]>
Commit | Line | Data |
---|---|---|
1 | // Scintilla source code edit control | |
2 | /** @file LexVB.cxx | |
3 | ** Lexer for Visual Basic and VBScript. | |
4 | **/ | |
5 | // Copyright 1998-2005 by Neil Hodgson <neilh@scintilla.org> | |
6 | // The License.txt file describes the conditions under which this software may be distributed. | |
7 | ||
8 | #include <stdlib.h> | |
9 | #include <string.h> | |
10 | #include <stdio.h> | |
11 | #include <stdarg.h> | |
12 | #include <assert.h> | |
13 | #include <ctype.h> | |
14 | ||
15 | #include "ILexer.h" | |
16 | #include "Scintilla.h" | |
17 | #include "SciLexer.h" | |
18 | ||
19 | #include "WordList.h" | |
20 | #include "LexAccessor.h" | |
21 | #include "Accessor.h" | |
22 | #include "StyleContext.h" | |
23 | #include "CharacterSet.h" | |
24 | #include "LexerModule.h" | |
25 | ||
26 | #ifdef SCI_NAMESPACE | |
27 | using namespace Scintilla; | |
28 | #endif | |
29 | ||
30 | // Internal state, highlighted as number | |
31 | #define SCE_B_FILENUMBER SCE_B_DEFAULT+100 | |
32 | ||
33 | ||
34 | static bool IsVBComment(Accessor &styler, int pos, int len) { | |
35 | return len > 0 && styler[pos] == '\''; | |
36 | } | |
37 | ||
38 | static inline bool IsTypeCharacter(int ch) { | |
39 | return ch == '%' || ch == '&' || ch == '@' || ch == '!' || ch == '#' || ch == '$'; | |
40 | } | |
41 | ||
42 | // Extended to accept accented characters | |
43 | static inline bool IsAWordChar(int ch) { | |
44 | return ch >= 0x80 || | |
45 | (isalnum(ch) || ch == '.' || ch == '_'); | |
46 | } | |
47 | ||
48 | static inline bool IsAWordStart(int ch) { | |
49 | return ch >= 0x80 || | |
50 | (isalpha(ch) || ch == '_'); | |
51 | } | |
52 | ||
53 | static inline bool IsANumberChar(int ch) { | |
54 | // Not exactly following number definition (several dots are seen as OK, etc.) | |
55 | // but probably enough in most cases. | |
56 | return (ch < 0x80) && | |
57 | (isdigit(ch) || toupper(ch) == 'E' || | |
58 | ch == '.' || ch == '-' || ch == '+'); | |
59 | } | |
60 | ||
61 | static void ColouriseVBDoc(unsigned int startPos, int length, int initStyle, | |
62 | WordList *keywordlists[], Accessor &styler, bool vbScriptSyntax) { | |
63 | ||
64 | WordList &keywords = *keywordlists[0]; | |
65 | WordList &keywords2 = *keywordlists[1]; | |
66 | WordList &keywords3 = *keywordlists[2]; | |
67 | WordList &keywords4 = *keywordlists[3]; | |
68 | ||
69 | styler.StartAt(startPos); | |
70 | ||
71 | int visibleChars = 0; | |
72 | int fileNbDigits = 0; | |
73 | ||
74 | // Do not leak onto next line | |
75 | if (initStyle == SCE_B_STRINGEOL || initStyle == SCE_B_COMMENT || initStyle == SCE_B_PREPROCESSOR) { | |
76 | initStyle = SCE_B_DEFAULT; | |
77 | } | |
78 | ||
79 | StyleContext sc(startPos, length, initStyle, styler); | |
80 | ||
81 | for (; sc.More(); sc.Forward()) { | |
82 | ||
83 | if (sc.state == SCE_B_OPERATOR) { | |
84 | sc.SetState(SCE_B_DEFAULT); | |
85 | } else if (sc.state == SCE_B_IDENTIFIER) { | |
86 | if (!IsAWordChar(sc.ch)) { | |
