1
0
forked from GitHub/gf-core
Commit Graph

664 Commits

Author SHA1 Message Date
Krasimir Angelov
7db9f51d78 "ma" command in the C shell 2017-08-30 16:35:57 +02:00
Krasimir Angelov
4a24bc98f2 the l command in the C shell now supports the same options as in the normal shell 2017-08-30 16:18:26 +02:00
Krasimir Angelov
1e4ab95e41 added option -output-format=java for producing code for embedded grammars in Java 2017-08-29 20:53:47 +02:00
Krasimir Angelov
b7b7a7c91c the embedded grammars now work with both the pure Haskell and the Haskell binding API 2017-08-29 18:44:50 +02:00
Thomas Hallgren
0a0eaa01bc Bump version requirements to base>=4.6, Cabal>=1.20
Cabal>=1.20 allows control over parallelism when compiling grammars from
Setup.hs and WebSetup.hs.

base>=4.6 allows conditional compilation with CPP to be eliminated from
a few modules.

base-4.6 corresponds to GHC 7.6.3, which is what you get in
Debian 8 (aka jessie, aka oldstable) from 2015.
2017-08-18 11:55:44 +02:00
Thomas Hallgren
8484dabd5b Disable detailed version info from darcs
TODO: get version info from git instead.
2017-08-11 15:33:21 +02:00
aarne
d51cbb0f1a added Arabic question mark to arabic and persian transliterations, as well as the zero-width non-joiner U+200C to persian" 2017-06-14 12:32:17 +00:00
aarne
ee2f3d085e vd command now reads local concrete configurations to deal with syncat words; TODO: dissolve clustered multiwords bound with + 2017-04-06 11:55:21 +00:00
aarne
1e33235144 fixed a bug in Eq instance generation for GADT and lexical categories 2017-03-22 22:25:21 +00:00
krasimir
f1240ef489 a better error message 2017-03-07 23:17:24 +00:00
krasimir
7361d22059 GF.Compile.Compute.ConcreteNew now handles Predef.Float 2017-03-07 19:50:43 +00:00
krasimir
2e5499ba97 bugfix in record subtyping checking 2017-03-07 17:53:56 +00:00
krasimir
5ec43f2f75 GF.Grammar.Printer now has a Terse mode which prints record types with lock fields with their corresponding abstract categories 2017-03-07 08:24:00 +00:00
krasimir
5a61ab5fcc fix for EPatt 2017-03-06 17:15:46 +00:00
krasimir
6c5cfa7750 type checking EPatt 2017-03-06 16:24:58 +00:00
krasimir
a0d6a4f9a4 GF.Compile.Compute.ConcreteNew now can handle EPattType 2017-03-06 15:44:25 +00:00
krasimir
fe4f38f6ee some missing patterns 2017-03-06 14:44:03 +00:00
krasimir
ad2a18592b added overload resolution in the experimental type checker 2017-03-06 14:09:16 +00:00
krasimir
2c1c2da89f some progress on the experimental type checker for the concrete syntax 2017-03-03 16:37:22 +00:00
aarne
34627e7427 corrected fg generation for lexical categories, to avoid a type error in generated code 2017-02-24 20:50:36 +00:00
hallgren
ae512e09e0 Fix compilation errors caused by changes in PGF2
These prevented GF from being built with C run-time support (-fc-runtime).
2017-02-21 13:21:59 +00:00
krasimir
db0f8b0dce improve the documentation for PGF2 2017-01-25 20:30:54 +00:00
aarne
8729339d26 BNF grammars (.bnf or .cf) can have rules without semicolons, as in the GF book examples. However, if the grammar has any multiline rules, all of them must have semicolons. 2016-09-05 21:15:44 +00:00
hallgren
12d2296991 GF cloud: more readable formatting of the GF version page
Factor out common CSS from gfse/editor.css into clouds.css.
2016-06-15 10:20:30 +00:00
hallgren
096b4cfcee PGF service & minibar: only show dependency diagrams if the labels are known
+ The PGF service now reads and caches dependency label configuration files.
+ The grammar info returned by command=grammar has a new boolean field 
  'hasDependencyLabels' to indicate if dependency labels were found for
  the grammar. Also, command=deptree will now fail if no labels are present.
+ The minibar only shows word dependency trees if labels are present.
+ Also changed the type of getDepLabels from [String] -> Labels to
  String -> Labels, since all uses were in the form "getDepLabels . lines".
2016-06-09 13:12:14 +00:00
hallgren
649ab21cec Changes for compatibility with GHC 8.0.1 2016-05-22 20:15:31 +00:00
hallgren
31df4a6c46 GF.Command.Commands2: adapt gf -cshell to the changes in the previous patch 2016-04-07 14:55:13 +00:00
hallgren
ab3cc77656 GF shell: fix a parsing problem with the cc command
This patch fixes a problem introduced last year when the GF shell was
refactored to allow more commands to be treated uniformly and be part
of pipes. The cc command was one of those commands, but unfortunately this
introduced a parsing problem, e.g.

	> cc "last"
	constant not found: last

	> cc "last"++"year"
	command not parsed: cc "last"++"year"

This happened because the generic command line parser in
GF.Command.{Abstract,Parse} assumes that all commands have an argument of
type PGF.Expr. Commands that expect other types of arguments have to
use PGF.showExpr combined with other conversion to the argument type they
expect. The cc command excpets a GF.Grammar.Term, and unfortunately not 
all terms survice the roundtrip through PGF.Expr, in part because of
an additional hack to allow strings to be roundtripped through PGF.Expr
without adding superfluous double quotes.

