IUBio GIL .. BIOSCI/Bionet News .. Biosequences .. Software .. FTP

"name read"

Detlef Wolf D.Wolf at dkfz-heidelberg.de
Tue Dec 13 14:48:29 EST 1994


a suggestion from a discussion in one of the IGD courses:

if one ACEDB is shared by (several) people not having write access
to the database they cannot store a keyset (the result of possibly
longer work) and continue on it in the next session.

suggestion: in the KeySet menu, analgous to "name dump"
there should be a function "name read".
It reads the contents of the .ace file ~/myKeySet
and makes it the active KeySet. It skips objects that do
not exist in the database.
The file ~/myKeySet can be generated by the "name dump" function.

What do you think about that?
Detlef

----------------------------------------------
Appendix:
this code works, but is not nice.
it should use ACEDB parsing routines and
could work on lexiques directly to be faster.

The KEYSET readKeySet(FILE* ksf) {
  KEYSET ks=0,kstmp; 
  char classname[128];
  char objname[256];
  char queryt[384];
  int i;
  
  kstmp=keySetCreate();
  classname[0]='\0';
  
  /* ignore first line */
  while ((i=fscanf(ksf,"%s : %s",classname,objname))!=2 && i!=EOF);
 
  if (strcmp(classname,"KeySet")) {
    printf("expected class KeySet, but found %s. Abort reading\n",classname);
    return(0);
  };
  
  while (fscanf(ksf,"%s : %s",classname,objname)==2) {
    sprintf(queryt,"FIND %s %s",classname,objname);
    kstmp=query(0,queryt);
    ks=keySetOR(kstmp,ks);
  };
  
  keySetDestroy(kstmp);
  return(ks);
}




More information about the Acedb mailing list

Send comments to us at archive@iubioarchive.bio.net