From fc2aa563ea22418e23abe89285c7a87c1ee90c5b Mon Sep 17 00:00:00 2001
From: Fredrik Tolf <fredrik@dolda2000.com>
Date: Thu, 14 Apr 2011 12:18:46 +0200
Subject: [PATCH 01/16] python: Documented the ashd.wsgidir module.

---
 python/ashd/wsgidir.py | 75 ++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 75 insertions(+)

diff --git a/python/ashd/wsgidir.py b/python/ashd/wsgidir.py
index 83d96a0..6080d91 100644
--- a/python/ashd/wsgidir.py
+++ b/python/ashd/wsgidir.py
@@ -1,7 +1,61 @@
+"""WSGI handler for serving chained WSGI modules from physical files
+
+The WSGI handler in this module examines the SCRIPT_FILENAME variable
+of the requests it handles -- that is, the physical file corresponding
+to the request, as determined by the webserver -- determining what to
+do with the request based on the extension of that file.
+
+By default, it handles files named `.wsgi' by compiling them into
+Python modules and using them, in turn, as chained WSGI handlers, but
+handlers for other extensions can be installed as well.
+
+When handling `.wsgi' files, the compiled modules are cached and
+reused until the file is modified, in which case the previous module
+is discarded and the new file contents are loaded into a new module in
+its place. When chaining such modules, an object named `wmain' is
+first looked for and called with no arguments if found. The object it
+returns is then used as the WSGI application object for that module,
+which is reused until the module is reloaded. If `wmain' is not found,
+an object named `application' is looked for instead. If found, it is
+used directly as the WSGI application object.
+
+This module itself contains both an `application' and a `wmain'
+object. If this module is used by ashd-wsgi(1) or scgi-wsgi(1) so that
+its wmain function is called, arguments can be specified to it to
+install handlers for other file extensions. Such arguments take the
+form `.EXT=MODULE.HANDLER', where EXT is the file extension to be
+handled, and the MODULE.HANDLER string is treated by splitting it
+along its last constituent dot. The part left of the dot is the name
+of a module which is imported, and the part right of the dot is the
+name of an object in that module, which should be a callable of three
+arguments. When files of the given extension are handled, that
+callable is called with the file's absolute path, the WSGI environment
+and the WSGI `start_response' function, in that order. For example,
+the argument `.fpy=my.module.foohandler' can be given to pass requests
+for `.fpy' files to the function `foohandler' in the module
+`my.module' (which must, of course, be importable). When writing such
+handler functions, you will probably want to use the getmod() function
+in this module.
+"""
+
 import os, threading, types
 import wsgiutil
 
+__all__ = ["application", "wmain", "getmod", "cachedmod"]
+
 class cachedmod(object):
+    """Cache entry for modules loaded by getmod()
+
+    Instances of this class are returned by the getmod()
+    function. They contain three data attributes:
+     * mod - The loaded module
+     * lock - A threading.Lock object, which can be used for
+       manipulating this instance in a thread-safe manner
+     * mtime - The time the file was last modified
+
+    Additional data attributes can be arbitrarily added for recording
+    any meta-data about the module.
+    """
     def __init__(self, mod, mtime):
         self.lock = threading.Lock()
         self.mod = mod
@@ -21,6 +75,18 @@ def mangle(path):
     return ret
 
 def getmod(path):
+    """Load the given file as a module, caching it appropriately
+
+    The given file is loaded and compiled into a Python module. The
+    compiled module is cached and returned upon subsequent requests
+    for the same file, unless the file has changed (as determined by
+    its mtime), in which case the cached module is discarded and the
+    new file contents are reloaded in its place.
+
+    The return value is an instance of the cachedmod class, which can
+    be used for locking purposes and for storing arbitrary meta-data
+    about the module. See its documentation for details.
+    """
     sb = os.stat(path)
     cachelock.acquire()
     try:
@@ -73,6 +139,10 @@ def addext(ext, handler):
     exts[ext] = getattr(mod, hname)
 
 def application(env, startreq):
+    """WSGI handler function
+
+    Handles WSGI requests as per the module documentation.
+    """
     if not "SCRIPT_FILENAME" in env:
         return wsgiutil.simpleerror(env, startreq, 500, "Internal Error", "The server is erroneously configured.")
     path = env["SCRIPT_FILENAME"]
@@ -86,6 +156,11 @@ def application(env, startreq):
     return(exts[ext](path, env, startreq))
 
 def wmain(*argv):
+    """Main function for ashd(7)-compatible WSGI handlers
+
+    Returns the `application' function. If any arguments are given,
+    they are parsed according to the module documentation.
+    """
     for arg in argv:
         if arg[0] == '.':
             p = arg.index('=')
-- 
2.11.0


From e66efe5f71c2bb6182055dffc9beefd0373ff3ad Mon Sep 17 00:00:00 2001
From: Fredrik Tolf <fredrik@dolda2000.com>
Date: Thu, 14 Apr 2011 12:20:47 +0200
Subject: [PATCH 02/16] python: Added __all__ to ashd.proto and ashd.util.

