Release 1.23.1

Author: daviesrob

Makefile

@@ -109,7 +109,7 @@ endif
 
 include config.mk
 
-PACKAGE_VERSION = 1.23
+PACKAGE_VERSION = 1.23.1
 
 # If building from a Git repository, replace $(PACKAGE_VERSION) with the Git
 # description of the working tree: either a release tag with the same value
@@ -231,7 +231,7 @@ bcftools_h = bcftools.h $(htslib_hts_defs_h) $(htslib_vcf_h) $(htslib_synced_bcf
 call_h = call.h $(htslib_vcf_h) $(htslib_synced_bcf_reader_h) vcmp.h
 variantkey_h = variantkey.h hex.h
 convert_h = convert.h $(htslib_vcf_h)
-tsv2vcf_h = tsv2vcf.h $(htslib_vcf_h)
+tsv2vcf_h = tsv2vcf.h $(htslib_vcf_h) $(bcftools_h)
 filter_h = filter.h $(htslib_vcf_h)
 gvcf_h = gvcf.h $(bcftools_h)
 khash_str2str_h = khash_str2str.h $(htslib_khash_h)

NEWS

@@ -1,3 +1,17 @@
+## Release 1.23.1 (18th March 2026)
+
+
+Changes affecting the whole of bcftools, or multiple commands:
+
+* This release bundles HTSlib 1.23.1, which includes many important
+  bug fixes.  Please see htslib/NEWS for details.
+
+* Fix silent output truncation due to missing checks for read errors in
+  the annotate, cnv, concat, convert, consensus, csq, filter, gtcheck,
+  isec, merge, norm, query and stats commands; and all plugins.  They
+  will now print an error and return a non-zero exit code if an error
+  is encountered when reading input data. (#2503)
+
 ## Release 1.23 (16th December 2025)
 
 

abuf.c

@@ -721,7 +721,7 @@ static inline int _is_acgtn(char *seq)
 {
     while ( *seq )
     {
-        char c = toupper(*seq);
+        char c = toupper_c(*seq);
         if ( c!='A' && c!='C' && c!='G' && c!='T' && c!='N' ) return 0;
         seq++;
     }

bam2bcf_indel.c

@@ -36,6 +36,9 @@ DEALINGS IN THE SOFTWARE.  */
 #include <htslib/ksort.h>
 KSORT_INIT_GENERIC(uint32_t)
 
+// Avoid having to include all of bcftools.h
+static inline char toupper_c(char c) { return toupper((unsigned char) c); }
+
 #define MINUS_CONST 0x10000000
 
 #define MAX_TYPES 64
@@ -89,8 +92,8 @@ inline int est_indelreg(int pos, const char *ref, int l, char *ins4)
     int i, j, max = 0, max_i = pos, score = 0;
     l = abs(l);
     for (i = pos + 1, j = 0; ref[i]; ++i, ++j) {
-        if (ins4) score += (toupper(ref[i]) != "ACGTN"[(int)ins4[j%l]])? -10 : 1;
-        else score += (toupper(ref[i]) != toupper(ref[pos+1+j%l]))? -10 : 1;
+        if (ins4) score += (toupper_c(ref[i]) != "ACGTN"[(int)ins4[j%l]])? -10 : 1;
+        else score += (toupper_c(ref[i]) != toupper_c(ref[pos+1+j%l]))? -10 : 1;
         if (score < 0) break;
         if (max < score) max = score, max_i = i;
     }

