2013-10-09	Georg Hopp	georg@steffers.org
	* A config/taskrambler.conf.in:
		add forgotten taskramble.conf.in
	* M Makefile.am:
	* M configure.ac:
	* A include/config/config.h:
	* A include/config/value.h:
	* M src/Makefile.am:
	* M src/application/Makefile.am:
	* M src/asset/Makefile.am:
	* M src/auth/Makefile.am:
	* M src/cbuf/Makefile.am:
	* M src/class/Makefile.am:
	* A src/config/Makefile.am:
	* A src/config/config.c:
	* A src/config/get.c:
	* A src/config/value.c:
	* A src/configtest.c:
	* M src/hash/Makefile.am:
	* M src/http/Makefile.am:
	* M src/http/worker/process.c:
	* M src/logger/Makefile.am:
	* M src/queue/Makefile.am:
	* M src/server/Makefile.am:
	* M src/session/Makefile.am:
	* M src/socket/Makefile.am:
	* M src/socket/listen.c:
	* M src/storage/Makefile.am:
	* M src/stream/Makefile.am:
	* M src/taskrambler.c:
	* A src/testconfig.cfg:
	* M src/tree/Makefile.am:
	* M src/user/Makefile.am:
	* M src/utils/daemonize.c:
	* M src/utils/mime_type.c:
		make install works now. So from now on taskrambler will not
		work when run from its build dir, because it can't find its
		assets then. To make installation work a basic config file
		reader class was implemented.
	* D .doxygen:
	* M .gitignore:
	* M docs/Doxyfile.in:
	* M docs/Makefile.am:
	* M include/auth/storage.h:
	* M include/queue.h:
	* M src/application/application.c:
	* M src/application/login.c:
	* M src/application/signup.c:
	* M src/auth/ldap.c:
	* M src/stream/write.c:
	* M src/user/save.c:
		some more changes for documentation generation

2013-10-06	Georg Hopp	georg@steffers.org
	* M assets/html/_download.html:
		add link to v0.1.6 download
	* M assets/html/_documentation.html:
		update documentation
	* M assets/style/common.css:
		raise dialogs above everyting
	* M docs/Makefile.am:
		and put it to extra_dist
	* A docs/Doxyfile.in:
		add forgotton Doxyfile.in
	* M include/application/application.h:
	* M include/hash/hash.h:
	* M src/application/Makefile.am:
	* M src/application/adapter/http/update.c:
	* M src/application/application.c:
	* A src/application/session_cleanup.c:
	* M src/application/session_get.c:
	* M src/application/session_start.c:
	* M src/application/session_stop.c:
	* D src/application/session_update.c:
	* M src/application/signup.c:
	* M src/hash/Makefile.am:
	* A src/hash/cleanup.c:
	* A src/hash/get_first.c:
	* M src/hash/hash.c:
		optimized session handling. closes #41

2013-10-05	Georg Hopp	georg@steffers.org
	* M configure.ac:
		there is a AC_SUBST for the coverage variables in gcov.m4 now
	* A docs/doxygen-1.8.4-r9999.ebuild:
	* A docs/doxygen-1.8.4-taskrambler.patch:
		added patch to doxygen to get identify my class definitions
		as structures
	* A assets/html/_author.html:
	* A assets/html/_documentation.html:
	* A assets/html/_download.html:
	* A assets/html/_main.html:
	* D assets/html/author.html:
	* A assets/html/author.html:
	* D assets/html/documentation.html:
	* A assets/html/documentation.html:
	* D assets/html/download.html:
	* A assets/html/download.html:
	* A assets/html/layout.html:
	* D assets/html/main.html:
	* A assets/html/main.html:
	* M assets/js/init.js:
		now use one layout file and load even the content as a partial
	* M src/utils/memory.c:
		use memSegment of when tree based optimizations where not
		used. That way all other features implemented here are still
		available
	* M TODO:
		update TODO
	* M m4/dragonegg.m4:
	* M m4/gcov.m4:
	* M m4/memopt.m4:
		add m4 to build via dragonegg and update gcov and memopt m4
	* M Makefile.am:
	* D Makefile.am.coverage:
	* M configure.ac
	* A docs/Makefile.am:
	* M tests/Makefile.am:
		actual coverage reports and api docs can be build via make
		again ... now project name and version are taken from
		configure ... ahh and configure holds version information
		from git.
	* D assets/coverage/:
	* D assets/doc/:
		remove generated documentations from repository.

2013-10-02	Georg Hopp	georg@steffers.org
	* M assets/style/taskrambler.css:
		status line always above title
	* M configure.ac:
	* M src/Makefile.am:
	* M src/application/Makefile.am:
	* M src/asset/Makefile.am:
	* M src/auth/Makefile.am:
	* M src/cbuf/Makefile.am:
	* M src/class/Makefile.am:
	* M src/hash/Makefile.am:
	* M src/http/Makefile.am:
	* M src/logger/Makefile.am:
	* M src/queue/Makefile.am:
	* M src/server/Makefile.am:
	* M src/session/Makefile.am:
	* M src/socket/Makefile.am:
	* M src/storage/Makefile.am:
	* M src/stream/Makefile.am:
	* M src/tree/Makefile.am:
	* M src/user/Makefile.am:
		use dragonegg config
	* M TODO:
		update TODO
	* A m4/dragonegg.m4:
	* M m4/memopt.m4:
		add m4 macros to build with dragonegg and make the memopt m4
		show a message if it is active
	* A assets/html/_footer.html:
	* A assets/html/_login.html:
	* A assets/html/_menu.html:
	* A assets/html/_signup.html:
	* A assets/html/_statusline.html:
	* A assets/html/_title.html:
	* M assets/html/author.html:
	* M assets/html/documentation.html:
	* M assets/html/download.html:
	* D assets/html/foo.html:
	* M assets/html/main.html:
	* D assets/html/robots.txt:
	* M assets/js/init.js:
	* M assets/style/taskrambler.css:
		load partials via javascript, so I do not need the server at
		all. closes #35

2013-10-01	Georg Hopp	georg@weird-web-workers.org
	* A assets/coverage/:
	* A assets/doc/:
		add content, coverage report an api doc
	* M src/server/run.c:
		accept as much as possible and stop processing as soon as
		possible
	* M src/server/read.c:
		fix return value...the situation should result in a
		connection close.
	* M src/application/adapter/http/update.c:
	* M src/http/writer/write.c:
		make write more fair...until now it was written as long a
		possible given no chance for reading anymore...

2013-09-30	Georg Hopp	georg@steffers.org
	* M src/stream/read.c:
	* M src/stream/write.c:
		fix the fix...with the last change taskrambler closes
		connections unexpectedly. Hopefully this fixes ssl handling.
	* M src/stream/read.c:
		and do the same when read
	* M src/server/close_conn.c:
	* M src/stream/write.c:
		improved ssl write error handling...this hopefully fixes
		occasionally crashes when using ssl

2013-09-29	Georg Hopp	georg@steffers.org
	* M assets/html/download.html:
		and add information about new download file
	* M assets/html/documentation.html:
		add informations about guest user
	* M assets/html/documentation.html:
		wrote some informations about the new version
	* M configure.ac:
		redo version bump
	* A m4/memopt.m4:
		m4 makros to select memory optimizations
	* M configure.ac:
		revert version to 0.1.5 after fixes from next have been
		merged...I have to obey my workflows to prevent such things
		in future.
	* M configure.ac:
		use silent build again
	* M src/auth/storage/hash_pw.c:
		fix use of uninitialized value
	* M src/application/login.c:
		fix remove of user. As email is in that case a part of
		credential it must be set to NULL before delete.
	* M configure.ac:
	* M include/utils/memory.h:
	* M src/Makefile.am:
	* M src/utils/memory.c:
		add facility to de/activate tree base memory optimizations
	* M Makefile.am:
		add run to extra dists
	* M configure.ac:
		version bump
	* M assets/html/author.html:
	* M assets/html/documentation.html:
	* M assets/html/download.html:
	* M assets/html/main.html:
	* M assets/js/init.js:
	* M assets/js/session.js:
	* M assets/style/common.css:
		make login, signup and logout available on all pages
	* M include/application/application.h:
	* M src/application/Makefile.am:
	* M src/application/adapter/http/update.c:
	* A src/application/logout.c:
		logout no longer tries to stop the session. Now simply the
		user is removed.