---
 python/ashd/proto.py | 2 ++
 python/ashd/util.py  | 2 ++
 2 files changed, 4 insertions(+)

diff --git a/python/ashd/proto.py b/python/ashd/proto.py
index c1ae1b0..4a48304 100644
--- a/python/ashd/proto.py
+++ b/python/ashd/proto.py
@@ -10,6 +10,8 @@ ashd.util module provides an easier-to-use interface.
 import os, socket
 import htlib
 
+__all__ = ["req", "recvreq", "sendreq"]
+
 class protoerr(Exception):
     pass
 
diff --git a/python/ashd/util.py b/python/ashd/util.py
index 2ac7dc0..0ff3878 100644
--- a/python/ashd/util.py
+++ b/python/ashd/util.py
@@ -7,6 +7,8 @@ handlers, wrapping the low-level ashd.proto module.
 import os, socket
 import proto
 
+__all__ = ["stdfork", "pchild", "respond", "serveloop"]
+
 def stdfork(argv, chinit = None):
     """Fork a persistent handler process using the `argv' argument
     list, as per the standard ashd(7) calling convention. For an
-- 
2.11.0


From 075a379e5dfba9584cc5f3b473ae343628fc160d Mon Sep 17 00:00:00 2001
From: Fredrik Tolf <fredrik@dolda2000.com>
Date: Thu, 14 Apr 2011 12:52:22 +0200
Subject: [PATCH 03/16] python: Fixed scgi-wsgi typo in the -T option.

---
 python/scgi-wsgi | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/python/scgi-wsgi b/python/scgi-wsgi
index 006f375..5ffcf6e 100755
--- a/python/scgi-wsgi
+++ b/python/scgi-wsgi
@@ -20,7 +20,7 @@ for o, a in opts:
         sk = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
         p = a.rfind(":")
         if p < 0:
-            bindhost = "hostname"
+            bindhost = "localhost"
             bindport = int(a)
         else:
             bindhost = a[:p]
-- 
2.11.0


From 338bee9d3d0ad8f236a17d9d1b8e7433e1d17c21 Mon Sep 17 00:00:00 2001
From: Fredrik Tolf <fredrik@dolda2000.com>
Date: Thu, 14 Apr 2011 13:00:15 +0200
Subject: [PATCH 04/16] python: Added (unused) documentation for ashd-wsgi and
 scgi-wsgi.

Unfortunately, I know not yet how to make distutils compile and install them.
---
 python/doc/.gitignore    |   3 ++
 python/doc/ashd-wsgi.doc | 107 +++++++++++++++++++++++++++++++++++++++++++++++
 python/doc/scgi-wsgi.doc |  63 ++++++++++++++++++++++++++++
 3 files changed, 173 insertions(+)
 create mode 100644 python/doc/.gitignore
 create mode 100644 python/doc/ashd-wsgi.doc
 create mode 100644 python/doc/scgi-wsgi.doc

