This patch provides the ability to receive a callback on the completion of a
request. The callback takes place immediately before the request's resources
are released.
evhttp_write_buffer() used by evhttp_send_reply_chunk() can take callback
executed when (part of) the buffer has been written. Using this callback to
schedule the next chunk avoids buffering large amounts of data in memory.
Basically tcp final handshake looks like this:
(C - client, S - server)
ACK[C] - FIN/ACK[S] - FIN/ACK[S] - ACK [C]
However there are servers, that didn't close connection like this,
while it is still _considered_ as valid, and using libevent http layer
we can do requests to such servers.
Modified handshake:
(C - client, S - server)
ACK[C] - RST/ACK[S] - RST/ACK[S]
And in this case we can't extract IP address from socket, because it is
already closed, and getpeername() will return: "transport endpoint is not connected".
So we need to store address that we are connecting to, after we know it,
and that is what this patch do.
I have reproduced it, however it have some extra packages.
(I will try to fix it)
https://github.com/azat/nfq-examples/blob/master/nfqnl_rst_fin.c
It is useful to know why you callback called with NULL (i.e. it failed),
for example if you set max_body with evhttp_connection_set_max_body_size()
you must know that it failed because of body was longer than this size.
(Commit message tweaked by Nick)
This is needed to be able to read the response code line especially
when acting as an http client using evhttp_make_request.
(patched by nickm to make the return value const)
Previously we used include-guards with names like _EVENT2_EVENT_H_.
But C reserves macros beginning with an underscore for use by the
system. This patch converts all include guards for files like
include/event2/<fname.h> to be of form EVENT2_<FNAME_H>_INCLUDED_,
and all Libevent 1.x headers in include/<fname.h> to be of the form
EVENT1_<FNAME_H>_INCLUDED_, and all internal libevent headers with
names like <fname.h> to the form <FNAME_H>_INCLUDED_.
FNAME_H is here derived from fname.h by replacing every
non-macro-usable character in fname.h with an underscore, and
putting every remaining character in uppercase.
This is an automatic conversion. The script that produced was made by
running the following script over all header files:
=====
#!/usr/bin/perl -w
# Run this on every .h file except config.h, sys/queue.h, WIN32/event2/event-config.h
use strict;
my %macros = ();
my %skipped = ();
FILE: for my $fn (@ARGV) {
my $f = $fn;
if ($fn !~ /^\.\//) {
$f = "./$fn";
}
if ($f eq './config.h' or
$f =~ m#/tree.h$# or
$f =~ m#/queue.h# or
$f =~ m#/event-config.h# or
$f =~ m#/evconfig-private.h#) {
$skipped{$fn} = 1;
next FILE;
}
$skipped{$fn} = 0;
open(F, $fn);
while (<F>) {
if (/^#ifndef ([A-Za-z0-9_]+)/) {
$macros{$fn} = $1;
next FILE;
}
}
}
print "#!/usr/bin/perl -w -i -p\n\n";
for my $fn (@ARGV) {
if (! exists $macros{$fn}) {
print "# No macro known for $fn!\n" if (!$skipped{$fn});
} else {
if ($macros{$fn} !~ /_H_?$/) {
print "# Weird macro for $fn...\n";
}
my $goodmacro = uc $fn;
$goodmacro =~ s#^\./##;
$goodmacro =~ s#INCLUDE/EVENT2#EVENT2#;
$goodmacro =~ s#INCLUDE/#EVENT1_#;
$goodmacro =~ s#TEST/##;
$goodmacro =~ s#[\/\-\.]#_#g;
print "s/(?<![A-Za-z0-9_])$macros{$fn}(?![A-Za-z0-9_])/${goodmacro}_INCLUDED_/g;\n"
}
}
=== And then running the script below that it generated over all
=== the .h files again
#!/usr/bin/perl -w -i -p
s/(?<![A-Za-z0-9_])_BUFFEREVENT_INTERNAL_H_(?![A-Za-z0-9_])/BUFFEREVENT_INTERNAL_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_CHANGELIST_H_(?![A-Za-z0-9_])/CHANGELIST_INTERNAL_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_DEFER_INTERNAL_H_(?![A-Za-z0-9_])/DEFER_INTERNAL_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVBUFFER_INTERNAL_H_(?![A-Za-z0-9_])/EVBUFFER_INTERNAL_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT_INTERNAL_H_(?![A-Za-z0-9_])/EVENT_INTERNAL_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVMAP_H_(?![A-Za-z0-9_])/EVMAP_INTERNAL_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVRPC_INTERNAL_H_(?![A-Za-z0-9_])/EVRPC_INTERNAL_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVSIGNAL_H_(?![A-Za-z0-9_])/EVSIGNAL_INTERNAL_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVTHREAD_INTERNAL_H_(?![A-Za-z0-9_])/EVTHREAD_INTERNAL_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT_HT_H(?![A-Za-z0-9_])/HT_INTERNAL_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_HTTP_INTERNAL_H_(?![A-Za-z0-9_])/HTTP_INTERNAL_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVDNS_H_(?![A-Za-z0-9_])/EVENT1_EVDNS_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT_H_(?![A-Za-z0-9_])/EVENT1_EVENT_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT2_BUFFER_H_(?![A-Za-z0-9_])/EVENT2_BUFFER_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT2_BUFFER_COMPAT_H_(?![A-Za-z0-9_])/EVENT2_BUFFER_COMPAT_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT2_BUFFEREVENT_H_(?![A-Za-z0-9_])/EVENT2_BUFFEREVENT_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT2_BUFFEREVENT_COMPAT_H_(?![A-Za-z0-9_])/EVENT2_BUFFEREVENT_COMPAT_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT2_BUFFEREVENT_SSL_H_(?![A-Za-z0-9_])/EVENT2_BUFFEREVENT_SSL_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT2_BUFFEREVENT_STRUCT_H_(?![A-Za-z0-9_])/EVENT2_BUFFEREVENT_STRUCT_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT2_DNS_H_(?![A-Za-z0-9_])/EVENT2_DNS_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT2_DNS_COMPAT_H_(?![A-Za-z0-9_])/EVENT2_DNS_COMPAT_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT2_DNS_STRUCT_H_(?![A-Za-z0-9_])/EVENT2_DNS_STRUCT_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT2_EVENT_H_(?![A-Za-z0-9_])/EVENT2_EVENT_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT2_EVENT_COMPAT_H_(?![A-Za-z0-9_])/EVENT2_EVENT_COMPAT_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT2_EVENT_STRUCT_H_(?![A-Za-z0-9_])/EVENT2_EVENT_STRUCT_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT2_HTTP_H_(?![A-Za-z0-9_])/EVENT2_HTTP_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT2_HTTP_COMPAT_H_(?![A-Za-z0-9_])/EVENT2_HTTP_COMPAT_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT2_HTTP_STRUCT_H_(?![A-Za-z0-9_])/EVENT2_HTTP_STRUCT_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT2_EVENT_KEYVALQ_STRUCT_H_(?![A-Za-z0-9_])/EVENT2_KEYVALQ_STRUCT_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT2_LISTENER_H_(?![A-Za-z0-9_])/EVENT2_LISTENER_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT2_RPC_H_(?![A-Za-z0-9_])/EVENT2_RPC_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT2_RPC_COMPAT_H_(?![A-Za-z0-9_])/EVENT2_RPC_COMPAT_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT2_RPC_STRUCT_H_(?![A-Za-z0-9_])/EVENT2_RPC_STRUCT_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT2_TAG_H_(?![A-Za-z0-9_])/EVENT2_TAG_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT2_TAG_COMPAT_H_(?![A-Za-z0-9_])/EVENT2_TAG_COMPAT_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT2_THREAD_H_(?![A-Za-z0-9_])/EVENT2_THREAD_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT2_UTIL_H_(?![A-Za-z0-9_])/EVENT2_UTIL_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVHTTP_H_(?![A-Za-z0-9_])/EVENT1_EVHTTP_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVRPC_H_(?![A-Za-z0-9_])/EVENT1_EVRPC_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVUTIL_H_(?![A-Za-z0-9_])/EVENT1_EVUTIL_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT_IOCP_INTERNAL_H(?![A-Za-z0-9_])/IOCP_INTERNAL_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT_IPV6_INTERNAL_H(?![A-Za-z0-9_])/IPV6_INTERNAL_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_LOG_H_(?![A-Za-z0-9_])/LOG_INTERNAL_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_MIN_HEAP_H_(?![A-Za-z0-9_])/MINHEAP_INTERNAL_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT_MM_INTERNAL_H(?![A-Za-z0-9_])/MM_INTERNAL_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_RATELIM_INTERNAL_H_(?![A-Za-z0-9_])/RATELIM_INTERNAL_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_STRLCPY_INTERNAL_H_(?![A-Za-z0-9_])/STRLCPY_INTERNAL_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_REGRESS_H_(?![A-Za-z0-9_])/REGRESS_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_TESTUTILS_H(?![A-Za-z0-9_])/REGRESS_TESTUTILS_H_INCLUDED_/g;
