Newer
Older
p1 = position_operator [tagset, vars] COMMA
p2 = position_operator [tagset, vars] {
op.reset(new Equals<Position>(p1, p2));
|
(symset_operator [tagset, vars]) =>
t1 = symset_operator [tagset, vars] COMMA
t2 = symset_operator [tagset, vars] {
op.reset(new Equals<TSet>(t1, t2));
}
)
|
(strset_operator [tagset, vars]) =>
s1 = strset_operator [tagset, vars] COMMA
s2 = strset_operator [tagset, vars] {
op.reset(new Equals<StrSet>(s1, s2));
}
)
|
(
b1 = bool_operator [tagset, vars] COMMA
b2 = bool_operator [tagset, vars] {
op.reset(new Equals<Bool>(b1, b2));
// ----------------------------------------------------------------------------
// In operator
in_operator
[const Corpus2::Tagset& tagset, Variables& vars]
returns [boost::shared_ptr<Function<Bool> > op]
boost::shared_ptr<Function<TSet> > t1, t2;
boost::shared_ptr<Function<StrSet> > s1, s2;
Paweł Kędzia
committed
}
:
"in" LPAREN
(symset_operator [tagset, vars]) =>
t1 = symset_operator [tagset, vars] COMMA
t2 = symset_operator [tagset, vars] {
op.reset(new IsSubsetOf<TSet>(t1, t2));
s1 = strset_operator [tagset, vars] COMMA
s2 = strset_operator [tagset, vars] {
op.reset(new IsSubsetOf<StrSet>(s1, s2));
;
// ----------------------------------------------------------------------------
// Inter operator
inter_operator
[const Corpus2::Tagset& tagset, Variables& vars]
returns [boost::shared_ptr<Function<Bool> > op]
boost::shared_ptr<Function<TSet> > t1, t2;
boost::shared_ptr<Function<StrSet> > s1, s2;
:
"inter" LPAREN
(symset_operator [tagset, vars]) =>
t1 = symset_operator [tagset, vars] COMMA
t2 = symset_operator [tagset, vars] {
op.reset(new Intersects<TSet>(t1, t2));
s1 = strset_operator [tagset, vars] COMMA
s2 = strset_operator [tagset, vars] {
op.reset(new Intersects<StrSet>(s1, s2));
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
// ----------------------------------------------------------------------------
// Iterations:
bool_iteration
[const Corpus2::Tagset& tagset, Variables& vars]
returns [boost::shared_ptr<Function<Bool> > ret]
{
int min_match = 0;
boost::shared_ptr<Function<Bool> > expr;
boost::shared_ptr<Function<Position> > lpos, rpos;
boost::shared_ptr<VariableAccessor<Position> > pacc;
}
: "only" LPAREN
lpos = position_operator [tagset, vars] COMMA
rpos = position_operator [tagset, vars] COMMA
pacc = position_variable_acc [vars] COMMA
expr = bool_operator [tagset, vars]
RPAREN {
ret.reset(new Only(lpos, rpos, *pacc, expr));
}
| "atleast" LPAREN
lpos = position_operator [tagset, vars] COMMA
rpos = position_operator [tagset, vars] COMMA
pacc = position_variable_acc [vars] COMMA
expr = bool_operator [tagset, vars] COMMA
min_match = number
RPAREN {
ret.reset(new AtLeast(lpos, rpos, *pacc, expr, min_match));
}
| "llook" LPAREN //note inverted rpos/lpos order
rpos = position_operator [tagset, vars] COMMA
lpos = position_operator [tagset, vars] COMMA
pacc = position_variable_acc [vars] COMMA
expr = bool_operator [tagset, vars]
RPAREN {
ret.reset(new LeftLook(lpos, rpos, *pacc, expr));
}
| "rlook" LPAREN
lpos = position_operator [tagset, vars] COMMA
rpos = position_operator [tagset, vars] COMMA
pacc = position_variable_acc [vars] COMMA
expr = bool_operator [tagset, vars]
RPAREN {
ret.reset(new RightLook(lpos, rpos, *pacc, expr));
}
;
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
// ----------------------------------------------------------------------------
// Agreement operator: agr, agrpp, wagr
bool_agreement
[const Corpus2::Tagset& tagset, Variables& vars]
returns [boost::shared_ptr<Function<Bool> > ret]
{
boost::shared_ptr<Function<TSet> > expr;
boost::shared_ptr<Function<Position> > lpos, rpos;
}
: "agr" LPAREN
lpos = position_operator [tagset, vars] COMMA
rpos = position_operator [tagset, vars] COMMA
expr = symset_operator [tagset, vars]
RPAREN {
ret.reset(new StrongAgreement(lpos, rpos, expr, tagset));
}
| "agrpp" LPAREN
lpos = position_operator [tagset, vars] COMMA
rpos = position_operator [tagset, vars] COMMA
expr = symset_operator [tagset, vars]
RPAREN {
ret.reset(new PointAgreement(lpos, rpos, expr, tagset));
}
| "wagr" LPAREN
lpos = position_operator [tagset, vars] COMMA
rpos = position_operator [tagset, vars] COMMA
expr = symset_operator [tagset, vars]
RPAREN {
ret.reset(new WeakAgreement(lpos, rpos, expr, tagset));
}
;
// ----------------------------------------------------------------------------
// Setvar operator
// Returns boost::shared_ptr<Function<Bool> >
