LinuxQuestions.org - parse bibtex with flex+bison

- Programming (https://www.linuxquestions.org/questions/programming-9/)

- - parse bibtex with flex+bison (https://www.linuxquestions.org/questions/programming-9/parse-bibtex-with-flex-bison-4175451265/)

parse bibtex with flex+bison

I want to parse bibtex file using flex/bison. A sample bibtex is:

Code:

@Book{a1,

author="amook",

Title="ASR",

Publisher="oxf",

Year="2010",

Add="UK",

Edition="1",

}

@Article{a2,

Author="Rudra Banerjee",

Title={FeNiMo},

Publisher={P{\"R}B},

Issue="12",

Page="36690",

Year="2011",

Add="UK",

Edition="1",

}

Now, I have written a flex code:

Code:

%{

#include <stdio.h>

#include <stdlib.h>

%}



%{

char yylval;

int YEAR,i;

//char array_author[1000];

%}

%x author

%x title

%x pub

%x year

%%

@                                  printf("\nNEWENTRY\n");

[a-zA-Z][a-zA-Z0-9]*                  {printf("%s",yytext);

                                        BEGIN(INITIAL);}

author=                        {BEGIN(author);}

<author>\"[a-zA-Z\/.]+\"          {printf("%s",yytext);

                                          BEGIN(INITIAL);}

title=                                {BEGIN(title);}

<title>\"[a-zA-Z\/.]+\"          {printf("%s",yytext);

                                          BEGIN(INITIAL);}

publisher=                        {BEGIN(pub);}

<pub>\"[a-zA-Z\/.]+\"                  {printf("%s",yytext);

                                          BEGIN(INITIAL);}

[a-zA-Z0-9\/.-]+=        printf("ENTRY TYPE ");

\"                      printf("QUOTE ");

\{                      printf("LCB ");

\}                      printf(" RCB");

;                      printf("SEMICOLON ");

\n                      printf("\n");

%%



int main(){

  yylex();

//char array_author[1000];

//printf("%d%s",&i,array_author[i]);

i++;

return 0;

}

while this is compiled and running, its not peeking up the fields. I would like to store the keys and corresponding vals to be used in a C code.
Some help please.

I'm rooting for you.

Edit later: Good, you have added code since my first reply.

Later: you forgot the Makefile, so here is one for test:

Code:

CFLAGS  += -m32 -std=c99 -g -W -Wall -Wextra

CPPFLAGS+= -D_GNU_SOURCE -D_XOPEN_SOURCE=500 -D_ALL_SOURCE

LDFLAGS += -m32 -g



clean:

        rm -f bibtex bibtex.o bibtex.c 2>/dev/null || true



bibtex.c: bibtex.l

        flex -o $@ $<



bibtex: LIBS += -lfl

bibtex: bibtex.o

        gcc -o $@ ${LDFLAGS} $^ ${LIBS}

It does something, I couldn't tell if it is the right thing or not:

Code:

$ ./bibtex <bibtex.in



NEWENTRY

BookLCB a1,

"amook",

ENTRY TYPE QUOTE ASRQUOTE ,

ENTRY TYPE QUOTE oxfQUOTE ,

ENTRY TYPE QUOTE 2010QUOTE ,

ENTRY TYPE QUOTE UKQUOTE ,

ENTRY TYPE QUOTE 1QUOTE ,

 RCB



NEWENTRY

ArticleLCB a2,

ENTRY TYPE QUOTE Rudra BanerjeeQUOTE ,

ENTRY TYPE LCB FeNiMo RCB,

ENTRY TYPE LCB PLCB \QUOTE R RCBB RCB,

ENTRY TYPE QUOTE 12QUOTE ,

ENTRY TYPE QUOTE 36690QUOTE ,

ENTRY TYPE QUOTE 2011QUOTE ,

ENTRY TYPE QUOTE UKQUOTE ,

ENTRY TYPE QUOTE 1QUOTE ,

 RCB

Quote:

Originally Posted by NevemTeve (Post 4897422)

It does something, I couldn't tell if it is the right thing or not:

Thanks for the makefile, mine was almost same, with flex is case-insensetive(flex -i).
With flex -i, one will get the output as

Quote:

ArticleLCB a2,
"Rudra Banerjee",
Title={FeNiMo},
Publisher={P{\"R}B},
Issue="12",
Page="36690",
Year="2011",
Add="UK",
ENTRY TYPE QUOTE 1QUOTE ,
RCB

What I want to do is to have those keys and vals stored in some way to be used in a C program.

i.e., may be something like which are commented in C block of the code.
Any further help please?

I'd start with something simpler, eg:

sample.in:

Code:

author="amook"

title="ASR"

author="Ian M. Banks"

title="The Player of Games"

sample.l:

Code:

%{

#include <stdio.h>

#include <stdlib.h>



#define ECHO    /* don't */



#define AUTHOR 256

#define TITLE  257



char yylval;



%}

%option noyywrap



%x author

%x title

%%

author=\"[^\"]*\"      { return AUTHOR; };

title=\"[^\"]*\"        { return TITLE; };

<<EOF>>                  return -1;

%%



typedef struct REC {

    char *f_author;

    char *f_title;

} REC;



static char *DupStr (const char *from)

{

    const char *p;

    size_t len;

    char *q;



    p= strchr (from, '=');

    if (p) from= p+1;

    len= strlen (from);

    if (len>=2 && from[0]=='"' && from[len-1]=='"') {

        ++from;

        len -= 2;

    }

    q= malloc (len+1);

    memcpy (q, from, len);

    q[len]= '\0';

    return q;

}



int main ()

{

    int rc;

    int leave= 0;

    REC rec= {NULL, NULL};



    while (!leave) {

        rc= yylex();



/*      printf ("yylex returned %d (%s)\n", rc, yytext);        */

        if (rc==EOF) {

            leave= 1; 



        } else if (rc==AUTHOR) {

            rec.f_author= DupStr (yytext);



        } else if (rc==TITLE) {

            rec.f_title= DupStr (yytext);



            printf ("Record found: %s: %s\n", rec.f_author, rec.f_title);

            free (rec.f_author); rec.f_author= NULL;

            free (rec.f_title);  rec.f_title= NULL;

        }

    }

    return 0;

}

output:

Code:

Record found: amook: ASR

Record found: Ian M. Banks: The Player of Games