diff --git a/python/doc/.gitignore b/python/doc/.gitignore
new file mode 100644
index 0000000..494c1f8
--- /dev/null
+++ b/python/doc/.gitignore
@@ -0,0 +1,3 @@
+/*.1
+/*.html
+/*.css
diff --git a/python/doc/ashd-wsgi.doc b/python/doc/ashd-wsgi.doc
new file mode 100644
index 0000000..25531ff
--- /dev/null
+++ b/python/doc/ashd-wsgi.doc
@@ -0,0 +1,107 @@
+ashd-wsgi(1)
+============
+
+NAME
+----
+ashd-wsgi - WSGI adapter for ashd(7)
+
+SYNOPSIS
+--------
+*ashd-wsgi* [*-hA*] [*-p* 'MODPATH'] 'HANDLER-MODULE' ['ARGS'...]
+
+DESCRIPTION
+-----------
+
+The *ashd-wsgi* handler translates *ashd*(7) requests to WSGI
+requests, and passes them to a specified Python handler module. The
+precise Python convention for doing so is described in the PROTOCOL
+section, below.
+
+*ashd-wsgi* is a persistent handler, as defined in *ashd*(7). It uses
+multithreaded dispatching in a single Python interpreter, which means
+that WSGI applications that use it need to be thread-safe, but that
+they can also share all Python data structures and global variables
+between requests.
+
+The Python module that *ashd-wsgi* comes with also contains a standard
+handler module, `ashd.wsgidir`, which serves individual WSGI
+applications directly from the files in which they reside and as such
+makes this program useful as a *dirplex*(1) handler. Please see its
+Python documentation for further details.
+
+*ashd-wsgi* requires the `ashd.proto` and `ashd.util` modules, which
+are only available for CPython. If you want to use some other Python
+implementation instead, you may want to use the *scgi-wsgi*(1) program
+instead, along with *callscgi*(1).
+
+OPTIONS
+-------
+
+*-h*::
+
+	Print a brief help message to standard output and exit.
+
+*-A*::
+
+	Use the convention used by Apache's mod_wsgi module to find
+	the WSGI application object. See the PROTOCOL section, below,
+	for details.
+
+*-p* 'MODPATH'::
+
+	Prepend 'MODPATH' to Python's `sys.path`; can be given multiple
+	times. Note that the working directory of *ashd-wsgi* is not
+	on Python's module path by default, so if you want to use a
+	module in that directory, you will need to specify "`-p .`".
+
+PROTOCOL
+--------
+
+When starting, *ashd-wsgi* will attempt to import the module named by
+'HANDLER-MODULE', look for an object named `wmain` in that module,
+call that object passing the 'ARGS' (as Python strings) as positional
+parameters, and use the returned object as the WSGI application
+object. If the *-A* option was specified, it will look for an object
+named `application` instead of `wmain`, and use that object directly
+as the WSGI application object.
+
+When calling the WSGI application, a new thread is started for each
+request, in which the WSGI application object is called. All requests
+run in the same interpreter, so it is guaranteed that data structures
+and global variables can be shared between requests.
+
+The WSGI environment is the standard CGI environment, including the
+`SCRIPT_FILENAME` variable whenever the `X-Ash-File` header was
+included in the request.
+
+EXAMPLES
+--------
+
+The following *dirplex*(1) configuration can be used for serving WSGI
+modules directly from the filesystem.
+
+--------
+child wsgidir
+  exec ashd-wsgi ashd.wsgidir
+match
+  filename *.wsgi
+  handler wsgidir
+--------
+
+Since *ashd-wsgi* is a persistent handler, it can be used directly as
+a root handler for *htparser*(1). For instance, if the directory
+`/srv/www/foo` contains a `wsgi.py` file, which declares a standard
+WSGI `application` object, it can be served with the following
+command:
+
+--------
+htparser plain:port=8080 -- ashd-wsgi -Ap /srv/www/foo wsgi
+--------
+
+AUTHOR
+------
+Fredrik Tolf <fredrik@dolda2000.com>
+
+SEE ALSO
+--------
+*scgi-wsgi*(1), *ashd*(7), <http://wsgi.org/>
diff --git a/python/doc/scgi-wsgi.doc b/python/doc/scgi-wsgi.doc
new file mode 100644
index 0000000..1aab621
--- /dev/null
+++ b/python/doc/scgi-wsgi.doc
@@ -0,0 +1,63 @@
+scgi-wsgi(1)
+============
+
+NAME
+----
+scgi-wsgi - WSGI adapter for SCGI
+
+SYNOPSIS
+--------
+*scgi-wsgi* [*-hA*] [*-p* 'MODPATH'] [*-T* \[HOST:]'PORT'] 'HANDLER-MODULE' ['ARGS'...]
+
+DESCRIPTION
+-----------
+
+The *scgi-wsgi* program translates SCGI requests to WSGI requests, and
+passes them to a specified Python module. It is mainly written to
+emulate the behavior of *ashd-wsgi*(1), but over SCGI instead of the
+native *ashd*(7) protocol, so please see its documentation for details
+of Python interoperation. Unlike *ashd-wsgi* which requires CPython,
+however, *scgi-wsgi* is written in pure Python using only the standard
+library, and so should be usable by any Python implementation. If
+using it under *ashd*(7), please see the documentation for
+*callscgi*(1) as well.
+
+Following *callscgi*(1) conventions, *scgi-wsgi* will, by default,
+accept connections on a socket passed on its standard input (a
+behavior which is, obviously, not available on all Python
+implementations). Use the *-T* option to listen to a TCP address
+instead.
+
+OPTIONS
+-------
+
+*-h*::
+
+	Print a brief help message to standard output and exit.
+
+*-A*::
+
+	Use the convention used by Apache's mod_wsgi module to find
+	the WSGI application object. See the PROTOCOL section of
+	*ashd-wsgi*(1) for details.
+
+*-p* 'MODPATH'::
+
+	Prepend 'MODPATH' to Python's `sys.path`; can be given multiple
+	times.
+
+*-T* \[HOST:]'PORT'::
+
+	Instead of using a listening socket passed on standard input
+	to accept SCGI connections, bind a TCP socket to the 'HOST'
+	address listening for connections on 'PORT' instead. If 'HOST'
+	is not given, `localhost` is used by default.
+
+AUTHOR
+------
+Fredrik Tolf <fredrik@dolda2000.com>
+
+SEE ALSO
+--------
+*ashd-wsgi*(1), *callscgi*(1), <http://wsgi.org/>,
+<http://www.python.ca/scgi/>
-- 
2.11.0


From 6586b2b37418f8cb54c5b82bec64bfb7af72725c Mon Sep 17 00:00:00 2001
From: Fredrik Tolf <fredrik@dolda2000.com>
Date: Thu, 14 Apr 2011 15:57:57 +0200
Subject: [PATCH 05/16] sendfile: Only allow GET and HEAD requests.

---
 src/sendfile.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/src/sendfile.c b/src/sendfile.c
index e6f6786..81fb67a 100644
--- a/src/sendfile.c
+++ b/src/sendfile.c
@@ -271,13 +271,20 @@ int main(int argc, char **argv)
 	simpleerror(1, 500, "Internal Error", "The server could not access its own data.");
 	exit(1);
     }
+    if(!strcasecmp(argv[optind], "get")) {
+	ishead = 0;
+    } else if(!strcasecmp(argv[optind], "head")) {
+	ishead = 1;
+    } else {
+	simpleerror(1, 405, "Method not allowed", "The requested method is not defined for this resource.");
+	return(0);
+    }
     if(contype == NULL)
 	contype = getmimetype(file, &sb);
     contype = ckctype(contype);
     
     checkcache(file, &sb);
     
-    ishead = !strcasecmp(argv[optind], "head");
     if((hdr = getenv("REQ_RANGE")) != NULL)
 	sendrange(fd, &sb, contype, hdr, ishead);
     else
-- 
2.11.0


From 3e11d7ede84e52974e41542ed60899ff340088d7 Mon Sep 17 00:00:00 2001
From: Fredrik Tolf <fredrik@dolda2000.com>
Date: Thu, 21 Apr 2011 04:46:21 +0200
Subject: [PATCH 06/16] python: Added a request limiter to ashd-wsgi.

