apertium -f line eats full stops for breakfast
add @c to r and l
apertium-transfer deosn't recognise lowercase lemmas
some more bound and emptiness checks, more and less verbose prints
some confusion with tokeni{s,z}e
wcin.get(c)) != NULL didn't compile so I removed null part.
apertium-validate-dictionary cannot count past 65535
Fix3ed?
test if MAX_COMBINATIONS can be increased without too much slow-down
some c/p&sed
bump MAX_COMBONATIONS to another magic number
I did some unscientific testing with apertium-deu & dewiki and it seems to me that...
I routinely run dewiki through apertium-deu if you need testings.
seddeds
dix classifier script
autostuff required files
bootstrap
free dictionary cc-by
puupankkie
stuffs
githubs
githubs
svn stuff
make github move more explizit
Miniminise does not work after fst2fst in my ve...
add stories and use est stuffs to fix fin
everything compiles
copy-pasta error
missing variables !
use apertium-tur-giella-fin
initial commit
new pair
add morphdb lexicons if someone wants to conver...
empty translation in bidix creates some mangled tags
Moved to...
adding The story
Yeah, while I made lexc's library version somewhat callable, it wasn't designed to...
GCI 2015–16 additions by porsi98.
It's noteworthy that leak count is all 0's for all the previous lines of the file,...
Ok, here's something I found with valgrind and Tino's help: $ head -n 167 europarl-v8.fi-en.norm.tok.en...
lt-proc memory leak on eng-fin with eng_feil
hfst-ospell-0.4.0 lost from SVN
note about github
warn abt move to github
Well having both automata sounds like a good option, I'd vote to have them both modes...
Trimmed lexicons don't show source of errors
is this not a duplicate of bug #313?
hfst python swig missing symbol in hfst-3.8.3
poses
proper mlu
tweak qst
kaa -> com
support vector machines
extra!
fixes
spat almost works
fixes
fix broken trimming again
all words in translation needed ever
remove harmful trimming
stuff
s/0/%0/
+a
-s
need to have lexicons mentioned when defined
ill
inverted
sks
poss dupe
some other cases
nouns categorisd
we don't rewrite-rule
Maybe related to bug #311 (or at least solution is in same part of code)?
I think a good solution is not easy for our use cases, since the problem is, this...
yeah, stderring that would work fine, the current format was just first I thought...
hfst-proc is missing or broken for new installations
It seems that for any non-toy example this leads to printing just copy of the string...
lookup collects huge weights before printing infinities
hfst-regexp2fst triggers stack-smashing
just FYI I've ran into same problem, omorfi-segment with interjection compounding...
try get rid of autowarns, wait for centos to fail
get rid of autowarning
FWIW, I’d be for such move now even though I’ve been wary of such given the learning...
$ hfst-strings2fst | hfst-compose -F proc-ol-differs.hfst | hfst-fst2txt Eurohpálaš...
ignores
Düsseldorf sessions
new fixes: symbols, some derivs
Well, the whole good point about $APERTIUM_PATH is that users who like the risky...
s/+'s/'s/
s/pri/pres/g
remove local eng dictionaries
use canonical apertium-eng instead of dir local
Version bump
Initial ebuild of HFST package
add emake install and dodoc to src_install
BUGS is not in distribution
Version bump.
Fix minorsyn errors
Version bump, drop non-existent betas
Version bump to sfst-1.4, add library installation