87 | // In Basic (except VBScript), a variable name or a function name | |
88 | // can end with a special character indicating the type of the value | |
89 | // held or returned. | |
90 | bool skipType = false; | |
91 | if (!vbScriptSyntax && IsTypeCharacter(sc.ch)) { | |
92 | sc.Forward(); // Skip it | |
93 | skipType = true; | |
94 | } | |
95 | if (sc.ch == ']') { | |
96 | sc.Forward(); | |
97 | } | |
98 | char s[100]; | |
99 | sc.GetCurrentLowered(s, sizeof(s)); | |
100 | if (skipType) { | |
101 | s[strlen(s) - 1] = '\0'; | |
102 | } | |
103 | if (strcmp(s, "rem") == 0) { | |
104 | sc.ChangeState(SCE_B_COMMENT); | |
105 | } else { | |
106 | if (keywords.InList(s)) { | |
107 | sc.ChangeState(SCE_B_KEYWORD); | |
108 | } else if (keywords2.InList(s)) { | |
109 | sc.ChangeState(SCE_B_KEYWORD2); | |
110 | } else if (keywords3.InList(s)) { | |
111 | sc.ChangeState(SCE_B_KEYWORD3); | |
112 | } else if (keywords4.InList(s)) { | |
113 | sc.ChangeState(SCE_B_KEYWORD4); | |
114 | } // Else, it is really an identifier... | |
115 | sc.SetState(SCE_B_DEFAULT); | |
116 | } | |
117 | } | |
118 | } else if (sc.state == SCE_B_NUMBER) { | |
119 | // We stop the number definition on non-numerical non-dot non-eE non-sign char | |
120 | // Also accepts A-F for hex. numbers | |
121 | if (!IsANumberChar(sc.ch) && !(tolower(sc.ch) >= 'a' && tolower(sc.ch) <= 'f')) { | |
122 | sc.SetState(SCE_B_DEFAULT); | |
123 | } | |
124 | } else if (sc.state == SCE_B_STRING) { | |
125 | // VB doubles quotes to preserve them, so just end this string | |
126 | // state now as a following quote will start again | |
127 | if (sc.ch == '\"') { | |
128 | if (sc.chNext == '\"') { | |
129 | sc.Forward(); | |
130 | } else { | |
131 | if (tolower(sc.chNext) == 'c') { | |
132 | sc.Forward(); | |
133 | } | |
134 | sc.ForwardSetState(SCE_B_DEFAULT); | |
135 | } | |
136 | } else if (sc.atLineEnd) { | |
137 | visibleChars = 0; | |
138 | sc.ChangeState(SCE_B_STRINGEOL); | |
139 | sc.ForwardSetState(SCE_B_DEFAULT); | |
140 | } | |
141 | } else if (sc.state == SCE_B_COMMENT) { | |
142 | if (sc.atLineEnd) { | |
143 | visibleChars = 0; | |
144 | sc.ForwardSetState(SCE_B_DEFAULT); | |
145 | } | |
146 | } else if (sc.state == SCE_B_PREPROCESSOR) { | |
147 | if (sc.atLineEnd) { | |
148 | visibleChars = 0; | |
149 | sc.ForwardSetState(SCE_B_DEFAULT); | |
150 | } | |
151 | } else if (sc.state == SCE_B_FILENUMBER) { | |
152 | if (IsADigit(sc.ch)) { | |
153 | fileNbDigits++; | |
154 | if (fileNbDigits > 3) { | |
155 | sc.ChangeState(SCE_B_DATE); | |
156 | } | |
157 | } else if (sc.ch == '\r' || sc.ch == '\n' || sc.ch == ',') { | |
158 | // Regular uses: Close #1; Put #1, ...; Get #1, ... etc. | |
159 | // Too bad if date is format #27, Oct, 2003# or something like that... | |
160 | // Use regular number state | |
161 | sc.ChangeState(SCE_B_NUMBER); | |
162 | sc.SetState(SCE_B_DEFAULT); | |
163 | } else if (sc.ch == '#') { | |
164 | sc.ChangeState(SCE_B_DATE); | |
165 | sc.ForwardSetState(SCE_B_DEFAULT); | |