# Weird macro for test/tinytest.h...
s/(?<![A-Za-z0-9_])TINYTEST_H_INCLUDED_(?![A-Za-z0-9_])/TINYTEST_H_INCLUDED_/g;
# No macro known for test/tinytest_local.h!
# Weird macro for test/tinytest_macros.h...
s/(?<![A-Za-z0-9_])TINYTEST_MACROS_H_INCLUDED_(?![A-Za-z0-9_])/TINYTEST_MACROS_H_INCLUDED_/g;
s/(?<![A-Za-z0-9_])_EVENT_UTIL_INTERNAL_H(?![A-Za-z0-9_])/UTIL_INTERNAL_H_INCLUDED_/g;
Applies the function specified in the first argument to all
evhttp_bound_sockets associated with a struct evhttp. The user
must not attempt to free or remove any connections, sockets or
listeners in the callback function.
If the EVHTTP_URI_NONCONFORMANT flag is passed in (which it is when
parsing URIs we get over the wire), then we relax our checks a lot.
Specifically, we do nothing to check for correct characters in the
path, query, and fragment parts of such a URI.
We could do much more here: we could relax our hostname requirements,
deal with spaces differently/better, trap some errors but not others,
etc. But this should solve the worst user-agent compatibility issues
for now; the other issues can wait for a later release.
evhttp needs to be mindful of all hostnames and addresses that clients
use to contact the main server and vhosts to know the difference between
proxy requests and non-proxy requests.
This patch defines enumerators for all HTTP methods that exist
(including PATCH introduced in RFC 5789).
It also makes them bit-masky (that's not a word, is it?), breaking
binary- but not source-code compatibility.
evhttp now stores a bitmask specifying for which methods requests to
dispatch and which ones to reject with "405 Method Not Allowed".
By default that's the ones we currently have (GET, POST, HEAD, PUT,
DELETE), thereby keeping functional compatibility (besides the minor
change that one of the other methods will now cause 405 instead of
400. But I believe that could even be considered a bug-fix).
evhttp is extended by evhttp_set_allowed_methods() with which the
user can change that bitmask.
no regressions here and my test-app still works. Haven't yet
actually tested any of the new methods.
What's obviously missing here is the special logic for the methods:
OPTIONS: We should be fine here - I believe our current dispatch
logic should work fine. Some convenience functions would be fine
though.
TRACE: I'm pretty certain we should never dispatch this to the
callbacks and simply implement the necessary functionality built-in.
CONNECT: Pretty straight-forward to implement (and considering the
framework in which we implement it very efficient too). Should
probably go built-in.
PATCH: Except for checking the RFC against our pre-dispatch logic
(there just might be some "MUST not have Some-Header" lurking
somewhere) there is nothing to be done here, this is completely up
to the user. Nothing to do.
We had to turn a couple of 32-bit size arguments into 64-bit arguments
or size_t arguments (since otherwise we would have had to do it post
2.0.x-stable, and that would be worse).
