about summary refs log tree commit diff stats
Commit message (Collapse)AuthorAgeFilesLines
* 3690Kartik K. Agaram2016-11-252-0/+2
| | | | | Attempting to write to a file in a non-existent directory silently writes to a black hole in Unix. Make that explicit.
* 3689Kartik K. Agaram2016-11-254-12/+8
|
* 3688Kartik K. Agaram2016-11-2512-10/+40
| | | | | | | Move my todos over the past couple of years into the codebase now that it might be going dormant. Surprising how few todos left undone!
* 3687Kartik K. Agaram2016-11-253-9/+8
|
* 3686Kartik K. Agaram2016-11-252-8/+8
|
* 3685Kartik K. Agaram2016-11-232-2/+0
|
* 3684Kartik K. Agaram2016-11-232-14/+16
| | | | | | | | | | | | | Trying to find examples where a layer includes complexity just so later layers can hook into it. Resolve_stack is the one big one I've found that isn't just a simple function-call indirection that may later get more complex. Conclusion of a conversation with Stephen Malina: Such examples should make one very nervous, because this sort of creep is how we end up with over-engineered abstractions (http://www.joelonsoftware.com/articles/fog0000000018.html). We need to step very carefully anytime we make the outsider reader's comprehension task harder..
* 3683Kartik K. Agaram2016-11-232-0/+52
|
* 3682Kartik K. Agaram2016-11-224-14/+18
| | | | | Clean up the flow of "mu --trace" followed by "mu browse-trace interactive".
* 3681Kartik K. Agaram2016-11-224-42/+44
| | | | Couple of fixes to example programs.
* 3680Kartik K. Agaram2016-11-171-16/+13
| | | | | Was there some reason I used a subshell? No reason recorded at bottom.
* 3679Kartik K. Agaram2016-11-177-17/+62
|
* 3678 - complain about string labelsKartik K. Agaram2016-11-152-1/+10
| | | | Thanks Jack Couch for running into this.
* 3677 - gracefully handle parse errors in scenariosKartik K. Agaram2016-11-151-3/+26
| | | | Thanks Jack Couch for running into this.
* 3676 - stop scenarios on error in transformKartik K. Agaram2016-11-151-1/+10
| | | | Thanks Jack Couch for running into this.
* 3675Kartik K. Agaram2016-11-155-12/+16
|
* 3674Kartik K. Agaram2016-11-132-2/+2
|
* 3673Kartik K. Agaram2016-11-135-15/+127
|
* 3672Kartik K. Agaram2016-11-131-3/+3
|
* 3671 - support text in '$print'Kartik K. Agaram2016-11-122-0/+6
|
* 3670 - parsing whole numbers from textKartik K. Agaram2016-11-111-0/+83
| | | | Thanks Ella Couch for contributing this function.
* 3669Kartik K. Agaram2016-11-113-7/+7
|
* 3668 - $read a word, stopping at whitespaceKartik K. Agaram2016-11-112-8/+30
| | | | | | | | | | | | Useful for programming contests like https://halite.io Doesn't suffer from C++'s usual buffered gotchas: it'll skip leading whitespace. Slow, though. Can be speeded up, though. - 20 minutes later But what's the point? Typewriter mode is actually harder to test than 'raw' console mode. Writing Mu programs in typewriter mode is just going to encourage us all to slack off on writing tests.
* 3667Kartik K. Agaram2016-11-1142-444/+745
|
* 3666Kartik K. Agaram2016-11-111-0/+2
| | | | Fix a memory leak.
