Closed olzama closed 5 months ago
In example 4319 from tbdb05:
4319@unknown@formal@none@1@S@En tercera posición aparece Portugal.@(1, 0, 1, <0:2>, 1, "en" "en", 0, "sp", "sp" 1.00000000) (2, 1, 2, <3:10>, 1, "z" "tercera", 0, "z", "z" -1.00000000) (3, 2, 3, <11:19>, 1, "posición" "posición", 0, "ncfs000", "ncfs000" 1.00000000) (4, 3, 4, <20:27>, 1, "aparecer" "aparece", 0, "vmip3s0", "vmip3s0" 0.99479167) (5, 4, 5, <28:36>, 1, "portugal" "portugal", 0, "np00g00", "np00g00" 1.00000000) (6, 5, 6, <36:37>, 1, "." ".", 0, "fp", "fp" 1.00000000)@@@1@5@@montse@25-apr-2010
the word tercera ends up being analyzed not as an ordinal number but as an adjective that is for some reason plural. I don't understand why that is and how that is connected to the z tag.
The issue with item 4319 is that it gets a cardinal number analysis (incorrectly) but furthermore, the cardinal number is for some reason constrained to be 3pl?
d_-_q-nu-card_lex := det_quant_eval &
[ SYNSEM [ LOCAL [ AGR.PNG.PN 3pl,
CAT.VAL.SPR < > ],
LKEYS.ALTKEYREL.PRED card_q_rel ] ]
"""
a)- determiners / indef. adjectives (Z)
!!! num + sust pueden formar una unidad inseparable, e.g. esas tristes dos pesetas,
esas famosas tres tardes con Teresa
e.g. (los) tres hermanos
""".
This should be fixed in commit ab40a9c
E.g. item 3800 it tbdb18, for primera