lexperl.cxx.svn-base

Notepad++ is a generic source code editor (it tries to be anyway) and Notepad replacement written in

				}
			} else if (state == SCE_PL_HERE_DELIM) {
				//
				// From perldata.pod:
				// ------------------
				// A line-oriented form of quoting is based on the shell ``here-doc''
				// syntax.
				// Following a << you specify a string to terminate the quoted material,
				// and all lines following the current line down to the terminating
				// string are the value of the item.
				// The terminating string may be either an identifier (a word),
				// or some quoted text.
				// If quoted, the type of quotes you use determines the treatment of
				// the text, just as in regular quoting.
				// An unquoted identifier works like double quotes.
				// There must be no space between the << and the identifier.
				// (If you put a space it will be treated as a null identifier,
				// which is valid, and matches the first empty line.)
				// (This is deprecated, -w warns of this syntax)
				// The terminating string must appear by itself (unquoted and with no
				// surrounding whitespace) on the terminating line.
				//
				// From Bash info:
				// ---------------
				// Specifier format is: <<[-]WORD
				// Optional '-' is for removal of leading tabs from here-doc.
				// Whitespace acceptable after <<[-] operator.
				//
				if (HereDoc.State == 0) { // '<<' encountered
                    bool gotspace = false;
                    unsigned int oldi = i;
                    if (chNext == ' ' || chNext == '\t') {
                        // skip whitespace; legal for quoted delimiters
                        gotspace = true;
                        do {
                            i++;
                            chNext = styler.SafeGetCharAt(i + 1);
                        } while ((i + 1 < lengthDoc) && (chNext == ' ' || chNext == '\t'));
                        chNext2 = styler.SafeGetCharAt(i + 2);
                    }
					HereDoc.State = 1;
					HereDoc.Quote = chNext;
					HereDoc.Quoted = false;
					HereDoc.DelimiterLength = 0;
					HereDoc.Delimiter[HereDoc.DelimiterLength] = '\0';
					if (chNext == '\'' || chNext == '"' || chNext == '`') {
                        // a quoted here-doc delimiter
						i++;
						ch = chNext;
						chNext = chNext2;
						HereDoc.Quoted = true;
					} else if (isspacechar(chNext) || isdigit(chNext) || chNext == '\\'
						|| chNext == '=' || chNext == '$' || chNext == '@'
                        || ((isalpha(chNext) || chNext == '_') && gotspace)) {
						// left shift << or <<= operator cases
                        // restore position if operator
                        i = oldi;
						styler.ColourTo(i, SCE_PL_OPERATOR);
						state = SCE_PL_DEFAULT;
						HereDoc.State = 0;
                        goto restartLexer;
					} else {
						// an unquoted here-doc delimiter, no special handling
                        // (cannot be prefixed by spaces/tabs), or
						// symbols terminates; deprecated zero-length delimiter
					}

