| 1 |
// Scintilla source code edit control |
| 2 |
/** @file LexR.cxx |
| 3 |
** Lexer for R, S, SPlus Statistics Program (Heavily derived from CPP Lexer). |
| 4 |
** |
| 5 |
**/ |
| 6 |
// Copyright 1998-2002 by Neil Hodgson <[email protected]> |
| 7 |
// The License.txt file describes the conditions under which this software may be distributed. |
| 8 |
|
| 9 |
#include <cassert> |
| 10 |
#include <cctype> |
| 11 |
|
| 12 |
#include <string> |
| 13 |
#include <string_view> |
| 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 |
using namespace Lexilla; |
| 27 |
|
| 28 |
namespace { |
| 29 |
|
| 30 |
inline bool IsAWordChar(int ch) noexcept { |
| 31 |
return (ch < 0x80) && (isalnum(ch) || ch == '.' || ch == '_'); |
| 32 |
} |
| 33 |
|
| 34 |
inline bool IsAWordStart(int ch) noexcept { |
| 35 |
return (ch < 0x80) && (isalnum(ch) || ch == '_'); |
| 36 |
} |
| 37 |
|
| 38 |
constexpr bool IsAnOperator(int ch) noexcept { |
| 39 |
// '.' left out as it is used to make up numbers |
| 40 |
if (ch == '-' || ch == '+' || ch == '!' || ch == '~' || |
| 41 |
ch == '?' || ch == ':' || ch == '*' || ch == '/' || |
| 42 |
ch == '^' || ch == '<' || ch == '>' || ch == '=' || |
| 43 |
ch == '&' || ch == '|' || ch == '$' || ch == '(' || |
| 44 |
ch == ')' || ch == '}' || ch == '{' || ch == '[' || |
| 45 |
ch == ']') |
| 46 |
return true; |
| 47 |
return false; |
| 48 |
} |
| 49 |
|
| 50 |
constexpr bool IsOctalOrHex(int ch, bool hex) noexcept { |
| 51 |
return IsAnOctalDigit(ch) || (hex && IsAHeXDigit(ch)); |
| 52 |
} |
| 53 |
|
| 54 |
// https://search.r-project.org/R/refmans/base/html/Quotes.html |
| 55 |
struct EscapeSequence { |
| 56 |
int outerState = SCE_R_DEFAULT; |
| 57 |
int digitsLeft = 0; |
| 58 |
bool hex = false; |
| 59 |
bool brace = false; |
| 60 |
|
| 61 |
// highlight any character as escape sequence, unrecognized escape sequence is syntax error. |
| 62 |
void resetEscapeState(int state, int chNext) noexcept { |
| 63 |
outerState = state; |
| 64 |
digitsLeft = 1; |
| 65 |
hex = true; |
| 66 |
brace = false; |
| 67 |
if (chNext == 'x') { |
| 68 |
digitsLeft = 3; |
| 69 |
} else if (chNext == 'u') { |
| 70 |
digitsLeft = 5; |
| 71 |
} else if (chNext == 'U') { |
| 72 |
digitsLeft = 9; |
| 73 |
} else if (IsAnOctalDigit(chNext)) { |
| 74 |
digitsLeft = 3; |
| 75 |
hex = false; |
| 76 |
} |
| 77 |
} |
| 78 |
bool atEscapeEnd(int ch) noexcept { |
| 79 |
--digitsLeft; |
| 80 |
return digitsLeft <= 0 || !IsOctalOrHex(ch, hex); |
| 81 |
} |
| 82 |
}; |
| 83 |
|
| 84 |
int CheckRawString(LexAccessor &styler, Sci_Position pos, int &dashCount) { |
| 85 |
dashCount = 0; |
| 86 |
while (true) { |
| 87 |
const char ch = styler.SafeGetCharAt(pos++); |
| 88 |
switch (ch) { |
| 89 |
case '-': |
| 90 |
++dashCount; |
| 91 |
break; |
| 92 |
case '(': |
| 93 |
return ')'; |
| 94 |
case '[': |
| 95 |
return ']'; |
| 96 |
case '{': |
| 97 |
return '}'; |
| 98 |
default: |
| 99 |
dashCount = 0; |
| 100 |
return 0; |
| 101 |
} |
| 102 |
} |