bam_sample.c

@@ -295,18 +295,18 @@ int bam_smpl_add_samples(bam_smpl_t *bsmpl, char *list, int is_file)
         while ( *ptr )
         {
             if ( *ptr=='\\' && !escaped ) { escaped = 1; ptr++; continue; }
-            if ( isspace(*ptr) && !escaped ) break;
+            if ( isspace_c(*ptr) && !escaped ) break;
             kputc(*ptr, &ori);
             escaped = 0;
             ptr++;
         }
         if ( *ptr )
         {
-            while ( *ptr && isspace(*ptr) ) ptr++;
+            while ( *ptr && isspace_c(*ptr) ) ptr++;
             while ( *ptr )
             {
                 if ( *ptr=='\\' && !escaped ) { escaped = 1; ptr++; continue; }
-                if ( isspace(*ptr) && !escaped ) break;
+                if ( isspace_c(*ptr) && !escaped ) break;
                 kputc(*ptr, &ren);
                 escaped = 0;
                 ptr++;
@@ -343,30 +343,30 @@ int bam_smpl_add_readgroups(bam_smpl_t *bsmpl, char *list, int is_file)
         while ( *ptr )
         {
             if ( *ptr=='\\' && !escaped ) { escaped = 1; ptr++; continue; }
-            if ( isspace(*ptr) && !escaped ) break;
+            if ( isspace_c(*ptr) && !escaped ) break;
             kputc(*ptr, &fld1);
             escaped = 0;
             ptr++;
         }
         if ( *ptr )
         {
-            while ( *ptr && isspace(*ptr) ) ptr++;
+            while ( *ptr && isspace_c(*ptr) ) ptr++;
             while ( *ptr )
             {
                 if ( *ptr=='\\' && !escaped ) { escaped = 1; ptr++; continue; }
-                if ( isspace(*ptr) && !escaped ) break;
+                if ( isspace_c(*ptr) && !escaped ) break;
                 kputc(*ptr, &fld2);
                 escaped = 0;
                 ptr++;
             }
         }
         if ( *ptr )
         {
-            while ( *ptr && isspace(*ptr) ) ptr++;
+            while ( *ptr && isspace_c(*ptr) ) ptr++;
             while ( *ptr )
             {
                 if ( *ptr=='\\' && !escaped ) { escaped = 1; ptr++; continue; }
-                if ( isspace(*ptr) && !escaped ) break;
+                if ( isspace_c(*ptr) && !escaped ) break;
                 kputc(*ptr, &fld3);
                 escaped = 0;
                 ptr++;

bcftools.h

@@ -31,6 +31,7 @@ THE SOFTWARE.  */
 #include <htslib/synced_bcf_reader.h>
 #include <htslib/kfunc.h>
 #include <math.h>
+#include <ctype.h>
 
 #define FT_TAB_TEXT 0       // custom tab-delimited text file
 #define FT_GZ 1
@@ -184,4 +185,20 @@ static inline int get_unseen_allele(bcf1_t *line)
     return 0;
 }
 
+// <ctype.h> wrappers, borrowed from htslib's textutils_internal.h
+// The <ctype.h> functions operate on ints such as are returned by fgetc(),
+// i.e., characters represented as unsigned-char-valued ints, or EOF.
+// To operate on plain chars (and to avoid warnings on some platforms),
+// technically one must cast to unsigned char everywhere (see CERT STR37-C)
+// or less painfully use these *_c() functions that operate on plain chars
+// (but not EOF, which must be considered separately where it is applicable).
+static inline int isalnum_c(char c) { return isalnum((unsigned char) c); }
+static inline int isalpha_c(char c) { return isalpha((unsigned char) c); }
+static inline int isdigit_c(char c) { return isdigit((unsigned char) c); }
+static inline int isprint_c(char c) { return isprint((unsigned char) c); }
+static inline int ispunct_c(char c) { return ispunct((unsigned char) c); }
+static inline int isspace_c(char c) { return isspace((unsigned char) c); }
+static inline char tolower_c(char c) { return tolower((unsigned char) c); }
+static inline char toupper_c(char c) { return toupper((unsigned char) c); }
+
 #endif

configure.ac

@@ -247,8 +247,8 @@ AS_IF([test "$enable_perl_filters" != "no" ], [dnl
     fi
     AS_CASE([$PERL_LIBS],
       [*redhat/redhat-hardened-ld*],[[
-        PERL_CCOPTS="`echo $PERL_CCOPTS | sed 's,-specs=[a-z/]*/redhat/redhat-hardened-[a-z0-9]* *,,g'`"
-        PERL_LIBS="`echo $PERL_LIBS | sed 's,-specs=[a-z/]*/redhat/redhat-hardened-[a-z0-9]* *,,g'`"]
+        PERL_CCOPTS="`echo $PERL_CCOPTS | sed 's,-specs=[a-z/]*/redhat/redhat-hardened-[a-z0-9-]* *,,g'`"
+        PERL_LIBS="`echo $PERL_LIBS | sed 's,-specs=[a-z/]*/redhat/redhat-hardened-[a-z0-9-]* *,,g'`"]
         AC_MSG_NOTICE([removing -specs=...redhat-hardened... from Perl options])])
     AC_SUBST([PERL_CCOPTS])
     AC_SUBST([PERL_LIBS])

