Balau - Unnamed repository; edit this file 'description' to name the repository.

Age	Commit message (Collapse)	Author
2011-12-05	Making the ref counter on Handles thread-safe. Not that it renders the usage ↵	Pixel
	of Handles thread-safe, but at least one can now transmit Handles over threads safely without risking a refcount bug.
2011-12-04	Refactored the Thread code a bit, and created the GlobalThread class, for ↵	Pixel
	threads that are created on startup.
2011-12-04	Making sure that if a test fails, the 'make' command stops	Pixel

2011-12-04	Adding at least one TAssert into test-Threads, and fixing a typo.	Pixel

2011-12-04	Reworked some things in the architecture, mainly exceptions and asserts.	Pixel
	-) Removed Assert() -) Added AAssert(), IAssert(), RAssert(), TAssert() and Failure() -) Reworked all asserts in the code, and added meaningful messages to them. -) Changed the way the startup code is generated; BALAU_STARTUP is no longer necessary.
2011-12-01	Git ftl...	Pixel

2011-12-01	Having libcoro to be pthread-compliant for darwin.	Pixel

2011-11-29	(very small) optimization in the error handler of the HTTP server.	Pixel

2011-11-29	Getting rid of the std::pair; they are ugly as hell.	Pixel

2011-11-29	Making things a bit more pretty:	Pixel
	-) Using an actual embedded style sheet within the error document (serifs are evil) -) Changing the ugly `' into `´, at least for the HTML output. -) Adding the title into the error page.
2011-11-29	Fixing Mustache's escaping (again).	Pixel

2011-11-29	Now using SimpleMustache to render the error messages.	Pixel

2011-11-29	Fixing a stupid bug in the escape() code of SimpleMustache.	Pixel

2011-11-29	Having a forceWrite that takes a string.	Pixel

2011-11-29	Removing a few operators that aren't used anymore, and adding a better ↵	Pixel
	=(String) operator.
2011-11-29	Being a bit more laxist with Mustache's values, and accepting almost ↵	Pixel
	anything as a variable; also adding a checkTemplate() method to assert a few things onto a template after loading it.
2011-11-29	The TaskManager no longer waits on the pop(); instead it waits in the libev ↵	Pixel
	loop; also the Queue<> template changed for a naive but std-less implementation.
2011-11-29	Adding detailled logs, with timestamp and threadid.	Pixel

2011-11-29	Adding the memory fence across the resolver thread, and actually using ↵	Pixel
	getsockopt(..., SO_ERROR) on the second spin of connect(). That'd get better results out of it.
2011-11-29	Adding a few more useful features to Buffers, such as getBuffer() and rewind().	Pixel

2011-11-29	Adding memory fences to Atomic.h; won't hurt that much...	Pixel

2011-11-28	Fixing a debug message.	Pixel

2011-11-28	Whoops: forgot to implement the escaping function.	Pixel

2011-11-26	Adding a few todo-like comments in the SimpleMustache code - not that I ↵	Pixel
	actually intend to implement them anyway...
2011-11-25	Fixing a few bugs and warnings.	Pixel

2011-11-25	Adding a small documentation.	Pixel

2011-11-25	Making values other than bool always true.	Pixel

2011-11-25	If a Buffer came from const, don't try to free its memory...	Pixel

2011-11-25	Adding a simple 'Mustache' template processor in C++.	Pixel

2011-11-25	Adding a 'HandlePrinter' Handle type, to easily print out the results of a ↵	Pixel
	handle, and some small helpers; more helpers to come.
2011-11-25	We need the buffered streams to be detacheable	Pixel

2011-11-25	Consts are good.	Pixel

2011-11-24	Making strings being able to read from their end.	Pixel

2011-11-23	Making sure that the mingw32 target isn't too broken (it is, but that's a ↵	Pixel
	step in the proper direction.)
2011-11-23	That would make way more sense to the fromConst Buffer to have its size set ↵	Pixel
	up, wouldn't it ?
2011-11-23	Adding the isA<>() template / method to the IO proxy, and optimizing the ↵	Pixel
	BStream using it, in case one never wants to use the readString() or peekByte() methods.
2011-11-21	GeneralException() will now trace the callstack and store this, for ↵	Pixel
	debugging purposes.
2011-11-19	Having the ability to create buffers from const memory.	Pixel

2011-11-19	Having String.set returning this.	Pixel

2011-11-19	Adding error 500s support.	Pixel

2011-11-19	Of course, spaces are optionnal...	Pixel

2011-11-19	Handling more Connection tags, separated with commas; and adding the 'split' ↵	Pixel
	method to String.
2011-11-19	Fixing a few very stupid bugs.	Pixel

2011-11-18	Starting to re-organize the HTTP code a bit. Sharing code between the server ↵	Pixel
	and a potential client makes sense. Also packing requests and responses into structures / classes is probably a good idea.
2011-11-18	Updating libev's timestamp after we (potentially) wait a long time for ↵	Pixel
	something to do.
2011-11-18	Making the debug build to increase libev's debugness too.	Pixel

2011-11-18	Got this one wrong; one need to get the refcount to 0 to delete itself.	Pixel

2011-11-17	Adding a few reasons why we'd get an HTTP 400 error, and fixing the ↵	Pixel
	Connection header (I hate RFCs, they never mentionned keep-alive in the list of potential headers)
2011-11-17	Look at that. It seems that, indeed, IPv6 CAN get very retarded.	Pixel

2011-11-17	HTTP server's first real test, alongside multiple taskmanager threads.	Pixel
	I'm not really sure I fully like the way I'm designing this, but I guess it could be solved with an HTTP/HTML helper class around the Action class. However, the HTTP server awfully need reference counting, so it doesn't go away before all of the workers disappear, which means a bit of a redesign of the Listener template.