Compare commits

...

5 Commits

Author SHA1 Message Date
9ace9989c4 Switch to version 3.0.0b27 2020-07-07 16:47:21 +02:00
a3ebe5f118 C: AVL trees: fixed a bug where storing the difference between 2 crc64
values in an int64 would mess trees up resulting in failed data
dereplication
2020-07-07 16:47:00 +02:00
9100e14899 obi uniq: quick fix for bug where some sequences are not correctly
dereplicated
2020-07-03 17:36:57 +02:00
ccda0661ce small help documentation improvement 2020-07-01 18:20:38 +02:00
aab59f2214 obi clean: fixed a memory bug, fixed the behaviour when no sample info,
and added checks warnings and error handling when sample info not
dereplicated
2020-07-01 18:17:47 +02:00
7 changed files with 42 additions and 11 deletions

View File

@ -36,8 +36,7 @@ def addOptions(parser):
dest="clean:sample-tag-name",
metavar="<SAMPLE TAG NAME>",
type=str,
default="merged_sample",
help="Name of the tag where sample counts are kept.")
help="Name of the tag where merged sample count informations are kept (typically generated by obi uniq, usually MERGED_sample, default: None).")
group.add_argument('--ratio', '-r',
action="store", dest="clean:ratio",
@ -107,6 +106,9 @@ def run(config):
command_line = " ".join(sys.argv[1:])
comments = View.print_config(config, "clean", command_line, input_dms_name=[i_dms_name], input_view_name=[i_view_name])
if 'sample-tag-name' not in config['clean']:
config['clean']['sample-tag-name'] = ""
if obi_clean(i_dms.name_with_full_path, tobytes(i_view_name), tobytes(config['clean']['sample-tag-name']), tobytes(o_view_name), comments, \
config['clean']['distance'], config['clean']['ratio'], config['clean']['heads-only'], config['clean']['thread-count']) < 0:
raise Exception("Error running obiclean")

View File

@ -22,7 +22,7 @@ def addOptions(parser):
group.add_argument('-s','--sequence',
action="store_true", dest="count:sequence",
default=False,
help="Prints only the number of sequence records.")
help="Prints only the number of sequence records (much faster, default: False).")
group.add_argument('-a','--all',
action="store_true", dest="count:all",

View File

@ -307,6 +307,7 @@ cdef uniq_sequences(View_NUC_SEQS view, View_NUC_SEQS o_view, ProgressBar pb, di
for x in categories :
catl.append(i_seq[x])
#unique_id = tuple(catl) + (i_seq_col[i],)
unique_id = tuple(catl) + (i_seq_col.get_line_idx(i),)
#unique_id = tuple(i_seq[x] for x in categories) + (seq_col.get_line_idx(i),) # The line that cython can't read properly
@ -453,6 +454,7 @@ cdef uniq_sequences(View_NUC_SEQS view, View_NUC_SEQS o_view, ProgressBar pb, di
merged_dict[mkey] = {}
for i_idx in merged_sequences:
PyErr_CheckSignals()
pb(total_treated)
i_id = i_id_col[i_idx]

View File

@ -1,5 +1,5 @@
major = 3
minor = 0
serial= '0b26'
serial= '0b27'
version ="%d.%d.%s" % (major,minor,serial)

View File

@ -246,7 +246,16 @@ int obi_clean(const char* dms_name,
// Open the sample column if there is one
if ((strcmp(sample_column_name, "") == 0) || (sample_column_name == NULL))
sample_column = NULL;
{
fprintf(stderr, "Info: No sample information provided, assuming one sample.\n");
sample_column = obi_view_get_column(i_view, COUNT_COLUMN);
if (sample_column == NULL)
{
obidebug(1, "\nError getting the COUNT column");
return -1;
}
sample_count = 1;
}
else
{
sample_column = obi_view_get_column(i_view, sample_column_name);
@ -255,6 +264,13 @@ int obi_clean(const char* dms_name,
obidebug(1, "\nError getting the sample column");
return -1;
}
sample_count = (sample_column->header)->nb_elements_per_line;
// Check that the sample column is a merged column with all sample informations
if (sample_count == 1)
{
obidebug(1, "\n\nError: If a sample column is provided, it must contain 'merged' sample counts as built by obi uniq with the -m option\n");
return -1;
}
}
// Create the output view, or a temporary one if heads only
@ -279,8 +295,6 @@ int obi_clean(const char* dms_name,
return -1;
}
sample_count = (sample_column->header)->nb_elements_per_line;
// Create the output columns
if (create_output_columns(o_view, sample_column, sample_count) < 0)
{
@ -549,7 +563,7 @@ int obi_clean(const char* dms_name,
if (heads_only)
{
line_selection = malloc((o_view->infos)->line_count * sizeof(index_t));
line_selection = malloc((((o_view->infos)->line_count) + 1) * sizeof(index_t));
if (line_selection == NULL)
{
obi_set_errno(OBI_MALLOC_ERROR);

View File

@ -52,7 +52,8 @@
*
* @param dms A pointer on an OBIDMS.
* @param i_view_name The name of the input view.
* @param sample_column_name The name of the OBI_STR column in the input view where the sample information is kept.
* @param sample_column_name The name of the column in the input view where the sample information is kept.
* Must be merged informations as built by the obi uniq tool (checked by the function).
* NULL or "" (empty string) if there is no sample information.
* @param o_view_name The name of the output view where the results should be written (should not already exist).
* @param o_view_comments The comments that should be associated with the output view.

View File

@ -2259,7 +2259,13 @@ index_t obi_avl_add(OBIDMS_avl_p avl, Obi_blob_p value)
parent = next;
// Compare the crc of the value with the crc of the current node
comp = (current_node->crc64) - crc;
//comp = (current_node->crc64) - crc;
if ((current_node->crc64) == crc)
comp = 0;
else if ((current_node->crc64) > crc)
comp = 1;
else
comp = -1;
if (comp == 0)
{ // check if really same value
@ -2354,7 +2360,13 @@ index_t obi_avl_find(OBIDMS_avl_p avl, Obi_blob_p value)
current_node = (avl->tree)+next;
// Compare the crc of the value with the crc of the current node
comp = (current_node->crc64) - crc;
//comp = (current_node->crc64) - crc;
if ((current_node->crc64) == crc)
comp = 0;
else if ((current_node->crc64) > crc)
comp = 1;
else
comp = -1;
if (comp == 0)
{ // Check if really same value