mu - Soul of a tiny new machine. More thorough tests → More comprehensible and rewrite-friendly software → More resilient society.

	Commit message (Collapse)	Author	Age	Files	Lines
*	.	Kartik K. Agaram	2021-08-10	1	-0/+2
\|
*	.	Kartik K. Agaram	2021-08-10	1	-2/+2
\|
*	.	Kartik K. Agaram	2021-08-10	1	-2/+2
\|
*	.	Kartik K. Agaram	2021-08-10	6	-2/+2
\|
*	slack: parse users	Kartik K. Agaram	2021-08-10	1	-3/+103
\|
*	slack: import some test data	Kartik K. Agaram	2021-08-10	2	-3/+6453
\|
*	.	Kartik K. Agaram	2021-08-10	1	-2/+2
\|
*	new tool: creating a data disk	Kartik K. Agaram	2021-08-10	2	-1/+19
\|
*	slack: fix abort after load	Kartik K. Agaram	2021-08-10	1	-1/+8
\| \| \| \|	I forgot about the null bytes that come from loading sectors.
*	slack: takes almost 3 minutes to load 55MB	Kartik K. Agaram	2021-08-10	1	-3/+5
\| \| \| \| \| \| \| \| \|	This is with KVM on a pretty fast Thinkpad. If I exclude images, 14MB takes 45s to load. There's also a bug in parsing at the moment; it aborts after loading all records.
*	.	Kartik K. Agaram	2021-08-10	3	-0/+104
\|
*	.	Kartik K. Agaram	2021-08-10	2	-9/+42
\|
*	slack: emit comment parent indices in converter	Kartik K. Agaram	2021-08-10	1	-8/+15
\| \| \| \| \| \| \|	They're easier to process when loading the data disk. In the process we lose a few more items because they're comments to items we were dropping earlier.
*	.	Kartik K. Agaram	2021-08-10	1	-3/+3
\|
*	undo my stupid format for post/comment ids	Kartik K. Agaram	2021-08-10	1	-9/+6
\| \| \| \|	Now items just have a field for parent id.
*	.	Kartik K. Agaram	2021-08-10	1	-2/+0
\|
*	.	Kartik K. Agaram	2021-08-10	1	-22/+16
\|
*	.	Kartik K. Agaram	2021-08-10	1	-6/+2
\|
*	.	Kartik K. Agaram	2021-08-10	1	-3/+1
\|
*	.	Kartik K. Agaram	2021-08-10	1	-1/+0
\|
*	.	Kartik K. Agaram	2021-08-10	1	-5/+3
\|
*	.	Kartik K. Agaram	2021-08-10	1	-7/+4
\| \| \| \| \|	Vestige from when we were scanning all files in the current directory rather than channels.json.
*	.	Kartik K. Agaram	2021-08-10	1	-4/+2
\|
*	.	Kartik K. Agaram	2021-08-10	1	-1/+1
\|
*	.	Kartik K. Agaram	2021-08-10	1	-6/+5
\|
*	.	Kartik K. Agaram	2021-08-10	1	-15/+9
\|
*	.	Kartik K. Agaram	2021-08-10	1	-2/+2
\|
*	.	Kartik K. Agaram	2021-08-10	1	-2/+2
\|
*	.	Kartik K. Agaram	2021-08-10	1	-2/+2
\|
*	.	Kartik K. Agaram	2021-08-10	1	-8/+1
\|
*	.	Kartik K. Agaram	2021-08-10	1	-27/+27
\|
*	.	Kartik K. Agaram	2021-08-10	1	-9/+0
\|
*	.	Kartik K. Agaram	2021-08-08	3	-5/+28
\|
*	.	Kartik K. Agaram	2021-08-07	1	-0/+1
\|
*	.	Kartik K. Agaram	2021-08-07	4	-10/+111
\|
*	.	Kartik K. Agaram	2021-08-06	1	-3/+9
\|
*	.	Kartik K. Agaram	2021-08-06	1	-8/+15
\|
*	beginnings of a Slack archive reader	Kartik K. Agaram	2021-08-06	2	-0/+105
\| \| \| \| \| \| \| \|	I'm hackily depending on Python (3.something) to prototype the disk image creator. But no non-std libs. Once the disk image is created, I've validated that it can be loaded from disk without too much latency (assuming KVM).
*	.	Kartik K. Agaram	2021-08-02	1	-0/+4
\|
*	better error handling on stray commas	Kartik K. Agaram	2021-08-01	2	-0/+47
\|
*	.	Kartik K. Agaram	2021-08-01	1	-13/+16
\|
*	.	Kartik K. Agaram	2021-08-01	1	-1/+3
\|
*	support non-line-oriented processing in next-word	Kartik K. Agaram	2021-07-29	16	-36/+73
\| \| \| \|	Immediately this simplifies support for comments in image data.
*	maintain aspect ratio when rendering images	Kartik K. Agaram	2021-07-29	3	-8/+43
\|
*	.	Kartik K. Agaram	2021-07-29	1	-0/+2
\|
*	downsampling photos makes the dithering seem worse	Kartik K. Agaram	2021-07-29	1	-2/+2
\| \| \| \|	We support 128px, so let's use the whole 128px.
*	.	Kartik K. Agaram	2021-07-29	1	-1/+1
\|
*	shell: reliable image rendering	Kartik K. Agaram	2021-07-29	1	-0/+1
\|
*	shell: rendering image definitions	Kartik K. Agaram	2021-07-29	1	-1/+68
\| \| \| \| \| \|	Not quite working yet. Only the very first rendering succeeds. After that any keypress triggers a second render which aborts. Image is getting corrupted in memory somehow.
*	shell: support loading 128x128px images	Kartik K. Agaram	2021-07-28	6	-10/+10
\| \| \| \| \| \|	I'm loading them in uncompressed ASCII format, and all streams and gap buffers all over the place need to get massively scaled up to 256KB capacity. But the tests don't yet run out of RAM, so I'll keep going.