1 ! Copyright (C) 2005, 2009 Daniel Ehrenberg, Doug Coleman.
2 ! See http://factorcode.org/license.txt for BSD license.
3 USING: namespaces math kernel sequences accessors fry circular
4 unicode.case unicode.categories locals combinators.short-circuit
5 make combinators io splitting math.parser math.ranges
6 generalizations sorting.functor math.order sorting.slots ;
9 TUPLE: sequence-parser sequence n ;
11 : <sequence-parser> ( sequence -- sequence-parser )
16 :: with-sequence-parser ( sequence-parser quot -- seq/f )
17 sequence-parser n>> :> n
18 sequence-parser quot call [
19 n sequence-parser (>>n) f
22 : offset ( sequence-parser offset -- char/f )
24 [ n>> + ] [ sequence>> ?nth ] bi ; inline
26 : current ( sequence-parser -- char/f ) 0 offset ; inline
28 : previous ( sequence-parser -- char/f ) -1 offset ; inline
30 : peek-next ( sequence-parser -- char/f ) 1 offset ; inline
32 : advance ( sequence-parser -- sequence-parser )
33 [ 1 + ] change-n ; inline
35 : advance* ( sequence-parser -- )
38 : next ( sequence-parser -- obj ) [ current ] [ advance* ] bi ;
40 : get+increment ( sequence-parser -- char/f )
41 [ current ] [ advance drop ] bi ; inline
43 :: skip-until ( sequence-parser quot: ( obj -- ? ) -- )
44 sequence-parser current [
45 sequence-parser quot call
46 [ sequence-parser advance quot skip-until ] unless
47 ] when ; inline recursive
49 : sequence-parse-end? ( sequence-parser -- ? ) current not ;
51 : take-until ( sequence-parser quot: ( obj -- ? ) -- sequence/f )
52 over sequence-parse-end? [
57 [ drop [ n>> ] [ sequence>> ] bi ] 2tri subseq f like
60 : take-while ( sequence-parser quot: ( obj -- ? ) -- sequence/f )
61 [ not ] compose take-until ; inline
63 : <safe-slice> ( from to seq -- slice/f )
66 [ [ drop ] 2dip length > ]
68 } 3|| [ 3drop f ] [ slice boa ] if ; inline
70 :: take-sequence ( sequence-parser sequence -- obj/f )
71 sequence-parser [ n>> dup sequence length + ] [ sequence>> ] bi
72 <safe-slice> sequence sequence= [
74 sequence-parser [ sequence length + ] change-n drop
79 : take-sequence* ( sequence-parser sequence -- )
82 :: take-until-sequence ( sequence-parser sequence -- sequence'/f )
83 sequence-parser n>> :> saved
84 sequence length <growing-circular> :> growing
87 current growing push-growing-circular
88 sequence growing sequence=
90 growing sequence sequence= [
92 growing length 1- - head
93 sequence-parser [ growing length - 1 + ] change-n drop
94 ! sequence-parser advance drop
96 saved sequence-parser (>>n)
100 :: take-until-sequence* ( sequence-parser sequence -- sequence'/f )
101 sequence-parser sequence take-until-sequence :> out
103 sequence-parser [ sequence length + ] change-n drop
106 : skip-whitespace ( sequence-parser -- sequence-parser )
107 [ [ current blank? not ] take-until drop ] keep ;
109 : skip-whitespace-eol ( sequence-parser -- sequence-parser )
110 [ [ current " \t\r" member? not ] take-until drop ] keep ;
112 : take-c-comment ( sequence-parser -- seq/f )
114 dup "/*" take-sequence [
115 "*/" take-until-sequence*
119 ] with-sequence-parser ;
121 : take-c++-comment ( sequence-parser -- seq/f )
123 dup "//" take-sequence [
126 { [ current CHAR: \n = ] [ sequence-parse-end? ] } 1||
134 ] with-sequence-parser ;
136 : skip-whitespace/comments ( sequence-parser -- sequence-parser )
