Alvis-NLPPlatform
view release on metacpan or search on metacpan
- Switching the Makefile.PL on Build.PL
- Correction in the Yatea wrapper in the handling of the output file.
0.4 - Correction in the function sigint handler : nlp_host and
nlp_port are now declared as global.
- Correction in the TermTagging : language switch was well
taken into account
- Correction in the management of the ".proc_id" file
- correction in the computing of the xml rendering time
(the variable is set to zero ;-)
- stderr when NLP tools are called, is redirected in a log file
- addition of a variable DEBUG defining a debug mode (temporary
files are not removed)
- alvis-nlp-standalone can read a file given in argument or on
the STDIN stream
- Documentation of the modules and scripts are gathered at the
end of each file
- Addition of DTD and XSD files in the documentation (etc
directory)
- Additional functionality: Loading files in various formats
(PDF, LaTeX, Word, etc.) before carrying out linguistic
lib/Alvis/NLPPlatform/patches/link-4.1a-WithWhiteSpace.diff view on Meta::CPAN
*** ./src/parse.c 2004-02-01 21:57:55.000000000 +0100
--- ../link-4.1a-WithWhiteSpace/src/parse.c 2006-11-02 11:32:06.000000000 +0100
***************
*** 441,447 ****
label = strip_off_label(input_string);
}
! sent = sentence_create(input_string, dict);
if (sent == NULL) {
if (verbosity > 0) fprintf(stderr, "%s\n", lperrmsg);
--- 441,447 ----
label = strip_off_label(input_string);
}
! sent = sentence_create(input_string, dict, opts);
if (sent == NULL) {
if (verbosity > 0) fprintf(stderr, "%s\n", lperrmsg);
diff -cr ./src/tokenize.c ../link-4.1a-WithWhiteSpace/src/tokenize.c
*** ./src/tokenize.c 2004-02-01 21:57:55.000000000 +0100
--- ../link-4.1a-WithWhiteSpace/src/tokenize.c 2006-11-02 11:32:07.000000000 +0100
***************
*** 15,20 ****
--- 15,25 ----
#define MAX_STRIP 10
+ /* If true, attempt to break words into word text and POS tag parts,
lib/Alvis/NLPPlatform/patches/link-4.1b-WithWhiteSpace.diff view on Meta::CPAN
*** ./src/parse.c 2005-01-12 18:09:54.000000000 +0100
--- ../link-4.1b-WithWhiteSpace/src/parse.c 2006-11-02 14:08:24.000000000 +0100
***************
*** 441,447 ****
label = strip_off_label(input_string);
}
! sent = sentence_create(input_string, dict);
if (sent == NULL) {
if (verbosity > 0) fprintf(stderr, "%s\n", lperrmsg);
--- 441,447 ----
label = strip_off_label(input_string);
}
! sent = sentence_create(input_string, dict, opts);
if (sent == NULL) {
if (verbosity > 0) fprintf(stderr, "%s\n", lperrmsg);
diff -rc ./src/tokenize.c ../link-4.1b-WithWhiteSpace/src/tokenize.c
*** ./src/tokenize.c 2005-01-12 18:09:54.000000000 +0100
--- ../link-4.1b-WithWhiteSpace/src/tokenize.c 2006-11-02 14:08:24.000000000 +0100
***************
*** 15,20 ****
--- 15,25 ----
#define MAX_STRIP 10
+ /* If true, attempt to break words into word text and POS tag parts,
( run in 1.913 second using v1.01-cache-2.11-cpan-49f99fa48dc )