miloignis/kraken
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
326 Commits 2 Branches 0 Tags
bb3ccf808b6cfd42247545e87ff81ae48be8b824
Commit Graph

230 Commits

Author SHA1 Message Date
Nathan Braswell
7b6e47544a Some more bugfixes, got regex working as well as the cpp version. (leaks memory like craaazy) 2015-06-14 18:13:52 -04:00
Nathan Braswell
36833ec263 fixed another oneString CGenerator bug. Most of these should really be removed, but I wanna commit this one now 2015-06-12 15:02:22 -04:00
Nathan Braswell
cd1b10a633 Fixed bug where no parameter function calls were not typechecked and function/struct name collision. Improved regex library to where it can do straight-line regexs 2015-06-12 14:16:28 -04:00
Nathan Braswell
d90cb4b6db Tons of stuff. Regex still a work in progress, along with related template member function scoping bugs 2015-06-09 20:02:02 -04:00
Nathan Braswell
47bc52f00c Some bugfixes, allow overloading of [] and add that to vector and string, work on regex. Need closures before that finishes.... 2015-06-08 21:47:02 -04:00
Nathan Braswell
69048ebc31 Allow copy_constructers to be called with other types (i.e. var a:string = "hi" calls fun copy_construct(it:char**):void 2015-06-07 19:54:08 -04:00
Nathan Braswell
7abab02fbf More all-rounded implementation of the new objecty features, several bugfixes, and updates to the standard library to behave. Vector still needs a bit more work for some operations 2015-06-05 00:34:24 -04:00
Nathan Braswell
6f9ceaa717 work on string, bug fixes, overloaded assignment operator. Still need to get overloaded copy_construct for declaration assignment 2015-06-01 01:43:23 -04:00
Nathan Braswell
bbcebf7c17 I'm pretty sure I missed cases and introduced bugs with the new CGenerator triplit system, but I finally got all normal tests passing, and it's almost 5 in the morning, so I'm calling it a night. We have destructors and copy constructors now\! I need to work out copy assignment, but again, another day 2015-05-30 04:43:01 -04:00
Nathan Braswell
b71bb84070 Now allow one-line functions without codeblocks\! (espically nice for lambdas) fun a():void println(4), or fun():void println(7) 2015-05-27 12:27:50 -04:00
Nathan Braswell
85834789e4 Some bugfixes and added templated member functions\! (even for templated objs\!) In this vein, added in_place and map functions to vector\! 2015-05-27 00:58:33 -04:00
Nathan Braswell
88bab7f0d7 Woo! Very basic lambdas! 2015-05-25 22:03:24 -04:00
Nathan Braswell
920a9ab81f ahh, I git added in the tests directory and thus missed the entire ASTTransformation. Sigh. 2015-05-25 12:20:37 -04:00
Nathan Braswell
7448a88f66 Fixed up function template inference to handle pointer to template types 2015-05-23 23:40:55 -04:00
Nathan Braswell
eebffb404e Ok, full functions as values support with types and fixes of earlier bugs. Now typedef the function pointer types when generateing C. 2015-05-22 22:30:25 -04:00
Nathan Braswell
6a144c9a8a Add initialization to NULL for Type that I missed last time 2015-05-18 17:14:12 -04:00
Nathan Braswell
911b25f78b Initial work for functions as values. Actually works, but you can't write out the type and thus can only declare function pointer variables with type inference. 2015-05-18 04:46:03 -04:00
Nathan Braswell
65fd2ed9b7 Changed from typedef to obj and def 2015-05-16 12:05:23 -04:00
Nathan Braswell
aafb52d108 A kinda ugly/hacky addition, but overloading () as operator() works now! 2015-05-16 00:09:50 -04:00
Nathan Braswell
795f8715ff Break, continue, and DEFER statements! Woooo 2015-05-15 15:19:55 -04:00
Nathan Braswell
5dcb5ba02a Turns out it works now, added more to the test and some comments explaining why 2015-05-15 13:03:45 -04:00
Nathan Braswell
a64624ba25 More complicated inference working, now just have to round out the edges 2015-05-15 12:29:44 -04:00
Nathan Braswell
f5e74ca7ce Template function instantiation type inference work - super simple inference works (i.e. id<T>(i:T) { return i; }), and the framework is good, just have to flesh out unifyType to handle more than the trivial case 2015-05-14 00:25:18 -04:00