139 { [ dup take-c-comment ] [ skip-whitespace/comments ] }
140 { [ dup take-c++-comment ] [ skip-whitespace/comments ] }
144 : take-define-identifier ( sequence-parser -- string )
145 skip-whitespace/comments
146 [ current { [ blank? ] [ CHAR: ( = ] } 1|| ] take-until ;
148 : take-rest-slice ( sequence-parser -- sequence/f )
149 [ sequence>> ] [ n>> ] bi
150 2dup [ length ] dip < [ 2drop f ] [ tail-slice ] if ; inline
152 : take-rest ( sequence-parser -- sequence )
153 [ take-rest-slice ] [ sequence>> like ] bi f like ;
155 : take-until-object ( sequence-parser obj -- sequence )
156 '[ current _ = ] take-until ;
158 : parse-sequence ( sequence quot -- )
159 [ <sequence-parser> ] dip call ; inline
161 :: take-quoted-string ( sequence-parser escape-char quote-char -- string )
162 sequence-parser n>> :> start-n
163 sequence-parser advance
166 [ { [ previous escape-char = ] [ current quote-char = ] } 1&& ]
167 [ current quote-char = not ]
169 ] take-while :> string
170 sequence-parser current quote-char = [
171 sequence-parser advance* string
173 start-n sequence-parser (>>n) f
176 : (take-token) ( sequence-parser -- string )
177 skip-whitespace [ current { [ blank? ] [ f = ] } 1|| ] take-until ;
179 :: take-token* ( sequence-parser escape-char quote-char -- string/f )
180 sequence-parser skip-whitespace
182 { quote-char [ escape-char quote-char take-quoted-string ] }
184 [ drop (take-token) ]
187 : take-token ( sequence-parser -- string/f )
188 CHAR: \ CHAR: " take-token* ;
190 : take-integer ( sequence-parser -- n/f )
191 [ current digit? ] take-while ;
193 :: take-n ( sequence-parser n -- seq/f )
194 n sequence-parser [ n>> + ] [ sequence>> length ] bi > [
195 sequence-parser take-rest
197 sequence-parser n>> dup n + sequence-parser sequence>> subseq
198 sequence-parser [ n + ] change-n drop
201 : c-identifier-begin? ( ch -- ? )
202 CHAR: a CHAR: z [a,b]
203 CHAR: A CHAR: Z [a,b]
204 { CHAR: _ } 3append member? ;
206 : c-identifier-ch? ( ch -- ? )
207 CHAR: a CHAR: z [a,b]
208 CHAR: A CHAR: Z [a,b]
209 CHAR: 0 CHAR: 9 [a,b]
210 { CHAR: _ } 4 nappend member? ;
212 : (take-c-identifier) ( sequence-parser -- string/f )
213 dup current c-identifier-begin? [
214 [ current c-identifier-ch? ] take-while
219 : take-c-identifier ( sequence-parser -- string/f )
220 [ (take-c-identifier) ] with-sequence-parser ;
222 << "length" [ length ] define-sorting >>
224 : sort-tokens ( seq -- seq' )
225 { length>=< <=> } sort-by ;
227 : take-first-matching ( sequence-parser seq -- seq )
229 '[ _ [ swap take-sequence ] with-sequence-parser ] find nip ;
232 : take-longest ( sequence-parser seq -- seq )
233 sort-tokens take-first-matching ;
235 : take-c-integer ( sequence-parser -- string/f )
239 { "ull" "uLL" "Ull" "ULL" "ll" "LL" "l" "L" "u" "U" }
240 take-longest [ append ] when*
244 ] with-sequence-parser ;
246 CONSTANT: c-punctuators
248 "[" "]" "(" ")" "{" "}" "." "->"
249 "++" "--" "&" "*" "+" "-" "~" "!"
250 "/" "%" "<<" ">>" "<" ">" "<=" ">=" "==" "!=" "^" "|" "&&" "||"
252 "=" "*=" "/=" "%=" "+=" "-=" "<<=" ">>=" "&=" "^=" "|="
254 "<:" ":>" "<%" "%>" "%:" "%:%:"
257 : take-c-punctuator ( sequence-parser -- string/f )
258 c-punctuators take-longest ;
260 : write-full ( sequence-parser -- ) sequence>> write ;
261 : write-rest ( sequence-parser -- ) take-rest write ;