about summary refs log tree commit diff stats
Commit message (Collapse)AuthorAgeFilesLines
* 6512Kartik Agaram2020-06-121-14762/+14758
|
* 6511 - start of error-checkingKartik Agaram2020-06-122-17/+84
| | | | | We now raise an error if a variable is declared on the stack with an initializer. And there are unit tests for this functionality.
* 6510Kartik Agaram2020-06-118-8/+8
|
* 6509 - mu.subx: exit-descriptors everywhereKartik Agaram2020-06-112-323/+285
|
* 6508 - support null exit-descriptorKartik Agaram2020-06-1017-30/+19
|
* 6507 - use syscall names everywhereKartik Agaram2020-06-1054-213/+108
|
* 6506Kartik Agaram2020-06-101-3/+0
|
* 6505Kartik Agaram2020-06-091-1/+1
|
* 6504Kartik Agaram2020-06-081-0/+1
|
* 6503Kartik Agaram2020-06-083-12/+4
|
* 6502Kartik Agaram2020-06-073-14753/+15233
|
* 6501Kartik Agaram2020-06-072-2/+2
| | | | Bugfix in support for CRLF line-endings.
* 6500Kartik Agaram2020-06-073-174/+174
| | | | Minor formatting tweaks.
* 6499Kartik Agaram2020-06-071-2/+2
|
* 6498Kartik Agaram2020-06-062-116/+116
| | | | | Switch bullet lists in Markdown files away from `*`; it's ambiguous with emphasis.
* 6497Kartik Agaram2020-06-061-19/+29
|
* 6496Kartik Agaram2020-06-065-1/+474
|
* 6495Kartik Agaram2020-06-062-89/+91
|
* 6494Kartik Agaram2020-06-064-0/+471
|
* 6493 - browse: '#'s inside lines aren't headingsKartik Agaram2020-06-064-11/+18
|
* 6492Kartik Agaram2020-06-061-2/+0
|
* 6491Kartik Agaram2020-06-064-0/+466
|
* 6490 - browse: some manual testsKartik Agaram2020-06-062-0/+16293
|
* 6489 - browse app: headersKartik Agaram2020-06-062-2/+94
| | | | | Again quite ugly. There's an increasing amount of state here, particularly the interplay between headers and soft newlines.
* 6488Kartik Agaram2020-06-054-0/+377
|
* 6487 - browse: support soft newlinesKartik Agaram2020-06-053-6/+55
| | | | | I'm not bothering with this for bold regions just yet. Might need rethinking, given how ugly this is.
* 6486Kartik Agaram2020-06-054-0/+332
|
* 6485Kartik Agaram2020-06-051-5/+9
|
* 6484Kartik Agaram2020-06-051-16/+6
| | | | Ooh, it's nice and composable if we just never render the delimiters. Perfect.
* 6483 - finally, bold markup in the paginatorKartik Agaram2020-06-051-1/+75
| | | | | | The state machines are still not composing perfectly. The initial asterisk gets added in one, and the trailing asterisk in another. I suppose "always render the terminator" is fairly regular.
* 6482Kartik Agaram2020-06-054-0/+261
|
* 6481Kartik Agaram2020-06-054-0/+0
|
* 6480Kartik Agaram2020-06-052-6/+9
|
* 6479Kartik Agaram2020-06-052-1/+59
| | | | | Fix a stray copy-paste when deciding whether to emit spills for registers (commit 6464).
* 6478Kartik Agaram2020-06-051-1/+1
| | | | Fix CI.
* 6477Kartik Agaram2020-06-052-36/+157
| | | | | | | I had a little "optimization" to avoid creating nested blocks if "they weren't needed". Except, of course, they were. Lose the optimization. Sometimes we create multiple jumps when a single one would suffice. Ignore that for now.
* 6476Kartik Agaram2020-06-054-0/+258
|
* 6475Kartik Agaram2020-06-054-0/+258
|
* 6474Kartik Agaram2020-06-0516-44/+44
|
* 6473Kartik Agaram2020-06-054-0/+238
|
* 6472Kartik Agaram2020-06-051-1/+10
|
* 6471Kartik Agaram2020-06-054-0/+230
|
* 6470Kartik Agaram2020-06-056-229/+13
| | | | No, we can stay with the previous prototype a little longer.
* 6469Kartik Agaram2020-06-053-0/+222
| | | | | | Let's start committing new prototype directories *before* we start hacking on them. Version control is useful right from the draft when programming in a language without any error messages.
* 6468Kartik Agaram2020-06-057-5/+230
|
* 6467Kartik Agaram2020-06-043-0/+86
|
* 6466Kartik Agaram2020-06-042-0/+48
|
* 6465Kartik Agaram2020-06-042-0/+2
|
* 6464 - support temporaries in fn output registersKartik Agaram2020-06-042-11/+262
| | | | | | | | The rule: emit spills for a register unless the output is written somewhere in the current block after the current instruction. Including in nested blocks. Let's see if this is right.
* 6463 - clean up some duplicationKartik Agaram2020-06-032-148/+115
| | | | | | Rather than have two ways to decide whether to emit push/pop instructions, just record for each var on the 'vars' stack whether we emitted a push for it, and reuse the decision to emit a pop.