---
 python/ashd-wsgi         | 34 ++++++++++++++++++++++++++++++----
 python/doc/ashd-wsgi.doc | 10 +++++++++-
 2 files changed, 39 insertions(+), 5 deletions(-)

diff --git a/python/ashd-wsgi b/python/ashd-wsgi
index 79aad4f..e43dbc0 100755
--- a/python/ashd-wsgi
+++ b/python/ashd-wsgi
@@ -1,13 +1,14 @@
 #!/usr/bin/python
 
-import sys, os, getopt, threading
+import sys, os, getopt, threading, time
 import ashd.proto, ashd.util
 
 def usage(out):
-    out.write("usage: ashd-wsgi [-hA] [-p MODPATH] HANDLER-MODULE [ARGS...]\n")
+    out.write("usage: ashd-wsgi [-hA] [-p MODPATH] [-l REQLIMIT] HANDLER-MODULE [ARGS...]\n")
 
+reqlimit = 0
 modwsgi_compat = False
-opts, args = getopt.getopt(sys.argv[1:], "+hAp:")
+opts, args = getopt.getopt(sys.argv[1:], "+hAp:l:")
 for o, a in opts:
     if o == "-h":
         usage(sys.stdout)
@@ -16,6 +17,8 @@ for o, a in opts:
         sys.path.insert(0, a)
     elif o == "-A":
         modwsgi_compat = True
+    elif o == "-l":
+        reqlimit = int(a)
 if len(args) < 1:
     usage(sys.stderr)
     sys.exit(1)
@@ -170,14 +173,37 @@ def dowsgi(req):
         if hasattr(respiter, "close"):
             respiter.close()
 
+flightlock = threading.Condition()
+inflight = 0
+
 class reqthread(threading.Thread):
     def __init__(self, req):
         super(reqthread, self).__init__(name = "Request handler")
         self.req = req.dup()
     
     def run(self):
+        global inflight
         try:
-            dowsgi(self.req)
+            flightlock.acquire()
+            try:
+                if reqlimit != 0:
+                    start = time.time()
+                    while inflight >= reqlimit:
+                        flightlock.wait(10)
+                        if time.time() - start > 10:
+                            os.abort()
+                inflight += 1
+            finally:
+                flightlock.release()
+            try:
+                dowsgi(self.req)
+            finally:
+                flightlock.acquire()
+                try:
+                    inflight -= 1
+                    flightlock.notify()
+                finally:
+                    flightlock.release()
         finally:
             self.req.close()
     
diff --git a/python/doc/ashd-wsgi.doc b/python/doc/ashd-wsgi.doc
index 25531ff..b73f90d 100644
--- a/python/doc/ashd-wsgi.doc
+++ b/python/doc/ashd-wsgi.doc
@@ -7,7 +7,7 @@ ashd-wsgi - WSGI adapter for ashd(7)
 
 SYNOPSIS
 --------
-*ashd-wsgi* [*-hA*] [*-p* 'MODPATH'] 'HANDLER-MODULE' ['ARGS'...]
+*ashd-wsgi* [*-hA*] [*-p* 'MODPATH'] [*-l* 'LIMIT'] 'HANDLER-MODULE' ['ARGS'...]
 
 DESCRIPTION
 -----------
@@ -54,6 +54,14 @@ OPTIONS
 	on Python's module path by default, so if you want to use a
 	module in that directory, you will need to specify "`-p .`".
 
+*-l* 'LIMIT'::
+
+	Allow at most 'LIMIT' requests to run concurrently. If a new
+	request is made when 'LIMIT' requests are executing, the new
+	request will wait up to ten seconds for one of them to
+	complete; if none does, the *ashd-wsgi* will assume that the
+	process is foobar and *abort*(3).
+
 PROTOCOL
 --------
 
-- 
2.11.0


From 254eb937b8c3298a191863e88e39f9a5035fabf9 Mon Sep 17 00:00:00 2001
From: Fredrik Tolf <fredrik@dolda2000.com>
Date: Wed, 27 Apr 2011 05:29:02 +0200
Subject: [PATCH 07/16] call[fs]cgi: Ensure less delay in closing stdin when
 shutting down.

---
 src/callfcgi.c | 1 +
 src/callscgi.c | 1 +
 2 files changed, 2 insertions(+)

diff --git a/src/callfcgi.c b/src/callfcgi.c
index 27c9c5e..9af5929 100644
--- a/src/callfcgi.c
+++ b/src/callfcgi.c
@@ -806,6 +806,7 @@ static void sigign(int sig)
 
 static void sigexit(int sig)
 {
+    shutdown(0, SHUT_RDWR);
     exit(0);
 }
 
diff --git a/src/callscgi.c b/src/callscgi.c
index 07064f3..062baa1 100644
--- a/src/callscgi.c
+++ b/src/callscgi.c
@@ -595,6 +595,7 @@ static void sigign(int sig)
 
 static void sigexit(int sig)
 {
+    shutdown(0, SHUT_RDWR);
     exit(0);
 }
 
-- 
2.11.0


From 9f974c1ff83d52fe3ba11040d46dfb2540bb0630 Mon Sep 17 00:00:00 2001
From: Fredrik Tolf <fredrik@dolda2000.com>
Date: Fri, 29 Apr 2011 03:10:05 +0200
Subject: [PATCH 08/16] doc: Fixed a couple of errors.

