summaryrefslogtreecommitdiff
path: root/src/runtime
AgeCommit message (Collapse)Author
2013-06-19fíx in the Python binding for compatibility with Python<2.7kr.angelov
2013-06-19fix an issue in the Python binding related to the 32 vs 64 bit compatibilitykr.angelov
2013-06-17fix for a couple of warnings that are generated when GNU lightning is usedkr.angelov
2013-06-17add the source code for GNU lightning in the source directory for the C runtimekr.angelov
2013-06-14bugfix in the python bindingkr.angelov
2013-05-31now the call Expr.unpack("? e1 e2") in Python returns a pair with None as ↵kr.angelov
the first element and a the list [e1,e2] as the second. This makes it easier to decompose partial abstract trees
2013-05-31improved error message in the Python bindingkr.angelov
2013-05-30bugfix for the grammar printer in the C runtimekr.angelov
2013-05-29a simple refactoring in the Python runtimekr.angelov
2013-05-29readPGF in the Python runtime now throws "No such file or directory" ↵kr.angelov
exception if the grammar is missing
2013-05-28added a test class for the Java API plus a small refinement in the ↵kr.angelov
implementation for the binding
2013-05-28an initial skeleton for building a Java binding to the C runtimekr.angelov
2013-05-21fix the encoding problem with unicode literals in the Python bindingkr.angelov
2013-05-07bugfix for bracketedLinearize which was causing crash if the tree cannot be ↵kr.angelov
linearized
2013-05-07bug fix in the management of memory pools in the statistical parserkr.angelov
2013-05-06the statistical parser is now using two memory pools: one for parsing and ↵kr.angelov
one for the output trees. This means that the memory for parsing can be released as soon as the needed abstract trees are retrieved, while the trees themselves are retained in the separate output pool
2013-05-05fix the leftcorner filtering after the addition of word completionkr.angelov
2013-05-04bugfix in the C runtime which was causing an infinite loop while linearizing ↵kr.angelov
partial trees
2013-05-03[haskell runtime] Remove trailing whitespaces in VisualizeTree.hsgregoire.detrez
2013-05-01word completion in the C runtime. The runtime/python/test.py example is now ↵kr.angelov
using readline with word completion
2013-04-26finally the statistical parser is able to return all possible abstract treeskr.angelov
2013-04-23a trivial refactoring of the reasoner in the C runtimekr.angelov
2013-04-22bug fix in pgf-translate which was hiding that there are more than one trees ↵kr.angelov
per sentence
2013-04-21reverse the direction of the arcs in the dependency treeskr.angelov
2013-04-19remove the dead code left behind by Peter Ljunglöf in VisualizeTreekr.angelov
2013-04-19the C runtime and the Python binding now have an API for parser evaluation. ↵kr.angelov
The API computes PARSEVAL and Exact Match for a given tree. As a side effect the abstract trees in Python are now compared for equality by value and not by reference
2013-04-18added API for computing bracketed strings from Python and Ckr.angelov
2013-04-16added a malt_tab format to the vd command in the GF shellkr.angelov
2013-04-16the generation of dependency trees in the Haskell runtime is now finally ↵kr.angelov
working with bracketed strings. This also fixes some errors in the old implementation
2013-04-15the compiler now sorts the list of functions per category in probability ↵kr.angelov
order. this ensures probability order search in the C runtime
2013-04-08PGF.hs: export function missingLinshallgren
Also in Commands.hs: be explicit about things imported from the PGF library that are not in the public API. Also a couple of haddock documentation fixes.
2013-04-05now the web service to the robust parser can to translations alsokr.angelov
2013-04-05a bugfix which was causing an infinite loop in the C linearizer for some ↵kr.angelov
sentences
2013-04-05a very simple linearization for partial abstract trees in the C runtimekr.angelov
2013-04-02Replace "CId" with "Language" in type signature for PGF.tabularLinearizesjohn.j.camilleri
2013-03-26haddock bug workaroundhallgren
2013-03-20the calculation of lexical_prob in the statistical parser doesn't work ↵kr.angelov
properly. It should be fixed but for now I just disabled the optimization
2013-03-20replace #if with #ifdef when checking for the optional bottom up filtering ↵kr.angelov
in the C runtime
2013-03-19the bottom up filtering in the C runtime is temporary disabled. It takes too ↵kr.angelov
much memory and even makes it impossible to load the Finnish and the German parsing grammars.
2013-03-14bug fix in the linearizer in the C runtimekr.angelov
2013-03-14update the pgf-service tool from the C runtime after the changes in the APIkr.angelov
2013-03-11bugfix in the statistical parserkr.angelov
2013-03-09ghc-7.6: add missing Num instance for BitsSergei Trofimovich
Fixes the following build failure: src/runtime/haskell/Data/Binary/IEEE754.lhs:256:17: Could not deduce (Num a) arising from a use of `mask' from the context (Bits a) bound by the type signature for clamp :: Bits a => BitCount -> a -> a
2013-02-26added gu_buf_flush in seq.c which removes all elements from a bufferkr.angelov
2013-02-19the parser in the C runtime should not crash if the start category is not ↵kr.angelov
defined
2013-02-19bugfix in the grammar reader in the C runtimekr.angelov
2013-02-13bugfix in the linearizer for the C runtimekr.angelov
2013-02-13Fix for a PGF portability problemhallgren
GF produced slightly different PGF files on 64-bit systems and 32-bit systems. This could cause problems when a PGF was produced on a 32-bit system and used on a 64-bit system. To fix this, the GF compiler and the Haskell PGF run-time library now reads and writes PGF files like the 32-bit version even when compiled on a 64-bit system. Note: the Haskell type Int is still used internally in GF, which could be 32 bits or 64 bits...
2013-02-12now the beam size for the statistical parser can be configured by using the ↵kr.angelov
flag beam_size in the top-level concrete module
2013-02-12the statistical parser now uses a baseline lexical estimation of the beam sizekr.angelov