1
0
forked from GitHub/gf-core
Commit Graph

192 Commits

Author SHA1 Message Date
krasimir
c571e6ef33 now every parse state keeps reference only to the concrete and the abstract syntaxes but not to the whole PGF 2010-12-14 17:20:53 +00:00
aarne
e7fbe756a3 work-around in Tree.expr2tree to make Paraphrase work 2010-12-14 14:36:05 +00:00
krasimir
7d58efb660 optimization in the parser for large lexicons. Now, the parser is slightly slower for grammars with a small lexicon but 3-4 times faster for the English Resource Grammar used in combination with Oxford Advanced Learners Dictionary 2010-12-14 08:46:22 +00:00
gdetrez
1ed5b52a03 In the haskell code, we make the binary representation of the doubles compliant with IEEE 754.
The default binary representation in haskell's Data.Binary package is
homemade and quite complicated. Making it compliant with IEEE 754 will
make it easyer for the java runtimes (and probably others) to load the
PGF.
2010-10-19 18:34:29 +00:00
hallgren
8bb09fd64a minibar/support.js: add an errorcallback parameter to ajax_http_get 2010-12-09 16:38:16 +00:00
krasimir
0f4927dd40 fixed typos in the documentation for PGF.Parse 2010-12-08 10:32:18 +00:00
krasimir
b6446ec36d simple refactoring in PGF.Macros and related 2010-12-06 14:19:51 +00:00
aarne
dd6e0b1730 moved PGF.ToApi to GF.Compile.ToAPI 2010-12-07 18:05:13 +00:00
hallgren
b39d69eeae minibar.js: restore start_miniabr function for backward compatibility 2010-12-07 13:00:36 +00:00
ra.monique
7fba8c338d no more IO in the syntax to API translator 2010-12-06 10:15:14 +00:00
aarne
00111b10c1 vt -api with conversion to resource API names 2010-12-06 09:53:04 +00:00
ra.monique
6268c2d7d9 added syntax to api translation module 2010-12-05 19:51:19 +00:00
hallgren
18e3687832 minibar: rewrite minibar.js in a more idiomatic JS OOP style
This provides easier access to minibar internals for people who want to extend
it, but probably also reveals too much of the internal machinery...
2010-12-01 16:37:14 +00:00
hallgren
b0027c3305 Let minibar keep its own copy of tree-btn.png 2010-12-01 16:28:27 +00:00
hallgren
a2b9a684c0 support.js: add function bind (JS OOP trick) 2010-12-01 16:20:28 +00:00
hallgren
8e6421a03e minibar/about.html: minor adjustments 2010-11-28 11:52:48 +00:00
krasimir
0b80b3ef0a command ai in the shell now shows the probability of the tree 2010-11-24 12:51:54 +00:00
hallgren
ee85111d38 minibar/about.html: document recent development 2010-11-23 12:26:32 +00:00
aarne
2acb125f6f fixed some bugs in translation quiz 2010-11-17 08:30:02 +00:00
krasimir
115b4213d5 operations in the abstract syntax 2010-11-12 19:37:19 +00:00
hallgren
41df4e00ff Some pgf-server API documentation 2010-11-09 21:02:07 +00:00
hallgren
b0594eb545 pgf_online.js: if a grammar list is supplied, you don't have to call switch_grammar. 2010-11-09 16:29:04 +00:00
hallgren
91ba2c01bf minibar.js: code simplification 2010-11-09 11:58:38 +00:00
hallgren
e07098e31f support.js: two new functions: commonPrefix & insertFirst 2010-11-05 15:41:06 +00:00
aarne
c5334613c5 trying to find some inefficiencies in Finnish resource 2010-10-25 19:56:47 +00:00
ramona.enache
1f8a0f0876 added giza Alignments with command ga and merged the rendering algorithm for graphviz and giza alignments 2010-10-31 13:39:01 +00:00
hallgren
524fb1a6bc minibar.js: bug fix: when not using group translation, and a particular language is chosen in the To: menu, the translation would not be displayed 2010-10-28 12:39:56 +00:00
hallgren
7fac5dc089 minibar.js: bug fix: the To: menu wasn't updated properly when switching grammars 2010-10-28 11:53:50 +00:00
hallgren
5fc0e059c5 minibar: restructured code and improved user interface
* Keyboard input and completion should now work much more smoothly:
  + When you press space, the current word will be completed (if incomplete)
    and a new magnet will be created. If there is more than one possible
    completion, no magnet is created, but the common prefix of the possible
    completions is added to the text box.
  + Instead of asking the server for possible completions every time a new
    letter is added to the curent word, minibar only ask for completions for
    whole words and then filters the list locally when more letters are entered,
    speeding things up when server responses are slow.