2013-09-28	Georg Hopp	georg@steffers.org
	* M src/auth/ldap.c:
		call to ldap_unbind is necessary to free memory even when
		ldap_bind failed.
	* M assets/html/example.html:
	* M assets/html/main.html:
	* M assets/js/init.js:
	* M assets/js/session.js:
	* M assets/style/common.css:
	* M assets/style/taskrambler.css:
		reflect current changes in main.html (and styles and
		javascript)
	* M include/application/application.h:
		update prototypes
	* M src/application/Makefile.am:
		there is no session_update method anymore...
		this will be done in session_get at least at the moment
	* M src/application/adapter/http/update.c:
		some changes on session handlings.
	* M src/application/login.c:
		create a real new user object when the fetch fails.
	* M src/application/session_stop.c:
		to stop a session is as simple as setting their livetime
		to 0. On the next request this session will be removed.
	* M src/storage/get.c:
		As gdbm_fetch changes the key I copy it before I try to get
		data with it.
	* A src/http/response/version.c:
		add forgotten version response.
	* A assets/image/fav128.png:
	* A assets/image/fav16.png:
	* A assets/image/fav256.png:
	* A assets/image/fav32.png:
	* A assets/image/fav64.png:
	* A assets/image/favicon.ico:
		add favicon...
	* M src/application/adapter/http/update.c:
	* M src/application/session_get.c:
		move session handling from adapter into application and some
		more stuff of the version thing...when the version thing
		works I will add the session livetime to every page.
	* M assets/html/main.html:
	* M assets/js/init.js:
	* M include/application/application.h:
	* M include/http/response.h:
	* M src/application/Makefile.am:
	* M src/application/application.c:
	* M src/http/Makefile.am:
		show taskrambler version in Frontend
	* M src/http/worker/add_common_header.c:
		get server name from config.h

2013-09-27	Georg Hopp	georg@steffers.org
	* M assets/html/example.html:
	* M assets/js/init.js:
	* M assets/style/common.css:
	* M src/application/adapter/http/update.c:
	* M src/application/signup.c:
		signup works...only logout has to be added then the new
		version is complete. refs #36

2013-09-26	Georg Hopp	georg@steffers.org
	* M include/queue.h:
	* M src/application/adapter/http/update.c:
	* M src/application/application.c:
		fix wrong handling of session queue and thus the memory leak.
	* A .gitattributes:
	* M README:
		experiment with attributes

2013-09-24	Georg Hopp	georg@steffers.org
	* M assets/html/author.html:
	* M assets/html/documentation.html:
	* M assets/html/download.html:
	* M assets/html/example.html:
		html fixes
	* M assets/html/author.html:
	* M assets/html/documentation.html:
	* M assets/html/download.html:
	* M assets/html/example.html:
	* M assets/html/main.html:
		add link tag to favicon on every page
	* M assets/favicon.ico
	* A assets/favicon.ico:

2013-09-22	Georg Hopp	georg@steffers.org
	* M tests/Makefile.am:
	* A tests/suppress/external.supp:
		add some valgrind suppressions for opanldap and gdbm stuff
	* M assets/js/session.js:
	* M src/application/adapter/http/update.c:
	* M src/application/application.c:
	* M src/application/login.c:
	* M src/auth/ldap.c:
	* M src/http/response.c:
	* M src/queue/get.c:
	* M src/session/session.c:
	* M src/storage/storage.c:
		login via ldap and gdbm now works. Signup stores the user
		account as well as the credentials but still returns a 500
		and I have a memory leak most likely within the queue code.
		To reproduce this one has to start the server and send two
		requests via telnet to the server...no keep-alive just a
		single GET. refs #36
	* M include/application/application.h:
	* M include/auth/storage.h:
	* D include/storage.h:
	* A include/storage/storage.h:
	* M include/user.h:
	* A signuptest.html:
	* M src/application/adapter/http/update.c:
	* M src/application/application.c:
	* M src/application/signup.c:
	* M src/auth/storage/storage.c:
	* M src/storage/get.c:
	* M src/storage/put.c:
	* M src/storage/storage.c:
	* M src/storage/update.c:
	* M src/taskrambler.c:
	* M src/user/load.c:
	* M src/user/save.c:
	* M src/user/user.c:
	* M src/utils/hash.c:
		now signup and login with storage of a hashed password works.
	* M include/auth/storage.h:
	* M src/Makefile.am:
	* M src/application/Makefile.am:
	* M src/application/adapter/http/update.c:
	* M src/application/application.c:
	* M src/application/signup.c:
	* M src/auth/Makefile.am:
	* M src/auth/storage/hash_pw.c:
	* M src/auth/storage/storage.c:
	* M src/storage/Makefile.am:
	* M src/storage/put.c:
	* M src/storage/update.c:
		everything builds again
	* D src/auth/storage/signup.c:
		removed signup from auth adapter as signup is no auth at all.
	* M include/application/application.h:
	* M include/auth.h:
	* A include/auth/storage.h:
	* M include/storage.h:
	* M src/application/adapter/http/update.c:
	* M src/application/application.c:
	* A src/application/signup.c:
	* A src/auth/storage/hash_pw.c:
	* A src/auth/storage/signup.c:
	* A src/auth/storage/storage.c:
	* M src/storage/put.c:
	* A src/storage/update.c:
	* M src/user/save.c:
	* M src/utils/hash.c:
		put most of the parts for simple signup/login in place...
		not tested now as it is late. Hopefully tomorrow I find the
		time to setup a small testform and bring this whole thing
		finally to work. refs #36

2013-09-21	Georg Hopp	georg@steffers.org
	* M assets/js/session.js:
	* M include/http/response.h:
	* M src/application/adapter/http/update.c:
	* M src/application/login.c:
	* M src/http/Makefile.am:
	* A src/http/response/user.c:
	* M src/usertest.c:
		use user class to load and get user informations.

2013-09-16	Georg Hopp	georg@steffers.org
	* M include/application/application.h:
	* M include/http/worker.h:
	* M include/session.h:
	* M src/Makefile.am:
	* M src/application/adapter/http/update.c:
	* M src/application/application.c:
	* M src/application/login.c:
	* M src/application/session_get.c:
	* M src/application/session_start.c:
	* M src/application/session_stop.c:
	* M src/application/session_update.c:
	* M src/http/Makefile.am:
	* M src/http/response/session.c:
	* M src/http/worker.c:
	* A src/http/worker/add_computed_header.c:
	* M src/http/worker/process.c:
	* M src/session/session.c:
	* M src/taskrambler.c:
	* A src/uuid.c:
		make sessions start on any request and clean them when they
		have timed out.
	* A run/.keep-me:
	* M src/storage/get.c:
	* M src/usertest.c:
		some more testing on the user class

2013-09-15	Georg Hopp	georg@steffers.org
	* M configure.ac:
	* A include/storage.h:
	* A include/user.h:
	* M src/Makefile.am:
	* A src/storage/Makefile.am:
	* A src/storage/get.c:
	* A src/storage/put.c:
	* A src/storage/storage.c:
	* A src/user/Makefile.am:
	* A src/user/load.c:
	* A src/user/save.c:
	* A src/user/user.c:
	* A src/usertest.c:
		add first user class and persistence experiments. refs #36
	* M src/utils/memory.c:
		initialize ref_count correctly
	* M src/utils/memory.c:
		add a ref_count to the memory segments...this way I could
		increase references and only insert the memory segment in the
		tree again if the ref_count is one.

