[developers] Unquoted MRS pred names in LOGON's terg

[Francis Bond]

G'day,

A spot of context --- we are using the new ERG in our Japanese-English
MT system.

> I am trying to update my automatically constructed transfer rules for
> jaen using the current terg in LOGON,
> but I get invalid type errors for rules containing some compound particles:
>
> Invalid type _AT+PRESENT_P_REL
>  Unifications specified are invalid or do not unify
>  Warning: An error occurred
>          (read-transfer-rules():
>           `imanotokoro_a_1--_at+present_p--edict-mtr' is incoherent.
>  )
>
> In lexdb.rev, the predicate name of "at present" is a symbol not a
> string. There appear to be several words
> like this (see the full list below). Can I assume this is a bug in the
> grammar, or am I missing something?

I think these are new, and thus not in the predicates.erg.tdl checked
in to LOGON.  As a temporary fix, add them to jaen/predicates.tdl.

> _aboveboard_p_rel
> _above_p_rel
> _abroad_p_rel
> _ahead_p_rel
> _aside_p_rel
> _at+all_a_1_rel
> _at+best_p_rel
> _at+present_p_rel
> _beforehand_p_rel
> _before_p_rel
> _behind_p_rel
> _below_p_rel
> _between_p_rel
> _down_p_rel
> _east_a_1_rel
> _en+route_p_rel
> _fore_p_rel
> _forward_p_rel
> _maximum_p_rel
> _meanwhile_p_rel
> _nearby_p_rel
> _north_a_1_rel
> _out_p_rel
> _since_p_rel
> _south_a_1_rel
> _southeast_a_1_rel
> _starboard_p_rel
> _thereafter_p_rel
> _together_p_rel
> _up_p_rel
> _upstairs_p_rel
> _west_a_1_rel
> _within_p_rel
> _without_p_rel

I think that uio/predicates.erg.tdl should be updated at some stage,
maybe when terg becomes the new ERG.


-- 
Francis Bond <http://www2.nict.go.jp/x/x161/en/member/bond/>
NICT Language Infrastructure Group