consensus.c

@@ -339,7 +339,7 @@ static void destroy_data(args_t *args)
 static void init_region(args_t *args, char *line)
 {
     char *ss, *se = line;
-    while ( *se && !isspace(*se) && *se!=':' ) se++;
+    while ( *se && !isspace_c(*se) && *se!=':' ) se++;
     hts_pos_t from = 0, to = 0;
     char tmp = 0, *tmp_ptr = NULL;
     if ( *se )
@@ -353,7 +353,7 @@ static void init_region(args_t *args, char *line)
             from--;
             ss = ++se;
             to = strtol(ss,&se,10);
-            if ( ss==se || (*se && !isspace(*se)) ) { from = 0; to = 0; }
+            if ( ss==se || (*se && !isspace_c(*se)) ) { from = 0; to = 0; }
             else to--;
         }
     }
@@ -411,6 +411,7 @@ static bcf1_t **next_vcf_line(args_t *args)
         }
         return &args->files->readers[0].buffer[0];
     }
+    if ( args->files->errnum ) error("Error: %s\n", bcf_sr_strerror(args->files->errnum));
     return NULL;
 }
 static void unread_vcf_line(args_t *args, bcf1_t **rec_ptr)
@@ -500,9 +501,9 @@ static void mark_ins(char *ref, char *alt, char mark)
 {
     int i, nref = strlen(ref), nalt = strlen(alt);
     if ( mark==TO_LOWER )
-        for (i=nref; i<nalt; i++) alt[i] = tolower(alt[i]);
+        for (i=nref; i<nalt; i++) alt[i] = tolower_c(alt[i]);
     else if ( mark==TO_UPPER )
-        for (i=nref; i<nalt; i++) alt[i] = toupper(alt[i]);
+        for (i=nref; i<nalt; i++) alt[i] = toupper_c(alt[i]);
     else if ( mark )
         for (i=nref; i<nalt; i++) alt[i] = mark;
 }
@@ -513,22 +514,17 @@ static void mark_snv(char *ref, char *alt, char mark)
     if ( mark==TO_LOWER )
     {
         for (i=0; i<n; i++)
-            if ( tolower(ref[i])!=tolower(alt[i]) ) alt[i] = tolower(alt[i]);
+            if ( tolower_c(ref[i])!=tolower_c(alt[i]) ) alt[i] = tolower_c(alt[i]);
     }
     else if ( mark==TO_UPPER)
     {
         for (i=0; i<n; i++)
-            if ( tolower(ref[i])!=tolower(alt[i]) ) alt[i] = toupper(alt[i]);
-    }
-    else if ( mark==TO_UPPER)
-    {
-        for (i=0; i<n; i++)
-            if ( tolower(ref[i])!=tolower(alt[i]) ) alt[i] = toupper(alt[i]);
+            if ( tolower_c(ref[i])!=tolower_c(alt[i]) ) alt[i] = toupper_c(alt[i]);
     }
     else if ( mark )
     {
         for (i=0; i<n; i++)
-            if ( tolower(ref[i])!=tolower(alt[i]) ) alt[i] = mark;
+            if ( tolower_c(ref[i])!=tolower_c(alt[i]) ) alt[i] = mark;
     }
 }
 static void iupac_init(args_t *args, bcf1_t *rec)
@@ -943,7 +939,7 @@ static void apply_variant(args_t *args, bcf1_t *rec)
         // one base overlap
 
         int fail = 1;
-        if ( args->prev_base_pos==rec->pos && toupper(ref_allele[0])==toupper(args->prev_base) )
+        if ( args->prev_base_pos==rec->pos && toupper_c(ref_allele[0])==toupper_c(args->prev_base) )
         {
             if ( rec->rlen==1 ) fail = 0;
             else if ( !strncasecmp(ref_allele+1,args->fa_buf.s+idx+1,rec->rlen-1) ) fail = 0;
@@ -992,11 +988,11 @@ static void apply_variant(args_t *args, bcf1_t *rec)
     }
 
     int safe_idx = idx<0 ? 0 : idx; // idx can be negative in case of overlapping deletion
-    args->fa_case = toupper(args->fa_buf.s[safe_idx])==args->fa_buf.s[safe_idx] ? TO_UPPER : TO_LOWER;
+    args->fa_case = toupper_c(args->fa_buf.s[safe_idx])==args->fa_buf.s[safe_idx] ? TO_UPPER : TO_LOWER;
     if ( args->fa_case==TO_UPPER )
-        for (i=0; i<alen; i++) alt_allele[i] = toupper(alt_allele[i]);
+        for (i=0; i<alen; i++) alt_allele[i] = toupper_c(alt_allele[i]);
     else
-        for (i=0; i<alen; i++) alt_allele[i] = tolower(alt_allele[i]);
+        for (i=0; i<alen; i++) alt_allele[i] = tolower_c(alt_allele[i]);
 
