fluencelabs/lalrpop
1
0
Fork 0
mirror of https://github.com/fluencelabs/lalrpop synced 2025-03-16 17:00:53 +00:00
Code Issues Projects Releases Wiki Activity
1,143 Commits 13 Branches 20 Tags
Commit Graph

1143 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Markus Westerlind
3173a41090 Only build mdbook on stable to fix CI 2020-08-05 15:57:37 +02:00
Markus Westerlind
6d2c21e6ac refactor: Avoid generating a panic for unwrapping symbols 
Merges the panic from the unwrap into the general symbol mismatch,
hopefully simplifying LLVMs work.

cc #543
 2020-07-29 09:15:24 +02:00
Markus Westerlind
d0004f35b3
Merge pull request #502 from TheBeggs/doc-link-fixes 
Doc link fixes
 2020-06-26 08:47:20 +02:00
Markus Westerlind
ce7af5a0dc
Merge branch 'master' into doc-link-fixes 2020-06-26 08:47:13 +02:00
Markus Westerlind
58fdec8ecf
Merge pull request #532 from ElEHsiang/update-calculator-to-2018 
Ship calculator .lalrpop files to rust 2018
 2020-06-22 10:12:11 +02:00
Yun Hsiang
793df8d6b4 Ship calculator .lalrpop files to rust 2018 
Signed-off-by: Yun Hsiang <hsiang023167@gmail.com>
 2020-06-08 22:37:35 +08:00
Markus Westerlind
dd87e0eef3
Merge pull request #529 from ndmitchell/patch-1 
Add From<E> for ParseError<L, T, E>
 2020-05-15 10:43:00 +02:00
Neil Mitchell
878a167c80
Add From<E> for ParseError<L, T, E> 2020-05-14 17:03:44 +01:00
Markus Westerlind
c1009df08a
Merge pull request #528 from Marwes/release 
Version 0.19
 2020-05-12 10:49:53 +02:00
Markus Westerlind
6400c14cb8 Fix version in clog command 2020-05-12 10:07:16 +02:00
Markus Westerlind
6632b80ecf Fix version in RELEASES.md 2020-05-12 10:06:38 +02:00
Markus Westerlind
91f2500f68 Sleep longer when publishing 2020-05-12 10:05:38 +02:00
Markus Westerlind
69aa2e09ce Bump version to 0.19 0.19.0 2020-05-11 23:49:25 +02:00
Markus Westerlind
274c918251
Merge pull request #527 from Marwes/span_empty 
fix: Don't include whitespace in the span with empty nonterminals
 2020-05-11 18:45:11 +02:00
Markus Westerlind
11a50e70f7 fix: Don't include whitespace in the span with empty nonterminals 
Empty non-terminals used to be spanned by the whitespace between the
previous and next token. If the non-terminal was the first thing in
another production this span would leak into that production, causing it
to have a to large span assigned.

This makes empty productions instead only point to the start of the
lookahead to avoid this problem.
 2020-05-11 00:28:14 +02:00
Markus Westerlind
672716e098
Merge pull request #525 from Marwes/symbol_type 
perf: Move the symbol mismatch panic into a colder path
 2020-05-02 17:58:21 +02:00
Markus Westerlind
e932f811a6 test: Shrink the maximum depth a bit to avoid stack overflows 2020-05-02 17:42:14 +02:00
Markus Westerlind
0c69e999a5 perf: Move the symbol mismatch panic into a colder path 
Makes it easier for llvm to inline the pop functions
 2020-05-02 17:42:14 +02:00
Markus Westerlind
29a4bea914
Merge pull request #524 from Marwes/refactor_gen 
perf: Emit the GOTO table as nested matches
 2020-05-02 17:40:23 +02:00
Markus Westerlind
8a47ed8c7c perf: Avoid subtracting in goto 2020-05-02 00:16:05 +02:00
Markus Westerlind
554b203531 Document the goto match expression 2020-05-02 00:03:26 +02:00
Markus Westerlind
6ae57343f3
Merge pull request #523 from Marwes/refactor_gen 
perf(parse_table): Avoid generating unused rows in the GOTO matrix
 2020-05-02 00:02:50 +02:00
Markus Westerlind
c5070af2ee perf: Emit the GOTO table as nested matches 
Since most of the goto table is elements that will never be hit we can
shrink the binary size by emitting it as nested matches instead
without losing any performance.
 2020-05-01 23:38:04 +02:00
Markus Westerlind
688b91930a perf(parse_table): Avoid generating unused rows in the matrix 
Only the states in the beginning of a reduction are actually used, the
states for all other rows will never be used (really, no zero elements
will be accessed in the GOTO table so we might be able to do even better
without compromising performance).

