/**
* This function takes two offset arrays:
*
* old_offsets=original text => input text
* new_offsets=input text => output text
*
* and it computes and prints in the given file the new offset file:
*
* original text => output text
*
* If old_offsets is NULL, new offsets are saved in the output file
* without any modification.
*/
void process_offsets(const vector_offset* first_offsets, const vector_offset* second_offsets, U_FILE* f) {
if (f == NULL || second_offsets == NULL) return;
if (first_offsets == NULL) {
save_offsets(f, second_offsets);
}
else {
vector_offset* processed_offset = process_offsets(first_offsets, second_offsets);
save_offsets(f, processed_offset);
free_vector_offset(processed_offset);
}
}
int main_Normalize(int argc,char* const argv[]) {
if (argc==1) {
usage();
return SUCCESS_RETURN_CODE;
}
int mode=KEEP_CARRIAGE_RETURN;
int separator_normalization=1;
char rules[FILENAME_MAX]="";
char input_offsets[FILENAME_MAX]="";
char output_offsets[FILENAME_MAX]="";
VersatileEncodingConfig vec=VEC_DEFAULT;
int convLFtoCRLF=1;
int val,index=-1;
bool only_verify_arguments = false;
UnitexGetOpt options;
while (EOF!=(val=options.parse_long(argc,argv,optstring_Normalize,lopts_Normalize,&index))) {
switch(val) {
case 'l': convLFtoCRLF=0; break;
case 'n': mode=REMOVE_CARRIAGE_RETURN; break;
case 'r': if (options.vars()->optarg[0]=='\0') {
error("You must specify a non empty replacement rule file name\n");
return USAGE_ERROR_CODE;
}
strcpy(rules,options.vars()->optarg);
break;
case 1: separator_normalization=0; break;
case 'k': if (options.vars()->optarg[0]=='\0') {
error("Empty input_encoding argument\n");
return USAGE_ERROR_CODE;
}
decode_reading_encoding_parameter(&(vec.mask_encoding_compatibility_input),options.vars()->optarg);
break;
case 'q': if (options.vars()->optarg[0]=='\0') {
error("Empty output_encoding argument\n");
return USAGE_ERROR_CODE;
}
decode_writing_encoding_parameter(&(vec.encoding_output),&(vec.bom_output),options.vars()->optarg);
break;
case '$': if (options.vars()->optarg[0]=='\0') {
error("You must specify a non empty input offset file name\n");
return USAGE_ERROR_CODE;
}
strcpy(input_offsets,options.vars()->optarg);
break;
case '@': if (options.vars()->optarg[0]=='\0') {
error("You must specify a non empty output offset file name\n");
return USAGE_ERROR_CODE;
}
strcpy(output_offsets,options.vars()->optarg);
break;
case 'V': only_verify_arguments = true;
break;
case 'h': usage();
return SUCCESS_RETURN_CODE;
case ':': index==-1 ? error("Missing argument for option -%c\n",options.vars()->optopt) :
error("Missing argument for option --%s\n",lopts_Normalize[index].name);
return USAGE_ERROR_CODE;
break;
case '?': index==-1 ? error("Invalid option -%c\n",options.vars()->optopt) :
error("Invalid option --%s\n",options.vars()->optarg);
return USAGE_ERROR_CODE;
break;
}
index=-1;
}
if (options.vars()->optind!=argc-1) {
error("Invalid arguments: rerun with --help\n");
return USAGE_ERROR_CODE;
}
if (only_verify_arguments) {
// freeing all allocated memory
return SUCCESS_RETURN_CODE;
}
vector_offset* v_input_offsets=NULL;
vector_offset* v_output_offsets=NULL;
U_FILE* f_output_offsets=NULL;
if (output_offsets[0]!='\0') {
/* We deal with offsets only if we have to produce output offsets */
if (input_offsets[0]!='\0') {
v_input_offsets=load_offsets(&vec,input_offsets);
}
f_output_offsets=u_fopen(&vec, output_offsets, U_WRITE);
if (f_output_offsets==NULL) {
error("Cannot create offset file %s\n",output_offsets);
return DEFAULT_ERROR_CODE;
}
v_output_offsets=new_vector_offset();
}
char tmp_file[FILENAME_MAX];
get_extension(argv[options.vars()->optind],tmp_file);
if (!strcmp(tmp_file, ".snt")) {
/* If the file to process has already the .snt extension, we temporary rename it to
* .snt.normalizing */
strcpy(tmp_file,argv[options.vars()->optind]);
strcat(tmp_file,".normalizing");
af_rename(argv[options.vars()->optind],tmp_file);
} else {
strcpy(tmp_file,argv[options.vars()->optind]);
}
/* We set the destination file */
char dest_file[FILENAME_MAX];
remove_extension(argv[options.vars()->optind],dest_file);
strcat(dest_file,".snt");
u_printf("Normalizing %s...\n",argv[options.vars()->optind]);
int return_value = normalize(tmp_file,
dest_file,
&vec,
mode,
convLFtoCRLF,
rules,
v_output_offsets,
separator_normalization);
u_printf("\n");
/* If we have used a temporary file, we delete it */
if (strcmp(tmp_file,argv[options.vars()->optind])) {
af_remove(tmp_file);
}
process_offsets(v_input_offsets,v_output_offsets,f_output_offsets);
u_fclose(f_output_offsets);
free_vector_offset(v_input_offsets);
free_vector_offset(v_output_offsets);
u_printf((return_value==SUCCESS_RETURN_CODE) ? "Done.\n" : "Unsuccessfull.\n");
return return_value;
}