2013-09-14	Georg Hopp	georg@steffers.org
	* M assets/js/init.js:
		fixed linking in example section
	* M assets/html/author.html:
	* M assets/html/documentation.html:
	* M assets/html/download.html:
	* M assets/html/example.html:
	* M assets/html/main.html:
	* M assets/style/taskrambler.css:
		make the pages look not so bad on my wifes ipad

2013-09-13	Georg Hopp	georg@steffers.org
	* M assets/html/author.html:
	* M assets/html/documentation.html:
	* M assets/html/download.html:
	* M assets/html/example.html:
	* M assets/html/main.html:
	* M assets/style/taskrambler.css:
		add copyright notice and contact-email to pages
	* M assets/html/download.html:
		fill the download page with some links.
	* M assets/html/author.html:
	* M assets/html/documentation.html:
	* M assets/html/download.html:
	* M assets/html/example.html:
	* M assets/html/main.html:
	* A assets/image/rambler-border-b.jpg:
	* A assets/image/rambler-border-bl.jpg:
	* A assets/image/rambler-border-br.jpg:
	* A assets/image/rambler-border-l.jpg:
	* A assets/image/rambler-border-r.jpg:
	* A assets/image/rambler-border-t.jpg:
	* A assets/image/rambler-border-tl.jpg:
	* A assets/image/rambler-border-tr.jpg:
	* M assets/js/init.js:
	* M assets/js/session.js:
	* M assets/style/taskrambler.css:
		some additions to the project page
	* A assets/html/author.html:
	* A assets/html/documentation.html:
	* A assets/html/download.html:
	* M assets/html/example.html:
	* M assets/html/main.html:
	* A assets/image/rambler-bg.jpg:
	* A assets/image/rambler-logo-small.jpg:
	* A assets/image/rambler-logo2-small.jpg:
	* A assets/other/OldNewspaperTypes.ttf:
	* A assets/other/old_typewriter.ttf:
	* A assets/style/taskrambler.css:
		added first parts of project homepage serverd by taskrambler
		itself.
	* A assets/html/main.html:
		rename taskrambler.html to main.html
	* A assets/html/example.html:
	* D assets/html/main.html:
		rename main.html to example.html

2013-09-12	Georg Hopp	georg@steffers.org
	* M TODO:
		update TODO. closes #24
	* M src/application/adapter/http/update.c:
	* M src/application/session_stop.c:
		now it works mostly as before...the only difference is that
		you really have to be logged in to see the rand value
	* M src/application/adapter/http/update.c:
	* M src/application/session_update.c:
	* M src/http/worker/process.c:
		most stuff works...
		the session timeout will not be reset correctly within some
		answers, anyway it mostly works. refs #24
	* M configure.ac:
	* M include/application/adapter/http.h:
	* M include/application/application.h:
	* M include/hash/hash.h:
	* M include/session.h:
	* M src/Makefile.am:
	* A src/application/Makefile.am:
	* M src/application/adapter/http/http.c:
	* M src/application/adapter/http/update.c:
	* M src/application/application.c:
	* A src/application/login.c:
	* A src/application/session_get.c:
	* A src/application/session_start.c:
	* A src/application/session_stop.c:
	* A src/application/session_update.c:
	* M src/hash/delete.c:
	* M src/hash/get.c:
	* M src/http/worker/process.c:
	* M src/session/Makefile.am:
	* D src/session/add.c:
	* D src/session/delete.c:
	* D src/session/get.c:
	* M src/session/session.c:
	* M src/taskrambler.c:
		code with abstraced application compiles again, but does not
		work correctly, start debugging. refs #24
	* A include/application/adapter/http.h:
	* A include/application/application.h:
	* M include/http/worker.h:
	* M include/interface/observer.h:
	* A src/application/adapter/http/http.c:
	* A src/application/adapter/http/update.c:
	* A src/application/application.c:
	* M src/http/worker.c:
	* M src/http/worker/add_common_header.c:
	* M src/http/worker/get_asset.c:
	* M src/http/worker/process.c:
	* M src/interface/observer.c:
	* M src/taskrambler.c:
		implemented an application class as well as an http adapter
		for it and use it to start application logic by http requests
		as well as creating a fitting repsonse. Not perfect, but a
		start. This code is not finish and will not work...in fact it
		won't even compile i think. refs #24
	* M src/asset/asset.c:
		only mmap a file if its size is larger than 0 bytes
	* M src/asset/asset.c:
		fix wrong constant used in access

2013-09-11	Georg Hopp	georg@steffers.org
	* M TODO:
		another thing to do
	* M include/auth.h:
		add comment
	* M TODO:
		order optional tasks to the botton of the TODO list.
	* M README.md:
		add some taskrambler specific informations in README.md
	* M README.md:
		changed title in README.md
	* M TODO:
		updated TODO with the next things that have to be done.

2013-09-10	Georg Hopp	georg@steffers.org
	* M src/stream/read.c:
	* M src/stream/write.c:
		fix error handling on SSL handles
	* M src/server/run.c:
		use any value less than 0 as event done for ssl accept.
	* M README.md:
		removed SSL deactivation notice...
	* M src/server/run.c:
	* M src/server/server.c:
	* M src/stream/read.c:
	* M src/taskrambler.c:
		checked SSL support, the only thing that was really broken
		was the error handling while reading data... well maybe
		writing has also bugs, but it works.
	* M README.md:
		add info of my redmine to README.md
	* M README.md:
		update README.md
	* M include/utils/http.h:
	* M src/asset/asset.c:
	* M src/http/worker/add_common_header.c:
	* M src/utils/http.c:
	* M src/http/worker/add_common_header.c:
		now every date header is formatted in GMT. Commment: We still
		have a very weak Etag implementation.
	* M src/asset/pool.c:
		now don't add a NULL asset to the hash.

2013-09-09	Georg Hopp	georg@steffers.org
	* M Makefile.am:
		add config and thus mime type mapping to distribution
	* M src/server/server.c:
		fix deactivation of SSL
	* M src/Makefile.am:
	* M src/asset/Makefile.am:
	* M src/auth/Makefile.am:
	* M src/cbuf/Makefile.am:
	* M src/class/Makefile.am:
	* M src/hash/Makefile.am:
	* M src/http/Makefile.am:
	* M src/logger/Makefile.am:
	* M src/queue/Makefile.am:
	* M src/server/Makefile.am:
	* M src/session/Makefile.am:
	* M src/socket/Makefile.am:
	* M src/stream/Makefile.am:
	* M src/tree/Makefile.am:
		add subdir-objects to each Makefile.am
	* M tests/Makefile.am:
	* D tests/stream/Makefile:
		don't use non automake makefiles under tests anymore
	* M src/server/run.c:
	* M src/server/server.c:
		disabled SSL support for now, cause it seems that it crashed
		the whole thing right now.
	* M include/asset.h:
	* M src/asset/asset.c:
	* M src/server/handle_accept.c:
	* M src/socket/accept.c:
		a fix in size of asset names and change in logging

2013-09-08	Georg Hopp	georg@steffers.org
	* M src/hash.c:
	* M src/tree/destroy.c:
	* M src/utils/memory.c:
		fix post order traversal used for destruction of existing tree
	* M src/tree/delete.c:
		fix the fix...