* Code restructuring:
  + The PGF server API has been moved to its own file: pgf_online.js. This
    allows it to be reused in other applicaitons without importing the entire
    minibar. It also allows minibar to be used with different server interfaces.
  + The minibar code has been rewritten to avoid storing state information
    in the document tree and accessing it by referring to named document
    elements. The code now also avoids using string literals contaning
    the names of top-level functions to specify event handlers for buttons
    and menus. (The code is no longer introspective, so alpha conversion will
    not change its meaning.)
2010-10-26 12:42:33 +00:00
hallgren
1e51a2e9e5 support.js: some enhancements and new functions
+ ajax_http_get(url,callback) now checks the HTTP response status and shows an
  error message if the status is not OK (is not 200)
+ New function node(tag,attrs,children) to create document nodes with arbitraty
  attributes and children. The functions node() and text() should now be enough
  to construct arbitrary document trees in an easy, functional way, and many of
  the more specialized functions could be removed.
+ When calling the function button(label,action,key), the action can be a
  function instead of a string, so you can use local/anonymous functions.
+ New function debug(str) adds text to the element with id=debug, if it
  exists.
2010-10-26 11:52:04 +00:00
krasimir
0ba1daf5eb when faced with hard unification problem the type checker should just postpone the decision instead of failing immediately. added test case as well 2010-10-25 15:47:33 +00:00
krasimir
aa6b07afdb when we print the bracketed string it is useful to print the attached metavariables as well 2010-10-25 12:21:12 +00:00
krasimir
822a70cf7a change the TcM monad to continuation passing style. The old monad caused stack overflow for large search spaces 2010-10-21 15:01:52 +00:00
krasimir
82edf7bebb support for proof search with high-order functions 2010-10-21 12:47:26 +00:00
krasimir
2991b7acef bugfix in the term generation 2010-10-20 13:36:39 +00:00
krasimir
d45b4ae426 fixed typo which broke the parsing with -openclass 2010-10-20 09:04:48 +00:00
krasimir
4ac1cf4970 in the type checker monad we should use lazy pattern matching because otherwise we force the exhaustive tree generation to be strict 2010-10-19 13:04:41 +00:00
krasimir
775e59dabe added explicit depth parameter to the parsing API and the corresponding command in the shell 2010-10-18 16:51:25 +00:00
krasimir
205ac48ac3 typo in PGF.TypeCheck which spoiled the previous patch 2010-10-18 16:50:55 +00:00
krasimir
702b4aad3b now we use the GF reasoner to fillin meta variables in the abstract trees generated from the parser 2010-10-18 15:55:14 +00:00
hallgren
57f0124f2a minibar.html: add a <noscript> element with a note that the page doesn't work unless JavaScript is enabled 2010-10-23 14:56:10 +00:00
hallgren
d803d2a925 minibar.js: show the type error messages that can occur for grammars with dependent types 2010-10-18 12:15:30 +00:00
hallgren
26da60d0a9 minibar.js: add an option to enable/disable the Random button 2010-10-17 16:48:21 +00:00
krasimir
05a52d6937 now since the type checking monad TcM is nondeterministic we can use the same monad in PGF.Forest.getAbsTrees 2010-10-14 14:28:40 +00:00
krasimir
7fc638e030 bugfix in PGF.VisualizeTree for handling trees with implicit arguments 2010-10-13 13:32:48 +00:00
krasimir
09ce38742c fix PGF.Generate.restart. The restart should be done only if there are some solutions found 2010-10-13 12:38:51 +00:00
krasimir
de0354f991 the exhaustive/random generator now knows how to handle computable functions in the types 2010-10-11 17:18:28 +00:00
krasimir
f305587a62 now the generation from template with meta-variables respects the dependent types 2010-10-11 09:59:57 +00:00
krasimir
7200244605 backtracking for random generation is not enough. we must restart the search after each solution 2010-10-02 13:34:58 +00:00
krasimir
cb8795c222 refactor the API for random generation again. Now PGF contains probabilities in the abstract syntax 2010-10-02 13:03:57 +00:00