bool gth_intermediate_output_is_correct(char *outputfilename,
GthSACollection *orig_sa_collection,
GthInput *input,
GtFile **outfp, GtError *err)
{
SACollectionData sa_collection_data;
GthSACollection *read_sa_collection;
GtFileMode file_mode;
bool rval;
#ifndef NDEBUG
GtUword numofgenomicfiles, numofreferencefiles;
#endif
gt_error_check(err);
gt_assert(outputfilename);
gt_assert(*outfp);
#ifndef NDEBUG
numofgenomicfiles = gth_input_num_of_gen_files(input);
numofreferencefiles = gth_input_num_of_ref_files(input);
#endif
/* init */
read_sa_collection = gth_sa_collection_new(GTH_DC_NONE);
sa_collection_data.sa_collection = read_sa_collection;
sa_collection_data.sa_filter = NULL;
sa_collection_data.stat = NULL;
/* store file mode */
file_mode = gt_file_mode(*outfp);
/* close output file */
gt_file_delete(*outfp);
/* open intermediate file again for reading */
*outfp = gt_file_xopen_file_mode(file_mode, outputfilename, "r");
gt_assert(*outfp);
/* read in the intermediate output */
if (gt_parse_intermediate_output(input, store_in_sa_collection,
&sa_collection_data, outputfilename, *outfp,
err)) {
fprintf(stderr, "error: %s\n", gt_error_get(err));
exit(EXIT_FAILURE);
}
/* array of genomic files did not grow */
gt_assert(numofgenomicfiles == gth_input_num_of_gen_files(input));
/* array of reference files did not grow */
gt_assert(numofreferencefiles == gth_input_num_of_ref_files(input));
/* compare the trees */
rval = gth_sa_collections_are_equal(orig_sa_collection, read_sa_collection);
/* free */
gth_sa_collection_delete(read_sa_collection);
return rval;
}
static int split_fasta_file(const char *filename, unsigned long max_filesize,
bool force, GtError *err)
{
GtFile *srcfp = NULL, *destfp = NULL;
GtStr *destfilename = NULL;
unsigned long filenum = 0, bytecount = 0, separator_pos;
int read_bytes, had_err = 0;
char buf[BUFSIZ];
gt_error_check(err);
gt_assert(filename && max_filesize);
/* open source file */
srcfp = gt_file_xopen(filename, "r");
gt_assert(srcfp);
/* read start characters */
if ((read_bytes = gt_file_xread(srcfp, buf, BUFSIZ)) == 0) {
gt_error_set(err, "file \"%s\" is empty", filename);
had_err = -1;
}
bytecount += read_bytes;
/* make sure the file is in fasta format */
if (!had_err && buf[0] != '>') {
gt_error_set(err, "file is not in FASTA format");
had_err = -1;
}
if (!had_err) {
/* open destination file */
destfilename = gt_str_new();
gt_str_append_cstr_nt(destfilename, filename,
gt_file_basename_length(filename));
gt_str_append_char(destfilename, '.');
gt_str_append_ulong(destfilename, ++filenum);
gt_str_append_cstr(destfilename,
gt_file_mode_suffix(gt_file_mode(srcfp)));
if (!(destfp = gt_outputfile_xopen_forcecheck(gt_str_get(destfilename), "w",
force, err))) {
had_err = -1;
}
if (!had_err)
gt_file_xwrite(destfp, buf, read_bytes);
while (!had_err &&
(read_bytes = gt_file_xread(srcfp, buf, BUFSIZ)) != 0) {
if (bytecount + read_bytes > max_filesize) {
int offset = bytecount < max_filesize ? max_filesize - bytecount : 0;
if ((separator_pos = buf_contains_separator(buf, offset, read_bytes))) {
separator_pos--;
gt_assert(separator_pos < read_bytes);
if (separator_pos)
gt_file_xwrite(destfp, buf, separator_pos);
/* close current file */
gt_file_delete(destfp);
/* open new file */
gt_str_reset(destfilename);
gt_str_append_cstr_nt(destfilename, filename,
gt_file_basename_length(filename));
gt_str_append_char(destfilename, '.');
gt_str_append_ulong(destfilename, ++filenum);
gt_str_append_cstr(destfilename,
gt_file_mode_suffix(gt_file_mode(srcfp)));
if (!(destfp =
gt_outputfile_xopen_forcecheck(gt_str_get(destfilename), "w",
force, err))) {
had_err = -1;
break;
}
bytecount = read_bytes - separator_pos; /* reset */
gt_assert(buf[separator_pos] == '>');
gt_file_xwrite(destfp, buf + separator_pos,
read_bytes - separator_pos);
continue;
}
}
bytecount += read_bytes;
gt_file_xwrite(destfp, buf, read_bytes);
}
}
/* free */
gt_str_delete(destfilename);
/* close current file */
gt_file_delete(destfp);
/* close source file */
gt_file_delete(srcfp);
return had_err;
}
