obtainrepo(self):
         """Obtain a repo unique to the caller.
 
         Internally we maintain a stack of cachedlocalrepo instances
         to be handed out. If one is available, we pop it and return it,
         ensuring it is up to date in the process. If one is not available,
         we clone the most recently used repo instance and return it.
 
         It is currently possible for the stack to grow without bounds
         if the server allows infinite threads. However, servers should
         have a thread limit, thus establishing our limit.
         """
         if self._repos:
             cached = self._repos.pop()
             r, created = cached.fetch()
         else:
             cached = self._lastrepo.copy()
             r, created = cached.fetch()
         if created:
             r = self._webifyrepo(r)
 
         self._lastrepo = cached
         self.mtime = cached.mtime
         try:
             yield r
         finally:
             self._repos.append(cached)
 
     def run(self):
         """Start a server from CGI environment.
 
         Modern servers should be using WSGI and should avoid this
         method, if possible.
         """
         if not encoding.environ.get('GATEWAY_INTERFACE',
                                     '').startswith("CGI/1."):
             raise RuntimeError("This function is only intended to be "
                                "called while running as a CGI script.")
         wsgicgi.launch(self)
 
     def __call__(self, env, respond):
         """Run the WSGI application.
 
         This may be called by multiple threads.
         """
         req = requestmod.wsgirequest(env, respond)
         return self.run_wsgi(req)
 
     def run_wsgi(self, wsgireq):
         """Internal method to run the WSGI application.
 
         This is typically only called by Mercurial. External consumers
         should be using instances of this class as the WSGI application.
         """
         with self._obtainrepo() as repo:
             profile = repo.ui.configbool('profiling', 'enabled')
             with profiling.profile(repo.ui, enabled=profile):
                 for r in self._runwsgi(wsgireq, repo):
                     yield r
 
     def _runwsgi(self, wsgireq, repo):
+        req = requestmod.parserequestfromenv(wsgireq.env)
         rctx = requestcontext(self, repo)
 
         # This state is global across all threads.
         encoding.encoding = rctx.config('web', 'encoding')
         rctx.repo.ui.environ = wsgireq.env
 
         if rctx.csp:
             # hgwebdir may have added CSP header. Since we generate our own,
             # replace it.
             wsgireq.headers = [h for h in wsgireq.headers
                                if h[0] != 'Content-Security-Policy']
             wsgireq.headers.append(('Content-Security-Policy', rctx.csp))
 
-        # work with CGI variables to create coherent structure
-        # use SCRIPT_NAME, PATH_INFO and QUERY_STRING as well as our REPO_NAME
-
-        wsgireq.url = wsgireq.env[r'SCRIPT_NAME']
-        if not wsgireq.url.endswith(r'/'):
-            wsgireq.url += r'/'
-        if wsgireq.env.get('REPO_NAME'):
-            wsgireq.url += wsgireq.env[r'REPO_NAME'] + r'/'
+        wsgireq.url = pycompat.sysstr(req.apppath)
 
         if r'PATH_INFO' in wsgireq.env:
             parts = wsgireq.env[r'PATH_INFO'].strip(r'/').split(r'/')
             repo_parts = wsgireq.env.get(r'REPO_NAME', r'').split(r'/')
             if parts[:len(repo_parts)] == repo_parts:
                 parts = parts[len(repo_parts):]
             query = r'/'.join(parts)
         else:
             query = wsgireq.env[r'QUERY_STRING'].partition(r'&')[0]
             query = query.partition(r';')[0]
 
         # Route it to a wire protocol handler if it looks like a wire protocol
         # request.
         protohandler = wireprotoserver.parsehttprequest(rctx, wsgireq, query,
                                                         self.check_perm)
 
         if protohandler:
             try:
                 if query:
                     raise ErrorResponse(HTTP_NOT_FOUND)
 
                 return protohandler['dispatch']()
             except ErrorResponse as inst:
                 return protohandler['handleerror'](inst)
 
         # translate user-visible url structure to internal structure
 
         args = query.split(r'/', 2)
         if 'cmd' not in wsgireq.form and args and args[0]:
             cmd = args.pop(0)
             style = cmd.rfind('-')
             if style != -1:
                 wsgireq.form['style'] = [cmd[:style]]
                 cmd = cmd[style + 1:]
 