* 3665Kartik K. Agaram2016-11-111-22/+22
|
* 3664Kartik K. Agaram2016-11-101-1/+0
|
* 3663 - fix a refcounting bug: '(type)' != 'type'Kartik K. Agaram2016-11-1016-65/+201
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This was a large commit, and most of it is a follow-up to commit 3309, undoing what is probably the final ill-considered optimization I added to s-expressions in Mu: I was always representing (a b c) as (a b . c), etc. That is now gone. Why did I need to take it out? The key problem was the error silently ignored in layer 30. That was causing size_of("(type)") to silently return garbage rather than loudly complain (assuming 'type' was a simple type). But to take it out I had to modify types_strictly_match (layer 21) to actually strictly match and not just do a prefix match. In the process of removing the prefix match, I had to make extracting recipe types from recipe headers more robust. So far it only matched the first element of each ingredient's type; these matched: (recipe address:number -> address:number) (recipe address -> address) I didn't notice because the dotted notation optimization was actually representing this as: (recipe address:number -> address number) --- One final little thing in this commit: I added an alias for 'assert' called 'assert_for_now', to indicate that I'm not sure something's really an invariant, that it might be triggered by (invalid) user programs, and so require more thought on error handling down the road. But this may well be an ill-posed distinction. It may be overwhelmingly uneconomic to continually distinguish between model invariants and error states for input. I'm starting to grow sympathetic to Google Analytics's recent approach of just banning assertions altogether. We'll see..
* 3662Kartik K. Agaram2016-11-101-2/+2
|
* 3661Kartik K. Agaram2016-11-101-22/+24
| | | | Another place I missed in commit 3309.
* 3660Kartik K. Agaram2016-11-101-8/+11
| | | | | Fix a place I forgot to update in commit 3309, moving to the new type_tree representation.
* 3659Kartik K. Agaram2016-11-101-1/+1
|
* 3658Kartik K. Agaram2016-11-101-23/+23
|
* 3657 - better error messageKartik K. Agaram2016-11-102-3/+15
| | | | Thanks Ella Couch for reporting this.
* 3656Kartik K. Agaram2016-11-1020-79/+92
| | | | | | | | | | | | | | Periodic cleanup to replace 'reply' with 'return' everywhere in the repo. I use 'reply' for students to help reinforce the metaphor of function calls as being like messages through a pipe. But that causes 'reply' to get into my muscle memory when writing Mu code for myself, and I worry that that makes Mu seem unnecessarily alien to anybody reading on Github. Perhaps I should just give it up? I'll try using 'return' with my next student.
* 3655Kartik K. Agaram2016-11-081-1/+1
|
* 3654Kartik K. Agaram2016-11-088-41/+68
| | | | | | | | Follow-up to commit 3321: move get_base_type() more thoroughly to layer 55. The notion of a base_type doesn't really make sense before we introduce type ingredients and shape-shifting containers, and it simplifies early layers a *lot* even including the cost of that *ugly* preamble in layer 55 to retrofit all the places.
* 3653Kartik K. Agaram2016-11-085-5/+27
| | | | | | | | | Don't crash on bad types. I need to be more careful in distinguishing between the two causes of constraint violations: bad input and internal bugs. Maybe I should create a second assert() to indicate "this shouldn't really be an assert, but I'm too lazy to think about it right now."
* 3652Kartik K. Agaram2016-11-084-14/+23
| | | | | size_of(type_tree*) is a mess; clean it up with an eye to the final tangled version.
* 3651Kartik K. Agaram2016-11-082-2/+29
| | | | | | I was under the impression that I only needed static array lengths for container members, but these are *payload* types for allocations. So we need to compute the size of a dynamic array.
* 3650 - first of a couple of holes in refcountingKartik K. Agaram2016-11-081-0/+58
| | | | We weren't checking within (static) array elements for addresses.
* 3649Kartik K. Agaram2016-11-081-6/+0
|
* 3648Kartik K. Agaram2016-11-081-1/+1
|
* 3647Kartik K. Agaram2016-11-081-2/+3
|
* 3646Kartik K. Agaram2016-11-071-22/+22
|
* 3645Kartik K. Agaram2016-11-074-40/+26
| | | | | | Extract a helper to compute the element type of an array. As a side effect, the hack for disambiguating array:address:number and array:number:3 is now in just one place.
* 3644Kartik K. Agaram2016-11-072-71/+80
| | | | Eject some array-related code out of the container layer.
* 3643Kartik K. Agaram2016-11-0714-31/+31
| | | | | Standardize on calling literate waypoints "Special-cases" rather than "Cases". Invariably there's a default path already present.
* 3642Kartik K. Agaram2016-11-0610-119/+167
|
* 3641Kartik K. Agaram2016-11-061-2/+3
|