Removed unneeded initializers in lm_connection_new

More styling

More style fixes

Style fixes

Fixed root Makefile

Fixed root Makefile

Moved Makefile.decl to build sub directory and created a Makefile.am.lm All Makefile's should include the new Makefile.am.lm which in turns includes common and test Makefiles.

More cleaning up in Makefiles

More cleaning up in Makefile

Further cleanup of Makefile.am

Updated the build system to use a build directory and a common marshal Makefile. Inspired by Richards commit to Devhelp in order to clean up the build system a bit. Makefile setup for generating signal marshal are now defined in build/Makefile.am.marshal which is included in the loudmough/Makefile.am file.

Removed old .cvsignore file

Removed unused defines in lm-connection.c

Went over all of the files and made sure they all were indented the same way. This means from now on use 4 space instead of tabs everywhere for indentation.

Make the emacs header changes to tests and examples

Change the tab width to 4 steps in the emacs headers

Change the emacs format string to use space instead of tab for indentation

Allow replies to non-iq stanzas. [#22] This changes so that any stanza type will check if there is a reply handler waiting for the incoming ID. When implementing for example XEP-0184 (Message Receipts) there will be returned message stanzas that needs to be given to a reply handler. committer: Mikael Hallendal <micke@imendio.com>

Allow replies to non-iq stanzas. [#22] This changes so that any stanza type will check if there is a reply handler waiting for the incoming ID. When implementing for example XEP-0184 (Message Receipts) there will be returned message stanzas that needs to be given to a reply handler. [#22 tagged:committed responsible:Hallski milestone:none state:resolved] committer: Mikael Hallendal <micke@imendio.com>

Updated copyright years

Use SASL before old style auth if both are supported. [#19] Fixes an issue when server provides both old style and SASL authentication mechanisms. Loudmouth currently only checked if old style auth were supported and assumed that SASL weren't in those cases. [#19 tagged:committed responsible:Hallski milestone:none state:resolved] committer: Mikael Hallendal <micke@imendio.com>

Use SASL before old style auth if both are supported. [#19] Fixes an issue when server provides both old style and SASL authentication mechanisms. Loudmouth currently only checked if old style auth were supported and assumed that SASL weren't in those cases. [#19 tagged:committed responsible:Hallski milestone:none state:resolved] committer: Mikael Hallendal <micke@imendio.com>

Fixed a crash LmFeaturePing. Missed sending the LmFeaturePing object as user data to the message handler.

Cleaned up the moved XMPP Ping code from LmConnection and hooked up the LmFeaturePing object. Removed the code from LmConnection and instead use the new LmFeaturePing object.

Added the code from LmConnection to send and receive the XMPP Ping to LmFeaturePing. Copied and refactored the code to fit into LmFeaturePing instead of LmConnection.

Added properties, signals and start/stop calls to LmFeaturePing.

Added stubs for LmFeaturePing which will hold the XMPP Ping feature. To keep lm-connection.c as clean as possible the XMPP Ping (XEP-0199) code will live in lm-feature-ping.[ch] instead.

Use XMPP Ping (XEP-0199) instead of white space ping for keep alive. [#15] This patch introduces the use of XMPP Pings (defined in XEP-0199) for pinging the server instead of just sending white space messages to keep the connection alive through NAT routers. It also adds the feature of disconnecting if three pings in a row are sent unanswered. [#15 tagged:committed responsible:Hallski state:resolved] committer: Mikael Hallendal <micke@imendio.com>

Fixed a crash when lm_connection_close was called. Check if the socket exists before trying to cancel the asyncns call on it.

Create connect_data in LmOldSocket, revert a segfault when working on LmResolver. Was a bit agressive when removing old code as I was putting the LmResolver code in place.

Added _lm_connection_get_server that returns the server string. The function uses the host part of the JID used, if it is not set the function will return a copy of the connect host. Also updated to use this internally in LmConnection as the same code was duplicated in a couple of places.

Added _lm_connection_get_context as internal function. Returns the GMainContext currently used by the LmConnection. Internal for now, will be made public if needed from application level.

Fixed compile error when asyncns is not available [#17] Fixed a typo in lm-resolver that breaks compilation when libasyncns is not available. [#17 tagged:committed responsible:Hallski state:resolved] committer: Mikael Hallendal <micke@imendio.com>

Bumped version to 1.4.2

Make lm_sha_hash thread safe. Fixes LM-64 Compute the checksum in a stack-local variable and the result is printed into a newly allocated string of 41 bytes which is returned afterwards. Patch from Andreas Köhler. committer: Mikael Hallendal <micke@imendio.com>

Make lm_sha_hash thread safe. Fixes LM-64 Compute the checksum in a stack-local variable and the result is printed into a newly allocated string of 41 bytes which is returned afterwards. Patch from Andreas Köhler. committer: Mikael Hallendal <micke@imendio.com>