---
 doc/dirplex.doc | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/doc/dirplex.doc b/doc/dirplex.doc
index d41c1ad..cc794ba 100644
--- a/doc/dirplex.doc
+++ b/doc/dirplex.doc
@@ -82,11 +82,11 @@ CONFIGURATION
 Configuration in *dirplex* comes from several sources. When *dirplex*
 starts, unless the *-N* option is given, it tries to find a global
 configuration file named `dirplex.rc`. It looks in all directories
-named by the *PATH* environment variable, appended with `../etc`. For
-example, then, if *PATH* is `/usr/local/bin:/bin:/usr/bin`, the
-directories `/usr/local/etc`, `/etc` and `/usr/etc` are searched for
-`dirplex.rc`, in that order. Only the first file found is used, should
-there exist several.
+named by the *PATH* environment variable, appended with
+`../etc/ashd`. For example, then, if *PATH* is
+`/usr/local/bin:/bin:/usr/bin`, the directories `/usr/local/etc/ashd`,
+`/etc/ashd` and `/usr/etc/ashd` are searched for `dirplex.rc`, in that
+order. Only the first file found is used, should there exist several.
 
 If the *-c* option is given to *dirplex*, it too specifies a
 configuration file to load. If the name given contains any slashes, it
@@ -158,9 +158,9 @@ The follow configuration directives are recognized:
 	['ARGS'...], specifying the program to execute and the
 	arguments to pass it. In addition to the specified arguments,
 	the HTTP method, raw URL and the rest string will be appended
-	added as described in *ashd*(7). If given in a `.htrc` file,
-	the program will be started in the same directory as the
-	`.htrc` file itself.
+	as described in *ashd*(7). If given in a `.htrc` file, the
+	program will be started in the same directory as the `.htrc`
+	file itself.
 
 *match* [*directory*]::
 
-- 
2.11.0


From 0370bd82128b41fdbd5fbcbcac35104c0ede60fa Mon Sep 17 00:00:00 2001
From: Fredrik Tolf <fredrik@dolda2000.com>
Date: Fri, 29 Apr 2011 03:20:45 +0200
Subject: [PATCH 09/16] htparser: chdir as part of chrooting.

---
 src/htparser.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/htparser.c b/src/htparser.c
index ceae319..5179a25 100644
--- a/src/htparser.c
+++ b/src/htparser.c
@@ -499,7 +499,7 @@ int main(int argc, char **argv)
     if(usesyslog)
 	opensyslog();
     if(root) {
-	if(chroot(root)) {
+	if(chdir(root) || chroot(root)) {
 	    flog(LOG_ERR, "could not chroot to %s: %s", root, strerror(errno));
 	    exit(1);
 	}
-- 
2.11.0


From aa7e44069384016c57c417904bb45f6fa554981a Mon Sep 17 00:00:00 2001
From: Fredrik Tolf <fredrik@dolda2000.com>
Date: Fri, 29 Apr 2011 05:39:35 +0200
Subject: [PATCH 10/16] doc: Made mention of the include directive.

---
 doc/dirplex.doc | 10 ++++++++++
 doc/patplex.doc |  6 +++---
 2 files changed, 13 insertions(+), 3 deletions(-)

diff --git a/doc/dirplex.doc b/doc/dirplex.doc
index cc794ba..7d1e41f 100644
--- a/doc/dirplex.doc
+++ b/doc/dirplex.doc
@@ -124,6 +124,16 @@ treated as comments and ignored.
 
 The follow configuration directives are recognized:
 
+*include* ['FILENAME'...]::
+
+	Read the the named files and act as if their contents stood in
+	place of the *include* stanza. A 'FILENAME' may be a glob
+	pattern, in which case all matching files are used, sorted by
+	their filenames. If a 'FILENAME' is a relative path, it is
+	treated relative to the directory containing the file from
+	which the *include* stanza was read, even if the inclusion has
+	been nested. Inclusions may be nested to any level.
+
 *index-file* ['FILENAME'...]::
 
 	The given 'FILENAMEs' are used for finding index files (see
diff --git a/doc/patplex.doc b/doc/patplex.doc
index dbe8e96..2d9acbd 100644
--- a/doc/patplex.doc
+++ b/doc/patplex.doc
@@ -46,9 +46,9 @@ Should the global and the given configuration files conflict, the
 directives from the given file take precedence.
 
 The configuration files follow the same general format as for
-*dirplex*(1), though the recognized stanzas differ. The *child* and
-*fchild* stanzas are also shared with *dirplex*(1), so see its manpage
-for a description thereof.
+*dirplex*(1), though the recognized stanzas differ. The *child*,
+*fchild* and *include* stanzas are also shared with *dirplex*(1), so
+see its manpage for a description thereof.
 
