/*
 *  Permission to use, copy, modify, distribute, and sell this software
 *  for any purpose and without fee, restriction or acknowledgement is
 *  hereby granted.  The author (James Knight of the Univ. of California,
 *  Davis) places it in the public domain.
 *
 *  This software is provided AS IS with no warranties of any kind.  The
 *  author shall have no liability with respect to the infringement of
 *  copyrights, trade secrets or any patents by this software or any part
 *  thereof.  In no event will the author be liable for any lost revenue
 *  or profits or other special, indirect and consequential damages.
 */

#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <ctype.h>
#include "seqio.h"


/*
 * This program extracts one of the subfields of the EMBL feature table
 * entries (of a certain type).  The way the program is configured right
 * now, the feature type and subfield type are fixed by constants, and
 * only subfields of the form
 *
 *     /subfield="..."
 *
 * can be extracted.  The program, however, does collapse subfields that
 * extend across multiple lines.  Since this is just a demo program, I
 * didn't try to extend the user interface.
 */

#define FORMAT   "EMBL"
#define FEATURE  "\nFT   CDS"
#define CONTLINE "\nFT   "
#define SUBFIELD "/note=\""

void extract_subfield(char *feature, SEQFILE *sfp, int num);


int main(int argc, char *argv[])
{
  int i, len, count;
  char *entry, *s, *feature;
  SEQFILE *sfp;

  for (i=1; i < argc; i++) {
    if ((sfp = seqfopen2(argv[i])) == NULL)
      continue;

    /*
     * Make sure the file is in the right format.
     */
    if (strcmp(seqfformat(sfp, 0), FORMAT) != 0) {
      fprintf(stderr, "%s:  Not a %s-format file.\n", seqffilename(sfp, 0),
              FORMAT);
      continue;
    }

    /*
     * Read the entries.
     */
    while ((entry = seqfgetentry(sfp, &len, 0)) != NULL) {
      count = 0;
      s = entry;
      while ((s = strstr(s, FEATURE)) != NULL) {
        /*
         * Find the end of the feature lines for that feature, and make
         * it NULL-terminated.
         */
        feature = ++s;
        while (*s != '\n') s++;
        while (strncmp(s, CONTLINE, 6) == 0 && isspace(s[6])) {
          s++;
          while (*s != '\n') s++;
        }
        *s = '\0';

        extract_subfield(feature, sfp, ++count);
        *s = '\n';
      }
    }

    seqfclose(sfp);
  }

  return 0;
}


void extract_subfield(char *feature, SEQFILE *sfp, int num)
{
  int contlinelen;
  char *s, *t, *id, *subfield, *fieldvalue;

  /*
   * Look for the subfield.
   */
  if ((subfield = strstr(feature, SUBFIELD)) == NULL)
    return;

  /*
   * If the subfield is there, find the strings between
   * the quotes, squeezing out any line breaks.
   */
  contlinelen = strlen(CONTLINE);

  fieldvalue = t = s = subfield + 7;
  while (*s && *s != '"') {
    if (*s == '\n') {
      s += contlinelen;        /* Skip the CONTLINE and then the spaces */
      while (*s != '\n' && isspace(*s))
        s++;
      *t++ = ' ';
    }
    else {
      if (t != s)
        *t = *s;
      t++;
      s++;
    }
  }
  *t = '\0';

  /*
   * Print 'em out.
   */
  id = seqfmainid(sfp, 0);
  printf("In %s - %s\n", (id == NULL ? "Unknown" : id), fieldvalue);
}