| 103 |
} |
| 104 |
|
| 105 |
void ColouriseRDoc(Sci_PositionU startPos, Sci_Position length, int initStyle, WordList *keywordlists[], |
| 106 |
Accessor &styler) { |
| 107 |
|
| 108 |
const WordList &keywords = *keywordlists[0]; |
| 109 |
const WordList &keywords2 = *keywordlists[1]; |
| 110 |
const WordList &keywords3 = *keywordlists[2]; |
| 111 |
// state for raw string |
| 112 |
int matchingDelimiter = 0; |
| 113 |
int dashCount = 0; |
| 114 |
|
| 115 |
// property lexer.r.escape.sequence |
| 116 |
// Set to 1 to enable highlighting of escape sequences in strings. |
| 117 |
const bool escapeSequence = styler.GetPropertyInt("lexer.r.escape.sequence", 0) != 0; |
| 118 |
EscapeSequence escapeSeq; |
| 119 |
|
| 120 |
StyleContext sc(startPos, length, initStyle, styler); |
| 121 |
if (sc.currentLine > 0) { |
| 122 |
const int lineState = styler.GetLineState(sc.currentLine - 1); |
| 123 |
matchingDelimiter = lineState & 0xff; |
| 124 |
dashCount = lineState >> 8; |
| 125 |
} |
| 126 |
|
| 127 |
while (sc.More()) { |
| 128 |
// Determine if the current state should terminate. |
| 129 |
switch (sc.state) { |
| 130 |
case SCE_R_OPERATOR: |
| 131 |
sc.SetState(SCE_R_DEFAULT); |
| 132 |
break; |
| 133 |
|
| 134 |
case SCE_R_NUMBER: |
| 135 |
// https://cran.r-project.org/doc/manuals/r-release/R-lang.html#Literal-constants |
| 136 |
if (AnyOf(sc.ch, 'e', 'E', 'p', 'P') && (IsADigit(sc.chNext) || sc.chNext == '+' || sc.chNext == '-')) { |
| 137 |
sc.Forward(); // exponent part |
| 138 |
} else if (!(IsAHeXDigit(sc.ch) || (sc.ch == '.' && IsADigit(sc.chNext)))) { |
| 139 |
if (AnyOf(sc.ch, 'L', 'i')) { |
| 140 |
sc.Forward(); // integer and complex qualifier |
| 141 |
} |
| 142 |
sc.SetState(SCE_R_DEFAULT); |
| 143 |
} |
| 144 |
break; |
| 145 |
|
| 146 |
case SCE_R_IDENTIFIER: |
| 147 |
if (!IsAWordChar(sc.ch)) { |
| 148 |
char s[100]; |
| 149 |
sc.GetCurrent(s, sizeof(s)); |
| 150 |
if (keywords.InList(s)) { |
| 151 |
sc.ChangeState(SCE_R_KWORD); |
| 152 |
} else if (keywords2.InList(s)) { |
| 153 |
sc.ChangeState(SCE_R_BASEKWORD); |
| 154 |
} else if (keywords3.InList(s)) { |
| 155 |
sc.ChangeState(SCE_R_OTHERKWORD); |
| 156 |
} |
| 157 |
sc.SetState(SCE_R_DEFAULT); |
| 158 |
} |
| 159 |
break; |
| 160 |
|
| 161 |
case SCE_R_COMMENT: |
| 162 |
if (sc.MatchLineEnd()) { |
| 163 |
sc.SetState(SCE_R_DEFAULT); |
| 164 |
} |
| 165 |
break; |
| 166 |
|
| 167 |
case SCE_R_STRING: |
| 168 |
case SCE_R_STRING2: |
| 169 |
case SCE_R_BACKTICKS: |
| 170 |
if (sc.ch == '\\') { |
| 171 |
if (escapeSequence) { |
| 172 |
escapeSeq.resetEscapeState(sc.state, sc.chNext); |
| 173 |
sc.SetState(SCE_R_ESCAPESEQUENCE); |
| 174 |
sc.Forward(); |
| 175 |
if (sc.chNext == '{' && AnyOf(sc.ch, 'u', 'U')) { |
| 176 |
escapeSeq.brace = true; |
| 177 |
sc.Forward(); |
| 178 |
} else if (sc.MatchLineEnd()) { |
| 179 |
// don't highlight line ending as escape sequence: |
| 180 |
// escapeSeq.outerState is lost when editing on next line. |
| 181 |