 *patplex* recognizes the *match* stanza, which takes no arguments, but
 must contain at least one follow-up line to specify match rules. All
-- 
2.11.0


From 90b0ba0f9d93e454cc08a566b718abdcbfd0d9f6 Mon Sep 17 00:00:00 2001
From: Fredrik Tolf <fredrik@dolda2000.com>
Date: Sat, 30 Apr 2011 08:45:29 +0200
Subject: [PATCH 11/16] Imposed some limits on request parts.

---
 lib/req.c      | 4 ++++
 src/htparser.c | 6 ++++++
 2 files changed, 10 insertions(+)

diff --git a/lib/req.c b/lib/req.c
index a3e7273..da8e3f0 100644
--- a/lib/req.c
+++ b/lib/req.c
@@ -106,12 +106,16 @@ int parseheaders(struct hthead *head, FILE *in)
 {
     int c, state;
     struct charbuf name, val;
+    size_t tsz;
     
     bufinit(name);
     bufinit(val);
     state = 0;
+    tsz = 0;
     while(1) {
 	c = fgetc(in);
+	if(++tsz >= 65536)
+	    goto fail;
     again:
 	if(state == 0) {
 	    if(c == '\r') {
diff --git a/src/htparser.c b/src/htparser.c
index 5179a25..1ed9175 100644
--- a/src/htparser.c
+++ b/src/htparser.c
@@ -77,6 +77,8 @@ static struct hthead *parsereq(FILE *in)
 	    goto fail;
 	} else {
 	    bufadd(method, c);
+	    if(method.d >= 128)
+		goto fail;
 	}
     }
     while(1) {
@@ -87,6 +89,8 @@ static struct hthead *parsereq(FILE *in)
 	    goto fail;
 	} else {
 	    bufadd(url, c);
+	    if(url.d >= 65536)
+		goto fail;
 	}
     }
     while(1) {
@@ -98,6 +102,8 @@ static struct hthead *parsereq(FILE *in)
 	    goto fail;
 	} else {
 	    bufadd(ver, c);
+	    if(ver.d >= 128)
+		goto fail;
 	}
     }
     bufadd(method, 0);
-- 
2.11.0


From e2b404ab58259bb538e2d785d6ebbe9e9263bbef Mon Sep 17 00:00:00 2001
From: Fredrik Tolf <fredrik@dolda2000.com>
Date: Sat, 30 Apr 2011 08:53:14 +0200
Subject: [PATCH 12/16] htparser: Implemented resource limits for response
 headers as well.

---
 src/htparser.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/src/htparser.c b/src/htparser.c
index 1ed9175..ba76d01 100644
--- a/src/htparser.c
+++ b/src/htparser.c
@@ -146,6 +146,8 @@ static struct hthead *parseresp(FILE *in)
 	    goto fail;
 	} else {
 	    bufadd(ver, c);
+	    if(ver.d >= 128)
+		goto fail;
 	}
     }
     while(1) {
@@ -156,6 +158,8 @@ static struct hthead *parseresp(FILE *in)
 	    goto fail;
 	} else {
 	    code = (code * 10) + (c - '0');
+	    if(code >= 10000)
+		goto fail;
 	}
     }
     while(1) {
@@ -167,6 +171,8 @@ static struct hthead *parseresp(FILE *in)
 	    goto fail;
 	} else {
 	    bufadd(msg, c);
+	    if(msg.d >= 512)
+		goto fail;
 	}
     }
     bufadd(msg, 0);
-- 
2.11.0


From 64a9096a7cd4acf3bbf4a7bf4a214f4b5cca7fdc Mon Sep 17 00:00:00 2001
From: Fredrik Tolf <fredrik@dolda2000.com>
Date: Sat, 30 Apr 2011 09:20:14 +0200
Subject: [PATCH 13/16] htparser: Handle absolute-URI requests.

---
 src/htparser.c | 47 +++++++++++++++++++++++++++++++++++++++++------
 1 file changed, 41 insertions(+), 6 deletions(-)

diff --git a/src/htparser.c b/src/htparser.c
index ba76d01..bc8f7fd 100644
--- a/src/htparser.c
+++ b/src/htparser.c
@@ -240,12 +240,50 @@ static int hasheader(struct hthead *head, char *name, char *val)
     return(!strcasecmp(hd, val));
 }
 
+static int canonreq(struct hthead *req)
+{
+    char *p, *p2, *r;
+    int n;
+
+    if(req->url[0] == '/') {
+	replrest(req, req->url + 1);
+	if((p = strchr(req->rest, '?')) != NULL)
+	    *p = 0;
+	return(1);
+    }
+    if((p = strstr(req->url, "://")) != NULL) {
+	n = p - req->url;
+	if(((n == 4) && !strncasecmp(req->url, "http", 4)) ||
+	   ((n == 5) && !strncasecmp(req->url, "https", 5))) {
+	    if(getheader(req, "host"))
+		return(0);
+	    p += 3;
+	    if((p2 = strchr(p, '/')) == NULL) {
+		headappheader(req, "Host", p);
+		free(req->url);
+		req->url = sstrdup("/");
+	    } else {
+		r = sstrdup(p2);
+		*(p2++) = 0;
+		headappheader(req, "Host", p);
+		free(req->url);
+		req->url = r;
+	    }
+	    replrest(req, req->url + 1);
+	    if((p = strchr(req->rest, '?')) != NULL)
+		*p = 0;
+	    return(1);
+	}
+    }
+    return(0);
+}
+
 void serve(FILE *in, struct conn *conn)
 {
     int pfds[2];
     FILE *out;
     struct hthead *req, *resp;
-    char *hd, *p;
+    char *hd;
     off_t dlen;
     
