| 1 | // Scintilla source code edit control |
| 2 | /** |
| 3 | * @file LexRegistry.cxx |
| 4 | * @date July 26 2014 |
| 5 | * @brief Lexer for Windows registration files(.reg) |
| 6 | * @author nkmathew |
| 7 | * |
| 8 | * The License.txt file describes the conditions under which this software may be |
| 9 | * distributed. |
| 10 | * |
| 11 | */ |
| 12 | |
| 13 | #include <cstdlib> |
| 14 | #include <cassert> |
| 15 | #include <cctype> |
| 16 | #include <cstdio> |
| 17 | |
| 18 | #include <string> |
| 19 | #include <string_view> |
| 20 | #include <vector> |
| 21 | #include <map> |
| 22 | #include <functional> |
| 23 | |
| 24 | #include "ILexer.h" |
| 25 | #include "Scintilla.h" |
| 26 | #include "SciLexer.h" |
| 27 | #include "WordList.h" |
| 28 | #include "LexAccessor.h" |
| 29 | #include "StyleContext.h" |
| 30 | #include "CharacterSet.h" |
| 31 | #include "LexerModule.h" |
| 32 | #include "OptionSet.h" |
| 33 | #include "DefaultLexer.h" |
| 34 | |
| 35 | using namespace Scintilla; |
| 36 | using namespace Lexilla; |
| 37 | |
| 38 | static const char *const RegistryWordListDesc[] = { |
| 39 | 0 |
| 40 | }; |
| 41 | |
| 42 | struct OptionsRegistry { |
| 43 | bool foldCompact; |
| 44 | bool fold; |
| 45 | OptionsRegistry() { |
| 46 | foldCompact = false; |
| 47 | fold = false; |
| 48 | } |
| 49 | }; |
| 50 | |
| 51 | struct OptionSetRegistry : public OptionSet<OptionsRegistry> { |
| 52 | OptionSetRegistry() { |
| 53 | DefineProperty("fold.compact" , &OptionsRegistry::foldCompact); |
| 54 | DefineProperty("fold" , &OptionsRegistry::fold); |
| 55 | DefineWordListSets(RegistryWordListDesc); |
| 56 | } |
| 57 | }; |
| 58 | |
| 59 | class LexerRegistry : public DefaultLexer { |
| 60 | OptionsRegistry options; |
| 61 | OptionSetRegistry optSetRegistry; |
| 62 | |
| 63 | static bool IsStringState(int state) { |
| 64 | return (state == SCE_REG_VALUENAME || state == SCE_REG_STRING); |
| 65 | } |
| 66 | |
| 67 | static bool IsKeyPathState(int state) { |
| 68 | return (state == SCE_REG_ADDEDKEY || state == SCE_REG_DELETEDKEY); |
| 69 | } |
| 70 | |
| 71 | static bool AtValueType(LexAccessor &styler, Sci_Position start) { |
| 72 | Sci_Position i = 0; |
| 73 | while (i < 10) { |
| 74 | i++; |
| 75 | char curr = styler.SafeGetCharAt(start+i, '\0'); |
| 76 | if (curr == ':') { |
| 77 | return true; |
| 78 | } else if (!curr) { |
| 79 | return false; |
| 80 | } |
| 81 | } |
| 82 | return false; |
| 83 | } |
| 84 | |
| 85 | static bool IsNextNonWhitespace(LexAccessor &styler, Sci_Position start, char ch) { |
| 86 | Sci_Position i = 0; |
| 87 | while (i < 100) { |
| 88 | i++; |
| 89 | char curr = styler.SafeGetCharAt(start+i, '\0'); |
| 90 | char next = styler.SafeGetCharAt(start+i+1, '\0'); |
| 91 | bool atEOL = (curr == '\r' && next != '\n') || (curr == '\n'); |
| 92 | if (curr == ch) { |
| 93 | return true; |
| 94 | } else if (!isspacechar(curr) || atEOL) { |