2013-09-07	Georg Hopp	georg@steffers.org
	* A src/hash.c:
		add a small test program for the hash code to find where it
		leaks memory.
	* M src/tree/delete.c:
		add one missing delete for a node, thus fixing a memory leak
		in there.
	* M src/rbtree2.c:
		now do all things within rbtree2 that are also done in rbtree
	* M src/utils/memory.c:
		now use at minimum page size for an allocated memory segment,
		and also store this size. This should use the memory more
		effective.

2013-09-06	Georg Hopp	georg@steffers.org
	* M configure.ac:
	* M include/hash/hash.h:
	* A include/tree.h:
	* M src/Makefile.am:
	* M src/asset/asset.c:
	* M src/asset/pool.c:
	* M src/hash/add.c:
	* M src/hash/delete.c:
	* M src/hash/each.c:
	* M src/hash/get.c:
	* M src/hash/hash.c:
	* M src/rbtree.c:
	* A src/rbtree2.c:
	* M src/server/run.c:
	* A src/tree/Makefile.am:
	* A src/tree/delete.c:
	* A src/tree/destroy.c:
	* A src/tree/find.c:
	* A src/tree/inOrderSuccessor.c:
	* A src/tree/insert.c:
	* A src/tree/rotateLeft.c:
	* A src/tree/rotateRight.c:
	* A src/tree/tree.c:
	* A src/tree/walk.c:
	* M src/utils/memory.c:
		use now my own tree implementation as base for my hashes....
		sadly this thing is leaking memory again...
		on the other hand I workarounded the problem of a sometimes
		occuring infinite loop within connect.

2013-09-04	Georg Hopp	georg@steffers.org
	* M TODO:
	* M include/http/response.h:
	* M src/asset/asset.c:
	* M src/asset/pool.c:
	* M src/hash/delete.c:
	* M src/hash/get.c:
	* M src/hash/interface/hashable.c:
	* M src/http/message.c:
	* M src/http/response/asset.c:
	* M src/http/worker/get_asset.c:
	* M src/http/writer/write.c:
	* M src/server/handle_accept.c:
	* M src/server/run.c:
		I thought I had found a bug in tdelete in glibc, but it
		wasn't. In fact my comparison function was broken.
	* M src/auth/ldap.c:
	* M src/hash/each.c:
	* M src/server/read.c:
	* M src/server/run.c:
	* M src/server/write.c:
	* M src/socket/socket.c:
		as I currently have no idea whats the problem I first fixed
		all warnings...included the ldap warnings that where caused
		by the use of deprecated ldap_simple_bind_s and ldap_unbind_s.

2013-09-03	Georg Hopp	georg@steffers.org
	* A src/socket/nonblock.c:
		added forgotten file
	* M src/server/run.c:
		I think the problem with keep-alive and concurrency is fixed..
		the handling in serverRun was still not ok for the non
		blocking sockets.
	* M TODO:
		added some first analyse of new problem
	* M TODO:
	* M include/socket.h:
	* M src/cbuf/read.c:
	* M src/http/parser/parse.c:
	* M src/http/worker/process.c:
	* M src/server/handle_accept.c:
	* M src/server/run.c:
	* M src/server/server.c:
	* M src/socket/Makefile.am:
	* M src/socket/accept.c:
	* M src/socket/socket.c:
	* M src/stream/read.c:
	* M src/taskrambler.c:
	* M include/http/worker.h:
		changed socket handling according to my definition...
		and make sockets nonblocking as the answer of poll is just a
		guess...
	* M include/http/worker.h:
	* M include/http/writer.h:
	* M src/http/worker.c:
		remove occurences of write circular buffer
	* M src/asset/pool.c:
		fix cleanup of NULL asset_pool
	* M include/http/writer.h:
	* M include/utils/memory.h:
	* M src/cbuf/Makefile.am:
	* D src/cbuf/write.c:
	* M src/http/worker.c:
	* M src/http/worker/get_asset.c:
	* M src/http/worker/process.c:
	* M src/http/writer.c:
	* M src/http/writer/write.c:
	* M src/server/run.c:
	* M src/server/write.c:
	* M src/stream/write.c:
	* M src/utils/memory.c:
		this is a huge one...removed cbuf for writing again....
		This is not needed any more because I use memory mapped io
		for these. Additionally sanitised write error handling
		somewhat...anyway, under huge load I still observer
		unexpected connection closes. But maybe this is related to
		interrupt while reading...(reading is not handled very well
		right now)
	* M TODO:
		documented another bug
	* A docs/socket_states.txt:
		added some thoughts about socket states while reading and
		writing. This is not handled very well right now in my code
		which result in connections to be closed to early.

2013-09-01	Georg Hopp	georg@steffers.org
	* M include/asset.h:
	* M src/asset/asset.c:
	* M src/asset/pool.c:
	* M src/hash/delete.c:
		now asset is by itself hashable...
		and the leaking issue seems to be fixed.

2013-08-29	Georg Hopp	georg@steffers.org
	* M assets/html/main.html:
	* M configure.ac:
	* M include/asset.h:
	* M include/http/message.h:
	* M include/http/worker.h:
	* M src/Makefile.am:
	* A src/asset/Makefile.am:
	* M src/asset/asset.c:
	* A src/asset/pool.c:
	* M src/cbuf/write.c:
	* M src/hash/delete.c:
	* M src/http/Makefile.am:
	* M src/http/message.c:
	* A src/http/response/500.c:
	* M src/http/response/asset.c
	* M src/http/worker.c:
	* M src/http/worker/get_asset.c:
	* D src/http/worker/get_mime_type.c:
	* M src/http/worker/process.c:
	* M src/http/writer/write.c:
	* M src/server/read.c:
	* M src/server/run.c:
	* M src/server/server.c:
	* M src/server/write.c:
	* M src/taskrambler.c:
		try to use memory mapped io for asset access, but this one
		leaks like a rotten barge
	* M include/http/message.h:
	* M include/http/response.h:
	* M src/http/response/304.c:
	* M src/http/response/403.c:
	* M src/http/response/404.c:
	* M src/http/response/login_form.c:
	* M src/http/response/randval.c:
	* M src/http/response/session.c:
		remove message type (PIPED or BUFFERED)
	* A include/asset.h:
	* A include/utils/mime_type.h:
	* A src/asset/asset.c:
	* A src/mmapfiletest2.c:
	* A src/utils/mime_type.c:
		added asset class and move mimetype handling in separate
		helper. Tested this with mmapfiletest2.c
	* A docs/idea_for_asset_access.md:
	* A src/mmapfiletest.c:
		added thought about memory mapped asset handling, as well
		as a small mmap file test program.

2013-08-28	Georg Hopp	georg@steffers.org
	* M assets/html/main.html:
	* M src/http/worker.c:
	* M src/http/worker/process.c:
		remove specialized asset loadings and use generic one.
	* M src/utils/memory.c:
		now allocate only a multiple of pagesize / this has been
		reverted later on.
	* A assets/other/.keep-me:
	* M src/http/worker.c:
	* M src/http/worker/process.c:
		now sending non html data from one of the other folders
		also works
	* A assets/html/foo.html:
	* A assets/html/robots.txt:
	* A config/mime.types:
	* M include/http/worker.h:
	* M src/http/Makefile.am:
	* M src/http/worker.c
	* A src/http/worker/get_mime_type.c:
	* M src/http/worker/process.c:
		add suppport for file extension based mime type detection.
	* A docs/mime.types:
		add a file with file extension to mime-type mappings
	* M src/http/message.c:
	* M src/http/response/asset.c:
	* M src/http/worker/get_asset.c:
	* M src/http/worker/process.c:
		now every file accessible under assets/html will be delivered.