             # avoid accepting e.g. style parameter as command
             if util.safehasattr(webcommands, cmd):
                 wsgireq.form['cmd'] = [cmd]
 
             if cmd == 'static':
                 wsgireq.form['file'] = ['/'.join(args)]
             else:
                 if args and args[0]:
                     node = args.pop(0).replace('%2F', '/')
                     wsgireq.form['node'] = [node]
                 if args:
                     wsgireq.form['file'] = args
 
             ua = wsgireq.env.get('HTTP_USER_AGENT', '')
             if cmd == 'rev' and 'mercurial' in ua:
                 wsgireq.form['style'] = ['raw']
 
             if cmd == 'archive':
                 fn = wsgireq.form['node'][0]
                 for type_, spec in rctx.archivespecs.iteritems():
                     ext = spec[2]
                     if fn.endswith(ext):
                         wsgireq.form['node'] = [fn[:-len(ext)]]
                         wsgireq.form['type'] = [type_]
         else:
             cmd = wsgireq.form.get('cmd', [''])[0]
 
         # process the web interface request
 
         try:
             tmpl = rctx.templater(wsgireq)
             ctype = tmpl('mimetype', encoding=encoding.encoding)
             ctype = templater.stringify(ctype)
 
             # check read permissions non-static content
             if cmd != 'static':
                 self.check_perm(rctx, wsgireq, None)
 
             if cmd == '':
                 wsgireq.form['cmd'] = [tmpl.cache['default']]
                 cmd = wsgireq.form['cmd'][0]
 
             # Don't enable caching if using a CSP nonce because then it wouldn't
             # be a nonce.
             if rctx.configbool('web', 'cache') and not rctx.nonce:
                 caching(self, wsgireq) # sets ETag header or raises NOT_MODIFIED
             if cmd not in webcommands.__all__:
                 msg = 'no such method: %s' % cmd
                 raise ErrorResponse(HTTP_BAD_REQUEST, msg)
             elif cmd == 'file' and 'raw' in wsgireq.form.get('style', []):
                 rctx.ctype = ctype
                 content = webcommands.rawfile(rctx, wsgireq, tmpl)
             else:
                 content = getattr(webcommands, cmd)(rctx, wsgireq, tmpl)
                 wsgireq.respond(HTTP_OK, ctype)
 
             return content
 
         except (error.LookupError, error.RepoLookupError) as err:
             wsgireq.respond(HTTP_NOT_FOUND, ctype)
             msg = pycompat.bytestr(err)
             if (util.safehasattr(err, 'name') and
                 not isinstance(err,  error.ManifestLookupError)):
                 msg = 'revision not found: %s' % err.name
             return tmpl('error', error=msg)
         except (error.RepoError, error.RevlogError) as inst:
             wsgireq.respond(HTTP_SERVER_ERROR, ctype)
             return tmpl('error', error=pycompat.bytestr(inst))
         except ErrorResponse as inst:
             wsgireq.respond(inst, ctype)
             if inst.code == HTTP_NOT_MODIFIED:
                 # Not allowed to return a body on a 304
                 return ['']
             return tmpl('error', error=pycompat.bytestr(inst))
 
     def check_perm(self, rctx, req, op):
         for permhook in permhooks:
             permhook(rctx, req, op)
 
 def getwebview(repo):
     """The 'web.view' config controls changeset filter to hgweb. Possible
     values are ``served``, ``visible`` and ``all``. Default is ``served``.
     The ``served`` filter only shows changesets that can be pulled from the
     hgweb instance.  The``visible`` filter includes secret changesets but
     still excludes "hidden" one.
 
     See the repoview module for details.
 
     The option has been around undocumented since Mercurial 2.5, but no
     user ever asked about it. So we better keep it undocumented for now."""
     # experimental config: web.view
     viewconfig = repo.ui.config('web', 'view', untrusted=True)
     if viewconfig == 'all':
         return repo.unfiltered()
     elif viewconfig in repoview.filtertable:
         return repo.filtered(viewconfig)
     else:
         return repo.filtered('served')

mercurial/hgweb/request.py

@@ -1,158 +1,280 b''
 # hgweb/request.py - An http request from either CGI or the standalone server.
 #
 # Copyright 21 May 2005 - (c) 2005 Jake Edge <[e-mail]>
 # Copyright 2005, 2006 Matt Mackall <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import cgi
 import errno
 import socket
+#import wsgiref.validate
 
 from .common import (
     ErrorResponse,
     HTTP_NOT_MODIFIED,
     statusmessage,
 )
 
