summaryrefslogtreecommitdiff
path: root/src
AgeCommit message (Collapse)Author
2007-12-08noemit as default in testgf3aarne
2007-12-08breaking points 1..6 in new Compileaarne
2007-12-08documenting implementation startedaarne
2007-12-08compilation line complete (but dysfunctional) for new GF internal formataarne
2007-12-08GFtoGFCC type checks (but is not correct)aarne
2007-12-08names resolve in new GFtoGFCCaarne
2007-12-08started modifying GFtoGFCC for new internal GF formataarne
2007-12-07a couple of fixes in new Optimizeaarne
2007-12-07restructured some of the new GF format; modules now in place up to gfo ↵aarne
generation
2007-12-07started adding GF back optimization (Factorize)aarne
2007-12-07optimize in the compilation chain for new formataarne
2007-12-07refresh compilation phase in the new formataarne
2007-12-06new type checker type checksaarne
2007-12-06in the middle of adapting CheckGrammaraarne
2007-12-06started grammar checking with new internal formataarne
2007-12-06restored work on Extend and Renameaarne
2007-12-05Expose AbsGFCC and DataGFCC in gf-embed.cabal, since code generated by ↵bringert
-haskell needs them.
2007-12-05Update GF-embed.cabal to use expose GF.GFCC.API. This pulls in a lot fewer ↵bringert
modules than the old version. Use Cabal configurations to compile on 6.8.1. This unfortunately means that users of GHC < 6.8 must upgrade to Cabal 1.2 or greater.
2007-12-04restored interface and instance - at least for a whileaarne
2007-12-04rebuild integrated in Extendaarne
2007-12-04building extensions in new source formataarne
2007-12-04unifying judgement information - now easyaarne
2007-12-04printing new source formataarne
2007-12-04some missing cases in SourceToGFaarne
2007-12-04test for new GF source formataarne
2007-12-04moved some modules to Devel.Grammaraarne
2007-12-04SourceToGF working though not completeaarne
2007-11-28Haskell GSyntax printer?: encode generated code as UTF-8. Current GHC ↵bringert
versions treat source files as UTF-8 encoded.
2007-11-12Added install-gf3 target. Use EXEEXT to get correct gf3 file name on Windows.bringert
2007-11-12Use configure to produce a gfc script with the correct path to gf3. NOTE: ↵bringert
you need to re-run autoconf after pulling this patch.
2007-12-03started parsing GF source into the new formataarne
2007-11-29some macros for terms, jments, modulesaarne
2007-11-28judgement construction after parsingaarne
2007-11-28lookup module separatedaarne
2007-11-28new definitions of term and judgement syntaxaarne
2007-11-28lookup modulesaarne
2007-11-27started experimental new module syntaxaarne
2007-11-15diagnostic error message in GrammarToHaskellaarne
2007-11-15dealing with Ints n in GrammarToGFCCaarne
2007-11-12made Devel versions of Compute, CheckGrammar, Optimize, etc, to leave GF 2.8 ↵aarne
untouched when experimenting
2007-11-12made compile from source use optimized modules internallyaarne
2007-11-12fixed a bug on not updating parameter table in record fields in GrammarToGFCCaarne
2007-11-12fixed bug with prawitz transform of course-of-values tables in Computeaarne
2007-11-12merged gfc and gfi to one binary, and a shell script for gfcaarne
2007-11-10printing grammarReference.js with gfc -jsrefaarne
2007-11-09post-compute in GrammarToGFCC - what else is needed?aarne
2007-11-09ResFin workaround; type annot for ghc 6.8.1aarne
2007-11-09Fix pre token bug in JS, both in gf and gfc. Pre tokens are still treated ↵bringert
like normal tokens, but at least the generated code should be correct now.
2007-11-09debugging gfcc bug; resulted in some type annotations to work around problemsaarne
2007-11-09changed js and hs file names in gfcc output to expected onesaarne
generated by cgit v1.2.3 (git 2.39.1) at 2026-06-18 07:41:54 +0000