| 1 | // Copyright 2008-2010 Sergiu Dotenco. The License.txt file describes the |
| 2 | // conditions under which this software may be distributed. |
| 3 | |
| 4 | /** |
| 5 | * @file LexBibTeX.cxx |
| 6 | * @brief General BibTeX coloring scheme. |
| 7 | * @author Sergiu Dotenco |
| 8 | * @date April 18, 2009 |
| 9 | */ |
| 10 | |
| 11 | #include <stdlib.h> |
| 12 | #include <string.h> |
| 13 | |
| 14 | #include <cassert> |
| 15 | #include <cctype> |
| 16 | |
| 17 | #include <string> |
| 18 | #include <string_view> |
| 19 | #include <algorithm> |
| 20 | #include <functional> |
| 21 | |
| 22 | #include "ILexer.h" |
| 23 | #include "Scintilla.h" |
| 24 | #include "SciLexer.h" |
| 25 | |
| 26 | #include "PropSetSimple.h" |
| 27 | #include "WordList.h" |
| 28 | #include "LexAccessor.h" |
| 29 | #include "Accessor.h" |
| 30 | #include "StyleContext.h" |
| 31 | #include "CharacterSet.h" |
| 32 | #include "LexerModule.h" |
| 33 | |
| 34 | using namespace Lexilla; |
| 35 | |
| 36 | namespace { |
| 37 | bool IsAlphabetic(unsigned int ch) |
| 38 | { |
| 39 | return IsASCII(ch) && std::isalpha(ch) != 0; |
| 40 | } |
| 41 | bool IsAlphaNumeric(char ch) |
| 42 | { |
| 43 | return IsASCII(ch) && std::isalnum(ch); |
| 44 | } |
| 45 | |
| 46 | bool EqualCaseInsensitive(const char* a, const char* b) |
| 47 | { |
| 48 | return CompareCaseInsensitive(a, b) == 0; |
| 49 | } |
| 50 | |
| 51 | bool EntryWithoutKey(const char* name) |
| 52 | { |
| 53 | return EqualCaseInsensitive(name,"string" ); |
| 54 | } |
| 55 | |
| 56 | char GetClosingBrace(char openbrace) |
| 57 | { |
| 58 | char result = openbrace; |
| 59 | |
| 60 | switch (openbrace) { |
| 61 | case '(': result = ')'; break; |
| 62 | case '{': result = '}'; break; |
| 63 | } |
| 64 | |
| 65 | return result; |
| 66 | } |
| 67 | |
| 68 | bool IsEntryStart(char prev, char ch) |
| 69 | { |
| 70 | return prev != '\\' && ch == '@'; |
| 71 | } |
| 72 | |
| 73 | bool IsEntryStart(const StyleContext& sc) |
| 74 | { |
| 75 | return IsEntryStart(sc.chPrev, sc.ch); |
| 76 | } |
| 77 | |
| 78 | void ColorizeBibTeX(Sci_PositionU start_pos, Sci_Position length, int /*init_style*/, WordList* keywordlists[], Accessor& styler) |
| 79 | { |
| 80 | WordList &EntryNames = *keywordlists[0]; |
| 81 | bool fold_compact = styler.GetPropertyInt("fold.compact" , 1) != 0; |
| 82 | |
| 83 | std::string buffer; |
| 84 | buffer.reserve(25); |
| 85 | |
| 86 | // We always colorize a section from the beginning, so let's |
| 87 | // search for the @ character which isn't escaped, i.e. \@ |
| 88 | while (start_pos > 0 && !IsEntryStart(styler.SafeGetCharAt(start_pos - 1), |
| 89 | styler.SafeGetCharAt(start_pos))) { |
| 90 | --start_pos; ++length; |
| 91 | } |
| 92 | |
| 93 | styler.StartAt(start_pos); |
| 94 | styler.StartSegment(start_pos); |
| 95 | |
| 96 | Sci_Position current_line = styler.GetLine(start_pos); |
| 97 | int prev_level = styler.LevelAt(current_line) & SC_FOLDLEVELNUMBERMASK; |
| 98 | int current_level = prev_level; |
| 99 | int visible_chars = 0; |
| 100 | |
| 101 | bool = false ; |
| 102 | StyleContext sc(start_pos, length, SCE_BIBTEX_DEFAULT, styler); |
| 103 | |
| 104 | bool going = sc.More(); // needed because of a fuzzy end of file state |
| 105 | char closing_brace = 0; |
| 106 | bool collect_entry_name = false; |
| 107 | |
| 108 | for (; going; sc.Forward()) { |
| 109 | if (!sc.More()) |
| 110 | going = false; // we need to go one behind the end of text |
| 111 | |
| 112 | if (in_comment) { |
| 113 | if (sc.atLineEnd) { |
| 114 | sc.SetState(SCE_BIBTEX_DEFAULT); |
