hgweb: expose input stream on parsed WSGI request object
authorGregory Szorc <gregory.szorc@gmail.com>
Sat, 10 Mar 2018 11:06:13 -0800
changeset 36863 da4e2f87167d
parent 36862 1f7d9024674c
child 36864 01f6bba64424
hgweb: expose input stream on parsed WSGI request object Our next step towards moving away from wsgirequest to our newer, friendlier parsedrequest type is input stream access. This commit exposes the input stream on the instance. Consumers in the HTTP protocol server switch to it. Because there were very few consumers of the input stream, we stopped storing a reference to the input stream on wsgirequest directly. All access now goes through parsedrequest. However, wsgirequest still may read from this stream as part of cgi.parse(). So we still need to create the stream from wsgirequest. Differential Revision: https://phab.mercurial-scm.org/D2771
mercurial/hgweb/hgwebdir_mod.py
mercurial/hgweb/request.py
mercurial/wireprotoserver.py
--- a/mercurial/hgweb/hgwebdir_mod.py	Sat Mar 10 10:56:10 2018 -0800
+++ b/mercurial/hgweb/hgwebdir_mod.py	Sat Mar 10 11:06:13 2018 -0800
@@ -291,7 +291,8 @@
                     # variable.
                     # TODO this is kind of hacky and we should have a better
                     # way of doing this than with REPO_NAME side-effects.
-                    wsgireq.req = requestmod.parserequestfromenv(wsgireq.env)
+                    wsgireq.req = requestmod.parserequestfromenv(
+                        wsgireq.env, wsgireq.req.bodyfh)
                     try:
                         # ensure caller gets private copy of ui
                         repo = hg.repository(self.ui.copy(), real)
--- a/mercurial/hgweb/request.py	Sat Mar 10 10:56:10 2018 -0800
+++ b/mercurial/hgweb/request.py	Sat Mar 10 11:06:13 2018 -0800
@@ -61,7 +61,10 @@
 
 @attr.s(frozen=True)
 class parsedrequest(object):
-    """Represents a parsed WSGI request / static HTTP request parameters."""
+    """Represents a parsed WSGI request.
+
+    Contains both parsed parameters as well as a handle on the input stream.
+    """
 
     # Request method.
     method = attr.ib()
@@ -91,8 +94,10 @@
     # wsgiref.headers.Headers instance. Operates like a dict with case
     # insensitive keys.
     headers = attr.ib()
+    # Request body input stream.
+    bodyfh = attr.ib()
 
-def parserequestfromenv(env):
+def parserequestfromenv(env, bodyfh):
     """Parse URL components from environment variables.
 
     WSGI defines request attributes via environment variables. This function
@@ -209,6 +214,12 @@
     if 'CONTENT_LENGTH' in env and 'HTTP_CONTENT_LENGTH' not in env:
         headers['Content-Length'] = env['CONTENT_LENGTH']
 
+    # TODO do this once we remove wsgirequest.inp, otherwise we could have
+    # multiple readers from the underlying input stream.
+    #bodyfh = env['wsgi.input']
+    #if 'Content-Length' in headers:
+    #    bodyfh = util.cappedreader(bodyfh, int(headers['Content-Length']))
+
     return parsedrequest(method=env['REQUEST_METHOD'],
                          url=fullurl, baseurl=baseurl,
                          advertisedurl=advertisedfullurl,
@@ -219,7 +230,8 @@
                          querystring=querystring,
                          querystringlist=querystringlist,
                          querystringdict=querystringdict,
-                         headers=headers)
+                         headers=headers,
+                         bodyfh=bodyfh)
 
 class wsgirequest(object):
     """Higher-level API for a WSGI request.
@@ -233,28 +245,27 @@
         if (version < (1, 0)) or (version >= (2, 0)):
             raise RuntimeError("Unknown and unsupported WSGI version %d.%d"
                                % version)
-        self.inp = wsgienv[r'wsgi.input']
+
+        inp = wsgienv[r'wsgi.input']
 
         if r'HTTP_CONTENT_LENGTH' in wsgienv:
-            self.inp = util.cappedreader(self.inp,
-                                         int(wsgienv[r'HTTP_CONTENT_LENGTH']))
+            inp = util.cappedreader(inp, int(wsgienv[r'HTTP_CONTENT_LENGTH']))
         elif r'CONTENT_LENGTH' in wsgienv:
-            self.inp = util.cappedreader(self.inp,
-                                         int(wsgienv[r'CONTENT_LENGTH']))
+            inp = util.cappedreader(inp, int(wsgienv[r'CONTENT_LENGTH']))
 
         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.form = normalize(cgi.parse(inp,
                                         self.env,
                                         keep_blank_values=1))
         self._start_response = start_response
         self.server_write = None
         self.headers = []
 
-        self.req = parserequestfromenv(wsgienv)
+        self.req = parserequestfromenv(wsgienv, inp)
 
     def respond(self, status, type, filename=None, body=None):
         if not isinstance(type, str):
@@ -315,7 +326,7 @@
                 # input stream doesn't overrun the actual request. So there's
                 # no guarantee that reading until EOF won't corrupt the stream
                 # state.
-                if not isinstance(self.inp, util.cappedreader):
+                if not isinstance(self.req.bodyfh, util.cappedreader):
                     close = True
                 else:
                     # We /could/ only drain certain HTTP response codes. But 200
@@ -329,9 +340,9 @@
                 self.headers.append((r'Connection', r'Close'))
 
             if drain:
-                assert isinstance(self.inp, util.cappedreader)
+                assert isinstance(self.req.bodyfh, util.cappedreader)
                 while True:
-                    chunk = self.inp.read(32768)
+                    chunk = self.req.bodyfh.read(32768)
                     if not chunk:
                         break
 
--- a/mercurial/wireprotoserver.py	Sat Mar 10 10:56:10 2018 -0800
+++ b/mercurial/wireprotoserver.py	Sat Mar 10 11:06:13 2018 -0800
@@ -83,7 +83,7 @@
         postlen = int(self._req.headers.get(b'X-HgArgs-Post', 0))
         if postlen:
             args.update(urlreq.parseqs(
-                self._wsgireq.inp.read(postlen), keep_blank_values=True))
+                self._req.bodyfh.read(postlen), keep_blank_values=True))
             return args
 
         argvalue = decodevaluefromheaders(self._req, b'X-HgArg')
@@ -97,7 +97,7 @@
         # If httppostargs is used, we need to read Content-Length
         # minus the amount that was consumed by args.
         length -= int(self._req.headers.get(b'X-HgArgs-Post', 0))
-        for s in util.filechunkiter(self._wsgireq.inp, limit=length):
+        for s in util.filechunkiter(self._req.bodyfh, limit=length):
             fp.write(s)
 
     @contextlib.contextmanager