sc.SetState(escapeSeq.outerState); |
| 182 |
} |
| 183 |
} else { |
| 184 |
sc.Forward(); // Skip all characters after the backslash |
| 185 |
} |
| 186 |
} else if ((sc.state == SCE_R_STRING && sc.ch == '\"') |
| 187 |
|| (sc.state == SCE_R_STRING2 && sc.ch == '\'') |
| 188 |
|| (sc.state == SCE_R_BACKTICKS && sc.ch == '`')) { |
| 189 |
sc.ForwardSetState(SCE_R_DEFAULT); |
| 190 |
} |
| 191 |
break; |
| 192 |
|
| 193 |
case SCE_R_ESCAPESEQUENCE: |
| 194 |
if (escapeSeq.atEscapeEnd(sc.ch)) { |
| 195 |
if (escapeSeq.brace && sc.ch == '}') { |
| 196 |
sc.Forward(); |
| 197 |
} |
| 198 |
sc.SetState(escapeSeq.outerState); |
| 199 |
continue; |
| 200 |
} |
| 201 |
break; |
| 202 |
|
| 203 |
case SCE_R_RAWSTRING: |
| 204 |
case SCE_R_RAWSTRING2: |
| 205 |
while (sc.ch == matchingDelimiter) { |
| 206 |
sc.Forward(); |
| 207 |
int count = dashCount; |
| 208 |
while (count != 0 && sc.ch == '-') { |
| 209 |
--count; |
| 210 |
sc.Forward(); |
| 211 |
} |
| 212 |
if (count == 0 && sc.ch == ((sc.state == SCE_R_RAWSTRING) ? '\"' : '\'')) { |
| 213 |
matchingDelimiter = 0; |
| 214 |
dashCount = 0; |
| 215 |
sc.ForwardSetState(SCE_R_DEFAULT); |
| 216 |
break; |
| 217 |
} |
| 218 |
} |
| 219 |
break; |
| 220 |
|
| 221 |
case SCE_R_INFIX: |
| 222 |
if (sc.ch == '%') { |
| 223 |
sc.ForwardSetState(SCE_R_DEFAULT); |
| 224 |
} else if (sc.atLineEnd) { |
| 225 |
sc.ChangeState(SCE_R_INFIXEOL); |
| 226 |
} |
| 227 |
break; |
| 228 |
|
| 229 |
case SCE_R_INFIXEOL: |
| 230 |
if (sc.atLineStart) { |
| 231 |
sc.SetState(SCE_R_DEFAULT); |
| 232 |
} |
| 233 |
break; |
| 234 |
} |
| 235 |
|
| 236 |
// Determine if a new state should be entered. |
| 237 |
if (sc.state == SCE_R_DEFAULT) { |
| 238 |
if (IsADigit(sc.ch) || (sc.ch == '.' && IsADigit(sc.chNext))) { |
| 239 |
sc.SetState(SCE_R_NUMBER); |
| 240 |
if (sc.ch == '0' && AnyOf(sc.chNext, 'x', 'X')) { |
| 241 |
sc.Forward(); |
| 242 |
} |
| 243 |
} else if (AnyOf(sc.ch, 'r', 'R') && AnyOf(sc.chNext, '\"', '\'')) { |
| 244 |
const int chNext = sc.chNext; |
| 245 |
matchingDelimiter = CheckRawString(styler, sc.currentPos + 2, dashCount); |
| 246 |
if (matchingDelimiter) { |
| 247 |
sc.SetState((chNext == '\"') ? SCE_R_RAWSTRING : SCE_R_RAWSTRING2); |
| 248 |
sc.Forward(dashCount + 2); |
| 249 |
} else { |
| 250 |
// syntax error |
| 251 |
sc.SetState(SCE_R_IDENTIFIER); |
| 252 |
sc.ForwardSetState((chNext == '\"') ? SCE_R_STRING : SCE_R_STRING2); |
| 253 |
} |
| 254 |
} else if (IsAWordStart(sc.ch) ) { |
| 255 |
sc.SetState(SCE_R_IDENTIFIER); |
| 256 |
} else if (sc.Match('#')) { |
| 257 |
sc.SetState(SCE_R_COMMENT); |
| 258 |
} else if (sc.ch == '\"') { |
| 259 |
sc.SetState(SCE_R_STRING); |
| 260 |
} else if (sc.ch == '%') { |
| 261 |
sc.SetState(SCE_R_INFIX); |
| 262 |
} else if (sc.ch == '\'') { |
| 263 |
sc.SetState(SCE_R_STRING2); |
| 264 |
} else if (sc.ch == '`') { |
| 265 |
sc.SetState(SCE_R_BACKTICKS); |
| 266 |
} else if (IsAnOperator(sc.ch)) { |
| 267 |
sc.SetState(SCE_R_OPERATOR); |
| 268 |
} |
| 269 |
