diff options
Diffstat (limited to 'libmscgen/mscgen_lexer.l')
| -rw-r--r-- | libmscgen/mscgen_lexer.l | 236 |
1 files changed, 236 insertions, 0 deletions
diff --git a/libmscgen/mscgen_lexer.l b/libmscgen/mscgen_lexer.l new file mode 100644 index 0000000..29d6aea --- /dev/null +++ b/libmscgen/mscgen_lexer.l @@ -0,0 +1,236 @@ +%{ +/*************************************************************************** + * + * $Id: lexer.l 184 2011-02-28 21:38:28Z Michael.McTernan $ + * + * Mscgen language lexer definition. + * Copyright (C) 2009 Michael C McTernan, Michael.McTernan.2001@cs.bris.ac.uk + * + * This file is part of msclib. + * + * Msc is free software; you can redistribute it and/or modify it + * under the terms of the GNU Lesser General Public License as published + * by the Free Software Foundation; either version 2.1 of the License, or + * (at your option) any later version. + * + * Msclib is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public License + * along with msclib; if not, write to the Free Software + * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + * + ***************************************************************************/ + +#include <stdio.h> +#include <string.h> +#include <ctype.h> +#include "mscgen_config.h" +#include "mscgen_msc.h" +#include "mscgen_bool.h" +#include "mscgen_safe.h" +#include "mscgen_lexer.h" +#include "mscgen_language.h" /* Token definitions from Yacc/Bison */ +/* Counter for error reporting */ +static unsigned long lex_linenum = 1; +static char *lex_line = NULL; +static Boolean lex_utf8 = FALSE; + +/* Local function prototypes */ +static void newline(const char *text, unsigned int n); +static char *trimQstring(char *s); + +%} + +/* Not used, so prevent compiler warning */ +%option never-interactive +%option noinput +%option noyywrap + +%x IN_COMMENT BODY +%% + +<INITIAL>{ +\xef\xbb\xbf lex_utf8 = TRUE; BEGIN(BODY); +(\r\n).* newline(yytext, 2); BEGIN(BODY); +(\r|\n).* newline(yytext, 1); BEGIN(BODY); +. unput(yytext[0]); BEGIN(BODY); +} + +<IN_COMMENT>{ +"*/" BEGIN(BODY); +[^*\n]+ +"*" +(\r\n).* newline(yytext, 2); +(\r|\n).* newline(yytext, 1); +} + +<BODY>{ + +"/*" BEGIN(IN_COMMENT); + +(\r\n).* newline(yytext, 2); +(\r|\n).* newline(yytext, 1); + +#.*$ /* Ignore lines after '#' */ +\/\/.*$ /* Ignore lines after '//' */ + +msc return TOK_MSC; +HSCALE|hscale yylval.optType = MSC_OPT_HSCALE; return TOK_OPT_HSCALE; +WIDTH|width yylval.optType = MSC_OPT_WIDTH; return TOK_OPT_WIDTH; +ARCGRADIENT|arcgradient yylval.optType = MSC_OPT_ARCGRADIENT; return TOK_OPT_ARCGRADIENT; +WORDWRAPARCS|wordwraparcs yylval.optType = MSC_OPT_WORDWRAPARCS; return TOK_OPT_WORDWRAPARCS; +URL|url yylval.attribType = MSC_ATTR_URL; return TOK_ATTR_URL; +LABEL|label yylval.attribType = MSC_ATTR_LABEL; return TOK_ATTR_LABEL; +IDURL|idurl yylval.attribType = MSC_ATTR_IDURL; return TOK_ATTR_IDURL; +ID|id yylval.attribType = MSC_ATTR_ID; return TOK_ATTR_ID; +LINECOLO(U?)R|linecolo(u?)r yylval.attribType = MSC_ATTR_LINE_COLOUR; return TOK_ATTR_LINE_COLOUR; +TEXTCOLO(U?)R|textcolo(u?)r yylval.attribType = MSC_ATTR_TEXT_COLOUR; return TOK_ATTR_TEXT_COLOUR; +TEXTBGCOLO(U?)R|textbgcolo(u?)r yylval.attribType = MSC_ATTR_TEXT_BGCOLOUR; return TOK_ATTR_TEXT_BGCOLOUR; +ARCLINECOLO(U?)R|arclinecolo(u?)r yylval.attribType = MSC_ATTR_ARC_LINE_COLOUR; return TOK_ATTR_ARC_LINE_COLOUR; +ARCTEXTCOLO(U?)R|arctextcolo(u?)r yylval.attribType = MSC_ATTR_ARC_TEXT_COLOUR; return TOK_ATTR_ARC_TEXT_COLOUR; +ARCTEXTBGCOLO(U?)R|arctextbgcolo(u?)r yylval.attribType = MSC_ATTR_ARC_TEXT_BGCOLOUR; return TOK_ATTR_ARC_TEXT_BGCOLOUR; +ARCSKIP|arcskip yylval.attribType = MSC_ATTR_ARC_SKIP; return TOK_ATTR_ARC_SKIP; +\.\.