Fixed a compiler warning with gcc 4.3. Fixes LM-127 Patch from Frederic Peters. committer: Mikael Hallendal <micke@imendio.com>

Fixed a compiler warning with gcc 4.3. Fixes LM-127 Patch from Frederic Peters. committer: Mikael Hallendal <micke@imendio.com>

Use GError instead of polution console with gnutls printouts. Fixes LM-126. Patch from Leonid Evdokimov. committer: Mikael Hallendal <micke@imendio.com>

Use GError instead of polution console with gnutls printouts. Fixes LM-126. Patch from Leonid Evdokimov. committer: Mikael Hallendal <micke@imendio.com>

Fixed problem with having NULL byte in SSL fingerprint, fixes LM-128. Use g_memdup instead of g_strdup. Patch submitted by Frank Zschockelt. committer: Mikael Hallendal <micke@imendio.com>

Use g_strndup instead of strndup in asyncns.c, fixes LM-129 Switched to use g_strndup and remove the internal version of strndup from asyncns.c. Fixes problem where strndup exists.

Fixed problem with having NULL byte in SSL fingerprint, fixes LM-128. Use g_memdup instead of g_strdup. Patch submitted by Frank Zschockelt. committer: Mikael Hallendal <micke@imendio.com>

Use g_strndup instead of strndup in asyncns.c, fixes LM-129 Switched to use g_strndup and remove the internal version of strndup from asyncns.c. Fixes problem where strndup exists.

Include netinet/in_systm.h in lm-sock.c, fixes LM-130. Added check for netinet/in_systm.h and include it in lm-sock.c if it exists. It fixes a build error on BSD.

Attempt to use STARTTLS in lm-send-async and test-lm examples.

Fixed the help string for --username in lm-send-async and test-lm examples The help string said that you should only use 'user' in 'user@foo.bar' while the code requires a full jid.

Include netinet/in_systm.h in lm-sock.c, fixes LM-130. Added check for netinet/in_systm.h and include it in lm-sock.c if it exists. It fixes a build error on BSD.

Attempt to use STARTTLS in lm-send-async and test-lm examples.

Fixed the help string for --username in lm-send-async and test-lm examples The help string said that you should only use 'user' in 'user@foo.bar' while the code requires a full jid.

Updated reference to GLib in README

Added comment about SSL in README

Added comment about SSL in README

Changed lm_resolver_create_srv_string to be internal to LmResolvers

Put a ref on the resolver while calling the callback. This makes sure that the object will still be around even if the resolver is unreffed in the callback.

First cut of integrating the new resolver in LmOldSocket

Added cancel operation for asyncns resolver

Fixed so that service lookup seems to work with new asyncns resolver.

More work on the asyncns srv lookup. Moved the srv_parse function to LmResolver to be used by both blocking and non-blocking versions.

Renamed the test to test-dns and fixed a bug in the asyncns resolver. test-dns.c now just tests whatever dns LmResolver::new* gives. Also checked that new_for_host works now with the asyncns resolver.

Copied in code from LmOldSocket to LmAsyncnsResolver

fixed typos that accidently got commited

Bumped version to 1.4.1

Changed so that normal XMPP "pings" are always sent when keep alive is enabled. The change to disable the periodic sending of spaces in order to ensure traffic over the socket made the LM clients disconnect periodically with specific gateways. The implementation looks correct but it might be the firewall software that is not handling it correctly. This commit reverts the behaviour to always send the spaces even if the platform supports the TCP keep alive settings.

The blocking dns resolver now works as expected. Fixed up the last bits and and use the blocking dns lookup in the simple test. Next step will be to make a second one that uses asyncns to do the job.

Added test-dns-blocking to ignore file

Removed soup dependencies from the build

Added small test program and simple tests seems to work. Added test-dns-blocking, this program will be removed later though. Also needed to export some internal functions in the library file to get the test program to build. These were lm_resolver_lookup and lm_blocking_resolver_get_type.

Moved MIN_PORT and MAX_PORT to lm-internals.h and renamed them. Now call the LM_MIN_PORT and LM_MAX_PORT to make it clear they are defined inside of Loudmouth. Also more work on the DNS cleanups.

Added GMainContext to LmResolver and use lm_misc_add_idle to add an idle in it. Since we need to use the correct GMainContext it is now being passed to LmResolver and later used in LmBlockingResolver.

Added protocol for srv lookups

More work on the blocking resolver

Implemented the idle "spawn" in LmBlockingResolver. In order to simulate the asynchronous API of LmResolver the synchronous LmBlockingResolver runs in an idle callback. That means the lm_resolver_lookup call will return just as it does in the asynchronous version and instead the result will come through the given callback.