166 | } else { | |
167 | sc.ChangeState(SCE_B_DATE); | |
168 | } | |
169 | if (sc.state != SCE_B_FILENUMBER) { | |
170 | fileNbDigits = 0; | |
171 | } | |
172 | } else if (sc.state == SCE_B_DATE) { | |
173 | if (sc.atLineEnd) { | |
174 | visibleChars = 0; | |
175 | sc.ChangeState(SCE_B_STRINGEOL); | |
176 | sc.ForwardSetState(SCE_B_DEFAULT); | |
177 | } else if (sc.ch == '#') { | |
178 | sc.ForwardSetState(SCE_B_DEFAULT); | |
179 | } | |
180 | } | |
181 | ||
182 | if (sc.state == SCE_B_DEFAULT) { | |
183 | if (sc.ch == '\'') { | |
184 | sc.SetState(SCE_B_COMMENT); | |
185 | } else if (sc.ch == '\"') { | |
186 | sc.SetState(SCE_B_STRING); | |
187 | } else if (sc.ch == '#' && visibleChars == 0) { | |
188 | // Preprocessor commands are alone on their line | |
189 | sc.SetState(SCE_B_PREPROCESSOR); | |
190 | } else if (sc.ch == '#') { | |
191 | // It can be a date literal, ending with #, or a file number, from 1 to 511 | |
192 | // The date literal depends on the locale, so anything can go between #'s. | |
193 | // Can be #January 1, 1993# or #1 Jan 93# or #05/11/2003#, etc. | |
194 | // So we set the FILENUMBER state, and switch to DATE if it isn't a file number | |
195 | sc.SetState(SCE_B_FILENUMBER); | |
196 | } else if (sc.ch == '&' && tolower(sc.chNext) == 'h') { | |
197 | // Hexadecimal number | |
198 | sc.SetState(SCE_B_NUMBER); | |
199 | sc.Forward(); | |
200 | } else if (sc.ch == '&' && tolower(sc.chNext) == 'o') { | |
201 | // Octal number | |
202 | sc.SetState(SCE_B_NUMBER); | |
203 | sc.Forward(); | |
204 | } else if (IsADigit(sc.ch) || (sc.ch == '.' && IsADigit(sc.chNext))) { | |
205 | sc.SetState(SCE_B_NUMBER); | |
206 | } else if (IsAWordStart(sc.ch) || (sc.ch == '[')) { | |
207 | sc.SetState(SCE_B_IDENTIFIER); | |
208 | } else if (isoperator(static_cast<char>(sc.ch)) || (sc.ch == '\\')) { // Integer division | |
209 | sc.SetState(SCE_B_OPERATOR); | |
210 | } | |
211 | } | |
212 | ||
213 | if (sc.atLineEnd) { | |
214 | visibleChars = 0; | |
215 | } | |
216 | if (!IsASpace(sc.ch)) { | |
217 | visibleChars++; | |
218 | } | |
219 | } | |
220 | ||
221 | if (sc.state == SCE_B_IDENTIFIER && !IsAWordChar(sc.ch)) { | |
222 | // In Basic (except VBScript), a variable name or a function name | |
223 | // can end with a special character indicating the type of the value | |
224 | // held or returned. | |
225 | bool skipType = false; | |
226 | if (!vbScriptSyntax && IsTypeCharacter(sc.ch)) { | |
227 | sc.Forward(); // Skip it | |
228 | skipType = true; | |
229 | } | |
230 | if (sc.ch == ']') { | |
231 | sc.Forward(); | |
232 | } | |
233 | char s[100]; | |
234 | sc.GetCurrentLowered(s, sizeof(s)); | |
235 | if (skipType) { | |
236 | s[strlen(s) - 1] = '\0'; | |
237 | } | |
238 | if (strcmp(s, "rem") == 0) { | |
239 | sc.ChangeState(SCE_B_COMMENT); | |
240 | } else { | |
241 | if (keywords.InList(s)) { | |
242 | sc.ChangeState(SCE_B_KEYWORD); | |
243 | } else if (keywords2.InList(s)) { | |
244 | sc.ChangeState(SCE_B_KEYWORD2); | |
245 | } else if (keywords3.InList(s)) { | |