				} else if (HereDoc.State == 1) { // collect the delimiter
                    backflag = BACK_NONE;
					if (HereDoc.Quoted) { // a quoted here-doc delimiter
						if (ch == HereDoc.Quote) { // closing quote => end of delimiter
							styler.ColourTo(i, state);
							state = SCE_PL_DEFAULT;
						} else {
							if (ch == '\\' && chNext == HereDoc.Quote) { // escaped quote
								i++;
								ch = chNext;
								chNext = chNext2;
							}
							HereDoc.Delimiter[HereDoc.DelimiterLength++] = ch;
							HereDoc.Delimiter[HereDoc.DelimiterLength] = '\0';
						}
					} else { // an unquoted here-doc delimiter
						if (isalnum(ch) || ch == '_') {
							HereDoc.Delimiter[HereDoc.DelimiterLength++] = ch;
							HereDoc.Delimiter[HereDoc.DelimiterLength] = '\0';
						} else {
							styler.ColourTo(i - 1, state);
							state = SCE_PL_DEFAULT;
							goto restartLexer;
						}
					}
					if (HereDoc.DelimiterLength >= HERE_DELIM_MAX - 1) {
						styler.ColourTo(i - 1, state);
						state = SCE_PL_ERROR;
						goto restartLexer;
					}
				}
			} else if (HereDoc.State == 2) {
				// state == SCE_PL_HERE_Q || state == SCE_PL_HERE_QQ || state == SCE_PL_HERE_QX
				if (isEOLChar(chPrev) && isMatch(styler, lengthDoc, i, HereDoc.Delimiter)) {
					i += HereDoc.DelimiterLength;
					chPrev = styler.SafeGetCharAt(i - 1);
					ch = styler.SafeGetCharAt(i);
					if (isEOLChar(ch)) {
						styler.ColourTo(i - 1, state);
						state = SCE_PL_DEFAULT;
                        backflag = BACK_NONE;
						HereDoc.State = 0;
						goto restartLexer;
					}
					chNext = styler.SafeGetCharAt(i + 1);
				}
			} else if (state == SCE_PL_POD
				|| state == SCE_PL_POD_VERB) {
				if (isEOLChar(chPrev)) {
					if (ch == ' ' || ch == '\t') {
						styler.ColourTo(i - 1, state);
						state = SCE_PL_POD_VERB;
					} else {
						styler.ColourTo(i - 1, state);
						state = SCE_PL_POD;
						if (ch == '=') {
							if (isMatch(styler, lengthDoc, i, "=cut")) {
								styler.ColourTo(i - 1 + 4, state);
								i += 4;
								state = SCE_PL_DEFAULT;
								ch = styler.SafeGetCharAt(i);
								//chNext = styler.SafeGetCharAt(i + 1);
								goto restartLexer;
							}
						}
					}
				}
			} else if (state == SCE_PL_SCALAR	// variable names
				|| state == SCE_PL_ARRAY
				|| state == SCE_PL_HASH
				|| state == SCE_PL_SYMBOLTABLE) {
				if (ch == ':' && chNext == ':') {	// skip ::
					i++;
					ch = chNext;
					chNext = chNext2;
				}
				else if (isEndVar(ch)) {
					if (i == (styler.GetStartSegment() + 1)) {
						// Special variable: $(, $_ etc.
						styler.ColourTo(i, state);
						state = SCE_PL_DEFAULT;
					} else {
						styler.ColourTo(i - 1, state);
						state = SCE_PL_DEFAULT;
						goto restartLexer;
					}
				}
			} else if (state == SCE_PL_REGEX
				|| state == SCE_PL_STRING_QR
				) {
				if (!Quote.Up && !isspacechar(ch)) {
					Quote.Open(ch);
				} else if (ch == '\\' && Quote.Up != '\\') {
					// SG: Is it save to skip *every* escaped char?
					i++;
					ch = chNext;
					chNext = styler.SafeGetCharAt(i + 1);
				} else {
					if (ch == Quote.Down /*&& chPrev != '\\'*/) {
						Quote.Count--;
