related to #20

src/common/kmc_api/kmer_api.h

@@ -309,11 +309,15 @@ class CKmerAPI
 	}
 
 	inline uint64 subkmer(uint32 pos, uint32 len) {
+		if (len == 0)
+			return 0;
 		auto lo_row = (pos + len - 1 + byte_alignment) >> 5;
 		auto hi_row = (pos + byte_alignment) >> 5;
 
 		auto start_pos = ((pos + len - 1 + byte_alignment) % 32) * 2;
-		uint64 mask = (1ull << (2 * len)) - 1;
+
+		//uint64 mask = (1ull << (2 * len)) - 1;
+		uint64 mask = ((1ull << len) << len) - 1; //I shift twice because len may be 32...
 		//subkmer is in single row
 		if (hi_row == lo_row) {
 			return (kmer_data[hi_row] >> (62 - start_pos)) & mask;

src/splash.py

@@ -33,12 +33,23 @@ def _split_lines(self, text, width):
                     )
 
 parser.add_argument("input_file", help="path to the file where input samples are defined, the format is: per each line {sample_name}<space>{path}, path is a fastq[.gz] file")
+def check_anchor_len(len):
+    ilen = int(len)
+    if ilen < 1 or ilen > 32:
+        raise argparse.ArgumentTypeError(f"anchor_len must be in [1;32]")
+    return ilen
+
+def check_target_len(len):
+    ilen = int(len)
+    if ilen < 1 or ilen > 32:
+        raise argparse.ArgumentTypeError(f"target_len must be in [1;32]")
+    return ilen
 
 group_base_configuration = parser.add_argument_group('Base configuration')
 group_base_configuration.add_argument("--outname_prefix", default="result", type=str, help="prefix of output file names")
-group_base_configuration.add_argument("--anchor_len", default=27, type=int, help="anchor length")
+group_base_configuration.add_argument("--anchor_len", default=27, type=check_anchor_len, help="anchor length")
 group_base_configuration.add_argument("--gap_len", default="0", type=str, help="gap length, if 'auto' it will be inferred from the data, in the opposite case it must be an int")
-group_base_configuration.add_argument("--target_len", default=27, type=int, help="target length")
+group_base_configuration.add_argument("--target_len", default=27, type=check_target_len,  help="target length")
 group_base_configuration.add_argument("--anchor_list", default="", type=str, help="list of accepted anchors, this is path to plain text file with one anchor per line without any header")
 group_base_configuration.add_argument("--pvals_correction_col_name", default="pval_opt", type=str, help="for which column correction should be applied")
 group_base_configuration.add_argument("--without_compactors", default=False, action='store_true', help="if used compactors will not be run")
@@ -498,6 +509,10 @@ def stage_1_task(id, input, out, err):
     cmd = f"{kmc_tools} -t{n_threads_stage_1_internal} transform {tmp_dir}/{sample_name} sort {tmp_dir}/{sample_name}.sorted"
     run_cmd(cmd, out, err)
 
+    # if we have small k kmc_tools will omit sorting because it is already sorted, lets just copy
+    if not os.path.exists(f"{tmp_dir}/{sample_name}.sorted.kmc_pre"):
+        shutil.copy(f"{tmp_dir}/{sample_name}.kmc_pre", f"{tmp_dir}/{sample_name}.sorted.kmc_pre")
+        shutil.copy(f"{tmp_dir}/{sample_name}.kmc_suf", f"{tmp_dir}/{sample_name}.sorted.kmc_suf")
     if clean_up:
         remove_kmc_output(f"{tmp_dir}/{sample_name}")