From f54da734eb7b69e8e34de505bd89a13479e230e0 Mon Sep 17 00:00:00 2001 From: Richard Purdie Date: Wed, 31 Aug 2005 10:47:56 +0000 Subject: Initial population git-svn-id: https://svn.o-hand.com/repos/poky@2 311d38ba-8fff-0310-9ca6-ca027cbcb966 --- bitbake/lib/bb/parse/parse_c/bitbakeparser.l | 288 +++++++++++++++++++++++++++ 1 file changed, 288 insertions(+) create mode 100644 bitbake/lib/bb/parse/parse_c/bitbakeparser.l (limited to 'bitbake/lib/bb/parse/parse_c/bitbakeparser.l') diff --git a/bitbake/lib/bb/parse/parse_c/bitbakeparser.l b/bitbake/lib/bb/parse/parse_c/bitbakeparser.l new file mode 100644 index 000000000..ee4ce1483 --- /dev/null +++ b/bitbake/lib/bb/parse/parse_c/bitbakeparser.l @@ -0,0 +1,288 @@ +/* bbf.flex + + written by Marc Singer + 6 January 2005 + + This program is free software; you can redistribute it and/or + modify it under the terms of the GNU General Public License as + published by the Free Software Foundation; either version 2 of the + License, or (at your option) any later version. + + This program 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 + General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software + Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 + USA. + + DESCRIPTION + ----------- + + flex lexer specification for a BitBake input file parser. + + Unfortunately, flex doesn't welcome comments within the rule sets. + I say unfortunately because this lexer is unreasonably complex and + comments would make the code much easier to comprehend. + + The BitBake grammar is not regular. In order to interpret all + of the available input files, the lexer maintains much state as it + parses. There are places where this lexer will emit tokens that + are invalid. The parser will tend to catch these. + + The lexer requires C++ at the moment. The only reason for this has + to do with a very small amount of managed state. Producing a C + lexer should be a reasonably easy task as long as the %reentrant + option is used. + + + NOTES + ----- + + o RVALUES. There are three kinds of RVALUES. There are unquoted + values, double quote enclosed strings, and single quote + strings. Quoted strings may contain unescaped quotes (of either + type), *and* any type may span more than one line by using a + continuation '\' at the end of the line. This requires us to + recognize all types of values with a single expression. + Moreover, the only reason to quote a value is to include + trailing or leading whitespace. Whitespace within a value is + preserved, ugh. + + o CLASSES. C_ patterns define classes. Classes ought not include + a repitition operator, instead letting the reference to the class + define the repitition count. + + C_SS - symbol start + C_SB - symbol body + C_SP - whitespace + +*/ + +%option never-interactive +%option yylineno +%option noyywrap +%option reentrant stack + + +%{ + +#include "token.h" +#include "lexer.h" +#include + +extern void *bbparseAlloc(void *(*mallocProc)(size_t)); +extern void bbparseFree(void *p, void (*freeProc)(void*)); +extern void *bbparseAlloc(void *(*mallocProc)(size_t)); +extern void *bbparse(void*, int, token_t, lex_t*); +extern void bbparseTrace(FILE *TraceFILE, char *zTracePrompt); + +//static const char* rgbInput; +//static size_t cbInput; + + +int lineError; +int errorParse; + +enum { + errorNone = 0, + errorUnexpectedInput, + errorUnsupportedFeature, +}; + +#define YY_EXTRA_TYPE lex_t* + + /* Read from buffer */ +#define YY_INPUT(buf,result,max_size) \ + { yyextra->input(buf, &result, max_size); } + +//#define YY_DECL static size_t yylex () + +#define ERROR(e) \ + do { lineError = yylineno; errorParse = e; yyterminate (); } while (0) + +static const char* fixup_escapes (const char* sz); + +%} + + +C_SP [ \t] +COMMENT #.*\n +OP_ASSIGN "=" +OP_IMMEDIATE ":=" +OP_PREPEND "=+" +OP_APPEND "+=" +OP_COND "?