| 115 | in_comment = false; |
| 116 | } |
| 117 | } |
| 118 | else { |
| 119 | // Found @entry |
| 120 | if (IsEntryStart(sc)) { |
| 121 | sc.SetState(SCE_BIBTEX_UNKNOWN_ENTRY); |
| 122 | sc.Forward(); |
| 123 | ++current_level; |
| 124 | |
| 125 | buffer.clear(); |
| 126 | collect_entry_name = true; |
| 127 | } |
| 128 | else if ((sc.state == SCE_BIBTEX_ENTRY || sc.state == SCE_BIBTEX_UNKNOWN_ENTRY) |
| 129 | && (sc.ch == '{' || sc.ch == '(')) { |
| 130 | // Entry name colorization done |
| 131 | // Found either a { or a ( after entry's name, e.g. @entry(...) @entry{...} |
| 132 | // Closing counterpart needs to be stored. |
| 133 | closing_brace = GetClosingBrace(sc.ch); |
| 134 | |
| 135 | sc.SetState(SCE_BIBTEX_DEFAULT); // Don't colorize { ( |
| 136 | |
| 137 | // @string doesn't have any key |
| 138 | if (EntryWithoutKey(buffer.c_str())) |
| 139 | sc.ForwardSetState(SCE_BIBTEX_PARAMETER); |
| 140 | else |
| 141 | sc.ForwardSetState(SCE_BIBTEX_KEY); // Key/label colorization |
| 142 | } |
| 143 | |
| 144 | // Need to handle the case where entry's key is empty |
| 145 | // e.g. @book{,...} |
| 146 | if (sc.state == SCE_BIBTEX_KEY && sc.ch == ',') { |
| 147 | // Key/label colorization done |
| 148 | sc.SetState(SCE_BIBTEX_DEFAULT); // Don't colorize the , |
| 149 | sc.ForwardSetState(SCE_BIBTEX_PARAMETER); // Parameter colorization |
| 150 | } |
| 151 | else if (sc.state == SCE_BIBTEX_PARAMETER && sc.ch == '=') { |
| 152 | sc.SetState(SCE_BIBTEX_DEFAULT); // Don't colorize the = |
| 153 | sc.ForwardSetState(SCE_BIBTEX_VALUE); // Parameter value colorization |
| 154 | |
| 155 | Sci_Position start = sc.currentPos; |
| 156 | |
| 157 | // We need to handle multiple situations: |
| 158 | // 1. name"one two {three}" |
| 159 | // 2. name={one {one two {two}} three} |
| 160 | // 3. year=2005 |
| 161 | |
| 162 | // Skip ", { until we encounter the first alphanumerical character |
| 163 | while (sc.More() && !(IsAlphaNumeric(sc.ch) || sc.ch == '"' || sc.ch == '{')) |
| 164 | sc.Forward(); |
| 165 | |
| 166 | if (sc.More()) { |
| 167 | // Store " or { |
| 168 | char ch = sc.ch; |
| 169 | |
| 170 | // Not interested in alphanumerical characters |
| 171 | if (IsAlphaNumeric(ch)) |
| 172 | ch = 0; |
| 173 | |
| 174 | int skipped = 0; |
| 175 | |
| 176 | if (ch) { |
| 177 | // Skip preceding " or { such as in name={{test}}. |
| 178 | // Remember how many characters have been skipped |
| 179 | // Make sure that empty values, i.e. "" are also handled correctly |
| 180 | while (sc.More() && (sc.ch == ch && (ch != '"' || skipped < 1))) { |
| 181 | sc.Forward(); |
| 182 | ++skipped; |
| 183 | } |
| 184 | } |
| 185 | |
| 186 | // Closing counterpart for " is the same character |
| 187 | if (ch == '{') |
| 188 | ch = '}'; |
| 189 | |
| 190 | // We have reached the parameter value |
| 191 | // In case the open character was a alnum char, skip until , is found |
| 192 | // otherwise until skipped == 0 |
| 193 | while (sc.More() && (skipped > 0 || (!ch && !(sc.ch == ',' || sc.ch == closing_brace)))) { |
| 194 | // Make sure the character isn't escaped |
| 195 | if (sc.chPrev != '\\') { |
| 196 | // Parameter value contains a { which is the 2nd case described above |
| 197 | if (sc.ch == '{') |
| 198 | ++skipped; // Remember it |
| 199 | else if (sc.ch == '}') |
| 200 | --skipped; |
| 201 | else if (skipped == 1 && sc.ch == ch && ch == '"') // Don't ignore cases like {"o} |
| 202 | skipped = 0; |
| 203 | } |
| 204 | |
| 205 | sc.Forward(); |
| 206 | } |
| 207 | } |
| 208 | |
| 209 | // Don't colorize the , |
| 210 | sc.SetState(SCE_BIBTEX_DEFAULT); |