\. yylval.arctype = MSC_ARC_DISCO; return TOK_SPECIAL_ARC; /* ... */ +--- yylval.arctype = MSC_ARC_DIVIDER; return TOK_SPECIAL_ARC; /* --- */ +\|\|\| yylval.arctype = MSC_ARC_SPACE; return TOK_SPECIAL_ARC; /* ||| */ +\<-\> yylval.arctype = MSC_ARC_SIGNAL; return TOK_REL_SIG_BI; /* <-> */ +-\> yylval.arctype = MSC_ARC_SIGNAL; return TOK_REL_SIG_TO; /* -> */ +\<- yylval.arctype = MSC_ARC_SIGNAL; return TOK_REL_SIG_FROM; /* <- */ +-- yylval.arctype = MSC_ARC_SIGNAL; return TOK_REL_SIG; /* -- */ +-[Xx] yylval.arctype = MSC_ARC_LOSS; return TOK_REL_LOSS_TO; /* -x */ +[Xx]- yylval.arctype = MSC_ARC_LOSS; return TOK_REL_LOSS_FROM; /* x- */ +\<=\> yylval.arctype = MSC_ARC_METHOD; return TOK_REL_METHOD_BI; /* <=> */ +=\> yylval.arctype = MSC_ARC_METHOD; return TOK_REL_METHOD_TO; /* => */ +\<= yylval.arctype = MSC_ARC_METHOD; return TOK_REL_METHOD_FROM; /* <= */ +== yylval.arctype = MSC_ARC_METHOD; return TOK_REL_METHOD; /* == */ +\<\<\>\> yylval.arctype = MSC_ARC_RETVAL; return TOK_REL_RETVAL_BI; /* <<>> */ +\>\> yylval.arctype = MSC_ARC_RETVAL; return TOK_REL_RETVAL_TO; /* >> */ +\<\< yylval.arctype = MSC_ARC_RETVAL; return TOK_REL_RETVAL_FROM; /* << */ +\.\. yylval.arctype = MSC_ARC_RETVAL; return TOK_REL_RETVAL; /* .. */ +\<:\> yylval.arctype = MSC_ARC_DOUBLE; return TOK_REL_DOUBLE_BI; /* <:> */ +:\> yylval.arctype = MSC_ARC_DOUBLE; return TOK_REL_DOUBLE_TO; /* :> */ +\<: yylval.arctype = MSC_ARC_DOUBLE; return TOK_REL_DOUBLE_FROM; /* <: */ +:: yylval.arctype = MSC_ARC_DOUBLE; return TOK_REL_DOUBLE; /* :: */ +\<\<=\>\> yylval.arctype = MSC_ARC_CALLBACK; return TOK_REL_CALLBACK_BI; /* <<=>> */ +=\>\> yylval.arctype = MSC_ARC_CALLBACK; return TOK_REL_CALLBACK_TO; /* =>> */ +\<\<= yylval.arctype = MSC_ARC_CALLBACK; return TOK_REL_CALLBACK_FROM; /* <<= */ +BOX|box yylval.arctype = MSC_ARC_BOX; return TOK_REL_BOX; /* box */ +ABOX|abox yylval.arctype = MSC_ARC_ABOX; return TOK_REL_ABOX; /* abox */ +RBOX|rbox yylval.arctype = MSC_ARC_RBOX; return TOK_REL_RBOX; /* rbox */ +NOTE|note yylval.arctype = MSC_ARC_NOTE; return TOK_REL_NOTE; /* note */ +[A-Za-z0-9_]+ yylval.string = strdup_s(yytext); return TOK_STRING; +\"(\\\"|[^\"])*\" yylval.string = trimQstring(strdup_s(yytext)); return TOK_QSTRING; += return TOK_EQUAL; +, return TOK_COMMA; +\; return TOK_SEMICOLON; +\{ return TOK_OCBRACKET; +\} return TOK_CCBRACKET; +\[ return TOK_OSBRACKET; +\] return TOK_CSBRACKET; +\* return TOK_ASTERISK; +[ \t]+ /* ignore whitespace */; + +} + + +<*>.|\n|\r return TOK_UNKNOWN; + +%% + +/* Handle a new line of input. + * This counts the line number and duplicates the string incase we need + * it for error reporting. The line is then returned back for parsing + * without the newline characters prefixed. + */ +static void newline(const char *text, unsigned int n) +{ + lex_linenum++; + if(lex_line != NULL) + { + free(lex_line); + } + + lex_line = strdup(text + n); + yyless(n); +} + + +/* Trim a multi-line quoted string. + * This allows the parsed input quoted strings to span multiple lines of + * input but be condensed to only a single line of output e.g. + * a->b [label="line 1 + * line 1 too"]; + * Will parse to a string such as"line1\n line1 too". This function + * will collapse the \n and whitespace into a single space. + */ +static char *trimQstring(char *const s) +{ + int i = 0, o = 0, skipmode = 0; + + /* Strip leading " */ + if(s[i] == '\"') + { + i++; + } + + /* Copy body, compacting whitespace after newline sequences */ + while(s[i] != '\0') + { + if(s[i] == '\r' || s[i] == '\n' || s[i] == '\f') + { + skipmode = 1; + } + else if(!skipmode || !isspace(s[i])) + { + if(skipmode) + { + s[o] = ' '; + o++; + } + + skipmode = 0; + s[o] = s[i]; + o++; + } + + i++; + } + + /* Null terminate */ + s[o] = '\0'; + + /* Remove trailing " */ + if(o >= 1 && s[o - 1] == '\"') + s[o-1] = '\0'; + + return s; +} + +unsigned long lex_getlinenum(void) +{ + return lex_linenum; +} + +char *lex_getline(void) +{ + return lex_line; +} + +void lex_destroy(void) +{ + if(lex_line != NULL) + { + free(lex_line); + lex_line = NULL; + } +} + +Boolean lex_getutf8(void) +{ + return lex_utf8; +} + +/* END OF FILE */ |