Mark Karpov
a8bb4d9e10
Merge pull request #86 from mrkkrp/recovery-feature
...
Implement the recovery feature
2016-02-18 12:25:28 +05:00
mrkkrp
4f5d73f230
Remove statement about performance
...
More recent benchmarks made after all changes show no considerable
difference. It's faster than Parsec for sure, so…
2016-02-18 13:22:38 +06:00
mrkkrp
e9df1ba30a
Correct ‘withRecovery’ so that it passes the tests
2016-02-18 13:09:19 +06:00
mrkkrp
4ff95dd2a8
Add tests for ‘withRecovery’ primitive
2016-02-18 02:07:08 +06:00
mrkkrp
1285699b74
Improve coverage of older primitives
2016-02-18 00:12:27 +06:00
mrkkrp
3baa263eab
Auto-backtracking for ‘tokens’ and friends
2016-02-17 23:32:48 +06:00
mrkkrp
4c5aae7098
Update ‘CHANGELOG.md’ and ‘README.md’
2016-02-12 22:12:30 +06:00
mrkkrp
3edbe9f54a
Implement the recovery feature
...
Close #80 .
2016-02-09 13:59:21 +06:00
Mark Karpov
b6048d3ec3
Merge pull request #84 from hvr/pr/ghc8-instances
...
Define MonadFail & Semigroup instances
2016-02-08 19:34:57 +05:00
Herbert Valerio Riedel
9cfb8e627a
Add GHC 8.0 to test-matrix
2016-02-08 11:41:02 +01:00
Herbert Valerio Riedel
ba9bd6a25e
Define MonadFail & Semigroup instances
...
This also enables the respective warnings flags in dev mode
to help megaparsec remain forward compatible.
The dependencies on `semigroup` and `fail` are conditional on
`impl(ghc >= 8)` and avoid CPP and conditionally defined instances
(which would result in an conditional API).
2016-02-08 11:38:10 +01:00
mrkkrp
174c02ce38
Add ‘< 0.6’ constraint to ‘transformers’
2016-02-08 13:37:30 +06:00
mrkkrp
e1be100bd6
Cosmetic whims
...
Can't help it…
2016-02-08 13:36:39 +06:00
Mark Karpov
fe9e4d736e
Merge pull request #82 from mrkkrp/state-on-failure
...
Support getting actual parser state on failure
2016-02-08 11:49:17 +05:00
mrkkrp
2bfc724d8d
Add tests for returned state on failure
2016-02-08 00:41:09 +06:00
mrkkrp
c2c33045cc
Fix the build for older GHCs
2016-02-07 20:50:27 +06:00
mrkkrp
4ba5bd7df1
Update ‘CHANGELOG.md’ file
2016-02-07 19:58:41 +06:00
mrkkrp
7727821c2a
Make some combinators more efficient
2016-02-07 19:42:11 +06:00
mrkkrp
4bde699b84
Support getting actual parser state on failure
...
Close #81 .
This solution is mostly OK as it passes tests and almost all benchmarks
show that there is no performance degradation.
The only function that bothers me is ‘pPlus’ (or ‘mplus’, or
‘<|>’). Benchmarks ‘choice/match’, ‘choice/nomatch’, and ‘manyTill’ show
about 44 % worse performance with current implementation of the feature
— this is not acceptable. All these functions are defined via ‘mplus’,
so it's necessary to find a way to improve that function.
Also, ‘mplus’ is tricky in that it combines different branches of
parsing. Previously, all logic describing how to combine failing
branches into one ‘ParseError’ were in ‘mergeError’ function. Now we
have to have ‘longestMatch’ function to choose right state as well,
because it's natural to expect that state on failure would correspond to
‘ParseError’. This should be done elegantly.
2016-02-06 17:59:50 +06:00
mrkkrp
e5ad380832
Remove redundant mode annotations
2016-01-25 12:56:41 +06:00
mrkkrp
f0e11c3be4
Add section “Related packages” to ‘README.md’
2016-01-24 20:57:04 +06:00
mrkkrp
9918dd5ad9
Add missing per-argument descriptions
2016-01-22 20:20:36 +06:00
mrkkrp
6830477a8d
Add “Stackage LTS” badge
2016-01-22 13:12:22 +06:00
mrkkrp
24b8e41af0
Remove non-informative phrases from docs
...
