1 ! Copyright (C) 2008 Slava Pestov.
2 ! See http://factorcode.org/license.txt for BSD license.
3 USING: kernel namespaces make xmode.rules xmode.tokens
4 xmode.marker.state xmode.marker.context xmode.utilities
5 xmode.catalog sequences math assocs combinators strings
6 regexp splitting unicode.case
7 combinators.short-circuit accessors ;
10 ! Next two words copied from parser-combinators
11 ! Just like head?, but they optionally ignore case
13 : string= ( str1 str2 ignore-case -- ? )
14 [ [ >upper ] bi@ ] when sequence= ;
16 : string-head? ( str1 str2 ignore-case -- ? )
21 [ length head-slice ] 2bi
25 ! Based on org.gjt.sp.jedit.syntax.TokenMarker
27 : current-keyword ( -- string )
28 last-offset get position get line get subseq ;
30 : keyword-number? ( keyword -- ? )
32 [ current-rule-set highlight-digits?>> ]
33 [ dup [ digit? ] any? ]
36 current-rule-set digit-re>>
37 dup [ dupd matches? ] [ drop f ] if
42 : mark-number ( keyword -- id )
43 keyword-number? DIGIT and ;
45 : mark-keyword ( keyword -- id )
46 current-rule-set keywords>> at ;
48 : add-remaining-token ( -- )
49 current-rule-set default>> prev-token, ;
53 dup mark-number [ ] [ mark-keyword ] ?if
54 [ prev-token, ] when* ;
56 : current-char ( -- char )
57 position get line get nth ;
59 GENERIC: match-position ( rule -- n )
61 M: mark-previous-rule match-position drop last-offset get ;
63 M: rule match-position drop position get ;
65 : can-match-here? ( matcher rule -- ? )
68 [ over at-line-start?>> over zero? implies ]
69 [ over at-whitespace-end?>> over whitespace-end get = implies ]
70 [ over at-word-start?>> over last-offset get = implies ]
73 : rest-of-line ( -- str )
74 line get position get tail-slice ;
76 GENERIC: text-matches? ( string text -- match-count/f )
81 M: string-matcher text-matches?
83 [ string>> ] [ ignore-case?>> ] bi string-head?
84 ] keep string>> length and ;
86 M: regexp text-matches?
87 [ >string ] dip match-head ;
89 : rule-start-matches? ( rule -- match-count/f )
90 dup start>> tuck swap can-match-here? [
91 rest-of-line swap text>> text-matches?
96 : rule-end-matches? ( rule -- match-count/f )
97 dup mark-following-rule? [
98 dup start>> swap can-match-here? 0 and
100 dup end>> tuck swap can-match-here? [
102 swap text>> context get end>> or
111 : get-always-rules ( vector/f ruleset -- vector/f )
112 f swap rules>> at ?push-all ;
114 : get-char-rules ( vector/f char ruleset -- vector/f )
115 [ ch>upper ] dip rules>> at ?push-all ;
117 : get-rules ( char ruleset -- seq )
118 [ f ] 2dip [ get-char-rules ] keep get-always-rules ;
120 GENERIC: handle-rule-start ( match-count rule -- )
122 GENERIC: handle-rule-end ( match-count rule -- )
124 : find-escape-rule ( -- rule )
126 in-rule-set>> escape-rule>> [ ] [
127 parent>> in-rule-set>>
128 dup [ escape-rule>> ] when
131 : check-escape-rule ( rule -- ? )
133 find-escape-rule dup [
134 dup rule-start-matches? dup [
135 swap handle-rule-start
136 delegate-end-escaped? [ not ] change
144 : check-every-rule ( -- ? )
145 current-char current-rule-set get-rules
146 [ rule-start-matches? ] map-find
147 dup [ handle-rule-start t ] [ 2drop f ] if ;
151 dup rule-end-matches?
152 dup [ swap handle-rule-end ] [ 2drop ] if
155 : rule-match-token* ( rule -- id )
157 { f [ dup body-token>> ] }
158 { t [ current-rule-set default>> ] }
162 M: escape-rule handle-rule-start
165 process-escape? get [
166 escaped? [ not ] change
167 position [ + ] change
170 M: seq-rule handle-rule-start
174 tuck body-token>> next-token,
175 delegate>> [ push-context ] when* ;
177 UNION: abstract-span-rule span-rule eol-span-rule ;
179 M: abstract-span-rule handle-rule-start
183 tuck rule-match-token* next-token,
185 dup context get (>>in-rule)
186 delegate>> push-context ;
188 M: span-rule handle-rule-end
191 M: mark-following-rule handle-rule-start
193 mark-token add-remaining-token
194 tuck rule-match-token* next-token,
195 f context get (>>end)
196 context get (>>in-rule) ;
198 M: mark-following-rule handle-rule-end
199 nip rule-match-token* prev-token,
200 f context get (>>in-rule) ;
202 M: mark-previous-rule handle-rule-start
205 dup body-token>> prev-token,
206 rule-match-token* next-token, ;
214 : check-end-delegate ( -- ? )
215 context get parent>> [
217 dup rule-end-matches? dup [
223 ] keep context get parent>> in-rule>>
224 rule-match-token* next-token,
226 seen-whitespace-end? on t
227 ] [ drop check-escape-rule ] if
231 : handle-no-word-break ( -- )
232 context get parent>> [
234 dup no-word-break?>> [
235 rule-match-token* prev-token,
245 add-remaining-token ;
247 : (check-word-break) ( -- )
250 1 current-rule-set default>> next-token, ;
252 : rule-set-empty? ( ruleset -- ? )
253 [ rules>> ] [ keywords>> ] bi
254 [ assoc-empty? ] bi@ and ;
256 : check-word-break ( -- ? )
257 current-char dup blank? [
260 seen-whitespace-end? get [
261 position get 1+ whitespace-end set
267 ! Micro-optimization with incorrect semantics; we keep
268 ! it here because jEdit mode files depend on it now...
269 current-rule-set rule-set-empty? [
275 current-rule-set rule-set-no-word-sep* member? [
281 seen-whitespace-end? on
284 delegate-end-escaped? off t ;
287 : mark-token-loop ( -- )
288 position get line get length < [
290 [ check-end-delegate ]
299 : mark-remaining ( -- )
300 line get length position set
303 : unwind-no-line-break ( -- )
304 context get parent>> [
313 : tokenize-line ( line-context line rules -- line-context' seq )