Compare commits

...

8 Commits

11 changed files with 46 additions and 15 deletions

View File

@ -175,6 +175,14 @@ def run(config):
o_dms_name = output[0].name
o_view_name = output[1]
# Open the taxonomy DMS
taxdms = open_uri(config['obi']['taxoURI'],
dms_only=True)
if taxdms is None:
raise Exception("Could not open taxonomy DMS")
tax_dms = taxdms[0]
tax_dms_name = taxdms[0].name
# Read taxonomy name
taxonomy_name = config['obi']['taxoURI'].split("/")[-1] # Robust in theory
@ -197,7 +205,8 @@ def run(config):
# TODO: primers in comments?
if obi_ecopcr(i_dms.name_with_full_path, tobytes(i_view_name), tobytes(taxonomy_name), \
if obi_ecopcr(i_dms.name_with_full_path, tobytes(i_view_name),
tax_dms.name_with_full_path, tobytes(taxonomy_name), \
o_dms.name_with_full_path, tobytes(o_view_name), comments, \
tobytes(config['ecopcr']['primer1']), tobytes(config['ecopcr']['primer2']), \
config['ecopcr']['error'], \

2
python/obitools3/commands/ngsfilter.pyx Normal file → Executable file
View File

@ -271,7 +271,7 @@ cdef tuple annotate(sequences, infos, no_tags, verbose=False):
sequences[0][REVERSE_QUALITY_COLUMN] = sequences[1].quality # used by alignpairedend tool
for seq in sequences:
if hasattr(seq, "quality_array"):
if hasattr(seq, "quality_array") and seq.quality_array is not None:
q = -reduce(lambda x,y:x+y,(math.log10(z) for z in seq.quality_array),0)/len(seq.quality_array)*10
seq[b'avg_quality']=q
q = -reduce(lambda x,y:x+y,(math.log10(z) for z in seq.quality_array[0:10]),0)

View File

@ -119,9 +119,12 @@ def mean(values, options):
def variance(v):
if len(v)==1:
return 0
return 0
s = reduce(lambda x,y:(x[0]+y,x[1]+y**2),v,(0.,0.))
return s[1]/(len(v)-1) - s[0]**2/len(v)/(len(v)-1)
var = round(s[1]/(len(v)-1) - s[0]**2/len(v)/(len(v)-1), 5) # round to go around shady python rounding stuff when var is actually 0
if var == -0.0: # then fix -0 to +0 if was rounded to -0
var = 0.0
return var
def varpop(values, options):

View File

@ -8,6 +8,7 @@ cdef extern from "obi_ecopcr.h" nogil:
int obi_ecopcr(const char* input_dms_name,
const char* i_view_name,
const char* tax_dms_name,
const char* taxonomy_name,
const char* output_dms_name,
const char* o_view_name,

View File

@ -93,6 +93,8 @@ cdef class Taxonomy(OBIWrapper) :
raise RuntimeError("Error : Cannot read taxonomy %s"
% tostr(name))
print("Taxonomy read", file=sys.stderr)
taxo = OBIWrapper.new_wrapper(Taxonomy, pointer)
dms.register(taxo)

View File

@ -345,7 +345,7 @@ cdef class View(OBIWrapper) :
nb_elements_per_line=new_nb_elements_per_line, elements_names=new_elements_names,
dict_column=(new_nb_elements_per_line>1), comments=old_column.comments, alias=column_name_b+tobytes('___new___'))
switch_to_dict = old_column.nb_elements_per_line == 1 and new_nb_elements_per_line > 1
switch_to_dict = not old_column.dict_column and new_nb_elements_per_line > 1
ori_key = old_column._elements_names[0]
for i in range(length) :

View File

@ -1,5 +1,5 @@
major = 3
minor = 0
serial= '1b11'
serial= '1b12'
version ="%d.%d.%s" % (major,minor,serial)

View File

@ -645,7 +645,8 @@ static int print_seq(Obiview_p i_view, Obiview_p o_view,
int obi_ecopcr(const char* i_dms_name,
const char* i_view_name,
const char* taxonomy_name, // TODO discuss that input dms assumed
const char* tax_dms_name,
const char* taxonomy_name,
const char* o_dms_name,
const char* o_view_name,
const char* o_view_comments,
@ -678,6 +679,7 @@ int obi_ecopcr(const char* i_dms_name,
OBIDMS_p i_dms = NULL;
OBIDMS_p o_dms = NULL;
OBIDMS_p tax_dms = NULL;
OBIDMS_taxonomy_p taxonomy = NULL;
Obiview_p i_view = NULL;
Obiview_p o_view = NULL;
@ -965,8 +967,16 @@ int obi_ecopcr(const char* i_dms_name,
return -1;
}
// Open taxonomy DMS
tax_dms = obi_open_dms(tax_dms_name, false);
if (tax_dms == NULL)
{
obidebug(1, "\nError opening the taxonomy DMS");
return -1;
}
// Open the taxonomy
taxonomy = obi_read_taxonomy(i_dms, taxonomy_name, false);
taxonomy = obi_read_taxonomy(tax_dms, taxonomy_name, false);
if (taxonomy == NULL)
{
obidebug(1, "\nError opening the taxonomy");

View File

@ -77,7 +77,8 @@
*
* @param i_dms_name The path to the input DMS.
* @param i_view_name The name of the input view.
* @param taxonomy_name The name of the taxonomy in the input DMS.
* @param tax_dms_name The path to the DMS containing the taxonomy.
* @param taxonomy_name The name of the taxonomy.
* @param o_dms_name The path to the output DMS.
* @param o_view_name The name of the output view.
* @param o_view_comments The comments to associate with the output view.
@ -106,6 +107,7 @@
*/
int obi_ecopcr(const char* i_dms_name,
const char* i_view_name,
const char* tax_dms_name,
const char* taxonomy_name,
const char* o_dms_name,
const char* o_view_name,

View File

@ -1417,7 +1417,7 @@ char* obi_dms_formatted_infos(OBIDMS_p dms, bool detailed)
char* view_name = NULL;
char* tax_name = NULL;
char* all_tax_dir_path = NULL;
int i;
int i, last_dot_pos;
struct dirent* dp;
Obiview_p view;
@ -1439,17 +1439,21 @@ char* obi_dms_formatted_infos(OBIDMS_p dms, bool detailed)
if ((dp->d_name)[0] == '.')
continue;
i=0;
while ((dp->d_name)[i] != '.')
while (i < strlen(dp->d_name))
{
if ((dp->d_name)[i] == '.')
last_dot_pos = i;
i++;
view_name = (char*) malloc((i+1) * sizeof(char));
}
view_name = (char*) malloc((last_dot_pos+1) * sizeof(char));
if (view_name == NULL)
{
obi_set_errno(OBI_MALLOC_ERROR);
obidebug(1, "\nError allocating memory for a view name when getting formatted DMS infos: file %s", dp->d_name);
return NULL;
}
strncpy(view_name, dp->d_name, i);
view_name[i] = '\0';
strncpy(view_name, dp->d_name, last_dot_pos);
view_name[last_dot_pos] = '\0';
view = obi_open_view(dms, view_name);
if (view == NULL)
{

View File

@ -2910,7 +2910,7 @@ int obi_clean_unfinished_views(OBIDMS_p dms)
if ((dp->d_name)[0] == '.')
continue;
i=0;
while ((dp->d_name)[i] != '.')
while (strncmp((dp->d_name)+i, ".obiview", 8))
i++;
relative_path = (char*) malloc(strlen(VIEW_DIR_NAME) + strlen(dp->d_name) + 2);
strcpy(relative_path, VIEW_DIR_NAME);