Commit 21d1b2ed by Celine Mercier

First implementation of taxonomy reading

parent a08def47
......@@ -20,3 +20,5 @@
../../../src/encode.c
../../../src/obidmscolumn_idx.h
../../../src/obidmscolumn_idx.c
../../../src/obidms_taxonomy.c
../../../src/obidms_taxonomy.h
......@@ -4,7 +4,7 @@ from .capi.obidms cimport OBIDMS_p
from .capi.obidmscolumn cimport OBIDMS_column_p
from .capi.obiview cimport Obiview_p
from .capi.obitypes cimport obiversion_t, OBIType_t, index_t
from ._obitaxo cimport OBI_Taxonomy
cdef class OBIDMS_column:
......@@ -86,6 +86,7 @@ cdef class OBIDMS:
cdef str dms_name
cpdef close(self)
cpdef OBI_Taxonomy open_taxonomy(self, str taxo_name)
cpdef OBIView open_view(self, str view_name)
cpdef OBIView new_view(self, str view_name, object view_to_clone=*, list line_selection=*, str view_type=*, str comments=*)
cpdef dict read_view_infos(self, str view_name)
......
......@@ -26,6 +26,8 @@ from ._obidms cimport OBIDMS, \
OBIView, \
OBIView_line
from ._obitaxo cimport OBI_Taxonomy
from ._obiseq cimport OBI_Nuc_Seq, OBI_Nuc_Seq_Stored
from ._obidmscolumn_int cimport OBIDMS_column_int, \
......@@ -243,7 +245,8 @@ cdef class OBIView :
cdef OBIDMS_column_p* column_pp
cdef OBIDMS_column_header_p header
cdef index_t* line_selection_p
cdef object col_capsule
self.dms = dms
if line_selection is not None :
......@@ -649,6 +652,10 @@ cdef class OBIDMS :
if (obi_close_dms(self.pointer)) < 0 :
raise Exception("Problem closing an OBIDMS")
cpdef OBI_Taxonomy open_taxonomy(self, str taxo_name) :
return OBI_Taxonomy(self, taxo_name)
cpdef OBIView open_view(self, str view_name) :
......
../../../src/obidms.h
../../../src/obidms.c
../../../src/obiview.h
../../../src/obiview.c
../../../src/obidmscolumn.h
../../../src/obidmscolumn.c
../../../src/obidmscolumndir.h
../../../src/obidmscolumndir.c
../../../src/obierrno.h
../../../src/obierrno.c
../../../src/obilittlebigman.h
../../../src/obilittlebigman.c
../../../src/obitypes.h
../../../src/obitypes.c
../../../src/private_at_functions.h
../../../src/private_at_functions.c
../../../src/obiavl.h
../../../src/obiavl.c
../../../src/encode.h
../../../src/encode.c
../../../src/obidmscolumn_idx.h
../../../src/obidmscolumn_idx.c
../../../src/obidms_taxonomy.c
../../../src/obidms_taxonomy.h
#cython: language_level=3
from .capi.obitaxonomy cimport ecotx_t, OBIDMS_taxonomy_p
from libc.stdint cimport int32_t
cdef class OBI_Taxonomy :
cdef str name
cdef OBIDMS_taxonomy_p pointer
cpdef close(self)
cdef class OBI_Taxon :
cdef ecotx_t* pointer
cdef int32_t taxid
cdef int32_t rank
cdef int32_t farest
cdef ecotx_t* parent
cdef str name
cpdef int32_t taxid(self)
cpdef int32_t rank(self)
cpdef int32_t farest(self)
cpdef OBI_Taxon parent(self)
#cython: language_level=3
from obitools3.utils cimport bytes2str, str2bytes
from .capi.obitaxonomy cimport obi_read_taxonomy, \
obi_close_taxonomy, \
obi_taxo_get_taxon_with_taxid
from ._obidms cimport OBIDMS
from cpython.pycapsule cimport PyCapsule_New, PyCapsule_GetPointer
cdef class OBI_Taxonomy :
def __init__(self, OBIDMS dms, str name) :
self.name = name
self.pointer = obi_read_taxonomy(dms.pointer, str2bytes(name), True) # TODO discuss
def __getitem__(self, object ref):
cdef ecotx_t* taxon_p
cdef object taxon_capsule
if type(ref) == int :
taxon_p = obi_taxo_get_taxon_with_taxid(self.pointer, ref)
taxon_capsule = PyCapsule_New(taxon_p, NULL, NULL)
return OBI_Taxon(taxon_capsule)
cpdef close(self) :
if (obi_close_taxonomy(self.pointer) < 0) :
raise Exception("Error closing the taxonomy")
cdef class OBI_Taxon : # dict subclass?