To solve the problem, this patch

 + makes room for arguments of type Term in the Argument type in
   GF.Command.Abstract.
   
 + makes a special case for the cc command in GF.Command.Parse, by
   calling the partial parser 'runPartial pTerm' recently added in
   GF.Grammar.Lexer and GF.Grammar.Parser. Care was taken so that
   that "|" and ";" can be used both inside terms and as separators between
   commands in the shell, e.g. things like the following now work:

       > cc ("a"|"b") | ps -lexcode
       variants { "a" ; "b" }

 + introduces a type CommandArgument that replaces [Expr] as the
   type of values passed between commands in pipes. It has room for
   values of type [Expr], [String] and Term, thus eliminating the need
   to roundtrip through the Expr type all the time.
   The hack to avoid adding superfluous quotes when strings are
   roundtripped through Expr has been left in place for now,
   but can probably be removed.
2016-04-07 13:40:05 +00:00
hallgren
65e675d8e2 Lexer.x & Parser.y: add a partial parser for terms
Lexer.x: Change the parser monad type P to allow the remaining input to
	 be returned after a partial parse. Add function

           runPartial :: P t -> String -> Either (Posn, String) (String, t)

Parser.y: Add a partial parser pTerm for nonterminal Exp1.
          Re-export runPartial.
2016-04-07 13:32:14 +00:00
hallgren
c8b7ebc163 Lexer.x: fix cyclic Functor instance
It looks like I introduced this cyclic definition in August 2014, but since
it isn't used, it hasn't been a problem...
2016-04-06 13:46:48 +00:00
krasimir
a393c1a246 fix the handling of separators in BNFC which are not nonempty 2016-03-22 13:13:35 +00:00
krasimir
ce70720859 CFGtoPGF is now extended to support context-free grammars with primitive parameters 2016-03-22 10:28:15 +00:00
krasimir
fbdf21d862 added missing module GF.Grammar.BNFC 2016-03-21 20:03:28 +00:00
krasimir
8f80e5451e fix comment in GF.Grammar.CFG 2016-03-21 19:33:14 +00:00
krasimir
c8ebe09315 initial support for BNFC syntax in context-free grammars for GF. Not all features are supported yet. Based on contribution from Gleb Lobanov 2016-03-21 13:27:44 +00:00
krasimir
da10bb33c7 bugfix: GF.Compile.CFGtoPGF now supports literal categories 2016-03-21 13:26:32 +00:00
krasimir
07cf4d6509 more progress on the typechecker 2016-03-15 14:16:17 +00:00
krasimir
c1671d43e2 current state of the experimental typechecker 2016-03-07 20:20:47 +00:00
krasimir
4ba9712f16 add missing cases in vapply in the partial evaluator 2016-03-03 06:59:50 +00:00
krasimir
508bc346e5 fix in the typechecker 2016-03-02 21:28:54 +00:00
krasimir
13ff91bbb9 a better interface between the type checker and the partial evaluator 2016-03-02 16:22:56 +00:00
krasimir
47eb774cdf the experimental type checker in GF.Compile.TypeCheck.ConcreteNew is now rewriten to use the complete evaluator in GF.Compile.Compute.ConcreteNew. The old sketchy implementation in GF.Compile.Compute.ConcreteNew1 is now removed. 2016-03-02 13:38:02 +00:00
leiss
df2901c9c0 add lexer and unlexer for Ancient Greek accent normalization 2016-02-23 16:30:39 +00:00
hallgren
f2b057c078 GF shell, cc command: try to compute pre{...} tokens in token sequences
This is implemented as a simple post-processing step after partial evaluation
to try compute pre{...} tokens in token sequences. Nothing is done to deal
with intervening free variants.

This was done in response to a query from René T on the gf-dev mailing list.
2015-12-02 16:41:18 +00:00
hallgren
5bb792be5e GF.Data.Str: change matchPrefix to agree with the GF book and the run-time system
The matchPrefix function is used in str2strings and the partial evaluator to
compute pre{...} tokens. But unlike the description in the GF book and
the implementation in the run-time system, matchPrefix looked at
the concatenation(!) of all following tokens and not just the next token
when deciding how to compute a pre{...} token.

This is a backwards incompatible change, but it is subtle and probably won't
cause any problems. In particular, the example grammars are unaffected.
2015-12-01 15:14:27 +00:00
hallgren
d6ae5e811b GF.Data.Str: unexport & comment out unused functions 2015-12-01 14:32:24 +00:00
aarne
0a38e137b6 vd -conll2latex now converts conll to latex. Without going through GF trees, but as a service to the dependency parser community. 2015-11-23 10:43:03 +00:00
aarne
d6a505169a added -output=latex to visialize_dependencies. This generates more familiar-looking output than the default graphviz, which can moreover be pasted into LaTeX documents. Some more work is needed to make long sentences look nice and fit on a page; a constant word length is now used to simplify computing the coordinates. 2015-11-17 09:11:10 +00:00
aarne
ce6557f1f2 the visualization commands (aw,vd,vp,vt) can now show multiple trees. Previously they only showed one tree even if there were for instance after ambiguous parsing. The reason was that dot (graphviz) ignored all graphs but the first one. Now the graphs are put into separate files. The 'convert' command from ImageMagick package is used to combine them to one pdf. If this is a problem, the old behaviour can be restored by the -number=1 option to the tree-generating command, which cuts away all trees but the one and doesn't require ImageMagick. 2015-11-05 16:47:41 +00:00
aarne
eb49b6ab56 improved documentation of vp -showdep 2015-11-05 08:23:33 +00:00