2013-08-27	Georg Hopp	georg@steffers.org
	* M src/rbtree.c:
	* M src/taskrambler.c:
	* M src/utils/memory.c:
		now valgrind shows no more violations...but cleanup does not
		work correcly. The memory usage depends now on the amount of
		paralell connections...ab show approximately 10% to 20%
		performance improvement / and the code is far from optimal.
	* M src/rbtree.c:
	* M src/utils/memory.c:
		tree based memory management does not segfault anymore, but
		reusage does not seem to work, as well as free
	* M TODO:
	* M src/rbtree.c:
		update todo and make my rbtree implementation use externally
		allocated elements.

2013-08-24	Georg Hopp	georg@steffers.org
	* M src/rbtree.c:
		rbtree insert and delete now as one function and working as
		expected...still has to be checked on sideeffects.
	* M src/binarytree.c:
		realize that delete does not work correct.
	* M src/rbtree.c:
		all insert cases merged in one insertElement function.

2013-08-23	Georg Hopp	georg@steffers.org
	* M src/rbtree.c:
		now the memory is freed outside the tree delete function....
	* M src/rbtree.c:
		now it seems that rb trees are working now...stiff a lot of
		optimization work to be done, as this is mostly one 2 one the
		code from wikipedia, with some modifiations to work with NULL
		leaf nodes.
	* M src/rbtree.c:
		inserting in red black tree now works as in my example.
	* M src/binarytree.c:
		change the way traversed elements are shown and add some
		comments
	* A src/rbtree.c:
		cope binarytree to play with an rbtree implementation
	* M src/binarytree.c:
		fix traversal
	* M src/binarytree.c:
		alternative approach on delete of single child nodes. This
		one really exchanges the nodes, thus preventing large copy on
		big data that might be in the tree...anyway, for small data i
		think the other approach is faster.
	* M src/binarytree.c:
		now use the parent in element in delete

2013-08-22	Georg Hopp	georg@steffers.org
	* M src/binarytree.c:
		fix insert and make another in-order traverse for checking
	* M src/binarytree.c:
		remove wrong comment
	* M src/binarytree.c:
		add parent in element and implement an iterative in-order
		traversal with it.
	* M docs/rbdelete.txt:
	* M docs/rbinsert.txt:
		further improvments on the rb examples
	* M docs/rbdelete.txt:
		more stuff on rbdelete
	* M docs/rbdelete.txt:
		rbdelete stuff
	* A docs/rbdelete.txt:
		added first exampled for rbtree delete stuff...
	* M docs/rbinsert.txt:
		small fix in rbinsert doc
	* A docs/rbinsert.txt:
		add small example on rbtree insert, just to see what I have
		to care about when implementing this
	* A src/binarytree.c:
		add first own binary tree testcode. Not ballanced right now.

2013-08-21	Georg Hopp	georg@steffers.org
	* M configure.ac:
	* M src/server/handle_accept.c:
	* M src/stream/read.c:
	* M src/stream/write.c:
	* M src/taskrambler.c:
	* M src/utils/daemonize.c:
	* M tests/loggerTest.c:
	* M tests/serverTest.c:
		fix some warnings
	* M configure.ac:
	* M m4/gcov.m4:
	* M src/Makefile.am:
	* M src/auth/Makefile.am:
	* M src/cbuf/Makefile.am:
	* M src/class/Makefile.am:
	* M src/hash/Makefile.am:
	* M src/http/Makefile.am:
	* M src/logger/Makefile.am:
	* M src/queue/Makefile.am:
	* M src/server/Makefile.am:
	* M src/session/Makefile.am:
	* M src/socket/Makefile.am:
	* M src/stream/Makefile.am:
		change build system to accept CFLAGS
	* M TODO:
	* M src/server/poll.c:
	* M src/server/read.c:
	* M src/server/run.c:
		probably fixed problem in main look under load.
		This still has to be tested.
	* M src/http/worker/process.c:
		code formatting

2013-08-20	Georg Hopp	georg@steffers.org
	* M src/taskrambler.c:
		add memCleanup before process end
	* M src/http/parser.c:
		change last FREE to MEM_FREE
	* M src/session/session.c:
		replace malloc in session/session.c with quick fit approach.
	* M src/http/worker/process.c:
		replace calloc http/worker/process.c with quick fit approach.
	* M src/http/worker.c:
		replace calloc http/worker.c with quick fit approach.
	* M src/http/parser/parse.c:
		replace calloc http/parser.c with quick fit approach.
	* M src/http/parser/p_request_vars.c:
	* M src/http/request.c:
		replace calloc http/request.c with quick fit approach.
	* M src/auth/credential.c:
		replace calloc auth/credentials.c with quick fit approach.
	* M src/auth/ldap.c:
		replace calloc auth/ldap.c with quick fit approach.
	* M src/http/header.c:
		replace calloc in http/cookie.c with quick fit approach.
	* M src/http/cookie.c:
		replace calloc in http/cookie.c with quick fit approach.
		It seems that domain and path are never set...have to check
		this (most likely its not implemented now
	* M src/hash/value.c:
		replace calloc in hash/value.c with quick fit approach
	* M src/server/server.c:
		replace calloc in server.c with quick fit approach
	* M src/http/response.c:
		replace calloc in response.c with quick fit approach
	* M src/http/message.c:
	* M src/http/parser/p_header.c:
	* M src/http/response/404.c:
	* M src/http/response/login_form.c:
	* M src/http/response/randval.c:
	* M src/http/response/session.c:
		use quick fit approach in message
	* M src/logger/interface/i_logger.c:
	* M src/utils/memory.c:
		fix the last issues with using quick fit with the class
		interface.
	* M src/class/interface/i_class.c:
	* M src/logger/interface/i_logger.c:
	* M src/utils/memory.c:
	* M tests/Makefile.am:
		added new memory management to the most fragile part in this
		matter...the class system...well, the server test gives me a
		memory corruption, but all other are running now....anyway
		most likely this is in an unusable state right now.
	* M include/cbuf.h:
	* M include/utils/memory.h:
	* M src/cbuf/cbuf.c:
	* M src/logger/interface/i_logger.c:
	* M src/utils/memory.c:
		simplified the quick fit interface
	* M tests/serverTest.c:
	* M tests/socketTest.c:
		change port for tests
	* M include/cbuf.h:
	* M src/cbuf/cbuf.c:
		add quick fit to cbuf
	* M src/logger/interface/i_logger.c:
	* M tests/Makefile.am:
		to test it use Best Fit memory management in logger.
	* M include/utils/memory.h:
	* M src/utils/memory.c:
		first version of Quick Fit memory management
	* M .gitignore:
		now ignore log tsr files under tests
	* M include/cbufpool.h:
		more comments regarding cbuf pools.
	* M include/http/writer.h:
		more comments regarding cbuf pools.
	* M src/utils/memory.c:
		fix the new functions so that they compile at least.
	* A include/cbufpool.h:
		added idea comment for a cbuf pool to not always
		reinitialize these all the time...
	* A src/cbufpool.c:
		added idea comment for a cbuf pool to not always
		reinitialize these all the time...
	* M src/utils/memory.c:
		started implementing the Quick Fit memory management
	* M src/utils/memory.c:
		added comment about idea of Quick Fit memory management
		system to improve performance

2013-08-12	Georg Hopp	georg@steffers.org
	* M .gitignore:
		add test-driver to git ignores
	* M src/taskrambler.c:
		put daemonize before our working child is forked.
	* M src/http/parser/parse.c:
	* M src/server/server.c:
		only code format changes
	* M src/utils/daemonize.c:
		make a more real daemonizing
	* M configure.ac:
		add build dir as macro definition to daemonze.c thus been
		able to chdir there so that all assets would be found.

