Feature print bed

column.c

@@ -16,6 +16,7 @@
  * external functions declaration
  */
 extern int compare_row_list(const void *p1, const void *p2);
+extern int split_ip(char ***tab, char *s, char *delim);
 
 /*
  * global variables declaration
@@ -77,6 +78,160 @@ void print_row(FILE *output, GTF_ROW *r, char delim, int add_chr) {
 	fprintf(output, "\n");
 }
 
+/*
+ * prints a list of comma (or other sep) separated attribute values from a GTF row.
+ *
+ *
+ * Parameters:
+ * 		row:	the GTF row from which to print attributes
+ * 		output:	where to print (a file or stdout)
+ *      keys: a comma separated list of keys to be selected.
+ *      sep:    the separator to use (e.g '|', ','...).
+ */
+void print_attr_with_sep(GTF_ROW *row, FILE *output, char delim, char *keys, char *sep, char *more_info) {
+	int n, i, j;
+	int found = 0;
+	char **token;
+	char *keys_concat = (char *)calloc(5000, sizeof(char));
+
+
+    n = split_ip(&token, strdup(keys), ",");
+
+    for(i=0; i<n; i++)
+    {
+        found = 0;
+
+        if(strcmp("seqid", token[i]) == 0)
+        {
+                    strcat(keys_concat, row->field[0]);
+                    if (i < (n-1)) strcat(keys_concat, sep);
+                    found = 1;
+        }
+        else if(strcmp("chrom", token[i]) == 0)
+        {
+                    strcat(keys_concat, row->field[0]);
+                    if (i < (n-1)) strcat(keys_concat, sep);
+                    found = 1;
+        }
+        else if(strcmp("source", token[i]) == 0)
+        {
+                    strcat(keys_concat, row->field[1]);
+                    if (i < (n-1)) strcat(keys_concat, sep);
+                    found = 1;
+        }
+        else if(strcmp("feature", token[i]) == 0)
+        {
+                    strcat(keys_concat, row->field[2]);
+                    if (i < (n-1)) strcat(keys_concat, sep);
+                    found = 1;
+        }
+        else if(strcmp("start", token[i]) == 0)
+        {
+                    strcat(keys_concat, row->field[3]);
+                    if (i < (n-1)) strcat(keys_concat, sep);
+                    found = 1;
+        }
+        else if(strcmp("end", token[i]) == 0)
+        {
+                    strcat(keys_concat, row->field[4]);
+                    if (i < (n-1)) strcat(keys_concat, sep);
+                    found = 1;
+        }
+        else if(strcmp("score", token[i]) == 0)
+        {
+                    strcat(keys_concat, row->field[5]);
+                    if (i < (n-1)) strcat(keys_concat, sep);
+                    found = 1;
+        }
+        else if(strcmp("strand", token[i]) == 0)
+        {
+                    strcat(keys_concat, row->field[6]);
+                    if (i < (n-1)) strcat(keys_concat, sep);
+                    found = 1;
+        }
+        else if(strcmp("phase", token[i]) == 0)
+        {
+                    strcat(keys_concat, row->field[7]);
+                    if (i < (n-1)) strcat(keys_concat, sep);
+                    found = 1;
+        }
+
+        if(found == 0)
+        {
+            if (row->attributes.nb != -1)
+            {
+                for(j=0; j< row->attributes.nb; j++)
+                {
+
+                    if(strcmp(row->attributes.attr[j]->key, token[i]) == 0)
+                    {
+                            strcat(keys_concat, row->attributes.attr[j]->value);
+                            if (i < (n-1)) strcat(keys_concat, sep);
+                            found = 1;
+
+                    }
+                }
+            }
+        }
+
+        if(found == 0){
+            if(found == 0){
+                strcat(keys_concat, "?");
+                if (i < (n-1)) strcat(keys_concat, sep);
+            }
+        }
+    }
+
+    //keys_concat = (char *)realloc(keys_concat, (strlen(keys_concat) + 1) * sizeof(char));
+
+
+    if(strlen(more_info)){
+        strcat(keys_concat, sep);
+        strcat(keys_concat, more_info);
+    }
+
+    if(keys_concat){
+       fprintf(output, "%s", keys_concat);
+    }
+    else
+    {
+        fprintf(output, "");
+    }
+
+    fprintf(output, "%c", delim);
+
+    free(keys_concat);
+}
+
+/*
+ * prints a GTF row in bed format to output with the given character delimiter
+ *
+ * Parameters:
+ * 		output:		where to print
+ * 		r:			the row to print
+ * 		delim:		the delimiter character
+ * 		add_chr:	boolean; if true(1), add "chr" at the begining of the row
+ *      keys:      a comma separated list of keys whose values will appear in the name
+ *                  (4th) colum of the bed.
+ *      sep:        The separator to use for names (4th column).
+ */
+void print_row_bed(FILE *output, GTF_ROW *r,  char delim, int add_chr, char *keys, char *sep, char *more_info){
+
+
+	if (add_chr) fprintf(output, "chr");
+	// print chr, start, end
+	print_string(r->field[0], output, column[0], delim);
+	fprintf(output,"%d%c", atoi(r->field[3]) - 1, delim);
+	print_string(r->field[4], output, column[4], delim);
+	// print requested columns with user-defined delim
+	print_attr_with_sep(r, output, delim, keys, sep, more_info);
+	// print score strand
+	print_string(r->field[5], output, column[5], delim);
+	fprintf(output,"%s", r->field[6]);
+	fprintf(output, "\n");
+}
+
+
 /*
  * Creates an empty index and add/link it in the index list of the column
  * in index_id.