     if ( args->mark_ins && len_diff>0 )
         mark_ins(ref_allele, alt_allele, args->mark_ins);
@@ -1080,9 +1076,9 @@ static void mask_region(args_t *args, char *seq, int len)
             if ( idx_start < 0 ) idx_start = 0;
             if ( idx_end >= len ) idx_end = len - 1;
             if ( mask->with==MASK_UC )
-                for (j=idx_start; j<=idx_end; j++) seq[j] = toupper(seq[j]);
+                for (j=idx_start; j<=idx_end; j++) seq[j] = toupper_c(seq[j]);
             else if ( mask->with==MASK_LC )
-                for (j=idx_start; j<=idx_end; j++) seq[j] = tolower(seq[j]);
+                for (j=idx_start; j<=idx_end; j++) seq[j] = tolower_c(seq[j]);
             else
                 for (j=idx_start; j<=idx_end; j++) seq[j] = mask->with;
         }
@@ -1124,7 +1120,7 @@ static void consensus(args_t *args)
         args->fa_src_pos += str.l;
 
         // determine if uppercase or lowercase is used in this fasta file
-        if ( args->fa_case==-1 ) args->fa_case = toupper(str.s[0])==str.s[0] ? 1 : 0;
+        if ( args->fa_case==-1 ) args->fa_case = toupper_c(str.s[0])==str.s[0] ? 1 : 0;
 
         if ( args->mask ) mask_region(args, str.s, str.l);
         kputs(str.s, &args->fa_buf);

convert.c

@@ -1461,7 +1461,7 @@ static int parse_subscript(char **p)
     char *q = *p;
     if ( *q!='{' ) return -1;
     q++;
-    while ( *q && *q!='}' && isdigit(*q) ) q++;
+    while ( *q && *q!='}' && isdigit_c(*q) ) q++;
     if ( *q!='}' ) return -1;
     int idx = atoi((*p)+1);
     *p = q+1;
@@ -1474,7 +1474,7 @@ static char *parse_tag(convert_t *convert, char *p, int is_gtf)
     if ( is_vcf_column ) p++;
 
     char *q = ++p;
-    while ( *q && (isalnum(*q) || *q=='_' || *q=='.') ) q++;
+    while ( *q && (isalnum_c(*q) || *q=='_' || *q=='.') ) q++;
     kstring_t str = {0,0,0};
     if ( q-p==0 ) error("Could not parse format string: %s\n", convert->format_str);
     kputsn(p, q-p, &str);
@@ -1517,7 +1517,7 @@ static char *parse_tag(convert_t *convert, char *p, int is_gtf)
             }
             p = ++q;
             str.l = 0;
-            while ( *q && (isalnum(*q) || *q=='_' || *q=='.') ) q++;
+            while ( *q && (isalnum_c(*q) || *q=='_' || *q=='.') ) q++;
             if ( q-p==0 ) error("Could not parse format string: %s\n", convert->format_str);
             kputsn(p, q-p, &str);
             fmt_t *fmt = register_tag(convert, str.s, is_gtf, T_INFO);
@@ -1567,7 +1567,7 @@ static char *parse_tag(convert_t *convert, char *p, int is_gtf)
             {
                 p = ++q;
                 str.l = 0;
-                while ( *q && (isalnum(*q) || *q=='_' || *q=='.') ) q++;
+                while ( *q && (isalnum_c(*q) || *q=='_' || *q=='.') ) q++;
                 if ( q-p==0 ) error("Could not parse format string: %s\n", convert->format_str);
                 kputsn(p, q-p, &str);
                 fmt_t *fmt = register_tag(convert, str.s, is_gtf, T_INFO);

csq.c

@@ -2844,7 +2844,7 @@ static int sanity_check_ref(args_t *args, gf_tscript_t *tr, bcf1_t *rec)
     int i = 0;
     while ( ref[i] && vcf[i] )
     {
-        if ( ref[i]!=vcf[i] && toupper(ref[i])!=toupper(vcf[i]) )
+        if ( ref[i]!=vcf[i] && toupper_c(ref[i])!=toupper_c(vcf[i]) )
         {
             if ( !args->force )
                 error("Error: the fasta reference does not match the VCF REF allele at %s:%"PRId64" .. fasta=%c vcf=%c\n",
@@ -3918,6 +3918,7 @@ int main_csq(int argc, char *argv[])
     {
         process(args, &args->sr->readers[0].buffer[0]);
     }
+    if ( args->sr->errnum ) error("Error: %s\n", bcf_sr_strerror(args->sr->errnum));
     process(args,NULL);
 
     destroy_data(args);