+from ..thirdparty import (
+    attr,
+)
 from .. import (
     pycompat,
     util,
 )
 
 shortcuts = {
     'cl': [('cmd', ['changelog']), ('rev', None)],
     'sl': [('cmd', ['shortlog']), ('rev', None)],
     'cs': [('cmd', ['changeset']), ('node', None)],
     'f': [('cmd', ['file']), ('filenode', None)],
     'fl': [('cmd', ['filelog']), ('filenode', None)],
     'fd': [('cmd', ['filediff']), ('node', None)],
     'fa': [('cmd', ['annotate']), ('filenode', None)],
     'mf': [('cmd', ['manifest']), ('manifest', None)],
     'ca': [('cmd', ['archive']), ('node', None)],
     'tags': [('cmd', ['tags'])],
     'tip': [('cmd', ['changeset']), ('node', ['tip'])],
     'static': [('cmd', ['static']), ('file', None)]
 }
 
 def normalize(form):
     # first expand the shortcuts
     for k in shortcuts:
         if k in form:
             for name, value in shortcuts[k]:
                 if value is None:
                     value = form[k]
                 form[name] = value
             del form[k]
     # And strip the values
     bytesform = {}
     for k, v in form.iteritems():
         bytesform[pycompat.bytesurl(k)] = [
             pycompat.bytesurl(i.strip()) for i in v]
     return bytesform
 
+@attr.s(frozen=True)
+class parsedrequest(object):
+    """Represents a parsed WSGI request / static HTTP request parameters."""
+
+    # Full URL for this request.
+    url = attr.ib()
+    # URL without any path components. Just <proto>://<host><port>.
+    baseurl = attr.ib()
+    # Advertised URL. Like ``url`` and ``baseurl`` but uses SERVER_NAME instead
+    # of HTTP: Host header for hostname. This is likely what clients used.
+    advertisedurl = attr.ib()
+    advertisedbaseurl = attr.ib()
+    # WSGI application path.
+    apppath = attr.ib()
+    # List of path parts to be used for dispatch.
+    dispatchparts = attr.ib()
+    # URL path component (no query string) used for dispatch.
+    dispatchpath = attr.ib()
+    # Raw query string (part after "?" in URL).
+    querystring = attr.ib()
+
+def parserequestfromenv(env):
+    """Parse URL components from environment variables.
+
+    WSGI defines request attributes via environment variables. This function
+    parses the environment variables into a data structure.
+    """
+    # PEP-0333 defines the WSGI spec and is a useful reference for this code.
+
+    # We first validate that the incoming object conforms with the WSGI spec.
+    # We only want to be dealing with spec-conforming WSGI implementations.
+    # TODO enable this once we fix internal violations.
+    #wsgiref.validate.check_environ(env)
+
+    # PEP-0333 states that environment keys and values are native strings
+    # (bytes on Python 2 and str on Python 3). The code points for the Unicode
+    # strings on Python 3 must be between \00000-\000FF. We deal with bytes
+    # in Mercurial, so mass convert string keys and values to bytes.
+    if pycompat.ispy3:
+        env = {k.encode('latin-1'): v for k, v in env.iteritems()}
+        env = {k: v.encode('latin-1') if isinstance(v, str) else v
+               for k, v in env.iteritems()}
+
+    # https://www.python.org/dev/peps/pep-0333/#environ-variables defines
+    # the environment variables.
+    # https://www.python.org/dev/peps/pep-0333/#url-reconstruction defines
+    # how URLs are reconstructed.
+    fullurl = env['wsgi.url_scheme'] + '://'
+    advertisedfullurl = fullurl
+
+    def addport(s):
+        if env['wsgi.url_scheme'] == 'https':
+            if env['SERVER_PORT'] != '443':
+                s += ':' + env['SERVER_PORT']
+        else:
+            if env['SERVER_PORT'] != '80':
+                s += ':' + env['SERVER_PORT']
+
+        return s
+
+    if env.get('HTTP_HOST'):
+        fullurl += env['HTTP_HOST']
+    else:
+        fullurl += env['SERVER_NAME']
+        fullurl = addport(fullurl)
+
+    advertisedfullurl += env['SERVER_NAME']
+    advertisedfullurl = addport(advertisedfullurl)
+
+    baseurl = fullurl
+    advertisedbaseurl = advertisedfullurl
+
+    fullurl += util.urlreq.quote(env.get('SCRIPT_NAME', ''))
+    advertisedfullurl += util.urlreq.quote(env.get('SCRIPT_NAME', ''))
+    fullurl += util.urlreq.quote(env.get('PATH_INFO', ''))
+    advertisedfullurl += util.urlreq.quote(env.get('PATH_INFO', ''))
+
+    if env.get('QUERY_STRING'):
+        fullurl += '?' + env['QUERY_STRING']
+        advertisedfullurl += '?' + env['QUERY_STRING']
+
+    # When dispatching requests, we look at the URL components (PATH_INFO
+    # and QUERY_STRING) after the application root (SCRIPT_NAME). But hgwebdir
+    # has the concept of "virtual" repositories. This is defined via REPO_NAME.
+    # If REPO_NAME is defined, we append it to SCRIPT_NAME to form a new app
+    # root. We also exclude its path components from PATH_INFO when resolving
+    # the dispatch path.
+
+    # TODO the use of trailing slashes in apppath is arguably wrong. We need it
+    # to appease low-level parts of hgweb_mod for now.
+    apppath = env['SCRIPT_NAME']
+    if not apppath.endswith('/'):
+        apppath += '/'
+
+    if env.get('REPO_NAME'):
+        apppath += env.get('REPO_NAME') + '/'
+
+    if 'PATH_INFO' in env:
+        dispatchparts = env['PATH_INFO'].strip('/').split('/')
+
+        # Strip out repo parts.
+        repoparts = env.get('REPO_NAME', '').split('/')
+        if dispatchparts[:len(repoparts)] == repoparts:
+            dispatchparts = dispatchparts[len(repoparts):]
+    else:
+        dispatchparts = []
+
+    dispatchpath = '/'.join(dispatchparts)
+
+    querystring = env.get('QUERY_STRING', '')
+
+    return parsedrequest(url=fullurl, baseurl=baseurl,
+                         advertisedurl=advertisedfullurl,
+                         advertisedbaseurl=advertisedbaseurl,
+                         apppath=apppath,
+                         dispatchparts=dispatchparts, dispatchpath=dispatchpath,
+                         querystring=querystring)
+
 class wsgirequest(object):
     """Higher-level API for a WSGI request.
 