command/add_attr_to_pos.c

@@ -18,7 +18,6 @@
 extern GTF_DATA *clone_gtf_data(GTF_DATA *gtf_data);
 extern void add_attribute(GTF_ROW *row, char *key, char *value);
 extern int split_ip(char ***tab, char *s, char *delim);
-extern char *trim_ip(char *);
 
 /*
  * global variables declaration

command/convert_to_ensembl.c

@@ -95,9 +95,9 @@ static void action_transcript(const void *nodep, const VISIT which, const int de
 							nbrow++;
 						}
 				}
-				asprintf(&(tr_row->field[3]), "%d", start);
-				asprintf(&(tr_row->field[4]), "%d", end);
-				if (ok) {
+				if (asprintf(&(tr_row->field[3]), "%d", start) > 0 &&
+					asprintf(&(tr_row->field[4]), "%d", end) > 0 &&
+					ok) {
 					if (!strcmp(gtf_d->data[datap->row[0]]->field[2], "gene")) {
 						tr_row->next = gtf_d->data[datap->row[0]]->next;
 						gtf_d->data[datap->row[0]]->next = tr_row;
@@ -186,14 +186,14 @@ static void action_gene(const void *nodep, const VISIT which, const int depth) {
 						}
 					}
 				}
-				asprintf(&(g_row->field[3]), "%d", start);
-				asprintf(&(g_row->field[4]), "%d", end);
-
-				g_row->next = gtf_d->data[datap->row[0]];
-				if (datap->row[0] != 0)
-					gtf_d->data[datap->row[0] - 1]->next = g_row;
-				else
-					gtf_d0 = g_row;
+				if (asprintf(&(g_row->field[3]), "%d", start) > 0 &&
+					asprintf(&(g_row->field[4]), "%d", end) > 0) {
+					g_row->next = gtf_d->data[datap->row[0]];
+					if (datap->row[0] != 0)
+						gtf_d->data[datap->row[0] - 1]->next = g_row;
+					else
+						gtf_d0 = g_row;
+				}
 			}
 			break;
 		case endorder:

command/get_sequences.c

@@ -84,7 +84,7 @@ void get_chunk(char *ret, FILE *fasta_file, long seqpos, int L, int N, int p, ch
 				fseek(fasta_file, -1, SEEK_CUR);
 				reste_row_file = L;
 			}
-		} while (reste_row & !eof);
+		} while (reste_row && !eof);
 	}
 }
 
@@ -188,8 +188,9 @@ FILE *get_fasta_file_index(FILE *fasta_file, char *index) {
 	FILE *ffi = NULL;
 	long pfasta;
 	char *buffer = NULL, *p_end_dir = NULL;
-	int maxLineSize = 0;
+	size_t maxLineSize = 0;
 	size_t size = 0;
+
 	int n;
 	unsigned long old_crc, crc;
 
@@ -215,7 +216,7 @@ FILE *get_fasta_file_index(FILE *fasta_file, char *index) {
 			free(buffer);
 			buffer = NULL;
 		}
-		fprintf(ffi, "%d\n", maxLineSize - 1);
+		fprintf(ffi, "%lu\n", maxLineSize - 1);
 		fprintf(ffi, "%lx\n", crc);
 		fflush(ffi);
 		rewind(ffi);
@@ -249,7 +250,7 @@ FILE *get_fasta_file_index(FILE *fasta_file, char *index) {
 				free(buffer);
 				buffer = NULL;
 			}
-			fprintf(ffi, "%d\n", maxLineSize - 1);
+			fprintf(ffi, "%lu\n", maxLineSize - 1);
 			fprintf(ffi, "%lx\n", crc);
 			fflush(ffi);
 		}
@@ -260,14 +261,15 @@ FILE *get_fasta_file_index(FILE *fasta_file, char *index) {
 }
 
 void print_fasta_sequence(SEQUENCE *seq) {
-	unsigned int k;
+	size_t k;
+	int l;
 	FEATURE *feat;
 
 	fprintf(stdout, "%s\n", seq->header);
 	for (k = 0; k < strlen(seq->sequence); k += 60)
 		fprintf(stdout, "%.60s\n", seq->sequence + k);
-	for (k = 0; k < seq->features->nb; k++) {
-		feat = seq->features->feature[k];
+	for (l = 0; l < seq->features->nb; l++) {
+		feat = seq->features->feature[l];
 		fprintf(stdout, "  %s : %d-%d (%d-%d)\n", feat->name, feat->start, feat->end, feat->tr_start, feat->tr_end);
 	}
 }

