| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
There were several places where we push a call on to a routine without
incrementing call-stack depth, which was used to compute the depth at
which to trace an instruction. So sometimes you ended up one depth lower
than you started a call with. Do this enough times and instructions that
should be traced at level 100 end up at level 0 and pop up as errors.
Solution: since call-stack depth is only used for tracing, include it in
the trace stream and make sure we reset it along with the trace stream.
Then catch all places where we forget to increment call-stack depth and
make sure we catch such places in the future.
When I first ran into this with Caleb I thought there must be some way
that we're writing some output into the warnings result. I didn't
recognize that the spurious output as part of the trace, just at the
wrong level.
|
| |
|
|
| |
Another bugfix. Thanks Jack and Caleb Couch.
|
| | |
|
| |
|
|
| |
Now we can collect all traces, just modulating the depth.
|
| |
|
|
|
|
|
| |
At the lowest level I'm reluctantly starting to see the need for errors
that stop the program in its tracks. Only way to avoid memory corruption
and security issues. But beyond that core I still want to be as lenient
as possible at higher levels of abstraction.
|
| |
|
|
|
|
|
|
| |
We still can't check ingredient types, and even this is still a run-time
check. We'll need to start tracking recipe signatures at some point.
I've had to introduce a hack called /skiptypecheck. Time to get generics
working.
|
| | |
|
| |
|
|
|
|
|
|
| |
Always show recipe name where error occurred. But don't show internal
'interactive' name for sandboxes, that's just confusing.
What started out as warnings are now ossifying into errors that halt all
execution. Is this how things went with C and Unix as well?
|
| |
|
|
|
| |
Bugfix to 2186. I hadn't taken care of 'reload' as cleanly as I had
'run-interactive'.
|
| |
|
|
|
|
|
|
|
|
|
| |
Turns out the default format for printing floating point numbers is
neither 'scientific' nor 'fixed' even though those are the only two
options offered. Reading the C++ standard I found out that the default
(modulo locale changes) is basically the same as the printf "%g" format.
And "%g" is basically the shorter of:
a) %f with trailing zeros trimmed
b) %e
So we'll just do %f and trim trailing zeros.
|
| |
|
|
|
| |
Exception: allow run-interactive to continue even if it encounters
errors in parsing its ingredient as mu code.
|
| | |
|
| |
|
|
|
| |
We don't want to add duplicate fields to the container everytime we hit
F4.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Finally terminate the experiment of keeping debug prints around. I'm
also going to give up on maintaining counts.
What we really need is two kinds of tracing:
a) For tests, just the domain-specific facts, organized by labels.
b) For debugging, just transient dumps to stdout.
b) only works if stdout is clean by default.
Hmm, I think this means 'stash' should be the transient kind of trace.
|
| | |
|
| | |
|
| |
|
|
|
| |
Right after I lectured Caleb about not just adding random deltas to make
a program correct, too.
|
| | |
|
| |
|
|
| |
Extend deadline for run-interactive.
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This time I'm getting rid of all the intricate SCREEN hacking. Now
run-interactive is idiomatic: it creates a recipe on the fly, runs it,
and returns various results and stats about it.
One big consequence of the recipe now running interactively inside a
local-scope: you can't just use numeric addresses anymore -- not without
using /raw. Fixed one of the tests to reflect this.
I realize now that I don't really want my students to be using raw
addresses in their sandboxes, even if that was my idiom for lower-level
scenarios.
Hopefully running sandboxes in separate (synchronous) routines will be
easy now.
|
| |
|
|
|
| |
Still trying to minimize the work around run-interactive so I can build
it with a sandboxed routine rather than a call in editor space.
|
| | |
|
| |
|
|
| |
Thanks Caleb Couch.
|
| |
|
|
|
| |
More friendly way to 'stash' stuff in the trace so that you can toggle
lines of code to see their stashed traces.
|
| |
|
|
|
|
|
|
|
|
| |
Still iterating on the right way to handle incorrect number of
ingredients. My first idea of creating null results doesn't really work
once they're used in later instructions. Just add a warning at one place
in the run loop, but otherwise only add products when there's something
to save in them.
Undoes some work around commit 1886.
|
| |
|
|
|
|
|
| |
Region to click on to edit is now reduced to just the menu bar for the
sandbox (excluding the 'x' for deleting the sandbox). The symmetry there
might be useful, but we'll see if the relative click area is
in line with how commonly the actions are performed.
|
| |
|
|
|
| |
Generalize a little layer as a first step to providing traces for other
layers.
|
| | |
|
| |
|
|
|
|
|
|
| |
For example:
x:number <- index y:address:array:number, 3
(forgetting to do a lookup)
Thanks Caleb Couch.
|
| |
|
|
| |
Should be a little bit more mnemonic.
|
| |
|
|
| |
First step to reducing typing burden. Next step: inferring types.
|
| | |
|
| | |
|
| |
|
|
| |
Also standardized warnings.
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Eventually we might be able to get rid of die entirely.
This is just a preliminary stab at a random error. In the process I ran
into two issues that have impeded debugging before:
a) Naming conflicts within scenarios are a real no-no. I need to warn on
them, but the rules are getting complicated:
Always print warnings on redefine
But not in interactive mode
Or in scenarios checking warning behavior
Unless the scenario recipe itself is overridden
b) Now that we've added collect_layers and a long time can go between
traces, debugging is a minefield because trace lines don't print to
screen immediately after they're created. Need to do something about
that. Maybe explicitly trigger collection by tracing '\n' or something.
These are the next two items on my todo list.
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Things I figured out:
- 'row' in render-screen doesn't perfectly track cursor-row in screen
- proximal cause was forgetting to add left:number to stop-printing
- trying to print to screen outside bounds was silently succeeding and
corrupting simulated memory
- if we silently ignore prints outside bounds things are fine
But why are prints outside screen bounds working? We should be accessing
screen data using 'index', and that's checking its bounds.
|
| |
|
|
|
| |
Debugging simulated-screen support is taking too long, and I suddenly
have a few higher priorities.
|
| |
|
|
| |
I'm writing to location 'screen' somehow that's not the raw location.
|
| | |
|
| | |
|