2013-08-09	Georg Hopp	georg@steffers.org
	* M Makefile.am.coverage:
	* M m4/gcov.m4:
		small changes in build system

2013-03-16	Georg Hopp	georg@steffers.org
	* M README.md:
		first README.md
	* M m4/gcov.m4:
		update gcov m4 to include version 1.10
	* A .gitignore:
	* A README.md:
		Initial commit

2012-03-29	Georg Hopp	georg@steffers.org
	* M configure.ac:
	* D include/http/message/queue.h:
	* M include/http/parser.h:
	* M include/http/writer.h:
	* A include/queue.h:
	* M src/Makefile.am:
	* M src/http/Makefile.am:
	* D src/http/message/queue.c:
	* D src/http/message/queue/get.c:
	* D src/http/message/queue/put.c:
	* M src/http/parser.c:
	* M src/http/parser/parse.c:
	* M src/http/worker/process.c:
	* M src/http/writer.c:
	* M src/http/writer/write.c:
	* A src/queue/Makefile.am:
	* A src/queue/get.c:
	* A src/queue/put.c:
	* A src/queue/queue.c:
		moved httpMessageQueue to a generalized Queue class as in
		fact it was nothing else except that it only stored
		HttpMessages.
	* M src/http/message/queue/get.c:
		fix message queue get. Problem was not correctly updated
		last value on empty queue
	* M src/http/worker/process.c:
		remove no longer needed variable
	* M include/http/message/queue.h:
	* M src/http/Makefile.am:
	* M src/http/message/queue.c:
	* A src/http/message/queue/get.c:
	* A src/http/message/queue/put.c:
	* M src/http/parser/parse.c:
	* M src/http/worker/process.c:
	* M src/http/writer/write.c:
		changed HttpMessageQueue to be a real queue and not a fixed
		size array
	* M .gitignore:
		ignore profile information files
	* M Makefile.am.coverage:
	* M tests/Makefile.am:
		change on coverage generation...i still don't get any branch
		info, and i am not sure what it would describe

2012-03-28	Georg Hopp	georg@steffers.org
	* M tests/Makefile.am:
	* M tests/stream/Makefile:
		fix in Makefiles for tests
	* M src/Makefile.am:
	* M tests/Makefile.am:
	* D tests/loggerTest:
	* A tests/stream/Makefile:
		latest additions to tests and code coverage

2012-03-27	Georg Hopp	georg@steffers.org
	* M src/stream/Makefile.am:
	* A src/stream/interface/reader.c:
	* D src/stream/interface/stream_reader.c:
	* D src/stream/interface/stream_writer.c:
	* A src/stream/interface/writer.c:
	* M tests/Makefile.am:
	* M tests/loggerTest:
	* A tests/mock/mock_worker.c:
	* A tests/mock/mock_worker.h:
	* M tests/serverTest.c:
		added first tests for server class
	* M .gitignore:
	* M tests/Makefile.am:
	* M tests/socketTest.c:
		some tests for socket class
	* M .gitignore:
	* M Makefile.am.coverage:
	* M include/class/class.h:
	* M src/class/Makefile.am:
	* D src/class/interface/class.c:
	* A src/class/interface/i_class.c:
	* M src/http/Makefile.am:
	* D src/http/interface/http_intro.c:
	* A src/http/interface/i_http_intro.c:
	* D src/http/parser/body.c:
	* D src/http/parser/header.c:
	* A src/http/parser/p_body.c:
	* A src/http/parser/p_header.c:
	* A src/http/parser/p_post_vars.c:
	* A src/http/parser/p_request_vars.c:
	* D src/http/parser/post_vars.c:
	* D src/http/parser/request_vars.c:
	* A src/http/worker/answer.c:
	* D src/http/worker/write.c:
	* M src/logger/Makefile.am:
	* A src/logger/interface/i_logger.c:
	* D src/logger/interface/logger.c:
	* M tests/Makefile.am:
	* M tests/classTest.c:
	* A tests/loggerTest:
	* M tests/loggerTest.c:
	* M tests/mock/mock_class.c:
	* A tests/mock/mock_logger.c:
	* A tests/mock/mock_logger.h:
	* M tests/runtest.c:
	* M tests/runtest.h:
		more tests...and some little structural changes to make then
		happen correctly with coverage reports. KEEP IN MIND:
		coverage reports lie to you. (But they are a good hint to
		see whats not testes enough

2012-03-26	Georg Hopp	georg@steffers.org
	* M .gitignore:
	* M Makefile.am.coverage:
	* M configure.ac:
	* M include/class/class.h:
	* M m4/gcov.m4:
	* M src/http/Makefile.am:
	* M src/logger/Makefile.am:
	* M tests/Makefile.am:
	* D tests/cclassTest.c:
	* A tests/classTest.c:
	* D tests/mock/class.c:
	* D tests/mock/class.h:
	* A tests/mock/mock_class.c:
	* A tests/mock/mock_class.h:
	* M tests/runtest.c:
	* M tests/runtest.h:
		started tests and coverage report.
	* M configure.ac:
	* D include/interface/stream_reader.h:
	* D include/interface/stream_writer.h:
	* M include/stream.h:
	* A include/stream/interface/reader.h:
	* A include/stream/interface/writer.h:
	* A include/stream/stream.h:
	* M src/Makefile.am:
	* M src/http/parser.c:
	* M src/http/worker.c:
	* M src/http/writer.c:
	* D src/interface/stream_reader.c:
	* D src/interface/stream_writer.c:
	* D src/server.c:
	* A src/server/Makefile.am:
	* M src/server/read.c:
	* A src/server/server.c:
	* M src/server/write.c:
	* D src/session.c:
	* A src/session/Makefile.am:
	* A src/session/session.c:
	* D src/socket.c:
	* A src/socket/Makefile.am:
	* A src/socket/socket.c:
	* D src/stream.c:
	* A src/stream/Makefile.am:
	* A src/stream/interface/stream_reader.c:
	* A src/stream/interface/stream_writer.c:
	* A src/stream/stream.c:
		Now all classes are moved in according archives. Coming
		closer to the original purpose of the class construct to
		build small independent reusable code fragments.
	* M .gitignore:
		now ignore archive files
	* D src/auth/libauth.a:
	* D src/cbuf/libcbuf.a:
	* D src/class/libclass.a:
	* D src/hash/libhash.a:
	* D src/http/libhttp.a:
	* D src/logger/liblogger.a:
	* M configure.ac:
		moved logger to separate archive
	* M include/http.h:
	* A include/http/interface/http_intro.h:
	* D include/interface/http_intro.h:
	* D include/interface/logger.h:
	* M include/logger.h:
	* A include/logger/interface/logger.h:
	* A include/logger/logger.h:
	* M src/Makefile.am:
	* M src/auth/libauth.a:
	* M src/cbuf/libcbuf.a:
	* M src/class/libclass.a:
	* M src/hash/libhash.a:
	* M src/http/Makefile.am:
	* A src/http/interface/http_intro.c:
	* M src/http/libhttp.a:
	* M src/http/message/header_size_get.c:
	* M src/http/message/header_to_string.c:
	* M src/http/parser/parse.c:
	* M src/http/request.c:
	* M src/http/response.c:
	* D src/interface/http_intro.c:
	* D src/interface/logger.c:
	* M src/interface/stream_reader.c
	* D src/logger.c:
	* A src/logger/Makefile.am:
	* A src/logger/interface/logger.c:
	* A src/logger/liblogger.a:
	* A src/logger/logger.c:
	* M src/logger/stderr.c:
	* M src/logger/syslog.c:
	* M src/server/handle_accept.c:
	* M src/server/poll.c:
	* M src/server/read.c:
	* M src/server/run.c:
	* M src/server/write.c:
	* M src/socket.c:
	* M src/socket/accept.c:
	* M src/socket/connect.c:
	* M src/socket/listen.c:
	* M src/taskrambler.c:
		moved logger to separate archive
	* M .gitignore:
	* M Makefile.am:
	* A Makefile.am.coverage:
	* M configure.ac:
	* M include/auth.h:
	* A include/auth/auth.h:
	* A include/auth/credential.h:
	* A include/auth/interface/auth.h:
	* D include/credential.h:
	* M include/hash.h:
	* A include/hash/hash.h:
	* A include/hash/interface/hashable.h:
	* A include/hash/value.h:
	* D include/hash_value.h:
	* A include/http.h:
	* M include/http/message.h:
	* D include/interface/auth.h:
	* D include/interface/hashable.h:
	* A m4/gcov.m4:
	* M src/Makefile.am:
	* A src/auth/Makefile.am:
	* A src/auth/credential.c:
	* A src/auth/interface/auth.c:
	* M src/auth/ldap.c:
	* A src/auth/libauth.a:
	* D src/cbuf.c:
	* A src/cbuf/Makefile.am:
	* A src/cbuf/cbuf.c:
	* A src/cbuf/libcbuf.a:
	* M src/class/Makefile.am:
	* M src/class/libclass.a:
	* D src/credential.c:
	* D src/hash.c:
	* A src/hash/Makefile.am:
	* M src/hash/add.c:
	* M src/hash/delete.c:
	* M src/hash/get.c:
	* A src/hash/hash.c:
	* A src/hash/interface/hashable.c:
	* A src/hash/libhash.a:
	* A src/hash/value.c:
	* D src/hash_value.c:
	* A src/http/Makefile.am:
	* M src/http/cookie.c:
	* M src/http/header.c:
	* A src/http/libhttp.a:
	* M src/http/message.c:
	* M src/http/parser/header.c:
	* M src/http/parser/post_vars.c:
	* M src/http/parser/request_vars.c:
	* M src/http/request.c:
	* M src/http/worker.c:
	* M src/http/worker/process.c:
	* D src/interface/auth.c:
	* D src/interface/hashable.c:
	* M src/taskrambler.c:
	* M tests/Makefile.am:
		huge changes in build system. Moved a lot of stuff to
		separate archives, not finished right now. Additionally add
		some support for coverage reports. Next step after build
		restructuring is to write tests.