| 95 | return false; |
| 96 | } |
| 97 | } |
| 98 | return false; |
| 99 | } |
| 100 | |
| 101 | // Looks for the equal sign at the end of the string |
| 102 | static bool AtValueName(LexAccessor &styler, Sci_Position start) { |
| 103 | bool atEOL = false; |
| 104 | Sci_Position i = 0; |
| 105 | bool escaped = false; |
| 106 | while (!atEOL) { |
| 107 | i++; |
| 108 | char curr = styler.SafeGetCharAt(start+i, '\0'); |
| 109 | char next = styler.SafeGetCharAt(start+i+1, '\0'); |
| 110 | atEOL = (curr == '\r' && next != '\n') || (curr == '\n'); |
| 111 | if (escaped) { |
| 112 | escaped = false; |
| 113 | continue; |
| 114 | } |
| 115 | escaped = curr == '\\'; |
| 116 | if (curr == '"') { |
| 117 | return IsNextNonWhitespace(styler, start+i, '='); |
| 118 | } else if (!curr) { |
| 119 | return false; |
| 120 | } |
| 121 | } |
| 122 | return false; |
| 123 | } |
| 124 | |
| 125 | static bool AtKeyPathEnd(LexAccessor &styler, Sci_Position start) { |
| 126 | bool atEOL = false; |
| 127 | Sci_Position i = 0; |
| 128 | while (!atEOL) { |
| 129 | i++; |
| 130 | char curr = styler.SafeGetCharAt(start+i, '\0'); |
| 131 | char next = styler.SafeGetCharAt(start+i+1, '\0'); |
| 132 | atEOL = (curr == '\r' && next != '\n') || (curr == '\n'); |
| 133 | if (curr == ']' || !curr) { |
| 134 | // There's still at least one or more square brackets ahead |
| 135 | return false; |
| 136 | } |
| 137 | } |
| 138 | return true; |
| 139 | } |
| 140 | |
| 141 | static bool AtGUID(LexAccessor &styler, Sci_Position start) { |
| 142 | int count = 8; |
| 143 | int portion = 0; |
| 144 | int offset = 1; |
| 145 | char digit = '\0'; |
| 146 | while (portion < 5) { |
| 147 | int i = 0; |
| 148 | while (i < count) { |
| 149 | digit = styler.SafeGetCharAt(start+offset); |
| 150 | if (!(isxdigit(digit) || digit == '-')) { |
| 151 | return false; |
| 152 | } |
| 153 | offset++; |
| 154 | i++; |
| 155 | } |
| 156 | portion++; |
| 157 | count = (portion == 4) ? 13 : 5; |
| 158 | } |
| 159 | digit = styler.SafeGetCharAt(start+offset); |
| 160 | if (digit == '}') { |
| 161 | return true; |
| 162 | } else { |
| 163 | return false; |
| 164 | } |
| 165 | } |
| 166 | |
| 167 | public: |
| 168 | LexerRegistry() : DefaultLexer("registry" , SCLEX_REGISTRY) {} |
| 169 | virtual ~LexerRegistry() {} |
| 170 | int SCI_METHOD Version() const override { |
| 171 | return lvRelease5; |
| 172 | } |
| 173 | void SCI_METHOD Release() override { |
| 174 | delete this; |
| 175 | } |
| 176 | const char *SCI_METHOD PropertyNames() override { |
| 177 | return optSetRegistry.PropertyNames(); |
| 178 | } |
| 179 | int SCI_METHOD PropertyType(const char *name) override { |
| 180 | return optSetRegistry.PropertyType(name); |
| 181 | } |
| 182 | const char *SCI_METHOD DescribeProperty(const char *name) override { |
| 183 | return optSetRegistry.DescribeProperty(name); |
| 184 | } |
