Commits · 05ce301f3571e3e690792a189cb927328163f0bc · submodule / libzmq

04 Nov, 2011 4 commits
- Merge branch 'master' of github.com:zeromq/libzmq · 05ce301f
  Martin Sustrik authored Nov 04, 2011
  
  05ce301f
- Merge pull request #220 from 250bpm/HEAD · 6cdd7204
  sustrik authored Nov 04, 2011
```
Refactoring
```
  6cdd7204
- Linking exception corrected · e9c3a227
  Martin Sustrik authored Nov 04, 2011
```
Correctly states 'copyright holders' instead of 'iMatix'.
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
```
  e9c3a227
- Identities re-introduced · a4843b65
  Martin Sustrik authored Nov 04, 2011
```
However, the "durable socket" behaviour wasn't re-added.
Identities are used solely for routing in REQ/REP pattern.
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
```
  a4843b65
02 Nov, 2011 1 commit
- ZMQ_IDENTITY option re-introduced · d20ea25b
  Martin Sustrik authored Nov 02, 2011
```
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
```
  d20ea25b
01 Nov, 2011 5 commits
- Copyright dates adjusted to reflect reality · 8e21d64c
  Martin Sustrik authored Nov 01, 2011
```
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
```
  8e21d64c
- Miru copyrights added · 8a0f5f76
  Martin Sustrik authored Nov 01, 2011
```
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
```
  8a0f5f76
- LABELs removed from the documentation · 1c071f54
  Martin Sustrik authored Nov 01, 2011
```
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
```
  1c071f54
- Tests print their name before running · 75635189
  Martin Sustrik authored Nov 01, 2011
```
This makes finding out which test have hung-up easier.
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
```
  75635189
- LABELS and COMMANDs removed · 7842c710
  Martin Sustrik authored Nov 01, 2011
```
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
```
  7842c710
31 Oct, 2011 5 commits
- VTCP transport removed · 626099aa
  Martin Sustrik authored Oct 31, 2011
```
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
```
  626099aa
- 250bpm copyrights added · ac7717b7
  Martin Sustrik authored Oct 31, 2011
```
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
```
  ac7717b7
- Revert the early dropping of request and replies for disconnected clients · 4b832ea3
  Martin Sustrik authored Oct 31, 2011
```
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
```
  4b832ea3
- ROUTER and DEALER are aliases for XREP and XREQ · 9227c389
  Martin Sustrik authored Oct 31, 2011
```
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
```
  9227c389
- New style ROUTER socket removed. · dee8b236
  Martin Sustrik authored Oct 31, 2011
```
Signed-off-by: Martin Sustrik <sustrik@turist.(none)>
```
  dee8b236
29 Oct, 2011 2 commits

Perry Kundert added to the AUTHORS file · 541b83bc
Martin Sustrik authored Oct 29, 2011
```
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
```
541b83bc

Corrected discarding of remainder of message when request ID invalid · 09574a61

Perry Kundert authored Oct 29, 2011

When zmq::req_t::xrecv detects that a response has no request ID
label, or the ID is the wrong size, it would return an EAGAIN, but
would not discard the remainder of the message. This could allow the
remainder of the message to incorrectly "leak" into a future response,
if it is crafted to look like a reply with a valid response ID.
Discard all remaining message blocks, if the ID is invalid in any way.

09574a61

27 Oct, 2011 8 commits
- Missing bracket added · 52bab422
  Bernd Prager authored Oct 27, 2011
```
Signed-off: Martin Sustrik <sustrik@250bpm.com>
```
  52bab422
- Ben Gray added to the AUTHORS file · 9b3e61a1
  Martin Sustrik authored Oct 27, 2011
```
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
```
  9b3e61a1
- Patch for issue LIBZMQ-275. Dealing with VSM in distribution when pipes fail to write. · 9e000c8f
  Ben Gray authored Oct 27, 2011
```
Signed-off-by: Ben Gray <ben@benjamg.com>
```
  9e000c8f
- Paul Betts added to the AUTHORS file · 68ab5f87
  Martin Sustrik authored Oct 27, 2011
```
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
```
  68ab5f87
- Enable exceptions raising on assert on Win32 · 1b706ac0
  Paul Betts authored Oct 27, 2011
```
This patch changes the Win32 version to call RaiseException instead of abort
(which eventually calls TerminateProcess). This allows crash dumps to be sent
correctly instead of the process disappearing.
Signed-off-by: Paul Betts <paul@paulbetts.org>
```
  1b706ac0
- Enable exceptions raising on assert on Win32 · a8362abf
  Martin Sustrik authored Oct 27, 2011
```
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
```
  a8362abf
- Merge branch 'master' of github.com:zeromq/libzmq · a70bea01
  Martin Sustrik authored Oct 27, 2011
  