By simply reordering the states so that states needed in the GOTO table
occupy the lower indices we can simply avoid generating the remaining
part of the GOTO table.

For the LALRPOP parser itself this is a reduction of about -40% (~100kB)
of the size of the parse tables (not counting the generated code)
 2020-04-30 01:15:59 +02:00
Markus Westerlind
cb1924632b refactor: Various simplification in parse_table generation 2020-04-30 01:15:55 +02:00
Markus Westerlind
dd57a2b187
Merge pull request #521 from crlf0710/master 
Bump dependency versions.
 2020-04-23 15:51:12 +02:00
Markus Westerlind
8d22d77032
Merge pull request #522 from Kingdread/fallible-actions-tutorial 
Tutorial: Add a section about fallible actions
 2020-04-23 15:50:52 +02:00
Daniel Schadt
5ba76e2587 Add a tutorial site on fallible actions 2020-04-20 12:40:38 +02:00
Daniel Schadt
314ee6afc4 Move sections 7 & 8 one back 
This opens up a new section before "Error recovery" to teach about
fallible actions (defined with `=>?`).
 2020-04-20 11:40:38 +02:00
Charles Lew
075150bc8e Bump dependency versions. 2020-04-19 13:06:36 +08:00
Markus Westerlind
0d2e5baf7a
Merge pull request #517 from enterprisey/clippy-fix-1 
Remove some lifetimes in codegen and clippy fix
 2020-04-13 21:49:36 +02:00
Daniel Glus
7796cf881e regenerate lalrpop grammar 2020-03-21 16:49:08 -04:00
Daniel Glus
5247b2512c Remove some lifetimes in codegen and clippy fix 2020-03-18 02:46:15 -04:00
Markus Westerlind
8a96e9646b
Merge pull request #514 from Marwes/fn_mut 
Miscanellous cleanup
 2020-03-10 19:46:34 +01:00
Markus Westerlind
037a37264c doc: Add the lexer feature to the examples 
The lexer feature bled into the examples due to `lalrpop-test` already
specifying it.

Fixes #513
 2020-03-10 16:06:15 +01:00
Markus Westerlind
a07f90289f cargo fix --edition --all 2020-03-10 16:00:06 +01:00
Markus Westerlind
165cf03bd1 cargo fmt 2020-03-10 15:56:07 +01:00
Markus Westerlind
8f73c9dcfb feat: Use FnMut/FnOnce in ParseErrors map functions 2020-03-10 15:53:32 +01:00
Markus Westerlind
634efcc3b7
Fix releases.md 2020-03-04 19:53:59 +01:00
Markus Westerlind
bcdfaf6e87
Merge pull request #512 from Marwes/version 
Version
 2020-03-04 19:53:13 +01:00
Markus Westerlind
dd974b0b62 Version 0.18.1 0.18.1 2020-03-04 17:13:43 +01:00
Markus Westerlind
a9a775ebb8 fix: Remove eprintln which I thought were removed 2020-03-04 17:13:08 +01:00
Markus Westerlind
452e1dc6d4
Merge pull request #511 from Marwes/version 
Version 0.18.0
 2020-03-04 17:00:32 +01:00
Markus Westerlind
c044587f1d Fix the version in RELEASE.md 0.18.0 2020-03-04 16:45:01 +01:00
Markus Westerlind
648185d3da Mention that the lexer feature is needed 2020-03-04 16:39:59 +01:00
Markus Westerlind
df3f428e37 Version 0.18.0 2020-03-04 16:37:23 +01:00
Markus Westerlind
45ae391f3e Add clog to version updating 2020-03-04 16:37:02 +01:00
Markus Westerlind
723678f364
Merge pull request #509 from Marwes/lexer_comment 
feat: Allow the tokenizer to contain custom skip regexes/literals
 2020-03-03 21:03:21 +01:00
Markus Westerlind
b22e892fc2 Update lrgrammar 2020-03-03 20:32:05 +01:00
Markus Westerlind
26932e85c0 Lower the required rust version 2020-03-03 20:30:43 +01:00