# header file for published Spanish lexicon

# Each line of the lexicon contains the following nine fields
# in the following order.  The first field contains a name, the second
# contains a regular expression that should match every instance of that
# field.  Fields are tab separated, here as well as in the lexicon.

# For full documentation of the lexicon, please read the file sp_lex.doc!

# Usage of this file:
#  simple syntax check:
#	parse-lex -s sp_lex.hdr sp_lex.v04
#  output full "parse" of fields:
#	parse-lex sp_lex.hdr sp_lex.v04
# The "parse-lex" tool is available via FTP from ftp.ldc.upenn.edu

HEADWORD	[-a-zαινσϊδλοφόρΡA-ZΙΝ]+

# morphological and stem information
let $morphpiece=[-a-zαινσϊδλοφόρΡA-ZΙΝ_]+((\+$tag)|(\|Adv))+
let $tag=(1P|2P|3P|Abrev|Acc|Acro|Adj|Adv|Affix|Apoc|Art|Aum|Card|Com|Cond|Conj|Continent|Cort|Dat|Def|Dem|Det|Dim|FSubj|Fem|For|Fut|Gen|IInd|ISubj|Imp|Indef|Inf|Interj|Interrog|Let|Lit|Loc|Lugar|MF|Masc|NP|Neut|Nom|Noun|Num|Obl|Onom|Org|PInd|PP|PSubj|PastPart|Pay|Perf|Pl|Pluperf|Poss|PostDet|PreDet|Prep|PresPart|Pron|Prop|Quant|Reas|Ref|Rel|SP|Sg|Soc|Sup|Temp|Titl|Usastate|Var|Verb|Zodiac)

MORPH	$morphpiece((//|\|)$morphpiece)*

# pronunciation
let $pronpiece=[aieouhpbBfvlmwtdDsSCJnyrRxNkgG9z]+
PRON	$pronpiece(//$pronpiece)*

# stress pattern(s): 1=stressed, 0=unstressed
STRESS	0*1[01]*(//0*1[01]*)*

# number of occurrences in Callhome transcripts
CALLH	[0-9]+

# number of occurrences in Madrid Radio transcripts
MADRAD	[0-9]+

# number of occurrences in AP newswire text
AP	[0-9]+

# number of occurrences in Reuters newswire text
REUT	[0-9]+

# number of occurrences in El Norte newswire text
NORTE	[0-9]+