command/print_bed.c

@@ -0,0 +1,30 @@
+/*
+ * write_bed.c
+ *
+ *  Created on: December 7, 2018
+ *      Author: puthier (based on Fafa code...)
+ *  Objective: print a gtf obj in bed format
+ */
+
+#include "libgtftk.h"
+
+extern void print_row_bed(FILE *output, GTF_ROW *r, char delim, int add_chr, char *keys, char *sep, char *more_info);
+
+__attribute__ ((visibility ("default")))
+void *print_bed(GTF_DATA *gtf_data, char *output, int add_chr, char *keys, char *sep, char *more_info) {
+	int i;
+	FILE *out = stdout;
+
+	if (gtf_data != NULL) {
+		if (*output != '-') out = fopen(output, "w");
+		if (out == NULL) out = stdout;
+		for (i = 0; i < gtf_data->size; i++){
+		    print_row_bed(out, gtf_data->data[i], '\t', add_chr, keys, sep, more_info);
+		}
+		if (out != stdout) {
+			fflush(out);
+			fclose(out);
+		}
+	}
+	return 0;
+}

config.h

@@ -0,0 +1,98 @@
+/* config.h.  Generated from config.h.in by configure.  */
+/* config.h.in.  Generated from configure.ac by autoheader.  */
+
+/* Define to 1 if you have the <dlfcn.h> header file. */
+#define HAVE_DLFCN_H 1
+
+/* Define to 1 if you have the <inttypes.h> header file. */
+#define HAVE_INTTYPES_H 1
+
+/* Define to 1 if you have the <limits.h> header file. */
+#define HAVE_LIMITS_H 1
+
+/* Define to 1 if your system has a GNU libc compatible `malloc' function, and
+   to 0 otherwise. */
+#define HAVE_MALLOC 1
+
+/* Define to 1 if you have the <memory.h> header file. */
+#define HAVE_MEMORY_H 1
+
+/* Define to 1 if you have the `mkdir' function. */
+#define HAVE_MKDIR 1
+
+/* Define to 1 if your system has a GNU libc compatible `realloc' function,
+   and to 0 otherwise. */
+#define HAVE_REALLOC 1
+
+/* Define to 1 if you have the <stdint.h> header file. */
+#define HAVE_STDINT_H 1
+
+/* Define to 1 if you have the <stdlib.h> header file. */
+#define HAVE_STDLIB_H 1
+
+/* Define to 1 if you have the `strchr' function. */
+#define HAVE_STRCHR 1
+
+/* Define to 1 if you have the `strdup' function. */
+#define HAVE_STRDUP 1
+
+/* Define to 1 if you have the <strings.h> header file. */
+#define HAVE_STRINGS_H 1
+
+/* Define to 1 if you have the <string.h> header file. */
+#define HAVE_STRING_H 1
+
+/* Define to 1 if you have the `strrchr' function. */
+#define HAVE_STRRCHR 1
+
+/* Define to 1 if you have the `strstr' function. */
+#define HAVE_STRSTR 1
+
+/* Define to 1 if you have the <sys/stat.h> header file. */
+#define HAVE_SYS_STAT_H 1
+
+/* Define to 1 if you have the <sys/types.h> header file. */
+#define HAVE_SYS_TYPES_H 1
+
+/* Define to 1 if you have the <unistd.h> header file. */
+#define HAVE_UNISTD_H 1
+
+/* Define to the sub-directory in which libtool stores uninstalled libraries.
+   */
+#define LT_OBJDIR ".libs/"
+
+/* Name of package */
+#define PACKAGE "libgtftk"
+
+/* Define to the address where bug reports for this package should be sent. */
+#define PACKAGE_BUGREPORT "fabrice.lopez@inserm.fr"
+
+/* Define to the full name of this package. */
+#define PACKAGE_NAME "libgtftk"
+
+/* Define to the full name and version of this package. */
+#define PACKAGE_STRING "libgtftk 0.9.3"
+
+/* Define to the one symbol short name of this package. */
+#define PACKAGE_TARNAME "libgtftk"
+
+/* Define to the home page for this package. */
+#define PACKAGE_URL ""
+
+/* Define to the version of this package. */
+#define PACKAGE_VERSION "0.9.3"
+
+/* Define to 1 if you have the ANSI C header files. */
+#define STDC_HEADERS 1
+
+/* Version number of package */
+#define VERSION "0.9.3"
+
+/* Define to rpl_malloc if the replacement function should be used. */
+/* #undef malloc */
+
+/* Define to rpl_realloc if the replacement function should be used. */
+/* #undef realloc */
+
+/* Define to `unsigned int' if <sys/types.h> does not define. */
+/* #undef size_t */