| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
| |
This speeds up the final test but not all together.
|
| |
|
|
| |
Something wrong with my profiling, though. Numbers aren't adding up.
|
| |
|
|
|
| |
Tests are getting slow so quickly that I'm tempted to push forward the
rewrite to C.
|
| | |
|
| |
|
|
|
|
|
|
| |
This allows the initial model of a trace to have just one new number:
the currently expanded row. We'll make it more flexible later if we feel
the need.
Still no collapsing lines, though. But we're getting there.
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
This bugfix is not tested yet, but that's in the works.
|
| |
|
|
| |
This just supports it, but we don't actually have collapse yet.
|
| |
|
|
|
|
|
|
|
|
| |
Expanding trace items, of course, doesn't work on anything but the last
item. But we've now caught automatic tests up to the state of
development.
These tests are hugely expensive. Previous .t files run in ~30s. Already
the trace browser tests clock in at 2 minutes. And we're just getting
started. Time to explore reimplementing the mu VM in C.
|
| |
|
|
|
|
| |
Don't prevent run-code from clobbering existing functions, but warn
because it makes traces easier to read if the different sections of a
test can be distinguished.
|
| | |
|
| | |
|
| | |
|
| | |
|
|
|
This was painless compared to the chessboard app. Still need to handle
input, though. Fingers crossed..
|