     out = NULL;
@@ -253,11 +291,8 @@ void serve(FILE *in, struct conn *conn)
     while(1) {
 	if((req = parsereq(in)) == NULL)
 	    break;
-	replrest(req, req->url);
-	if(req->rest[0] == '/')
-	    replrest(req, req->rest + 1);
-	if((p = strchr(req->rest, '?')) != NULL)
-	    *p = 0;
+	if(!canonreq(req))
+	    break;
 	
 	if((conn->initreq != NULL) && conn->initreq(conn, req))
 	    break;
-- 
2.11.0


From 01bb89b0e602cfcc259f90388eb8e06834b60c48 Mon Sep 17 00:00:00 2001
From: Fredrik Tolf <fredrik@dolda2000.com>
Date: Sun, 1 May 2011 20:57:23 +0200
Subject: [PATCH 14/16] doc: Fixed markup error.

---
 doc/htparser.doc | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/doc/htparser.doc b/doc/htparser.doc
index 757f8cd..320e549 100644
--- a/doc/htparser.doc
+++ b/doc/htparser.doc
@@ -37,7 +37,7 @@ PORT SPECIFICATION
 Currently, the available 'HANDLERs' are *plain* and *ssl*, for
 handling plain TCP connections and SSL/TLS-protected connections,
 respectively. For details regarding the arguments that each handler
-accept, simply run *htparser* with 'HANDLER':*help*. For example, the
+accept, simply run *htparser* with 'HANDLER'*:help*. For example, the
 command "`htparser ssl:help`" will display help for the *ssl* handler to
 standard output and then exit.
 
-- 
2.11.0


From 16c2bec346ae486bd09d0b18ab276b4c89005cad Mon Sep 17 00:00:00 2001
From: Fredrik Tolf <fredrik@dolda2000.com>
Date: Sun, 1 May 2011 22:04:18 +0200
Subject: [PATCH 15/16] doc: Fixed up various parts of the documentation a bit.

---
 doc/ashd.doc     |  6 +++---
 doc/callcgi.doc  |  2 +-
 doc/dirplex.doc  | 50 ++++++++++++++++++++++++++++++--------------------
 doc/htparser.doc | 19 +++++++++++++------
 4 files changed, 47 insertions(+), 30 deletions(-)

diff --git a/doc/ashd.doc b/doc/ashd.doc
index eb081b7..f2790ba 100644
--- a/doc/ashd.doc
+++ b/doc/ashd.doc
@@ -15,9 +15,9 @@ technically. If you want a brief overview, please see the homepage at
 The basic premise of ashd is that of standard Unix philosophy; it
 consists of a number of different programs, each specialized to one
 precise task, passing HTTP requests around to each other in a manner
-akin to standard Unix pipelines. This document describes the protocols
-and conventions used between such programs that allows them to
-interoperate.
+akin to standard Unix pipelines. This document describes the set of
+protocols and conventions used between such programs that allows them
+to interoperate.
 
 REQUESTS
 --------
diff --git a/doc/callcgi.doc b/doc/callcgi.doc
index 4564279..2a61e10 100644
--- a/doc/callcgi.doc
+++ b/doc/callcgi.doc
@@ -47,4 +47,4 @@ Fredrik Tolf <fredrik@dolda2000.com>
 
 SEE ALSO
 --------
-*dirplex*(1), *ashd*(7), RFC 3875
+*dirplex*(1), *ashd*(7), *callscgi*(1), *callfcgi*(1), RFC 3875
diff --git a/doc/dirplex.doc b/doc/dirplex.doc
index 7d1e41f..362a5d7 100644
--- a/doc/dirplex.doc
+++ b/doc/dirplex.doc
@@ -126,7 +126,7 @@ The follow configuration directives are recognized:
 
 *include* ['FILENAME'...]::
 
-	Read the the named files and act as if their contents stood in
+	Read the named files and act as if their contents stood in
 	place of the *include* stanza. A 'FILENAME' may be a glob
 	pattern, in which case all matching files are used, sorted by
 	their filenames. If a 'FILENAME' is a relative path, it is
@@ -163,7 +163,7 @@ The follow configuration directives are recognized:
 *fchild* 'NAME'::
 
 	Declares a named, transient request handler (see *ashd*(7) for
-	a more detailed description of persistent handlers). It must
+	a more detailed description of transient handlers). It must
 	contain exactly one follow-up line, *exec* 'PROGRAM'
 	['ARGS'...], specifying the program to execute and the
 	arguments to pass it. In addition to the specified arguments,
@@ -187,7 +187,8 @@ The follow configuration directives are recognized:
 	remaining rest string, to the specified 'HANDLER', which must
 	by a named request handler specified either in the same
 	`.htrc` file or elsewhere. The *capture* directive accepts no
-	follow-up lines.
+	follow-up lines. Note that the `X-Ash-File` header is not
+	added to requests passed via *capture* directives.
 
 MATCHING
 --------
@@ -228,7 +229,7 @@ The following rules are recognized:
 *default*::
 
 	Matches if and only if no *match* stanza without a *default*
-	rule has matched.
+	rule matches (in any configuration file).
 