// ----------------------------------------------------------------------------
setvar_operator
[const Corpus2::Tagset& tagset, Variables& vars]
returns [boost::shared_ptr<Function<Bool> > ret]
: "setvar" LPAREN
(
ret = position_setvar [tagset, vars]
| ret = bool_setvar [tagset, vars]
| ret = strset_setvar [tagset, vars]
| ret = symset_setvar [tagset, vars]
)
RPAREN
;
// ----------------------------------------------------------------------------
// Setvar for position
position_setvar
[const Corpus2::Tagset& tagset, Variables& vars]
returns [boost::shared_ptr<Function<Bool> > op]
boost::shared_ptr<Function<Position> > ret_op;
boost::shared_ptr<VariableAccessor<Position> > ret_acc;
: ret_acc = position_variable_acc [vars]
COMMA
ret_op = position_operator [tagset, vars] {
op.reset(new VarSetter<Position>(*ret_acc, ret_op));
// ----------------------------------------------------------------------------
// Setvar for bool
bool_setvar
[const Corpus2::Tagset& tagset, Variables& vars]
returns [boost::shared_ptr<Function<Bool> > op]
boost::shared_ptr<Function<Bool> > ret_op;
boost::shared_ptr<VariableAccessor<Bool> > ret_acc;
: ret_acc = bool_variable_acc [vars]
COMMA
ret_op = bool_operator [tagset, vars] {
op.reset(new VarSetter<Bool>(*ret_acc, ret_op));
// ----------------------------------------------------------------------------
// Setvar for strset
strset_setvar
[const Corpus2::Tagset& tagset, Variables& vars]
returns [boost::shared_ptr<Function<Bool> > op]
boost::shared_ptr<Function<StrSet> > ret_op;
boost::shared_ptr<VariableAccessor<StrSet> > ret_acc;
: ret_acc = strset_variable_acc [vars]
COMMA
ret_op = strset_operator [tagset, vars] {
op.reset(new VarSetter<StrSet>(*ret_acc, ret_op));
Paweł Kędzia
committed
}
// ----------------------------------------------------------------------------
// Setvar for symset
symset_setvar
[const Corpus2::Tagset& tagset, Variables& vars]
returns [boost::shared_ptr<Function<Bool> > op]
boost::shared_ptr<Function<TSet> > ret_op;
boost::shared_ptr<VariableAccessor<TSet> > ret_acc;
: ret_acc = symset_variable_acc [vars]
COMMA
ret_op = symset_operator [tagset, vars] {
op.reset(new VarSetter<TSet>(*ret_acc, ret_op));
}
;
///////////////////////////////////////////////////////////////////////////////
///////////////////////////////////////////////////////////////////////////////
///////////////////////////////////////////////////////////////////////////////
///////////////////////////////////////////////////////////////////////////////
class ANTLRLexer extends Lexer;
options {
exportVocab = ANTLRExpr;
charVocabulary = '\3'..'\377';
: '"'! (~('"' | '\n' | '\r'))* '"'!
| '\''! (~('\'' | '\n' | '\r'))* '\''!
: ('-'|'+') (' '!|'\t'!)* ('0'..'9')+
UNSIGNED_INT
options {
paraphrase = "Unsigned integer";
}
: ('0'..'9')+
;
STR_PREFIX
options {
paraphrase = "String prefix";
}
Paweł Kędzia
committed
: "$s:"
;
TST_PREFIX
options {
Paweł Kędzia
committed
: "$t:"
;
BOOL_PREFIX
options {
paraphrase = "Bool prefix";
}
Paweł Kędzia
committed
: "$b:"
Paweł Kędzia
committed
paraphrase = "Position prefix";
Paweł Kędzia
committed
: '$'
RBRACKET
options {
paraphrase = "']'";
}
: ']'
LPAREN
options {
paraphrase = "'('";
}
: '('
RPAREN
options {
paraphrase = "')'";
}
: ')'
LCURLY
options {
paraphrase = "'{'";
}
: '{'
RCURLY
options {
paraphrase = "'}'";
}
: '}'
AT_MARK
options {
paraphrase = "'@'";
}
: '@'
: ('a'..'z' | 'A'..'Z' | '_') ('a'..'z' | 'A'..'Z' | '_' | '0'..'9')*
| '`' ('a'..'z' | 'A'..'Z' | '_') ('a'..'z' | 'A'..'Z' | '_' | '0'..'9')* '`'
| '\t'
| '\f'
|
( "\r\n"
| '\r'
| '\n'
) { newline(); }
) { $setType(antlr::Token::SKIP); }
paraphrase = "Single line comment";
}
: "//" (~('\n'|'\r'))* { $setType(antlr::Token::SKIP); }
;
ML_COMMENT
options {
paraphrase = "Multi line comment";
( // TODO: test it and add reference to the site it's taken from!
/* This actually works OK despite the ambiguity that
'\r' '\n' can be matched in one alternative or by matching
'\r' in one iteration and '\n' in another.. But
this is really matched just by one rule per (...)*
loop iteration, so it's OK.
This is exactly how they do it all over the web - just
turn off the warning for this particular token.*/
options {
generateAmbigWarnings = false;
}
: { LA(2)!='/' }? '*'
| '\r' '\n' { newline(); }
| '\r' { newline(); }
| '\n' { newline(); }
)*
{ $setType(antlr::Token::SKIP); }
//DSEPARATOR
//options {
// paraphrase = "':-'";
//}
// : ":-"
//;