     WSGI applications are invoked with 2 arguments. They are used to
     instantiate instances of this class, which provides higher-level APIs
     for obtaining request parameters, writing HTTP output, etc.
     """
     def __init__(self, wsgienv, start_response):
         version = wsgienv[r'wsgi.version']
         if (version < (1, 0)) or (version >= (2, 0)):
             raise RuntimeError("Unknown and unsupported WSGI version %d.%d"
                                % version)
         self.inp = wsgienv[r'wsgi.input']
         self.err = wsgienv[r'wsgi.errors']
         self.threaded = wsgienv[r'wsgi.multithread']
         self.multiprocess = wsgienv[r'wsgi.multiprocess']
         self.run_once = wsgienv[r'wsgi.run_once']
         self.env = wsgienv
         self.form = normalize(cgi.parse(self.inp,
                                         self.env,
                                         keep_blank_values=1))
         self._start_response = start_response
         self.server_write = None
         self.headers = []
 
     def __iter__(self):
         return iter([])
 
     def read(self, count=-1):
         return self.inp.read(count)
 
     def drain(self):
         '''need to read all data from request, httplib is half-duplex'''
         length = int(self.env.get('CONTENT_LENGTH') or 0)
         for s in util.filechunkiter(self.inp, limit=length):
             pass
 
     def respond(self, status, type, filename=None, body=None):
         if not isinstance(type, str):
             type = pycompat.sysstr(type)
         if self._start_response is not None:
             self.headers.append((r'Content-Type', type))
             if filename:
                 filename = (filename.rpartition('/')[-1]
                             .replace('\\', '\\\\').replace('"', '\\"'))
                 self.headers.append(('Content-Disposition',
                                      'inline; filename="%s"' % filename))
             if body is not None:
                 self.headers.append((r'Content-Length', str(len(body))))
 
             for k, v in self.headers:
                 if not isinstance(v, str):
                     raise TypeError('header value must be string: %r' % (v,))
 
             if isinstance(status, ErrorResponse):
                 self.headers.extend(status.headers)
                 if status.code == HTTP_NOT_MODIFIED:
                     # RFC 2616 Section 10.3.5: 304 Not Modified has cases where
                     # it MUST NOT include any headers other than these and no
                     # body
                     self.headers = [(k, v) for (k, v) in self.headers if
                                     k in ('Date', 'ETag', 'Expires',
                                           'Cache-Control', 'Vary')]
                 status = statusmessage(status.code, pycompat.bytestr(status))
             elif status == 200:
                 status = '200 Script output follows'
             elif isinstance(status, int):
                 status = statusmessage(status)
 
             self.server_write = self._start_response(
                 pycompat.sysstr(status), self.headers)
             self._start_response = None
             self.headers = []
         if body is not None:
             self.write(body)
             self.server_write = None
 
     def write(self, thing):
         if thing:
             try:
                 self.server_write(thing)
             except socket.error as inst:
                 if inst[0] != errno.ECONNRESET:
                     raise
 
     def writelines(self, lines):
         for line in lines:
             self.write(line)
 
     def flush(self):
         return None
 
     def close(self):
         return None
 
 def wsgiapplication(app_maker):
     '''For compatibility with old CGI scripts. A plain hgweb() or hgwebdir()
     can and should now be used as a WSGI application.'''
     application = app_maker()
     def run_wsgi(env, respond):
         return application(env, respond)
     return run_wsgi