Nathan Braswell
d70fa4ebbb Removed/commented out some destructor stuff. It'll come back for defer 2015-05-10 02:48:05 -04:00
Nathan Braswell
44d310c092 Got simple type inference for declaration statements working with test 2015-05-10 02:18:59 -04:00
Nathan Braswell
87e1853713 Wooo! Fixed up remaining bugs in new syntax! 2015-05-09 06:24:56 -04:00
Nathan Braswell
c22cadeed7 The template part of the new syntax! Gotta go transform all of stdlib and the tests now and see if any other bugs pop up 2015-05-09 03:13:40 -04:00
Nathan Braswell
08431aa748 New syntax starting to work! Not templates yet, that's next 2015-04-14 16:08:36 -04:00
Nathan Braswell
e3aa531856 Wooo actual scoping and better C interop 2015-04-10 00:37:31 -04:00
Nathan Braswell
e37836aea5 Working toward new C inline style 2015-04-04 01:32:40 -04:00
Nathan Braswell
47982114af Only makes output directory once 2015-03-24 19:33:31 -04:00
Nathan Braswell
414113f954 Wooo passthrough and more cleanup! Remove 3rd pass! 2015-03-24 18:29:21 -04:00
Nathan Braswell
48dc1f8e4d Fixed - and + as unary operators, negative numbers, etc 2015-03-24 15:47:08 -04:00
Nathan Braswell
b01078e7a0 If my measurements are right, WAYYYY better now. Missed a huge memory leak, now like 0.34147GB, I think 2015-03-23 15:52:25 -04:00
Nathan Braswell
6fa06f2b7e Removed LALR parser and an associated member from Symbol, which is used so much that actually got us down to ~3.42GB! 2015-03-23 15:28:03 -04:00
Nathan Braswell
2c4dbc60d1 Fixes for state generation to reduce memory usage - adding in optional semicolons balooned our memory usage to somewhere under 8 gigs, with some simple refactoring we're back down to a bit over 4. Needs to be smaller, but it's an improvement 2015-03-23 14:35:28 -04:00
Nathan Braswell
f8e82b5302 Added initial support for omitting semicolons! (the parser can now interpret either semicolons or linebreaks as the end of statements, but it can still interpret line breaks as white space too, so multi line statements still work.) You will need semicolons to break up multiple statements on the same line, however 2015-03-22 22:43:33 -04:00
Nathan Braswell
af2a28ae4f Fix chained indirect templates: template<T> |T| fun(|vector<T>| a) { return a.at(0); }, etc 2015-03-17 21:06:31 -04:00
Nathan Braswell
111e5a373e != didn't work either - missed it in the builtins at the top of ASTTransformation.cpp 2015-03-16 06:24:59 -04:00
Nathan Braswell
78d16c5510 So Chris found that if/else statements using blocks didn't work because they generated C like if(a) {b}; else {c}; -- so that's fixed now, with test :) 2015-03-16 05:59:56 -04:00
Nathan Braswell
15c4a4bf0e Better compiler invocation style and put all generated files (including dot files) in output folder 2015-03-15 18:41:55 -04:00
Nathan Braswell
f1b0d46ca0 WOW. NodeTree deletes duplicate nodes. So x+x didn't work.... 2015-03-14 16:53:00 -04:00
Nathan Braswell
e04cf03858 little less 2015-03-14 03:32:32 -04:00
Nathan Braswell
50cd9969b6 Ehh, not handleing syntax nodes is to common to make it an error right now 2015-03-14 02:50:18 -04:00
Nathan Braswell
7a865b82e9 Use std:cerr for errors 2015-03-14 02:42:07 -04:00
Nathan Braswell
6a311fb237 WOOO compiles all in one file! Dependencies resolved! Next up, C name mangeling for scoping 2015-03-11 01:58:10 -04:00
Nathan Braswell
9e9b4371da Fixed up testing etc 2015-01-09 14:28:07 -05:00
Nathan Braswell
ad0e90f74a Fixed up vector and make template searching functions use std::set to remove false positives on multiple matching templates 2015-01-07 03:03:14 -05:00
Nathan Braswell
b66e8cbaa5 Merge branch 'master' of https://www.github.com/Limvot/kraken 2015-01-07 02:29:52 -05:00
Nathan Braswell
aaca71a211 Got the new scoping working! Still some odd stuff happening to certian templates, and I think vector is having problems with new/traits. Really need to get canonnical filenames and what not worked out 2014-12-30 01:22:09 -05:00