} |
| 270 |
|
| 271 |
if (sc.atLineEnd) { |
| 272 |
const int lineState = matchingDelimiter | (dashCount << 8); |
| 273 |
styler.SetLineState(sc.currentLine, lineState); |
| 274 |
} |
| 275 |
sc.Forward(); |
| 276 |
} |
| 277 |
sc.Complete(); |
| 278 |
} |
| 279 |
|
| 280 |
// Store both the current line's fold level and the next lines in the |
| 281 |
// level store to make it easy to pick up with each increment |
| 282 |
// and to make it possible to fiddle the current level for "} else {". |
| 283 |
void FoldRDoc(Sci_PositionU startPos, Sci_Position length, int, WordList *[], |
| 284 |
Accessor &styler) { |
| 285 |
const bool foldCompact = styler.GetPropertyInt("fold.compact", 1) != 0; |
| 286 |
const bool foldAtElse = styler.GetPropertyInt("fold.at.else", 0) != 0; |
| 287 |
const Sci_PositionU endPos = startPos + length; |
| 288 |
int visibleChars = 0; |
| 289 |
Sci_Position lineCurrent = styler.GetLine(startPos); |
| 290 |
int levelCurrent = SC_FOLDLEVELBASE; |
| 291 |
if (lineCurrent > 0) |
| 292 |
levelCurrent = styler.LevelAt(lineCurrent-1) >> 16; |
| 293 |
int levelMinCurrent = levelCurrent; |
| 294 |
int levelNext = levelCurrent; |
| 295 |
char chNext = styler[startPos]; |
| 296 |
int styleNext = styler.StyleAt(startPos); |
| 297 |
for (Sci_PositionU i = startPos; i < endPos; i++) { |
| 298 |
const char ch = chNext; |
| 299 |
chNext = styler.SafeGetCharAt(i + 1); |
| 300 |
const int style = styleNext; |
| 301 |
styleNext = styler.StyleAt(i + 1); |
| 302 |
const bool atEOL = (ch == '\r' && chNext != '\n') || (ch == '\n'); |
| 303 |
if (style == SCE_R_OPERATOR) { |
| 304 |
if (ch == '{') { |
| 305 |
// Measure the minimum before a '{' to allow |
| 306 |
// folding on "} else {" |
| 307 |
if (levelMinCurrent > levelNext) { |
| 308 |
levelMinCurrent = levelNext; |
| 309 |
} |
| 310 |
levelNext++; |
| 311 |
} else if (ch == '}') { |
| 312 |
levelNext--; |
| 313 |
} |
| 314 |
} |
| 315 |
if (atEOL) { |
| 316 |
int levelUse = levelCurrent; |
| 317 |
if (foldAtElse) { |
| 318 |
levelUse = levelMinCurrent; |
| 319 |
} |
| 320 |
int lev = levelUse | levelNext << 16; |
| 321 |
if (visibleChars == 0 && foldCompact) |
| 322 |
lev |= SC_FOLDLEVELWHITEFLAG; |
| 323 |
if (levelUse < levelNext) |
| 324 |
lev |= SC_FOLDLEVELHEADERFLAG; |
| 325 |
if (lev != styler.LevelAt(lineCurrent)) { |
| 326 |
styler.SetLevel(lineCurrent, lev); |
| 327 |
} |
| 328 |
lineCurrent++; |
| 329 |
levelCurrent = levelNext; |
| 330 |
levelMinCurrent = levelCurrent; |
| 331 |
visibleChars = 0; |
| 332 |
} |
| 333 |
if (!isspacechar(ch)) |
| 334 |
visibleChars++; |
| 335 |
} |
| 336 |
} |
| 337 |
|
| 338 |
|
| 339 |
const char * const RWordLists[] = { |
| 340 |
"Language Keywords", |
| 341 |
"Base / Default package function", |
| 342 |
"Other Package Functions", |
| 343 |
"Unused", |
| 344 |
"Unused", |
| 345 |
nullptr, |
| 346 |
}; |
| 347 |
|
| 348 |
} |
| 349 |
|
| 350 |
extern const LexerModule lmR(SCLEX_R, ColouriseRDoc, "r", FoldRDoc, RWordLists); |