 *local*::
 
@@ -259,26 +260,26 @@ following actions are recognized:
 404 RESPONSES
 -------------
 
-Any of the following cases will result in a 404 response being sent to
-the client.
+A HTTP 404 response is sent to the client if
 
- * Failure of the mapping procedure to find a matching physical file.
- * Presence of a path element during mapping that begins with a dot.
- * A path element which, after URL unescaping, contains slashes.
- * The mapping procedure finding a file which is neither a directory
-   nor a regular file.
- * Presence of a non-final but empty path element during mapping.
- * A physical file having been found which is not being matched by any
+ * The mapping procedure fails to find a matching physical file;
+ * A path element is encountered during mapping which, after URL
+   unescaping, either begins with a dot or contains slashes;
+ * The mapping procedure finds a file which is neither a directory nor
+   a regular file;
+ * An empty, non-final path element is encountered during mapping; or
+ * The mapping procedure results in a file which is not matched by any
    *match* stanza.
 
-*dirplex* will send a built-in 404 response by default, but any
+By default, *dirplex* will send a built-in 404 response, but any
 `.htrc` file or global configuration may define a request handler
 named `.notfound` to customize the behavior. Note that, unlike
 successful requests, such a handler will not be passed the
 `X-Ash-File` header.
 
 The built-in `.notfound` handler can also be used in *match* or
-*capture* stanzas.
+*capture* stanzas (for example, to restrict access to certain files or
+directories).
 
 EXAMPLES
 --------
@@ -296,8 +297,16 @@ scripts can be used with the following configuration, using the
 *callcgi*(1) program.
 
 --------
+# To use plain CGI, which uses more resources per handled request,
+# but less static resources:
 fchild php
   exec callcgi -p php-cgi
+
+# To use FastCGI, which keeps PHP running at all times, but uses less
+# resources per handled request:
+child php
+  exec callfcgi multifscgi 5 php-cgi
+
 match
   filename *.php
   handler php
@@ -312,11 +321,12 @@ match directory
   fork htls
 --------
 
-If you want an entire directory to be dedicated to some external SCGI
-script engine, you can use the *callscgi*(1) program to serve it as
-follows. Note that *callscgi*, and therefore the script engine itself,
-is started in the directory itself, so that arbitrary code modules or
-data files can be put directly in that directory and easily found.
+The following configuration can be placed in a `.htrc` file in order
+to dedicate the directory containing that file to some external SCGI
+script engine. Note that *callscgi*, and therefore the script engine
+itself, is started in the directory itself, so that arbitrary code
+modules or data files can be put directly in that directory and easily
+found.
 
 --------
 child foo
diff --git a/doc/htparser.doc b/doc/htparser.doc
index 320e549..698e559 100644
--- a/doc/htparser.doc
+++ b/doc/htparser.doc
@@ -54,9 +54,9 @@ OPTIONS
 *-S*::
 
 	Log messages to *syslog*(3) instead of standard error. Also
-	sets the environment ASHD_USESYSLOG environment variable in
-	the root handler process, which indicates to the standard ashd
-	programs to do the same thing.
+	sets the ASHD_USESYSLOG environment variable in the root
+	handler process, which indicates to the standard ashd programs
+	to do the same thing.
 
 *-f*::
 
@@ -85,8 +85,8 @@ EXAMPLES
 
 `htparser plain -- dirplex /srv/www`::
 
-	This simple invocation will simply listen for HTTP requests on
-	port 80 and use *dirplex*(1) to serve files from the /srv/www
+	This simple invocation will listen for HTTP requests on port
+	80 and use *dirplex*(1) to serve files from the /srv/www
 	directory.
 
 `htparser plain:port=8080 -- dirplex /srv/www`::
@@ -97,7 +97,7 @@ EXAMPLES
 `htparser plain ssl:cert=/etc/ssl/private/web.pem -- dirplex /srv/www`::
 
 	The same as above, but will listen on port 443 for SSL
-	connections as well. The file `/etc/ssl/privte/web.pem` needs
+	connections as well. The file `/etc/ssl/private/web.pem` needs
 	to contain both the server certificate and its private key.
 
 `htparser plain -- sudo -u www-user dirplex /srv/www`::
@@ -116,6 +116,13 @@ EXAMPLES
 	file system, so that it can start other handler programs as
 	needed.
 
+`htparser -f plain -- errlogger -n ashd dirplex /srv/www`::
+
+	The same as the first example, but will daemonize and use the
+	*errlogger*(1) program to ensure that any errors or other
+	messages written by any handler program to its stderr are
+	recorded in the *syslog*(3).
+
 X-ASH HEADERS
 -------------
 
-- 
2.11.0


From 82e84769bb12dd531f44f83498f4ef2a146da63e Mon Sep 17 00:00:00 2001
From: Fredrik Tolf <fredrik@dolda2000.com>
Date: Sun, 1 May 2011 22:08:21 +0200
Subject: [PATCH 16/16] Updated changelog.

---
 ChangeLog | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/ChangeLog b/ChangeLog
index 3b03efb..a8decb6 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,8 @@
+Version 0.8:
+
+	* Various bug fixes, protocol compliance fixes, tunings,
+	  documentation improvements and other minor improvements.
+
 Version 0.7:
 
 	* Added some configuration examples.
-- 
2.11.0