| 185 | Sci_Position SCI_METHOD PropertySet(const char *key, const char *val) override { |
| 186 | if (optSetRegistry.PropertySet(&options, key, val)) { |
| 187 | return 0; |
| 188 | } |
| 189 | return -1; |
| 190 | } |
| 191 | const char * SCI_METHOD PropertyGet(const char *key) override { |
| 192 | return optSetRegistry.PropertyGet(key); |
| 193 | } |
| 194 | |
| 195 | Sci_Position SCI_METHOD WordListSet(int, const char *) override { |
| 196 | return -1; |
| 197 | } |
| 198 | void *SCI_METHOD PrivateCall(int, void *) override { |
| 199 | return 0; |
| 200 | } |
| 201 | static ILexer5 *LexerFactoryRegistry() { |
| 202 | return new LexerRegistry; |
| 203 | } |
| 204 | const char *SCI_METHOD DescribeWordListSets() override { |
| 205 | return optSetRegistry.DescribeWordListSets(); |
| 206 | } |
| 207 | void SCI_METHOD Lex(Sci_PositionU startPos, |
| 208 | Sci_Position length, |
| 209 | int initStyle, |
| 210 | IDocument *pAccess) override; |
| 211 | void SCI_METHOD Fold(Sci_PositionU startPos, |
| 212 | Sci_Position length, |
| 213 | int initStyle, |
| 214 | IDocument *pAccess) override; |
| 215 | }; |
| 216 | |
| 217 | void SCI_METHOD LexerRegistry::Lex(Sci_PositionU startPos, |
| 218 | Sci_Position length, |
| 219 | int initStyle, |
| 220 | IDocument *pAccess) { |
| 221 | int beforeGUID = SCE_REG_DEFAULT; |
| 222 | int beforeEscape = SCE_REG_DEFAULT; |
| 223 | CharacterSet setOperators = CharacterSet(CharacterSet::setNone, "-,.=:\\@()" ); |
| 224 | LexAccessor styler(pAccess); |
| 225 | StyleContext context(startPos, length, initStyle, styler); |
| 226 | bool highlight = true; |
| 227 | bool afterEqualSign = false; |
| 228 | while (context.More()) { |
| 229 | if (context.atLineStart) { |
| 230 | Sci_Position currPos = static_cast<Sci_Position>(context.currentPos); |
| 231 | bool continued = styler[currPos-3] == '\\'; |
| 232 | highlight = continued ? true : false; |
| 233 | } |
| 234 | switch (context.state) { |
| 235 | case SCE_REG_COMMENT: |
| 236 | if (context.atLineEnd) { |
| 237 | context.SetState(SCE_REG_DEFAULT); |
| 238 | } |
| 239 | break; |
| 240 | case SCE_REG_VALUENAME: |
| 241 | case SCE_REG_STRING: { |
| 242 | Sci_Position currPos = static_cast<Sci_Position>(context.currentPos); |
| 243 | if (context.ch == '"') { |
| 244 | context.ForwardSetState(SCE_REG_DEFAULT); |
| 245 | } else if (context.ch == '\\') { |
| 246 | beforeEscape = context.state; |
| 247 | context.SetState(SCE_REG_ESCAPED); |
| 248 | context.Forward(); |
| 249 | } else if (context.ch == '{') { |
| 250 | if (AtGUID(styler, currPos)) { |
| 251 | beforeGUID = context.state; |
| 252 | context.SetState(SCE_REG_STRING_GUID); |
| 253 | } |
| 254 | } |
| 255 | if (context.state == SCE_REG_STRING && |
| 256 | context.ch == '%' && |
| 257 | (isdigit(context.chNext) || context.chNext == '*')) { |
| 258 | context.SetState(SCE_REG_PARAMETER); |
| 259 | } |
| 260 | } |
| 261 | break; |
| 262 | case SCE_REG_PARAMETER: |