2012-02-23 17:40:00 +0100 Georg Hopp 

	* add cookie header again (HEAD, origin/master, origin/HEAD, master)

2012-02-23 16:41:36 +0100 Georg Hopp 

	* optmize header search

2012-02-23 13:02:23 +0100 Georg Hopp 

	* changed all string operation within header handling with fixed length mem operations, preventing multiple iterations over these strings. In theory this should improve performance in reality it seems that it is worse...CHECK WHY

2012-02-23 00:05:25 +0100 Georg Hopp 

	* fix initialization of search value

2012-02-22 21:50:21 +0100 Georg Hopp 

	* ed

2012-02-22 21:49:52 +0100 Georg Hopp 

	* structural changes for worker/process. @TODO actually i have no idea why this happens.

2012-02-22 12:19:40 +0100 Georg Hopp 

	* fix memory problems occured with latest changes

2012-02-22 09:03:40 +0100 Georg Hopp 

	* fixed bug in keep-alive check arised by implementation if #10

2012-02-22 08:51:05 +0100 Georg Hopp 

	* add forgotten jquery assets

2012-02-22 08:48:43 +0100 Georg Hopp 

	* closes #10: values for header ids are now stored in a char ** making multiple values for one id possible. Additionally added a jquery action that delivers the jquery java script and use it on the me action

2012-02-21 13:01:38 +0100 Georg Hopp 

	* now when a constructor returns -1 the new call will in turn call the destructor effectively freeing all resources. ATTENTION: now the destructor has to be aware that it might be called with a not completely initialized object. To make this more ease there is the FREE makro with the corresponding ffree that does NULL pointer checking and the destructor checks for NULL pointer too. Additionally the handle_accept now handles _SC_OPEN_MAX - 10 connections. The 10 are reserved for internal usage.

2012-02-21 09:45:01 +0100 Georg Hopp 

	* now a child is spawned and writes random values in a shared memory segment. These values will be shown in the me action

2012-02-20 21:36:55 +0100 Georg Hopp 

	* some code cleanups...no changes in the logic

2012-02-20 18:08:23 +0100 Georg Hopp 

	* move sdbm implementation in one file.

2012-02-20 17:16:44 +0100 Georg Hopp 

	* changed /**/ single line comments to //

2012-02-20 14:55:46 +0100 Georg Hopp 

	* start documenting this whole stuff...well at least add a copyright information in each file

2012-02-20 10:10:29 +0100 Georg Hopp 

	* first very crude, not complete, experimental 304 test implementation

2012-02-20 07:55:06 +0100 Georg Hopp 

	* disconnect on invalid request line

2012-02-19 20:12:40 +0100 Georg Hopp 

	* now incomplete requests should no longer block the complete server. Tested with \'echo -en "GET / HTTP\r\nConn" | nc -w 600 localhost 11212\' and then doing requests from my browser. @TODO: cleanup those stuff, check if a not correctly response reading would block the server.

2012-02-19 18:28:30 +0100 Georg Hopp 

	* increase writebuffer size a lot.

2012-02-19 18:15:55 +0100 Georg Hopp 

	* fixed the non keep-alive performance issue as well as i lower memory usage by using a single read and write circular buffer for every connection. @TODO: i noticed a server hang while getting large data (my image) with non keep-alive connections. Additionally an incomplete keep-alive request might stop the server now as the lock on the read buffer will not be released.

2012-02-19 15:41:48 +0100 Georg Hopp 

	* another try with the shmen trick...this time use MAP_ANONYMOUS ... as GNU extension.

2012-02-19 14:33:42 +0100 Georg Hopp 

	* Merge remote branch 'origin/master'

2012-02-19 12:13:52 +0100 Georg Hopp 

	* added missing header file to repo

2012-02-19 11:35:15 +0100 Georg Hopp 

	* another try with a shared memory based ringbuffer...this performs well for keep-alive sessions but is much slower without. actually i am not sure why but most likely the shared memory setup is quite expensive. @TODO: make a profiling.

2012-02-18 21:08:32 +0100 Georg Hopp 

	* fix inf loop. @TODO: This whole handling has to be cleaned.

2012-02-18 20:50:01 +0100 Georg Hopp 

	* this change hopefully makes the shm trick work on amd64

2012-02-18 20:12:27 +0100 Georg Hopp 

	* lots of changes but primarily change the request parser to use a ringbuffer. The ringbuffer is implemented using the shared memory trick.

2012-02-15 12:30:33 +0100 Georg Hopp 

	* some more cleanups in the server code. Removing not needed header includes

2012-02-15 12:17:39 +0100 Georg Hopp 

	* Merge branch 'master' of 192.168.100.2:/var/lib/git/server

2012-02-15 12:17:00 +0100 Georg Hopp 

	* now the separated http worker works. Changed some size_t to ssize_t as i use -1 and -2 es error indicator in my server and fixed caculation of remainig buffer size in reader

2012-02-15 09:38:32 +0100 Georg Hopp 

	* separated the server completely from the http processing

2012-02-15 06:19:52 +0100 Georg Hopp 

	* add subject/observer interface

2012-02-15 06:19:52 +0100 Georg Hopp 

	* add subject/observer interface

2012-02-15 04:55:46 +0100 Georg Hopp 

	* fix infinite busy loop in run

2012-02-15 04:44:38 +0100 Georg Hopp 

	* dynamically get and free buffer for response write pipe now

2012-02-14 21:32:38 +0100 Georg Hopp 

	* increase write buffer

2012-02-13 21:27:47 +0100 Georg Hopp 

	* use one dynamic buffer less and save at least one write on small responses

2012-02-13 18:25:36 +0100 Georg Hopp 

	* removed generated docs

2012-02-13 17:55:52 +0100 Georg Hopp 

	* fixed bug in new response handling

2012-02-13 17:29:35 +0100 Georg Hopp 

	* better response handling but still buggy with stream piping

2012-02-13 09:46:39 +0100 Georg Hopp 

	* now load image from actual server

2012-02-13 09:39:21 +0100 Georg Hopp 

	* first working version of content delivery from file....very crude... @TODO: rewrite complete response handline.