						if (Quote.Count == 0) {
							Quote.Rep--;
							if (Quote.Up == Quote.Down) {
								Quote.Count++;
							}
						}
						if (!isalpha(chNext)) {
							if (Quote.Rep <= 0) {
								styler.ColourTo(i, state);
								state = SCE_PL_DEFAULT;
								ch = ' ';
							}
						}
					} else if (ch == Quote.Up /*&& chPrev != '\\'*/) {
						Quote.Count++;
					} else if (!isascii(chNext) || !isalpha(chNext)) {
						if (Quote.Rep <= 0) {
							styler.ColourTo(i, state);
							state = SCE_PL_DEFAULT;
							ch = ' ';
						}
					}
				}
			} else if (state == SCE_PL_REGSUBST) {
				if (!Quote.Up && !isspacechar(ch)) {
					Quote.Open(ch);
				} else if (ch == '\\' && Quote.Up != '\\') {
					// SG: Is it save to skip *every* escaped char?
					i++;
					ch = chNext;
					chNext = styler.SafeGetCharAt(i + 1);
				} else {
					if (Quote.Count == 0 && Quote.Rep == 1) {
						/* We matched something like s(...) or tr{...}
						* and are looking for the next matcher characters,
						* which could be either bracketed ({...}) or non-bracketed
						* (/.../).
						*
						* Number-signs are problematic.  If they occur after
						* the close of the first part, treat them like
						* a Quote.Up char, even if they actually start comments.
						*
						* If we find an alnum, we end the regsubst, and punt.
						*
						* Eric Promislow   [e-mail]  Aug 9,2000
						*/
						if (isspacechar(ch)) {
							// Keep going
						}
						else if (!isascii(ch) || isalnum(ch)) {
							styler.ColourTo(i, state);
							state = SCE_PL_DEFAULT;
							ch = ' ';
						} else {
							Quote.Open(ch);
						}
					} else if (ch == Quote.Down /*&& chPrev != '\\'*/) {
						Quote.Count--;
						if (Quote.Count == 0) {
							Quote.Rep--;
						}
						if (!isascii(chNext) || !isalpha(chNext)) {
							if (Quote.Rep <= 0) {
								styler.ColourTo(i, state);
								state = SCE_PL_DEFAULT;
								ch = ' ';
							}
						}
						if (Quote.Up == Quote.Down) {
							Quote.Count++;
						}
					} else if (ch == Quote.Up /*&& chPrev != '\\'*/) {
						Quote.Count++;
					} else if (!isascii(chNext) || !isalpha(chNext)) {
						if (Quote.Rep <= 0) {
							styler.ColourTo(i, state);
							state = SCE_PL_DEFAULT;
							ch = ' ';
						}
					}
				}
			} else if (state == SCE_PL_STRING_Q
				|| state == SCE_PL_STRING_QQ
				|| state == SCE_PL_STRING_QX
				|| state == SCE_PL_STRING_QW
				|| state == SCE_PL_STRING
				|| state == SCE_PL_CHARACTER
				|| state == SCE_PL_BACKTICKS
				) {
				if (!Quote.Down && !isspacechar(ch)) {
					Quote.Open(ch);
				} else if (ch == '\\' && Quote.Up != '\\') {
					i++;
					ch = chNext;
					chNext = styler.SafeGetCharAt(i + 1);
				} else if (ch == Quote.Down) {
					Quote.Count--;
					if (Quote.Count == 0) {
						Quote.Rep--;
						if (Quote.Rep <= 0) {
							styler.ColourTo(i, state);
							state = SCE_PL_DEFAULT;
							ch = ' ';
						}
						if (Quote.Up == Quote.Down) {
							Quote.Count++;
						}
					}
				} else if (ch == Quote.Up) {
					Quote.Count++;
				}
			}
		}
		if (state == SCE_PL_ERROR) {
			break;
		}
		chPrev = ch;
	}
	styler.ColourTo(lengthDoc - 1, state);
}