| 263 | context.ForwardSetState(SCE_REG_STRING); |
| 264 | if (context.ch == '"') { |
| 265 | context.ForwardSetState(SCE_REG_DEFAULT); |
| 266 | } |
| 267 | break; |
| 268 | case SCE_REG_VALUETYPE: |
| 269 | if (context.ch == ':') { |
| 270 | context.SetState(SCE_REG_DEFAULT); |
| 271 | afterEqualSign = false; |
| 272 | } |
| 273 | break; |
| 274 | case SCE_REG_HEXDIGIT: |
| 275 | case SCE_REG_OPERATOR: |
| 276 | context.SetState(SCE_REG_DEFAULT); |
| 277 | break; |
| 278 | case SCE_REG_DELETEDKEY: |
| 279 | case SCE_REG_ADDEDKEY: { |
| 280 | Sci_Position currPos = static_cast<Sci_Position>(context.currentPos); |
| 281 | if (context.ch == ']' && AtKeyPathEnd(styler, currPos)) { |
| 282 | context.ForwardSetState(SCE_REG_DEFAULT); |
| 283 | } else if (context.ch == '{') { |
| 284 | if (AtGUID(styler, currPos)) { |
| 285 | beforeGUID = context.state; |
| 286 | context.SetState(SCE_REG_KEYPATH_GUID); |
| 287 | } |
| 288 | } |
| 289 | } |
| 290 | break; |
| 291 | case SCE_REG_ESCAPED: |
| 292 | if (context.ch == '"') { |
| 293 | context.SetState(beforeEscape); |
| 294 | context.ForwardSetState(SCE_REG_DEFAULT); |
| 295 | } else if (context.ch == '\\') { |
| 296 | context.Forward(); |
| 297 | } else { |
| 298 | context.SetState(beforeEscape); |
| 299 | beforeEscape = SCE_REG_DEFAULT; |
| 300 | } |
| 301 | break; |
| 302 | case SCE_REG_STRING_GUID: |
| 303 | case SCE_REG_KEYPATH_GUID: { |
| 304 | if (context.ch == '}') { |
| 305 | context.ForwardSetState(beforeGUID); |
| 306 | beforeGUID = SCE_REG_DEFAULT; |
| 307 | } |
| 308 | Sci_Position currPos = static_cast<Sci_Position>(context.currentPos); |
| 309 | if (context.ch == '"' && IsStringState(context.state)) { |
| 310 | context.ForwardSetState(SCE_REG_DEFAULT); |
| 311 | } else if (context.ch == ']' && |
| 312 | AtKeyPathEnd(styler, currPos) && |
| 313 | IsKeyPathState(context.state)) { |
| 314 | context.ForwardSetState(SCE_REG_DEFAULT); |
| 315 | } else if (context.ch == '\\' && IsStringState(context.state)) { |
| 316 | beforeEscape = context.state; |
| 317 | context.SetState(SCE_REG_ESCAPED); |
| 318 | context.Forward(); |
| 319 | } |
| 320 | } |
| 321 | break; |
| 322 | } |
| 323 | // Determine if a new state should be entered. |
| 324 | if (context.state == SCE_REG_DEFAULT) { |
| 325 | Sci_Position currPos = static_cast<Sci_Position>(context.currentPos); |
| 326 | if (context.ch == ';') { |
| 327 | context.SetState(SCE_REG_COMMENT); |
| 328 | } else if (context.ch == '"') { |
| 329 | if (AtValueName(styler, currPos)) { |
| 330 | context.SetState(SCE_REG_VALUENAME); |
| 331 | } else { |
| 332 | context.SetState(SCE_REG_STRING); |
| 333 | } |
| 334 | } else if (context.ch == '[') { |
| 335 | if (IsNextNonWhitespace(styler, currPos, '-')) { |
| 336 | context.SetState(SCE_REG_DELETEDKEY); |
| 337 | } else { |
| 338 | context.SetState(SCE_REG_ADDEDKEY); |
| 339 | } |
| 340 | } else if (context.ch == '=') { |