246 | sc.ChangeState(SCE_B_KEYWORD3); | |
247 | } else if (keywords4.InList(s)) { | |
248 | sc.ChangeState(SCE_B_KEYWORD4); | |
249 | } // Else, it is really an identifier... | |
250 | sc.SetState(SCE_B_DEFAULT); | |
251 | } | |
252 | } | |
253 | ||
254 | sc.Complete(); | |
255 | } | |
256 | ||
257 | static void FoldVBDoc(unsigned int startPos, int length, int, | |
258 | WordList *[], Accessor &styler) { | |
259 | int endPos = startPos + length; | |
260 | ||
261 | // Backtrack to previous line in case need to fix its fold status | |
262 | int lineCurrent = styler.GetLine(startPos); | |
263 | if (startPos > 0) { | |
264 | if (lineCurrent > 0) { | |
265 | lineCurrent--; | |
266 | startPos = styler.LineStart(lineCurrent); | |
267 | } | |
268 | } | |
269 | int spaceFlags = 0; | |
270 | int indentCurrent = styler.IndentAmount(lineCurrent, &spaceFlags, IsVBComment); | |
271 | char chNext = styler[startPos]; | |
272 | for (int i = startPos; i < endPos; i++) { | |
273 | char ch = chNext; | |
274 | chNext = styler.SafeGetCharAt(i + 1); | |
275 | ||
276 | if ((ch == '\r' && chNext != '\n') || (ch == '\n') || (i == endPos)) { | |
277 | int lev = indentCurrent; | |
278 | int indentNext = styler.IndentAmount(lineCurrent + 1, &spaceFlags, IsVBComment); | |
279 | if (!(indentCurrent & SC_FOLDLEVELWHITEFLAG)) { | |
280 | // Only non whitespace lines can be headers | |
281 | if ((indentCurrent & SC_FOLDLEVELNUMBERMASK) < (indentNext & SC_FOLDLEVELNUMBERMASK)) { | |
282 | lev |= SC_FOLDLEVELHEADERFLAG; | |
283 | } else if (indentNext & SC_FOLDLEVELWHITEFLAG) { | |
284 | // Line after is blank so check the next - maybe should continue further? | |
285 | int spaceFlags2 = 0; | |
286 | int indentNext2 = styler.IndentAmount(lineCurrent + 2, &spaceFlags2, IsVBComment); | |
287 | if ((indentCurrent & SC_FOLDLEVELNUMBERMASK) < (indentNext2 & SC_FOLDLEVELNUMBERMASK)) { | |
288 | lev |= SC_FOLDLEVELHEADERFLAG; | |
289 | } | |
290 | } | |
291 | } | |
292 | indentCurrent = indentNext; | |
293 | styler.SetLevel(lineCurrent, lev); | |
294 | lineCurrent++; | |
295 | } | |
296 | } | |
297 | } | |
298 | ||
299 | static void ColouriseVBNetDoc(unsigned int startPos, int length, int initStyle, | |
300 | WordList *keywordlists[], Accessor &styler) { | |
301 | ColouriseVBDoc(startPos, length, initStyle, keywordlists, styler, false); | |
302 | } | |
303 | ||
304 | static void ColouriseVBScriptDoc(unsigned int startPos, int length, int initStyle, | |
305 | WordList *keywordlists[], Accessor &styler) { | |
306 | ColouriseVBDoc(startPos, length, initStyle, keywordlists, styler, true); | |
307 | } | |
308 | ||
309 | static const char * const vbWordListDesc[] = { | |
310 | "Keywords", | |
311 | "user1", | |
312 | "user2", | |
313 | "user3", | |
314 | 0 | |
315 | }; | |
316 | ||
317 | LexerModule lmVB(SCLEX_VB, ColouriseVBNetDoc, "vb", FoldVBDoc, vbWordListDesc); | |
318 | LexerModule lmVBScript(SCLEX_VBSCRIPT, ColouriseVBScriptDoc, "vbscript", FoldVBDoc, vbWordListDesc); | |
319 |