2012-02-12 20:39:12 +0100 Georg Hopp 

	* more generalizing of response writing (implemented a response writer...now it should be possible to implement a stream writer for images

2012-02-12 12:43:56 +0100 Georg Hopp 

	* make http request and response childs of a common parent http message

2012-02-12 04:13:54 +0100 Georg Hopp 

	* remove now obsoleted header_sort

2012-02-12 04:05:38 +0100 Georg Hopp 

	* change response to tree based header storage and make everything work.

2012-02-12 00:05:13 +0100 Georg Hopp 

	* changed header hashing to use btree (GNU only). @TODO: make this conditional for other systems. Removed the qsort calls on server->fds making O(2nlogn) to O(n)

2012-02-11 13:52:32 +0100 Georg Hopp 

	* daemonize testserver now

2012-02-11 12:47:01 +0100 Georg Hopp 

	* fix seaks and hangs after adding response object (mostly not related with the response object but how i integated it into serverRun

2012-02-10 19:57:57 +0100 Georg Hopp 

	* started a response handler and changed serverRun to use it for its response

2012-02-10 12:42:04 +0100 Georg Hopp 

	* fixed bug at server destructor

2012-02-10 09:59:41 +0100 Georg Hopp 

	* reset keep_live flag on connection close

2012-02-10 09:52:27 +0100 Georg Hopp 

	* made a first cruel handling for keep-alive and non keep-alive requests. @TODO: this MUST BE cleaned

2012-02-10 08:14:31 +0100 Georg Hopp 

	* now only use keep-alive....

2012-02-10 06:36:43 +0100 Georg Hopp 

	* moved request_parser.h and request_queue.h in separeate request subfolder

2012-02-10 06:22:39 +0100 Georg Hopp 

	* fix rather nasty reentrance bug

2012-02-10 05:52:50 +0100 Georg Hopp 

	* fix bug that arose in rewrite of header get and results in an ugly memory leak, as well as no headers would be found any more

2012-02-10 00:27:51 +0100 Georg Hopp 

	* fix memory leak created while changing things

2012-02-09 22:39:08 +0100 Georg Hopp 

	* updated docs

2012-02-09 22:34:32 +0100 Georg Hopp 

	* start split of request parser

2012-02-09 11:44:17 +0100 Georg Hopp 

	* no more request body debig output

2012-02-09 11:32:28 +0100 Georg Hopp 

	* add missing header_get to repo and build header hash only from lowercase letters now as it seems header identifier should be case insensitive

2012-02-09 09:34:21 +0100 Georg Hopp 

	* access to headers via hash, read body (actually only with content-length header should also look for content-encoding)

2012-02-08 16:51:49 +0100 Georg Hopp 

	* fix handling of remote close - i should have another eye on this...there still seems to be something wrong.

2012-02-08 15:04:52 +0100 Georg Hopp 

	* fixed some warnings

2012-02-08 13:12:59 +0100 Georg Hopp 

	* changed documentation

2012-02-08 13:05:23 +0100 Georg Hopp 

	* added first generated documentation

2012-02-08 12:14:44 +0100 Georg Hopp 

	* oops commit...forgot to add request_queue.c

2012-02-08 11:52:30 +0100 Georg Hopp 

	* found the file handle lost...made a first workaround and added an todo.

2012-02-08 10:21:04 +0100 Georg Hopp 

	* changed request handling. @TODO: I still seem to have the problem that the file handles are not closed and freed correctly as the service refuses connections after about a 1000.

2012-02-07 14:20:00 +0100 Georg Hopp 

	* now stuff seems to work correct even if read does not provide a complete request (tested with telnet)

2012-02-07 13:41:49 +0100 Georg Hopp 

	* now each HttpRequestParser initializes its own request queue and enqueus completed requests there. The server now gets the queue and prints completed requests.

2012-02-07 11:12:30 +0100 Georg Hopp 

	* started filling out a request object with the parser

2012-02-07 09:29:59 +0100 Georg Hopp 

	* porformance improvement in parsing process (no longer do alloc and free on each line)

2012-02-07 08:52:18 +0100 Georg Hopp 

	* basic request parsing (line by line) implemented

2012-02-06 16:08:13 +0100 Georg Hopp 

	* split server implementation for readability

2012-02-06 11:35:40 +0100 Georg Hopp 

	* free reader (HttpRequestParser) when connection is closed

2012-02-06 11:20:00 +0100 Georg Hopp 

	* add StreamReader interface, modify HttpRequestParser and Server to use it

2012-02-06 11:15:00 +0100 Georg Hopp 

	* add missing include to stdarg.h

2012-02-06 10:45:33 +0100 Georg Hopp 

	* implement clone selector

2012-02-06 10:43:59 +0100 Georg Hopp 

	* add ability to call interface methods with return value

2012-02-06 02:37:55 +0100 Georg Hopp 

	* make build system work again

2012-02-06 02:37:24 +0100 Georg Hopp 

	* remove inline stuff for now ... add carefully again later perhaps

2012-02-06 00:57:26 +0100 Georg Hopp 

	* and also mod conigure.ac

2012-02-06 00:55:44 +0100 Georg Hopp 

	* makefile modification for new class stuff

2012-02-05 22:55:16 +0100 Georg Hopp 

	* changed class tool. Now multiple interface per class are supported as well as simple inheritence.

2012-02-05 22:47:10 +0100 Georg Hopp 

	* some latest work

2012-02-05 22:44:59 +0100 Georg Hopp 

	* added some documentation

2012-02-05 22:42:37 +0100 Georg Hopp 

	* changes related to server code

2012-01-19 16:41:41 +0100 Georg Hopp 

	* added some valueable thought about cclass and how this structure might evolve to a real class

2012-01-18 07:52:07 +0100 Georg Hopp 

	* add testserver and did some fixes not shown by my incomplete tests

2012-01-17 15:40:07 +0100 Georg Hopp 

	* more notes

2012-01-17 15:04:33 +0100 Georg Hopp 

	* add some thought

2012-01-17 14:49:49 +0100 Georg Hopp 

	* changed from select(UNIX) to poll(POSIX)

2012-01-16 18:39:01 +0100 Georg Hopp 

	* work on server_run

2012-01-16 17:05:57 +0100 Georg Hopp 

	* move test under docs dir

2012-01-16 17:05:08 +0100 Georg Hopp 

	* simply copy signal handling code from gameserver project

2012-01-16 16:04:11 +0100 Georg Hopp 

	* more work on socket handling stuff... @TODO think about renaming it to connection as it only handles TCP sockets

2012-01-16 13:48:05 +0100 Georg Hopp 

	* add info text about file handle passing and ported more stuff from my old server structure

2012-01-16 08:05:15 +0100 Georg Hopp 

	* reflect changes in configure.ac

2012-01-13 22:46:45 +0100 Georg Hopp 

	* add daemonize code from other project. (Might be integrated in a future class but i am not sure right now

2012-01-13 22:16:17 +0100 Georg Hopp 

	* logger now works and has some basic testing

2012-01-13 22:15:03 +0100 Georg Hopp 

	* change cclass so that the internal structure is no longer visible by the rest of the code

2012-01-13 16:06:02 +0100 Georg Hopp 

	* some fixes on first logger tests

2012-01-13 15:54:22 +0100 Georg Hopp 

	* initial checkin