X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=src%2Ffastq-grep.c;h=4d1603f7ced45f3d34e28d00e60204689031d4a8;hb=d6b81119e55cf77c1e4c17b9e9026abc01e22b96;hp=ca072ebc05fd234cab16e1190aa9a2ac1c9e7716;hpb=9537fe433ae5b59162a2f0a0834398c9fb4b7907;p=fastq-tools.git diff --git a/src/fastq-grep.c b/src/fastq-grep.c index ca072eb..4d1603f 100644 --- a/src/fastq-grep.c +++ b/src/fastq-grep.c @@ -1,22 +1,23 @@ - -/* - * fastq-grep: regular expression searches of the sequences within a fastq file +/* + * This file is part of fastq-tools. + * + * Copyright (c) 2011 by Daniel C. Jones * - * Febuary 2011 / Daniel Jones + * fastq-grep : + * Regular expression searches of the sequences within a FASTQ file. * */ +#include "common.h" +#include "parse.h" #include #include #include #include #include -#include "kseq.h" -KSEQ_INIT(gzFile, gzread) - #if defined(MSDOS) || defined(OS2) || defined(WIN32) || defined(__CYGWIN__) # include @@ -27,119 +28,149 @@ KSEQ_INIT(gzFile, gzread) #endif +static const char* prog_name = "fastq-grep"; + + void print_help() { - fprintf( stderr, + fprintf(stdout, "fastq-grep [OPTION]... PATTERN [FILE]...\n" "Search for PATTERN in the read sequences in each FILE or standard input.\n" "PATTERN, by default, is a perl compatible regular expression.\n\n" "Options:\n" -" -h, --help print this message\n" +" -i, --id match the read id (by default, sequence is matched)\n" " -v, --invert-match select nonmatching entries\n" +" -m, --mismatches=FILE output mismatching entries to the given file\n" +" -c, --count output only the number of matching sequences\n" +" -h, --help print this message\n" +" -V, --version output version information and exit\n" ); } static int invert_flag; -static int help_flag; -static int zout_flag; +static int count_flag; +static int id_flag; -void print_fastq_entry( FILE* fout, kseq_t* seq ) -{ - fprintf(fout, "@%s\n%s\n+%s\n%s\n", - seq->name.s, - seq->seq.s, - seq->comment.s, - seq->qual.s ); -} - - -void fastq_grep( gzFile fin, FILE* fout, pcre* re ) +void fastq_grep(FILE* fin, FILE* fout, FILE* mismatch_file, pcre* re) { int rc; int ovector[3]; + size_t count = 0; - kseq_t* seq; - seq = kseq_init(fin); + fastq_t* fqf = fastq_create(fin); + seq_t* seq = seq_create(); + + while (fastq_read(fqf, seq)) { - while (kseq_read(seq) >= 0) { rc = pcre_exec(re, /* pattern */ - NULL, /* extre data */ - seq->seq.s, /* subject */ - seq->seq.l, /* subject length */ + NULL, /* extra data */ + id_flag ? seq->id1.s : seq->seq.s, + id_flag ? seq->id1.n : seq->seq.n, 0, /* subject offset */ 0, /* options */ ovector, /* output vector */ 3 ); /* output vector length */ - if ((invert_flag && rc == PCRE_ERROR_NOMATCH) || rc >= 0) { - print_fastq_entry( fout, seq ); + if ((invert_flag && rc == PCRE_ERROR_NOMATCH) || (!invert_flag && rc >= 0)) { + if (count_flag) count++; + else fastq_print(fout, seq); + } + else if (mismatch_file) { + fastq_print(mismatch_file, seq); } } - kseq_destroy(seq); + seq_free(seq); + fastq_free(fqf); + + if (count_flag) fprintf(fout, "%zu\n", count); } int main(int argc, char* argv[]) { + SET_BINARY_MODE(stdin); + SET_BINARY_MODE(stdout); + const char* pat; pcre* re; const char* pat_error; int pat_error_offset; FILE* fin; - gzFile gzfin; - invert_flag = 0; - help_flag = 0; - zout_flag = 0; + invert_flag = 0; + count_flag = 0; + id_flag = 0; int opt; int opt_idx; + FILE* mismatch_file = NULL; static struct option long_options[] = { - {"help", no_argument, &help_flag, 1}, + {"id", no_argument, &id_flag, 1}, {"invert-match", no_argument, &invert_flag, 1}, + {"mismatches", required_argument, NULL, 'm'}, + {"count", no_argument, &count_flag, 1}, + {"help", no_argument, NULL, 'h'}, + {"version", no_argument, NULL, 'V'}, {0, 0, 0, 0} }; while (1) { - opt = getopt_long(argc, argv, "hv", long_options, &opt_idx); + opt = getopt_long(argc, argv, "ivmchV", long_options, &opt_idx); - if( opt == -1 ) break; + if (opt == -1) break; switch (opt) { case 0: if (long_options[opt_idx].flag != 0) break; if (optarg) { - /* TODO */ } break; - case 'h': - help_flag = 1; + case 'i': + id_flag = 1; break; case 'v': invert_flag = 1; break; + case 'm': + mismatch_file = fopen(optarg, "w"); + if (mismatch_file == NULL) { + fprintf(stderr, "No such file '%s'.\n", optarg); + return 1; + } + break; + + case 'c': + count_flag = 1; + break; + + case 'h': + print_help(); + return 0; + + case 'V': + print_version(stdout, prog_name); + return 0; + + case '?': + return 1; + default: abort(); } } - if (help_flag) { - print_help(); - return 0; - } - if (optind >= argc) { fprintf(stderr, "A pattern must be specified.\n"); return 1; @@ -157,15 +188,7 @@ int main(int argc, char* argv[]) if (optind >= argc || (argc - optind == 1 && strcmp(argv[optind],"-") == 0)) { - gzfin = gzdopen( fileno(stdin), "rb" ); - if (gzfin == NULL) { - fprintf(stderr, "Malformed file 'stdin'.\n"); - return 1; - } - - fastq_grep(gzfin, stdout, re); - - gzclose(gzfin); + fastq_grep(stdin, stdout, mismatch_file, re); } else { for (; optind < argc; optind++) { @@ -175,19 +198,14 @@ int main(int argc, char* argv[]) continue; } - gzfin = gzdopen(fileno(fin), "rb"); - if (gzfin == NULL) { - fprintf(stderr, "Malformed file '%s'.\n", argv[optind]); - continue; - } - - fastq_grep(gzfin, stdout, re); + fastq_grep(fin, stdout, mismatch_file, re); - gzclose(gzfin); + fclose(fin); } } pcre_free(re); + if (mismatch_file) fclose(mismatch_file); return 0; }