| 341 | afterEqualSign = true; |
| 342 | highlight = true; |
| 343 | } else if (afterEqualSign) { |
| 344 | bool wordStart = isalpha(context.ch) && !isalpha(context.chPrev); |
| 345 | if (wordStart && AtValueType(styler, currPos)) { |
| 346 | context.SetState(SCE_REG_VALUETYPE); |
| 347 | } |
| 348 | } else if (isxdigit(context.ch) && highlight) { |
| 349 | context.SetState(SCE_REG_HEXDIGIT); |
| 350 | } |
| 351 | highlight = (context.ch == '@') ? true : highlight; |
| 352 | if (setOperators.Contains(context.ch) && highlight) { |
| 353 | context.SetState(SCE_REG_OPERATOR); |
| 354 | } |
| 355 | } |
| 356 | context.Forward(); |
| 357 | } |
| 358 | context.Complete(); |
| 359 | } |
| 360 | |
| 361 | // Folding similar to that of FoldPropsDoc in LexOthers |
| 362 | void SCI_METHOD LexerRegistry::Fold(Sci_PositionU startPos, |
| 363 | Sci_Position length, |
| 364 | int, |
| 365 | IDocument *pAccess) { |
| 366 | if (!options.fold) { |
| 367 | return; |
| 368 | } |
| 369 | LexAccessor styler(pAccess); |
| 370 | Sci_Position currLine = styler.GetLine(startPos); |
| 371 | int visibleChars = 0; |
| 372 | Sci_PositionU endPos = startPos + length; |
| 373 | bool atKeyPath = false; |
| 374 | for (Sci_PositionU i = startPos; i < endPos; i++) { |
| 375 | atKeyPath = IsKeyPathState(styler.StyleAt(i)) ? true : atKeyPath; |
| 376 | char curr = styler.SafeGetCharAt(i); |
| 377 | char next = styler.SafeGetCharAt(i+1); |
| 378 | bool atEOL = (curr == '\r' && next != '\n') || (curr == '\n'); |
| 379 | if (atEOL || i == (endPos-1)) { |
| 380 | int level = SC_FOLDLEVELBASE; |
| 381 | if (currLine > 0) { |
| 382 | int prevLevel = styler.LevelAt(currLine-1); |
| 383 | if (prevLevel & SC_FOLDLEVELHEADERFLAG) { |
| 384 | level += 1; |
| 385 | } else { |
| 386 | level = prevLevel; |
| 387 | } |
| 388 | } |
| 389 | if (!visibleChars && options.foldCompact) { |
| 390 | level |= SC_FOLDLEVELWHITEFLAG; |
| 391 | } else if (atKeyPath) { |
| 392 | level = SC_FOLDLEVELBASE | SC_FOLDLEVELHEADERFLAG; |
| 393 | } |
| 394 | if (level != styler.LevelAt(currLine)) { |
| 395 | styler.SetLevel(currLine, level); |
| 396 | } |
| 397 | currLine++; |
| 398 | visibleChars = 0; |
| 399 | atKeyPath = false; |
| 400 | } |
| 401 | if (!isspacechar(curr)) { |
| 402 | visibleChars++; |
| 403 | } |
| 404 | } |
| 405 | |
| 406 | // Make the folding reach the last line in the file |
| 407 | int level = SC_FOLDLEVELBASE; |
| 408 | if (currLine > 0) { |
| 409 | int prevLevel = styler.LevelAt(currLine-1); |
| 410 | if (prevLevel & SC_FOLDLEVELHEADERFLAG) { |
| 411 | level += 1; |
| 412 | } else { |
| 413 | level = prevLevel; |
| 414 | } |
| 415 | } |
| 416 | styler.SetLevel(currLine, level); |
| 417 | } |
| 418 | |
| 419 | LexerModule lmRegistry(SCLEX_REGISTRY, |
| 420 | LexerRegistry::LexerFactoryRegistry, |
| 421 | "registry" , |
| 422 | RegistryWordListDesc); |
| 423 | |
| 424 | |