Nathan Braswell
|
b62c3e729f
|
Some more work, and a --parse-only option to support the new kraken.vim vim plugin that adds Syntastic support (and syntax highlighting)
|
2015-07-03 18:34:46 -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
|
03770028ad
|
Fixed some bugs in Parser::firstSet and added a bit of caching. It still doesn't work quite right, though, there's some problem with nullable left recursion. However, it's better than it was, and I need to go to bed. More work later.
|
2014-06-30 01:57:50 -07:00 |
|
Nathan Braswell
|
005659b7b7
|
Added caching of the RNGLR table. It is automatically regenerated whenever the grammer changes. Right now it has dropped compiling the test file from 30 seconds to less than one second.
|
2013-12-31 02:53:52 -06:00 |
|
Nathan Braswell
|
77f2b0a3e5
|
Lot's of work on a CGenerator, AST and language improvements
|
2013-11-01 02:52:18 -04:00 |
|
Nathan Braswell
|
b9ffe33d0b
|
Made Symbol always stack, not heap, allocated. Finally fixed bugs with ASTTransformation.
|
2013-10-02 03:15:20 -04:00 |
|
Nathan Braswell
|
35a0677d07
|
Still in progress
|
2013-08-22 15:41:30 -04:00 |
|
Nathan Braswell
|
9460bacf1c
|
Fixed some problems with grammer loading of null rules and rules non-terminals and terminals that had the same name
|
2013-08-04 14:24:20 -04:00 |
|
Nathan Braswell
|
3fba45591b
|
Changed Parser to be a virtual base class, inherited by LALRParser
|
2013-07-30 01:42:51 -04:00 |
|
Nathan Braswell
|
6d7b38a03b
|
More work towards RNGLR. First, NodeTree is now a template. Second, I've started writing the actual GLR parser and GSS and other things, but am still in the first write process.
|
2013-07-28 19:45:08 -04:00 |
|
Nathan Braswell
|
726ead0455
|
Pull out table to it's own Table class in prep for adding RNGLR algorithm.
|
2013-07-16 11:15:58 -04:00 |
|
Nathan Braswell
|
f84657f1ed
|
Fixed a lot of bugs. Actually gets through the entire experimental grammer. (A largeish experimental grammer for Kraken written to continue testing and to really start language development.
|
2013-07-10 23:50:53 -04:00 |
|
Nathan Braswell
|
1c85e3693e
|
Added support for null rules, works well.
|
2013-07-09 02:45:59 -04:00 |
|
Nathan Braswell
|
85da0bf646
|
Lexer/RegExs now work on a very basic level. Regular expressions have not been fully implemented, but the structure is there. It creates trivial regexs that only accept a specified sequence, no actual regular expression power.
|
2013-07-02 01:47:42 -04:00 |
|
Nathan Braswell
|
6a2977d12a
|
Redid lookahead in a much much saner way. Now should be a legitimate parser somewhere between LALR(1) and LR(1).
|
2013-06-26 14:27:28 -04:00 |
|
Nathan Braswell
|
a307e1f143
|
Changed followset to work on states, fixed many bugs. Kraken now should have a fairly legitimate LALR(1) parser. (null productions have not yet been added.)
|
2013-06-23 05:06:38 -04:00 |
|
Nathan Braswell
|
20b059ca45
|
Transfer computer
|
2013-06-21 14:16:16 -04:00 |
|
Nathan Braswell
|
66b88cb5d7
|
Added followSets(Symbol*) function to compute follow sets.
|
2013-06-13 23:43:12 -04:00 |
|
Nathan Braswell
|
da6e5f3de9
|
Added a firstSet(Symbol*) function that takes in a Symbol* and returns a vector of Symbol* s that are the first set of the first Symbol*.
|
2013-06-13 19:11:31 -04:00 |
|
Nathan Braswell
|
0c651a5d9e
|
Added EOF symbol, making the getTable() saner for ACCEPT and lookahead support in ParseRule
|
2013-06-13 14:25:10 -04:00 |
|
Nathan Braswell
|
949dbc532a
|
Fixed lots of bugs, added much more sane and efficient ParseAction table instead of recalculating every time. Fixed lots of bugs and inefficencies. Some temporary hacks in the table, mostly having to do with not having an EOF Symbol yet.
|
2013-06-04 19:50:16 -04:00 |
|
Nathan Braswell
|
0c4af245bf
|
Cleaned up some debug messages, parseInput(string) now returns a parse tree of NodeTree*. This is done by having each non-terminal Symbol have it's own subtree and calling a method to combine subtrees and terminals on each reduce. The output is now the DOT version of the parse tree.
|
2013-05-30 19:49:19 -04:00 |
|
Nathan Braswell
|
69713ffe1a
|
Actually parses LR(0) grammers and sentences\! Only two problems is very inefficint gotoTable() and actionTable(), and it REJECTS the sentence after completing the Goal->Sentence(*) production. However, need to go to bed now. Does practially work though\! Yay\!
|
2013-05-29 20:43:35 -04:00 |
|
Nathan Braswell
|
315dc55409
|
Modifed set creation to use a State object. Set creation works
|
2013-05-26 22:12:47 -04:00 |
|
Nathan Braswell
|
1aec4ee6a7
|
Added first go at creating stateSets. This is a commit to save work, as it still segfaults.:
|
2013-05-24 00:00:41 -04:00 |
|
Nathan Braswell
|
e4f83cbb67
|
Added main loop of parser.
|
2013-05-23 01:35:54 -04:00 |
|
Nathan Braswell
|
200cff6ab4
|
Separate the toString and toDOT methods.
|
2013-05-20 22:59:57 -04:00 |
|
Nathan Braswell
|
d2698cf203
|
Implemented grammer loading
|
2013-05-20 19:34:15 -04:00 |
|