about summary refs log tree commit diff stats
Commit message (Collapse)AuthorAgeFilesLines
...
* 631 - temporary: hook into underlying arc 'eval'Kartik K. Agaram2015-01-262-0/+21
|
* 630 - routines can now allocate unlimited memoryKartik K. Agaram2015-01-262-13/+11
|
* 629Kartik K. Agaram2015-01-262-15/+45
|
* 628Kartik K. Agaram2015-01-261-0/+27
|
* 627Kartik K. Agaram2015-01-262-4/+29
|
* 626 - start eliminating the memory-per-routine limitKartik K. Agaram2015-01-262-5/+31
|
* 625 - guard against errors with input-output argsKartik K. Agaram2015-01-252-16/+45
| | | | | | | | | | | (Another attempt at 623.) I dunno, this may all be a wild goose chase. I haven't been disciplined in tagging in-out arguments in 'read-move' and its helpers. Maybe I should just drop those 'nochange' oargs in 'read' and 'write'. Maybe I should reserve output args only for return values that callers might actually care about, and use more conventional metadata like 'const' or 'unique' or 'inout' on other args.
* 624 - no, that's all wrong, scratch 623Kartik K. Agaram2015-01-257-76/+46
|
* 623 - 'nochange' to guard against race conditionsKartik K. Agaram2015-01-258-55/+93
| | | | | | | | | I dunno, this may all be a wild goose chase. I haven't been disciplined in tagging in-out arguments in 'read-move' and its helpers. Maybe I should just drop those 'nochange' oargs in 'read' and 'write'. Maybe I should reserve output args only for return values that callers might actually care about, and use more conventional metadata like 'const' or 'unique' or 'inout' on other args.
* 621 - fix race condition in buffered-stdin.muKartik K. Agaram2015-01-252-19/+24
|
* 620 - fix broken examples and tests againKartik K. Agaram2015-01-254-26/+63
| | | | | | | | | I just did this in 611; what's the point of all this if tests can't stay passing? I don't understand why buffered-stdin.mu needs to preempt itself. stdin and buffered-stdin somehow end up sharing a single circular buffer, that's probably causing a race condition.
* 619 - chessboard now working rightKartik K. Agaram2015-01-252-5/+28
| | | | | | 2 bugs found: a) Have to slurp the newlines. b) Have to clear the line-buffer at the start of each line.
* 618Kartik K. Agaram2015-01-252-0/+0
|
* 617Kartik K. Agaram2015-01-251-92/+0
|
* 616 - buffered-stdin handles backspaceKartik K. Agaram2015-01-254-2/+46
| | | | Chessboard now *almost* handles backspace perfectly.
* 615 - eliminate scheduler non-determinismKartik K. Agaram2015-01-241-2/+5
|
* 614Kartik K. Agaram2015-01-241-0/+1
|
* 613 - support for interrupting with ctrl-cKartik K. Agaram2015-01-241-0/+3
|
* 612 - support for the backspace keyKartik K. Agaram2015-01-241-0/+7
|
* 611 - all tests passing againKartik K. Agaram2015-01-242-15/+9
|
* 610 - all examples working againKartik K. Agaram2015-01-243-7/+3
|
* 609Kartik K. Agaram2015-01-241-7/+7
| | | | Shove the complexity of reading newlines as low as possible.
* 608Kartik K. Agaram2015-01-241-3/+2
|
* 607Kartik K. Agaram2015-01-241-3/+9
| | | | | | Shove the complexity of printing newlines as low as possible. Definite instability in that one trace.
* 605 - example program for line-oriented inputKartik K. Agaram2015-01-243-2/+45
| | | | | | | | | | We need the hack of printing characters typed to screen right after we see them. But only when it's the real screen. And there's no way to write a test for that because it explicitly shouldn't happen for fake terminals :( Never mind, we'll be able to test it when we provide some mechanism for suppressing print. The equivalent of 'stty -echo'.
* 604 - new function to buffer stdinKartik K. Agaram2015-01-242-5/+129
| | | | | Still surprisingly hard to debug. We might be barking up the wrong tree. Or we might just need to lump it. System software is hard.
* 603Kartik K. Agaram2015-01-241-9/+12
|
* 602Kartik K. Agaram2015-01-241-6/+9
|
* 601 - fix callers of 'read-key'Kartik K. Agaram2015-01-243-3/+4
|
* 600 - fake keyboardKartik K. Agaram2015-01-222-9/+61
| | | | | Use asynchronous channels like 'stdin' for most tests. Use the synchronous fakes for testing low-level stdin helpers.
* 599Kartik K. Agaram2015-01-211-1/+0
|
* 598 - clear up a minor mysteryKartik K. Agaram2015-01-213-4/+11
|
* 597Kartik K. Agaram2015-01-212-8/+14
|
* 596Kartik K. Agaram2015-01-212-78/+48
|
* 595 - tests can now interrupt and restart routinesKartik K. Agaram2015-01-214-832/+70
| | | | Also cleaned up much of the mess in 594.
* 594 - random snapshot after a lengthy debugKartik K. Agaram2015-01-214-13/+860
| | | | | | I'd caused a bug in Arc: https://github.com/arclanguage/anarki/commit/97b3464256 Almost done interrupting and restarting routines.
* 593 - clean up the raw stdin addressKartik K. Agaram2015-01-182-34/+39
|
* 592Kartik K. Agaram2015-01-181-2/+5
|
* 591Kartik K. Agaram2015-01-182-6/+60
| | | | | I don't remember what exactly we wanted to test in the second move. But we can write it next time we need it.
* 590Kartik K. Agaram2015-01-181-0/+2
|
* 589Kartik K. Agaram2015-01-181-3/+0
|
* 588Kartik K. Agaram2015-01-182-2/+2
| | | | | Now that there isn't a stdout channel we can postpone the test showing the need to flush-stdout before clearing screen.
* 587Kartik K. Agaram2015-01-182-5/+10
|
* 586Kartik K. Agaram2015-01-182-28/+23
| | | | Dump yet another trying-to-be-polymorphic function.
* 585Kartik K. Agaram2015-01-181-7/+7
|
* 584Kartik K. Agaram2015-01-181-18/+0
|
* 583Kartik K. Agaram2015-01-181-1/+0
|
* 582 - first tests for printing to screenKartik K. Agaram2015-01-174-19/+167
|
* 580Kartik K. Agaram2015-01-171-1/+1
|
* 578 - switch to non-polymorphic 'print' functionsKartik K. Agaram2015-01-1711-64/+103
| | | | | Also clean up various prints from last few commits. As a convention, for debugging we always print directly to host.
generated by cgit-pink 1.4.1-2-gfad0 (git 2.36.2.497.gbbea4dcf42) at 2024-06-23 19:49:57 +0000