  a70bea01
- Bug in kqueue poller fixed (issue 261) · b3cda2ad
  Martin Sustrik authored Oct 27, 2011
```
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
```
  b3cda2ad
26 Oct, 2011 1 commit
- Added compile-time test for SOCK_CLOEXEC · 6c1b50cf
  Mikko Koppanen authored Oct 26, 2011
```
Signed-off-by: Mikko Koppanen <mkoppanen@php.net>
```
  6c1b50cf
06 Oct, 2011 1 commit

Default HWMs are set to 1000 · d31792e6

Martin Sustrik authored Oct 06, 2011

This patch is meant to prevent users from running out of memory
when using 0MQ in the default configuration.
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>

d31792e6

04 Oct, 2011 2 commits

New upstream OpenPGM maintenance release 5.1.118. · f5f1a8b1
Steven McCoy authored Oct 04, 2011
```
Signed-off-by: Steven McCoy <steven.mccoy@miru.hk>
```
f5f1a8b1

Re-add PGM rate limiting · cc5d3160

Steven McCoy authored Oct 04, 2011

It was accidently lost in commit fcfad568, equivalent
to commits 2ddb106 & b62dd40 in version 2.1.  Changes semantics
from TXW_MAX_RATE to ODATA_MAX_RATE.
Signed-off-by: Steven McCoy <steven.mccoy@miru.hk>

cc5d3160

29 Sep, 2011 3 commits
- Use single port for creating signalers on Windows · 81da391e
  Martin Sustrik authored Sep 29, 2011
```
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
```
  81da391e
- Bug in subscription matching fixed (issue 263) · 7a10bbe7
  Martin Sustrik authored Sep 29, 2011
```
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
```
  7a10bbe7
- Make sure new ROUTER socket honours POLLIN for cmd messages · 34b114d1
  Jon Dyte authored Sep 29, 2011
```
Signed-off-by: Jon Dyte <jon@totient.co.uk>
```
  34b114d1
28 Sep, 2011 1 commit
- Assert fixed in XREP & ROUTER when sending to terminating pipe (issue 258) · 8485a5e5
  Martin Sustrik authored Sep 28, 2011
```
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
```
  8485a5e5
26 Sep, 2011 1 commit
- Bug in matching algorithm fixed · d726120e
  Martin Sustrik authored Sep 26, 2011
```
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
```
  d726120e
18 Sep, 2011 1 commit
- MSVC build fixed · 25cc25e9
  Martin Sustrik authored Sep 18, 2011
```
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
```
  25cc25e9
16 Sep, 2011 3 commits

More bugs in mtrie fixed · e170136a

Martin Sustrik authored Sep 16, 2011

Aside of fixing couple of corner cases this patch turns the 'match'
function in mtrie from recursive to iterative.
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>

e170136a

Bug in mtrie fixed · 5936379b
Martin Sustrik authored Sep 16, 2011
```
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>
```
5936379b

Check message syntax in REQ asynchronously · 06bdf2c4

Martin Sustrik authored Sep 16, 2011

This patch adds support for checking messages as they arrive
(as opposed to when they are recv'd by the user) and drop
the connection if they are malformed.

It also uses this new feature to check for validity of inbound
messages in REQ socket.
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>

06bdf2c4

15 Sep, 2011 1 commit

Session class separated into socket-type-specific sessions · f78d9b6b

Martin Sustrik authored Sep 15, 2011

This is a preliminary patch allowing for socket-type-specific
functionality in the I/O thread. For example, message format
can be checked asynchronously and misbehaved connections dropped
straight away.
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>

f78d9b6b

14 Sep, 2011 1 commit

Minor optimisation in message distribution algorithm · 78b02d14

Martin Sustrik authored Sep 14, 2011

If several of the outbound pipes become passive while sending
a single message, the refcount on the message is adjusted
once only, not multiple times. It's an atomic operation so
the cost is not negligible.
Signed-off-by: Martin Sustrik <sustrik@250bpm.com>

78b02d14