static bool IsCommentLine(int line, Accessor &styler) {
	int pos = styler.LineStart(line);
	int eol_pos = styler.LineStart(line + 1) - 1;
	for (int i = pos; i < eol_pos; i++) {
		char ch = styler[i];
        int style = styler.StyleAt(i);
		if (ch == '#' && style == SCE_PL_COMMENTLINE)
			return true;
		else if (ch != ' ' && ch != '\t')
			return false;
	}
	return false;
}

static void FoldPerlDoc(unsigned int startPos, int length, int, WordList *[],
                            Accessor &styler) {
	bool foldComment = styler.GetPropertyInt("fold.comment") != 0;
	bool foldCompact = styler.GetPropertyInt("fold.compact", 1) != 0;
	// Custom folding of POD and packages
	bool foldPOD = styler.GetPropertyInt("fold.perl.pod", 1) != 0;
	bool foldPackage = styler.GetPropertyInt("fold.perl.package", 1) != 0;
	unsigned int endPos = startPos + length;
	int visibleChars = 0;
	int lineCurrent = styler.GetLine(startPos);
	int levelPrev = SC_FOLDLEVELBASE;
	if (lineCurrent > 0)
		levelPrev = styler.LevelAt(lineCurrent - 1) >> 16;
	int levelCurrent = levelPrev;
	char chNext = styler[startPos];
	char chPrev = styler.SafeGetCharAt(startPos - 1);
	int styleNext = styler.StyleAt(startPos);
	// Used at end of line to determine if the line was a package definition
	bool isPackageLine = false;
	bool isPodHeading = false;
	for (unsigned int i = startPos; i < endPos; i++) {
		char ch = chNext;
		chNext = styler.SafeGetCharAt(i + 1);
		int style = styleNext;
		styleNext = styler.StyleAt(i + 1);
		bool atEOL = (ch == '\r' && chNext != '\n') || (ch == '\n');
		bool atLineStart = isEOLChar(chPrev) || i == 0;
        // Comment folding
		if (foldComment && atEOL && IsCommentLine(lineCurrent, styler))
        {
            if (!IsCommentLine(lineCurrent - 1, styler)
                && IsCommentLine(lineCurrent + 1, styler))
                levelCurrent++;
            else if (IsCommentLine(lineCurrent - 1, styler)
                     && !IsCommentLine(lineCurrent+1, styler))
                levelCurrent--;
        }
		if (style == SCE_C_OPERATOR) {
			if (ch == '{') {
				levelCurrent++;
			} else if (ch == '}') {
				levelCurrent--;
			}
		}
		// Custom POD folding
		if (foldPOD && atLineStart) {
			int stylePrevCh = (i) ? styler.StyleAt(i - 1):SCE_PL_DEFAULT;
			if (style == SCE_PL_POD) {
				if (stylePrevCh != SCE_PL_POD && stylePrevCh != SCE_PL_POD_VERB)
					levelCurrent++;
				else if (styler.Match(i, "=cut"))
					levelCurrent--;
				else if (styler.Match(i, "=head"))
					isPodHeading = true;
			} else if (style == SCE_PL_DATASECTION) {
                if (ch == '=' && isalpha(chNext) && levelCurrent == SC_FOLDLEVELBASE)
                    levelCurrent++;
                else if (styler.Match(i, "=cut") && levelCurrent > SC_FOLDLEVELBASE)
                    levelCurrent--;
                else if (styler.Match(i, "=head"))
					isPodHeading = true;
                // if package used or unclosed brace, level > SC_FOLDLEVELBASE!
                // reset needed as level test is vs. SC_FOLDLEVELBASE
                else if (styler.Match(i, "__END__"))
                    levelCurrent = SC_FOLDLEVELBASE;
            }
		}
		// Custom package folding
		if (foldPackage && atLineStart) {
			if (style == SCE_PL_WORD && styler.Match(i, "package")) {
				isPackageLine = true;
			}
		}

		if (atEOL) {
			int lev = levelPrev;
			if (isPodHeading) {
                lev = levelPrev - 1;
                lev |= SC_FOLDLEVELHEADERFLAG;
                isPodHeading = false;
			}
			// Check if line was a package declaration
			// because packages need "special" treatment
			if (isPackageLine) {
				lev = SC_FOLDLEVELBASE | SC_FOLDLEVELHEADERFLAG;
				levelCurrent = SC_FOLDLEVELBASE + 1;
				isPackageLine = false;
			}
            lev |= levelCurrent << 16;
			if (visibleChars == 0 && foldCompact)
				lev |= SC_FOLDLEVELWHITEFLAG;
			if ((levelCurrent > levelPrev) && (visibleChars > 0))
				lev |= SC_FOLDLEVELHEADERFLAG;
			if (lev != styler.LevelAt(lineCurrent)) {
				styler.SetLevel(lineCurrent, lev);
			}
			lineCurrent++;
			levelPrev = levelCurrent;
			visibleChars = 0;
		}
		if (!isspacechar(ch))
			visibleChars++;
		chPrev = ch;
	}
	// Fill in the real level of the next line, keeping the current flags as they will be filled in later
	int flagsNext = styler.LevelAt(lineCurrent) & ~SC_FOLDLEVELNUMBERMASK;
	styler.SetLevel(lineCurrent, levelPrev | flagsNext);
}

static const char * const perlWordListDesc[] = {
	"Keywords",
	0
};

LexerModule lmPerl(SCLEX_PERL, ColourisePerlDoc, "perl", FoldPerlDoc, perlWordListDesc);