Added LmBlockingResolver which will handle the case when we do not have libasyncns. The blocking resolver will be blocking while waiting for response from a web server.

Added LmAsyncnsResolver

Fixed some spaces

Added type property to LmResolver

Changed so that we create a resolver with various arguments depending on type. Instead of having the lookup_host and lookup_srv virtual methods there is now only one lookup method and the resolver is created using new_for_host or new_for_srv instead. The host or domain/srv values are set as properties on the base class for the sub classes to read out.

Made LmResolver an object for subclassing instead of an interface

Added LmResolver interface. The LmResolver interface will be implemented in one blocking fashion and one implementation that uses Libasyncns. Designing this as an interface will make it easy to implement an asynchronous version using platform specific technologies for Windows and Mac OS X. It will also make it easier to create a mock implementation to simulate various error codes to test the layers using the resolver.

Some style fixes

Refactored some of the SSL code in LmConnection. Added functions to wrap LmSSL functions in LmOldSocket to minimize the number of calls from LmConnection to LmSSL directly. Also split out function to register a STARTTLS handler if it is not already registered and we are using STARTTLS.

Moved the call to _lm_ssl_close to LmOldSocket

Split out old_socket_do_write from lm_old_socket_do_write. The reason for having the buffer setup split out of lm_old_socket_do_write was that it's also used internally in LmOldSocket. Split out the part that should not care about buffered output into old_socket_do_write. Also renamed lm_old_socket_do_write to lm_old_socket_write.

Some more cleanups

Moved the ifdefs for HAVE_ASYNCNS to the LmOldSocket code. This cleans up the code a bit as it leaves the ifdefs to internal in LmOldSocket.

Updated _asyncns_cancel to follow coding style

Made the output buffer functions private to LmOldSocket.

Moved output buffer creation/addition to LmOldSocket where it belongs.

slight style fix

Some more work on the soup socket

Initital start on implementing the various write functions in LmSoupSocket

Call connect on SoupSocket in LmSoupSocket::connect

Renamed the implementation of the LmSocket interface functions to _soup_socket.. The naming policy of Loudmouth says that the static functions for LmSoupSocket should be named soup_socket_... but in this case that makes the name collide with functions in Libsoup. Prefixed the functions with _ to work around that.

Added prototypes for LmSocket interface in LmSoupSocket

Changed the LmSocket signals slightly

Added Libsoup as a dependency in configure.

Worked on the LmSocket interface and added a LmSoupSocket class. LmSoupSocket will be a possibly temporary implementation of the LmSocket interface on top of the SoupSocket API in order to get something going quickly. Depending on how it works out it might be made the default implementation. Added signals and worked a bit to model the LmSocket API after the SoupSocket API as it seems quite similar to what I had in mind.

Made LmSocket into an interface and hooked up implementation of it in LmTcpSocket. This design allows for the unit test framework to hook into it's mock socket object in order to short cut the network communication.

Added basic API for the new LmSocket

Renamed LmSocket to LmOldSocket to make room for a new socket abstraction layer.

Include lm-internals.h instead of lm-socket.h in lm-xmpp-writer.h

Style fixes

Fixed a left over DUMMY -> SIMPLE_IO

Removed vtable from LmSimpleIO

LmSimpleIO now implements the LmXmppWriter interface. Also fixed copyright years and removed signal definition from LmSimpleIO as it will never define any own signals.

Added LmSimpleIO. LmSimpleIO is meant to be the very simple implementation of the current LmConnection (send code) + lm_message_node_xml + LmParser. After this is properly refactored I can start looking into writing a more robust version of these using expat or LibXML.

Added lm-marshal.list and lm-marshal-main.c Also added lm-dummy.[ch] for a copy/paste GObject.

Added LmXmppWriter interface

Depend on GObject and initialize GType

Added Title to the inlined documentation

Fixed the SECTION parts of the inlined docs so that gtk-doc finds them. Got help from Stefan Kost on IRC to figure out that SECTION needs to be the same as <FILE> in the loudmouth-sections.txt file.

Ignore the tmpl directory

Removed the old tmpl files

Added inlined documentation from generated through the Gtk-doc migration script. Started moving the documentation from the tmpl file to code inline to not have it spread out over to different places and files.

Cleaned declarations of static functions in LmConnection

Moved _lm_connection_old_auth from public header file. This is an internal function and only used in lm-connection.c so made this static and removed from the public header file.

Fixed header file style in internal header file

Changed so that normal XMPP "pings" are always sent when keep alive is enabled. The change to disable the periodic sending of spaces in order to ensure traffic over the socket made the LM clients disconnect periodically with specific gateways. The implementation looks correct but it might be the firewall software that is not handling it correctly. This commit reverts the behaviour to always send the spaces even if the platform supports the TCP keep alive settings.

Fixed indentation