Parsec's legacy.
2016-01-22 13:00:58 +06:00
mrkkrp
c2f5504725
Fix a typo ‘ParserT’ → ‘ParsecT’
2016-01-22 12:47:15 +06:00
mrkkrp
f69661a930
Drop some upper bounds (‘transformers’ and ‘text’)
2016-01-13 15:00:15 +06:00
mrkkrp
611f2a4e7e
Minor improvements for new combinators
...
In particular, if input has no newline at the end, we need to treat it
specially, because otherwise we will get confusing “incorrect
indentation” message.
2016-01-10 23:23:26 +06:00
mrkkrp
a8bd0e4b10
Add direct link to tutorials from ‘README.md’
2016-01-10 13:54:33 +06:00
Mark Karpov
0cc4565dfb
Merge pull request #77 from stackbuilders/update-stack-lts-4.0
...
Update stack to latest LTS 4.0
2016-01-10 00:29:28 +05:00
mrkkrp
6f102ce1fe
Fix signature of ‘indentBlock’
2016-01-09 23:21:04 +06:00
Justin Leitgeb
dfb119d2f9
Update stack to latest LTS 4.0
2016-01-09 09:13:36 -08:00
mrkkrp
c4b2489d77
Disable Haddock build for Cabal 1.16
2016-01-09 20:20:23 +06:00
mrkkrp
7f505384df
Version bump (4.3.0)
2016-01-09 19:15:12 +06:00
mrkkrp
e9bf0e1773
Specify in which version some functions were added
2016-01-09 19:12:22 +06:00
mrkkrp
e5508941c8
Add year 2016 to copyright notices
2016-01-09 18:56:33 +06:00
mrkkrp
108cf2e6ff
Add Herbert Valerio Riedel to list of authors
2016-01-09 18:51:01 +06:00
Mark Karpov
61dc477eea
Merge pull request #76 from mrkkrp/advanced-indent
...
Advanced combinators for indentation-sensitive parsing
2016-01-09 17:42:03 +05:00
mrkkrp
b8e3029041
Tests for new combinators and some refactoring
2016-01-09 17:55:54 +06:00
mrkkrp
bd85deb38c
Finish new indentation-sensitive combinators
2016-01-09 17:38:31 +06:00
mrkkrp
618f5dbcb6
Updated ‘CHANGELOG.md’ to reflect changes in 4.3.0
2016-01-07 19:30:37 +06:00
mrkkrp
8984f5823a
Improved indentation-sensitive combinators
...
Close #54 .
These should be good enough (according to manual testing). We'll need to
add automated testing to make sure, though.
2016-01-02 20:30:36 +06:00
mrkkrp
718b0f358a
Don't use hints with custom error messages
...
Close #75 .
Now accumulated hints are not used with ‘ParseError’ records that have
only custom messages in them (created with ‘Message’ constructor, as
opposed to ‘Unexpected’ or ‘Expected’). This strips “expected” line from
custom error messages where it's unlikely to be relevant anyway.
2016-01-02 20:19:40 +06:00
mrkkrp
028b775885
Use ‘const’
2016-01-02 20:18:55 +06:00
mrkkrp
c73a2239b9
‘readFromFile’ → ‘parseFromFile’
2016-01-02 20:18:18 +06:00
mrkkrp
e94b39fc0d
Initial design of new indentation-sensitive helpers
2016-01-01 23:00:38 +06:00
mrkkrp
f317936613
Cosmetic whims in ‘Text.Megaparsec.Lexer’
2016-01-01 23:00:38 +06:00
mrkkrp
25cfeefd07
Fix some typos in ‘README.md’
2015-12-30 15:23:13 +06:00
mrkkrp
27f0b706b7
only use ‘-Werror’ on travis
2015-12-23 18:42:00 +06:00
mrkkrp
f699958d12
add ‘-Werror’ GHC option
2015-12-22 12:55:05 +06:00
mrkkrp
b6774d6d29
test newer GHC and container-based infrastructure
2015-12-17 00:11:10 +06:00
