apertium-tat.tat.mb.rlx

# Constraint grammar disambiguation rules for translating Tatar metrical books.
# Invoked after apertium-tat.tat.rlx to reduce ambiguity to 0 in a way suitable
# for this particular domain.

# ========== #
# Delimiters #
# ========== #

DELIMITERS = "<.>" "<!>" "<?>" "<...>" "<¶>" ;
SOFT-DELIMITERS = "<,>" ;

SUBREADINGS = LTR ;

# ============= #
# Tags and sets #
# ============= #

SETS

LIST BOS = (>>>) ;
LIST EOS = (<<<) ;

# First-level/Parts-of-speech tags
# ================================

LIST A = adj ;
LIST Adv = adv ;
LIST Pron = prn ;
LIST N = n ;
LIST Prop = np ;
LIST V = v ;
LIST Vaux = vaux ;
LIST Cop = cop ;
LIST Det = det ;
LIST CC = cnjcoo ;
LIST CS = cnjsub ;
LIST Interj = ij ;
LIST Num = num ;
LIST Post = post ;
LIST Postadv = postadv ;
LIST Mod_ass = mod_ass ;
LIST Sent = sent ;
LIST Cm = cm ;
LIST Rquot = rquot ;

# POS sub-categories
# ==================

LIST Pers = pers ;
LIST Dem = dem;
LIST Interr = itg ;
LIST Cog = cog ;
LIST Ant = ant ;

# "Syntactic" tags
# ================

LIST Advl = advl ;
LIST Attr = attr ;
LIST Subst = subst ;

# Morphosyntactic properties   # if you find yourself embracing the same tag in
# ==========================   # brackets over and over again, you may want to
                               # add it here

LIST Sg = sg ;
LIST Pl = pl ;

LIST Nom = nom ;
LIST Gen = gen ;
LIST Dat = dat ;
LIST Acc = acc ;
LIST Abl = abl ;
LIST Loc = loc ;

LIST P1 = p1 ;
LIST P2 = p2 ;
LIST P3 = p3 ;

LIST PersonalPossessives = px1sg px2sg px3sp px1pl px2pl ;

LIST Pres = pres ;
LIST Inf = inf ;
LIST TV = tv ;
LIST Opt = opt ;
LIST Imp = imp ;

LIST Coop = coop ;
LIST Frm = frm ;

# Postpositions
# =============

LIST PostGoverningDat = "күрә" ;

# Verb sets
# =========

LIST FiniteVerb = pres past ifi  fut fut2 imp opt;

LIST Gerund = ger ger_ppot ger_past ger_perf ger_impf ger_abs ;
LIST Ger6 = ger6 ;
LIST Ger_past = ger_past ;

LIST Participle = prc_perf prc_impf prc_cond prc_vol prc_plan ;
LIST Prc_impf = prc_impf ;

LIST VerbalAdverb = gna_perf gna_cond gna_until gna_after ;
LIST Gna_cond = gna_cond ;

LIST Gpr_fut = gpr_fut ;
LIST Gpr_past = gpr_past ;

LIST Bar-etc = ("бар" adj) ("юк" adj) ("күп" adj) ;

# All possible word categories
# ============================

SET WORD = N | V | A | Post | Postadv | Pron | Det | Adv | CC | CS | Interj |
           Num | ("\?") ;

SET PRE-N =  A | Det | Postadv | Num | (n gen) | (prn gen) | CC ;

SET MARK =  Cm | ("\\") | ("\;") ;

SET WORDMARK = WORD | MARK ;

SET N-MOD = A | Det | Num | (n gen) | (prn gen) ;

SET ADJ-MOD = Postadv | Adv ;

SET NPMARK = N-MOD | ADJ-MOD ;

SET NOMINAL = N | Prop | Pron | Subst | Gerund ;

# Categories which cannot be part of a noun phrase
# ================================================

SET NPNH = WORDMARK - PRE-N ;
SET NPNHA = WORDMARK - PRE-N - Adv ;
SET NOT-ADV = WORDMARK - Adv ;

# Barriers
# ========

SET S-BOUNDARY = CS | Interr | EOS;


SECTION

"<атасы>" SELECT N ;
"<Атасы>" SELECT N ;
"<яшендә>" SELECT N + ("яшь");
"<исем>" SELECT N + ("исем");
"<йөз>" SELECT Num;
SELECT Prop; # IF (0 ("[:upper:]+[:lower:]*"r));