CG-3 Release v1.0.0 (r12200)

[Tino Didriksen]

A new release of CG-3 has been tagged v1.0.0.12200, in preparation for the NoDaLiDa 2017 Constraint Grammar Workshop ( https://visl.sdu.dk/nodalida2017.html ) which takes place on Monday.

Finally calling it version 1.0. Doesn't change anything formally, but CG-3 is stable enough and widely enough used that it's about time for the 1.0 tag.

New features:
- Cmdline flag --dump-ast to output an XML dump of the parser representation. Useful for 3rd party transformation tools and highlighters
- Rule SPLITCOHORT that splits a cohort into multiple, while allowing the creation of an internal dependency tree. See https://visl.sdu.dk/cg3/chunked/rules.html#splitcohort
- Added a Bag of Tags feature. See https://visl.sdu.dk/cg3/chunked/contexts.html#test-bag-of-tags
- Added rule flag REPEAT. See https://visl.sdu.dk/cg3/chunked/rules.html#rule-options-repeat
- Added mathematical set difference set operator as \. See https://visl.sdu.dk/cg3/chunked/sets.html#set-operator-difference
- Renamed the old set difference operator (-) to the except operator
- Cmdline flag --trace now optionally takes a range of rules to break execution at
- ADDCOHORT now takes WITHCHILD to further narrow down where to insert the cohort
- New tool cg-strictify to aid in making your existing grammars STRICT-TAGS compliant. See https://visl.sdu.dk/cg3/chunked/cmdreference.html#cg-strictify

Changes:
- Now requires C++11 to build, but will test for and use C++14 and C++17 where available
- Regex captures are now done on a per-reading basis, which eliminates weird gotchas about matches
- Linking from a position override template will now go from the min/max edges of cohorts the template reached, rather than just the final cohort it touched
- OR'ed templates will now backtrack instead of trying only first match
- Input stream initial SETVAR commands are now honored immediately, so that variables can be used in DELIMITERS
- Context modifier A now works for almost all rule types
- KEEPORDER will now automatically be added for easily detectable cases where it is needed
- SUBSTITUTE will add the replacement tags for each found contiguous tag to be removed
- Better detection of endless loops
- Non-scanning contexts with modifers O and o will now throw an error as the writer probably meant 0
- Binary grammars can now be dumped as text, but in the internal mangled structure
- MOVE WITHCHILD changed to more logically interact with moving into and out of trees
- cg-conv no longer creates readings for plain text input. Cmdline flag --add-tags added to get these back where desired
- Numeric tags are now double-precision floating point
- cg-conv no longer multiplies weights in FST input by default

Fixed Bugs:
- Fixed removing a cohort that owned temporarily enclosed cohorts
- Fixed bug where e.g. <C:NN> was considered numerical with value 0
- Fixed a lot of bugs related to cohort moving, adding, and removing
- Fixed segfault when parsing tags longer than 256 bytes
- Fixed modifier pS acting as if it was p*

Main site is https://visl.sdu.dk/cg3.html
Google Group is https://groups.google.com/group/constraint-grammar
Source snapshots available at https://visl.sdu.dk/download/cg3/
Windows binary is at https://apertium.projectjj.com/win32/nightly/
OS X binary is at https://apertium.projectjj.com/osx/nightly/
RHEL/Fedora/CentOS/OpenSUSE packages are at https://apertium.projectjj.com/rpm/howto.txt
Debian/Ubuntu packages are at https://apertium.projectjj.com/apt/howto.txt

-- Tino Didriksen
CG-3 Developer

[Eckhard Bick]

Nice!

Jeg har bare lige eet spørgsmål - hvad mener du med "binary grammars as
text in mangled structure", and hvorfor er det nødvendigt? Øger det ikke
risikoen for dekompilering?

-- Eckhard

> --
> You received this message because you are subscribed to the Google
> Groups "Constraint Grammar" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to constraint-gram...@googlegroups.com
> <mailto:constraint-gram...@googlegroups.com>.
> To post to this group, send email to
> constrain...@googlegroups.com
> <mailto:constrain...@googlegroups.com>.
> Visit this group at https://groups.google.com/group/constraint-grammar.
> For more options, visit https://groups.google.com/d/optout.


--
Eckhard Bick,
cand.med., dr.phil.
University of Southern Denmark
e-mail: eckhar...@gmail.com
web: http://beta.visl.sdu.dk

[Tyler Ulrich]

Despite having moved on to another project, I'd like to congratulate you on this accomplishment. It's a significant contribution to computational linguistics.

To unsubscribe from this group and stop receiving emails from it, send an email to constraint-grammar+unsubscribe@googlegroups.com <mailto:constraint-grammar+unsubsc...@googlegroups.com>.
To post to this group, send email to constraint-grammar@googlegroups.com <mailto:constraint-grammar@googlegroups.com>.

Visit this group at https://groups.google.com/group/constraint-grammar.
For more options, visit https://groups.google.com/d/optout.

--
Eckhard Bick,
cand.med., dr.phil.
University of Southern Denmark
e-mail: eckhar...@gmail.com
web: http://beta.visl.sdu.dk

--
You received this message because you are subscribed to the Google Groups "Constraint Grammar" group.

To unsubscribe from this group and stop receiving emails from it, send an email to constraint-grammar+unsubscribe@googlegroups.com.
To post to this group, send email to constraint-grammar@googlegroups.com.

[Paul Meurer]

Hi Tino,

it's nice to see that so much development is going on!

I have made some minor additions for my own use; perhaps you could consider integrating them into the official source.

I need them to be able to show deleted readings in an interface that uses libcg3.

The additions are:

cg3.h :

size_t cg3_cohort_numdelreadings(cg3_cohort *cohort);
cg3_reading *cg3_cohort_getdelreading(cg3_cohort *cohort, size_t which);
size_t cg3_reading_gettrace_ruletype(cg3_reading *reading_, size_t which);

libcg3.cpp :

size_t cg3_cohort_numdelreadings(cg3_cohort *cohort_) {
Cohort *cohort = static_cast<Cohort*>(cohort_);
return cohort->deleted.size();
}

cg3_reading *cg3_cohort_getdelreading(cg3_cohort *cohort_, size_t which) {
Cohort *cohort = static_cast<Cohort*>(cohort_);
// return cohort->deleted[which];
ReadingList::iterator it = cohort->deleted.begin();
std::advance(it, which);
return *it;
}

size_t cg3_reading_gettrace_ruletype(cg3_reading *reading_, size_t which) {
Reading *reading = static_cast<Reading*>(reading_);
Grammar *grammar = reading->parent->parent->parent->parent->grammar;
const Rule *r = grammar->rule_by_number[reading->hit_by[which]];
return r->type;
}

Best,
Paul Meurer

> --
> You received this message because you are subscribed to the Google Groups "Constraint Grammar" group.

> To unsubscribe from this group and stop receiving emails from it, send an email to constraint-gram...@googlegroups.com.
> To post to this group, send email to constrain...@googlegroups.com.

> Visit this group at https://groups.google.com/group/constraint-grammar.
> For more options, visit https://groups.google.com/d/optout.

--

Paul Meurer, researcher
Uni Computing, Uni Research AS, Bergen, Norway
http://www.computing.uni.no/units/clu