=" +B_OPEN "{" +B_CLOSE "}" + +K_ADDTASK "addtask" +K_ADDHANDLER "addhandler" +K_AFTER "after" +K_BEFORE "before" +K_DEF "def" +K_INCLUDE "include" +K_INHERIT "inherit" +K_PYTHON "python" +K_FAKEROOT "fakeroot" +K_EXPORT "export" +K_EXPORT_FUNC "EXPORT_FUNCTIONS" + +STRING \"([^\n\r]|"\\\n")*\" +SSTRING \'([^\n\r]|"\\\n")*\' +VALUE ([^'" \t\n])|([^'" \t\n]([^\n]|(\\\n))*[^'" \t\n]) + +C_SS [a-zA-Z_] +C_SB [a-zA-Z0-9_+-.] +REF $\{{C_SS}{C_SB}*\} +SYMBOL {C_SS}{C_SB}* +VARIABLE $?{C_SS}({C_SB}*|{REF})*(\[[a-zA-Z0-9_]*\])? +FILENAME ([a-zA-Z_./]|{REF})(([-+a-zA-Z0-9_./]*)|{REF})* + +PROC \({C_SP}*\) + +%s S_DEF +%s S_DEF_ARGS +%s S_DEF_BODY +%s S_FUNC +%s S_INCLUDE +%s S_INHERIT +%s S_PROC +%s S_RVALUE +%s S_TASK + +%% + +{OP_APPEND} { BEGIN S_RVALUE; + yyextra->accept (T_OP_APPEND); } +{OP_PREPEND} { BEGIN S_RVALUE; + yyextra->accept (T_OP_PREPEND); } +{OP_IMMEDIATE} { BEGIN S_RVALUE; + yyextra->accept (T_OP_IMMEDIATE); } +{OP_ASSIGN} { BEGIN S_RVALUE; + yyextra->accept (T_OP_ASSIGN); } +{OP_COND} { BEGIN S_RVALUE; + yyextra->accept (T_OP_COND); } + +\\\n{C_SP}* { } +{STRING} { BEGIN INITIAL; + size_t cb = yyleng; + while (cb && isspace (yytext[cb - 1])) + --cb; + yytext[cb - 1] = 0; + yyextra->accept (T_STRING, yytext + 1); } +{SSTRING} { BEGIN INITIAL; + size_t cb = yyleng; + while (cb && isspace (yytext[cb - 1])) + --cb; + yytext[cb - 1] = 0; + yyextra->accept (T_STRING, yytext + 1); } + +{VALUE} { ERROR (errorUnexpectedInput); } +{C_SP}*\n+ { BEGIN INITIAL; + yyextra->accept (T_STRING, NULL); } + +{K_INCLUDE} { BEGIN S_INCLUDE; + yyextra->accept (T_INCLUDE); } +{K_INHERIT} { BEGIN S_INHERIT; + yyextra->accept (T_INHERIT); } +{K_ADDTASK} { BEGIN S_TASK; + yyextra->accept (T_ADDTASK); } +{K_ADDHANDLER} { yyextra->accept (T_ADDHANDLER); } +{K_EXPORT_FUNC} { BEGIN S_FUNC; + yyextra->accept (T_EXPORT_FUNC); } +{K_BEFORE} { yyextra->accept (T_BEFORE); } +{K_AFTER} { yyextra->accept (T_AFTER); } +{K_EXPORT} { yyextra->accept (T_EXPORT); } + +{K_FAKEROOT} { yyextra->accept (T_FAKEROOT); } +{K_PYTHON} { yyextra->accept (T_PYTHON); } +{PROC}{C_SP}*{B_OPEN}{C_SP}*\n* { BEGIN S_PROC; + yyextra->accept (T_PROC_OPEN); } +{B_CLOSE}{C_SP}*\n* { BEGIN INITIAL; + yyextra->accept (T_PROC_CLOSE); } +([^}][^\n]*)?\n* { yyextra->accept (T_PROC_BODY, yytext); } + +{K_DEF} { BEGIN S_DEF; } +{SYMBOL} { BEGIN S_DEF_ARGS; + yyextra->accept (T_SYMBOL, yytext); } +[^\n:]*: { yyextra->accept (T_DEF_ARGS, yytext); } +{C_SP}*\n { BEGIN S_DEF_BODY; } +{C_SP}+[^\n]*\n { yyextra->accept (T_DEF_BODY, yytext); } +\n { yyextra->accept (T_DEF_BODY, yytext); } +. { BEGIN INITIAL; unput (yytext[0]); } + +{COMMENT} { } + +{SYMBOL} { yyextra->accept (T_SYMBOL, yytext); } +{VARIABLE} { yyextra->accept (T_VARIABLE, yytext); } + +{SYMBOL} { yyextra->accept (T_TSYMBOL, yytext); } +{SYMBOL} { yyextra->accept (T_FSYMBOL, yytext); } +{SYMBOL} { yyextra->accept (T_ISYMBOL, yytext); } +{FILENAME} { BEGIN INITIAL; + yyextra->accept (T_ISYMBOL, yytext); } + +\n { BEGIN INITIAL; } +\n { BEGIN INITIAL; } +\n { BEGIN INITIAL; } + +[ \t\r\n] /* Insignificant whitespace */ + +. { ERROR (errorUnexpectedInput); } + + /* Check for premature termination */ +<> { return T_EOF; } + +%% + +void lex_t::accept (int token, const char* sz) +{ + token_t t; + memset (&t, 0, sizeof (t)); + t.copyString(sz); + + /* tell lemon to parse the token */ + parse (parser, token, t, this); +} + +int lex_t::line ()const +{ + return yyget_lineno (scanner); +} + +const char* lex_t::filename ()const +{ + return m_fileName; +} + +void parse (MappedFile* mf) +{ + void* parser = bbparseAlloc (malloc); + yyscan_t scanner; + lex_t lex; + + yylex_init (&scanner); + + lex.parser = parser; + lex.scanner = scanner; + lex.mf = mf; + lex.rgbInput = mf->m_rgb; + lex.cbInput = mf->m_cb; + lex.parse = bbparse; + yyset_extra (&lex, scanner); + + + int result = yylex (scanner); + + lex.accept (0); + bbparseTrace (NULL, NULL); + + if (result != T_EOF) + WARNING ("premature end of file\n"); + + yylex_destroy (scanner); + bbparseFree (parser, free); +} -- cgit v1.2.3