Nathan Braswell
|
d72cbdcedb
|
Serilization and caching the table works!
|
2015-08-26 03:45:34 -04:00 |
|
Nathan Braswell
|
6a62f03fb4
|
Move away from fully_reduces_to_null to rule_position, fixed some bugs. Have not fixed all, still looks up unkown key-value for the full kraken parse test.
|
2015-08-13 01:48:35 -04:00 |
|
Nathan Braswell
|
4b6693ac1c
|
Fixed add_children to actually add children correctly and only do the ambiguity at the right time. Still have the looking in map with non-existant key problem, but it is actually parsing nicely.
|
2015-08-12 23:15:41 -04:00 |
|
Nathan Braswell
|
8321b35a03
|
moving computers. Parser still not quite right
|
2015-08-12 17:41:26 -04:00 |
|
Nathan Braswell
|
58ab3e311f
|
Starting to really parse! Fixed the bad syntax trees (reduction copy_construct wasn't actually copying label because param and dest had same name) and fixed the krakenGrammer and CGenerator to allow escaped quotes in regular strings. add_children needs to be finished so the resulting trees don't go nuts with ambiguity, and some situations still cause vector access errors, which is odd. Also added the crazy simple write_file to io and the test_grammer.krak now outputs a syntax_tree.dot
|
2015-08-11 01:07:16 -04:00 |
|
Nathan Braswell
|
2777ca10f1
|
Parser is mostly working as a recognizer now, though the grammer2 still causes vector out of bounds. Sigh. Also, it segfaults on printing the tree, even with no null leaves. Somehow internal data from a symbol becomes null.
|
2015-08-09 04:18:31 -04:00 |
|
Nathan Braswell
|
674e7e6538
|
More work, finishing the parse_input and lots of reducer
|
2015-08-06 17:38:41 -04:00 |
|
Nathan Braswell
|
1f119af8ad
|
more work
|
2015-08-06 02:42:40 -04:00 |
|
Nathan Braswell
|
dec9b7d0bd
|
some tests failing because things have been made reference in vector, but grammer actually generates the state set for the real grammer in 2 minutes or so after a day of profiling and bugfixing, so this is gonna be committed.
|
2015-08-05 03:43:34 -04:00 |
|
Nathan Braswell
|
e1dbe08c0a
|
Optimization of string and vector with references and less functional code, bugfix of closing over references
|
2015-08-04 14:57:56 -04:00 |
|
Nathan Braswell
|
c987459a77
|
Gonna checkpoint here at a lot of good progress. Wrote generate_state_automaton, closure, and goto, and they compile and run, though don't work. (they don't crash yet though).
|
2015-08-04 01:07:33 -04:00 |
|
Nathan Braswell
|
a4f2febfd5
|
some bug fixes, templated operator method overloading
|
2015-08-03 18:37:42 -04:00 |
|
Nathan Braswell
|
517f9b79fb
|
moving computers
|
2015-08-03 14:38:17 -04:00 |
|
Nathan Braswell
|
f431b0305a
|
some more work on the first set. I think it might actually be right now
|
2015-07-16 23:32:27 -04:00 |
|
Nathan Braswell
|
51adf491fa
|
fix reference type inference
|
2015-07-15 13:56:57 -04:00 |
|
Nathan Braswell
|
06f36f2a87
|
Return by reference and pass by reference working with objects. Closures might present problems, however
|
2015-07-15 00:53:53 -04:00 |
|
Nathan Braswell
|
8c490908d4
|
Saving work pre-references
|
2015-07-13 12:16:30 -04:00 |
|
Nathan Braswell
|
f3cdea068e
|
Added mem::safe_recursive_clone, and while it works for regex, it's actually slower then remaking it. Hmmmm, maybe because some of the stdlib is inefficent
|
2015-07-07 00:46:00 -04:00 |
|
Nathan Braswell
|
92f5c63c9a
|
Grammer loading works and test file can lex itself. Got rid of regex memory saftey for speed, will reintroduce next
|
2015-07-06 13:48:19 -04:00 |
|
Nathan Braswell
|
501331e37b
|
Grammer file parser works great
|
2015-07-06 12:49:29 -04:00 |
|
Nathan Braswell
|
cfadf7cebb
|
Fix nested closures and related bugs
|
2015-07-05 02:34:45 -04:00 |
|
Nathan Braswell
|
d2b12fea35
|
Fix vector addEnd so that it increments size AFTER the resize and resize() doesn't copy_construct an invalid item, which only SOMETIMES failes
|
2015-07-04 12:59:05 -04:00 |
|
Nathan Braswell
|
54721b4284
|
More work on grammer and standard library! It can kinda load grammer now! Kinda. Marcus ran into the function pointer returns pointer ambiguity, so that'll have to be done tomorrow.
|
2015-07-04 03:21:36 -04:00 |
|