| 211 | |
| 212 | // Skip until the , or entry's closing closing_brace is found |
| 213 | // since this parameter might be the last one |
| 214 | while (sc.More() && !(sc.ch == ',' || sc.ch == closing_brace)) |
| 215 | sc.Forward(); |
| 216 | |
| 217 | int state = SCE_BIBTEX_PARAMETER; // The might be more parameters |
| 218 | |
| 219 | // We've reached the closing closing_brace for the bib entry |
| 220 | // in case no " or {} has been used to enclose the value, |
| 221 | // as in 3rd case described above |
| 222 | if (sc.ch == closing_brace) { |
| 223 | --current_level; |
| 224 | // Make sure the text between entries is not colored |
| 225 | // using parameter's style |
| 226 | state = SCE_BIBTEX_DEFAULT; |
| 227 | } |
| 228 | |
| 229 | Sci_Position end = sc.currentPos; |
| 230 | current_line = styler.GetLine(end); |
| 231 | |
| 232 | // We have possibly skipped some lines, so the folding levels |
| 233 | // have to be adjusted separately |
| 234 | for (Sci_Position i = styler.GetLine(start); i <= styler.GetLine(end); ++i) |
| 235 | styler.SetLevel(i, prev_level); |
| 236 | |
| 237 | sc.ForwardSetState(state); |
| 238 | } |
| 239 | |
| 240 | if (sc.state == SCE_BIBTEX_PARAMETER && sc.ch == closing_brace) { |
| 241 | sc.SetState(SCE_BIBTEX_DEFAULT); |
| 242 | --current_level; |
| 243 | } |
| 244 | |
| 245 | // Non escaped % found which represents a comment until the end of the line |
| 246 | if (sc.chPrev != '\\' && sc.ch == '%') { |
| 247 | in_comment = true; |
| 248 | sc.SetState(SCE_BIBTEX_COMMENT); |
| 249 | } |
| 250 | } |
| 251 | |
| 252 | if (sc.state == SCE_BIBTEX_UNKNOWN_ENTRY || sc.state == SCE_BIBTEX_ENTRY) { |
| 253 | if (!IsAlphabetic(sc.ch) && collect_entry_name) |
| 254 | collect_entry_name = false; |
| 255 | |
| 256 | if (collect_entry_name) { |
| 257 | buffer += static_cast<char>(tolower(sc.ch)); |
| 258 | if (EntryNames.InList(buffer.c_str())) |
| 259 | sc.ChangeState(SCE_BIBTEX_ENTRY); |
| 260 | else |
| 261 | sc.ChangeState(SCE_BIBTEX_UNKNOWN_ENTRY); |
| 262 | } |
| 263 | } |
| 264 | |
| 265 | if (sc.atLineEnd) { |
| 266 | int level = prev_level; |
| 267 | |
| 268 | if (visible_chars == 0 && fold_compact) |
| 269 | level |= SC_FOLDLEVELWHITEFLAG; |
| 270 | |
| 271 | if ((current_level > prev_level)) |
| 272 | level |= SC_FOLDLEVELHEADERFLAG; |
| 273 | // else if (current_level < prev_level) |
| 274 | // level |= SC_FOLDLEVELBOXFOOTERFLAG; // Deprecated |
| 275 | |
| 276 | if (level != styler.LevelAt(current_line)) { |
| 277 | styler.SetLevel(current_line, level); |
| 278 | } |
| 279 | |
| 280 | ++current_line; |
| 281 | prev_level = current_level; |
| 282 | visible_chars = 0; |
| 283 | } |
| 284 | |
| 285 | if (!isspacechar(sc.ch)) |
| 286 | ++visible_chars; |
| 287 | } |
| 288 | |
| 289 | sc.Complete(); |
| 290 | |
| 291 | // Fill in the real level of the next line, keeping the current flags as they will be filled in later |
| 292 | int flagsNext = styler.LevelAt(current_line) & ~SC_FOLDLEVELNUMBERMASK; |
| 293 | styler.SetLevel(current_line, prev_level | flagsNext); |
| 294 | } |
| 295 | } |
| 296 | static const char * const BibTeXWordLists[] = { |
| 297 | "Entry Names" , |
| 298 | 0, |
| 299 | }; |
| 300 | |
| 301 | |
| 302 | LexerModule lmBibTeX(SCLEX_BIBTEX, ColorizeBibTeX, "bib" , 0, BibTeXWordLists); |
| 303 | |
| 304 | // Entry Names |
| 305 | // article, book, booklet, conference, inbook, |
| 306 | // incollection, inproceedings, manual, mastersthesis, |
| 307 | // misc, phdthesis, proceedings, techreport, unpublished, |
| 308 | // string, url |
| 309 | |
| 310 | |