SWARM 2.1.8: Fix rare bug, do not show alignment parameters for d=1

README.md

@@ -42,6 +42,8 @@ Table of Content
 * [Third-party pipelines](#pipelines)
 * [Alternatives](#alternatives)
 * [New features](#features)
+  * [version 2.1.8](#version218)
+  * [version 2.1.7](#version217)
   * [version 2.1.6](#version216)
   * [version 2.1.5](#version215)
   * [version 2.1.4](#version214)
@@ -434,6 +436,19 @@ methods, here are some links:
 <a name="features"/>
 ## New features##
 
+<a name="version218"/>
+### version 2.1.8 ###
+
+**swarm** 2.1.8 fixes a rare bug triggered when clustering extremely
+short undereplicated sequences. Also, alignment parameters are not
+shown when d=1.
+
+<a name="version217"/>
+### version 2.1.7 ###
+
+**swarm** 2.1.7 fixes more problems with seed output. Ignore CR
+  characters in FASTA files. Improved help and error messsages.
+
 <a name="version216"/>
 ### version 2.1.6 ###
 
@@ -459,10 +474,11 @@ with the `-w` option when d>1.
 <a name="version212"/>
 ### version 2.1.2 ###
 
-**swarm** 2.1.2 adds the -a (--append-abundance) option to set a default
-abundance value to be used when abundance information is missing from
-the input file. If this option is not specified, missing abundance information
-will result in a fatal error. The error message in that case is improved.
+**swarm** 2.1.2 adds the -a (--append-abundance) option to set a
+default abundance value to be used when abundance information is
+missing from the input file. If this option is not specified, missing
+abundance information will result in a fatal error. The error message
+in that case is improved.
 
 <a name="version211"/>
 ### version 2.1.1 ###

man/swarm.1

@@ -1,5 +1,5 @@
 .\" ============================================================================
-.TH swarm 1 "February 24, 2016" "version 2.1.7" "USER COMMANDS"
+.TH swarm 1 "March 11, 2016" "version 2.1.8" "USER COMMANDS"
 .\" ============================================================================
 .SH NAME
 swarm \(em find clusters of nearly-identical nucleotide amplicons
@@ -329,6 +329,11 @@ New features and important modifications of \fBswarm\fR (short lived
 or minor bug releases are not mentioned):
 .RS
 .TP
+.BR v2.1.8\~ "released March 11, 2016"
+Version 2.1.8 fixes a rare bug triggered when clustering extremely
+short undereplicated sequences. Also, alignment parameters are not
+shown when \fId\rR=1.
+.TP
 .BR v2.1.7\~ "released February 24, 2016"
 Version 2.1.7 fixes a bug in the output of seeds with the \-w option
 when \fId\fR > 1 that was not properly fixed in version 2.1.6. It also

src/algod1.cc

@@ -119,6 +119,8 @@ static int * global_hits_data = 0;
 static int global_hits_alloc = 0;
 static int global_hits_count = 0;
 
+static unsigned long threads_used = 0;
+
 inline unsigned int hash_getindex(unsigned long hash)
 {
 #ifdef POWEROFTWO
@@ -315,7 +317,7 @@ void generate_variants(unsigned long thread,
 
 #if 1
   /* identical non-variant */
-  if (thread == threads - 1)
+  if (thread == threads_used - 1)
     find_variant_matches(thread, varseq, seqlen, seed);
 #endif
 
@@ -470,16 +472,17 @@ void process_seed(int seed, int subseed)
 {
   unsigned long seqlen = db_getsequencelen(subseed);
 
-  unsigned long thr = threads;
-  if (thr > seqlen + 1)
-    thr = seqlen+1;
+  threads_used = threads;
+  if (threads_used > seqlen + 1)
+    threads_used = seqlen+1;
 
   /* prepare work for the threads */
   unsigned long start = 0;
-  for(unsigned long t=0; t<thr; t++)
+  for(unsigned long t=0; t<threads_used; t++)
     {
       struct thread_info_s * tip = ti + t;
-      unsigned long length = (seqlen - start + thr - t) / (thr - t);
+      unsigned long length =
+        (seqlen - start + threads_used - t) / (threads_used - t);
       tip->seed = subseed;
       tip->mut_start = start;
       tip->mut_length = length;
@@ -491,8 +494,8 @@ void process_seed(int seed, int subseed)
       pthread_mutex_unlock(&tip->workmutex);
     }
 
-  /* wait for theads to finish their work */
-  for(unsigned int t=0; t<thr; t++)
+  /* wait for threads to finish their work */
+  for(unsigned int t=0; t<threads_used; t++)
     {
       struct thread_info_s * tip = ti + t;
       pthread_mutex_lock(&tip->workmutex);
@@ -503,7 +506,7 @@ void process_seed(int seed, int subseed)
 
   /* join hits from the threads */
 
-  for(unsigned int t=0; t<thr; t++)
+  for(unsigned int t=0; t<threads_used; t++)
     {
       if (global_hits_count + ti[t].hits_count > global_hits_alloc)
         {

src/swarm.cc

@@ -173,9 +173,13 @@ void args_show()
     fprintf(logfile, "Uclust file:       %s\n", uclustfilename);
   fprintf(logfile, "Resolution (d):    %ld\n", resolution);
   fprintf(logfile, "Threads:           %lu\n", threads);
+
+  if (resolution > 1)
+    {
   fprintf(logfile, "Scores:            match: %ld, mismatch: %ld\n", matchscore, mismatchscore);
   fprintf(logfile, "Gap penalties:     opening: %ld, extension: %ld\n", gapopen, gapextend);
   fprintf(logfile, "Converted costs:   mismatch: %ld, gap opening: %ld, gap extension: %ld\n", penalty_mismatch, penalty_gapopen, penalty_gapextend);
+    }
   fprintf(logfile, "Break OTUs:        %s\n", opt_no_otu_breaking ? "No" : "Yes");
   if (opt_fastidious)
     fprintf(logfile, "Fastidious:        Yes, with boundary %ld\n", opt_boundary);

src/swarm.h

@@ -52,7 +52,7 @@
 #define LINE_MAX 2048
 #endif
 
-#define SWARM_VERSION "2.1.7"
+#define SWARM_VERSION "2.1.8"
 #define WIDTH 32
 #define WIDTH_SHIFT 5
 #define BLOCKWIDTH 32