def __init__(self, object taxon_capsule) :
cdef ecotx_t* taxon
taxon = <ecotx_t*> PyCapsule_GetPointer(taxon_capsule, NULL)
self.pointer = taxon
self.taxid = taxon.taxid
self.rank = taxon.rank
self.farest = taxon.farest
self.parent = taxon.parent
self.name = bytes2str(taxon.name)
cpdef int32_t taxid(self):
return self.taxid
cpdef int32_t rank(self):
return self.rank
cpdef int32_t farest(self):
return self.farest
cpdef OBI_Taxon parent(self):
cdef object parent_capsule
parent_capsule = PyCapsule_New(self.parent, NULL, NULL)
return OBI_Taxon(parent_capsule)
#cython: language_level=3
from .obitypes cimport const_char_p
from .obidms cimport OBIDMS_p
from libc.stdint cimport int32_t
cdef extern from "obidms_taxonomy.h" nogil:
struct OBIDMS_taxonomy_t
ctypedef OBIDMS_taxonomy_t* OBIDMS_taxonomy_p
struct ecotxnode :
int32_t taxid
int32_t rank
int32_t farest
ecotxnode* parent
char* name
ctypedef ecotxnode ecotx_t
OBIDMS_taxonomy_p obi_read_taxonomy(OBIDMS_p dms, const_char_p taxonomy_name, bint read_alternative_names)
int obi_close_taxonomy(OBIDMS_taxonomy_p taxonomy)
ecotx_t* obi_taxo_get_parent_at_rank(ecotx_t* taxon, int32_t rankidx)
ecotx_t* obi_taxo_get_taxon_with_taxid(OBIDMS_taxonomy_p taxonomy, int32_t taxid)
bint obi_taxo_is_taxon_under_taxid(ecotx_t* taxon, int32_t other_taxid)
ecotx_t* obi_taxo_get_species(ecotx_t* taxon, OBIDMS_taxonomy_p taxonomy)
ecotx_t* obi_taxo_get_genus(ecotx_t* taxon, OBIDMS_taxonomy_p taxonomy)
ecotx_t* obi_taxo_get_family(ecotx_t* taxon, OBIDMS_taxonomy_p taxonomy)
ecotx_t* obi_taxo_get_kingdom(ecotx_t* taxon, OBIDMS_taxonomy_p taxonomy)
ecotx_t* obi_taxo_get_superkingdom(ecotx_t* taxon, OBIDMS_taxonomy_p taxonomy)
......@@ -29,6 +29,8 @@
*/
#define AVL_TREES_DIR_NAME "AVL_trees" /**< The name of the AVL trees directory.
*/
#define TAXONOMY_DIR_NAME "TAXONOMY" /**< The name of the taxonomy directory.
*/
#define MAX_NB_OPENED_COLUMNS (100) /**< The maximum number of columns open at the same time.
*/
#define MAX_NB_OPENED_AVL_TREES (100) /**< The maximum number of AVL trees open at the same time.
......
/********************************************************************
* OBIDMS taxonomy headeer file *
********************************************************************/
/**
* @file obidms_taxonomy.h
* @author Celine Mercier (celine.mercier@metabarcoding.org)
* @date March 2nd 2016
* @brief Header file for the functions handling the reading of binary taxonomy files.
*/
#include <stdlib.h>
#include <stdio.h>
#include <stdbool.h>
#include "obidms.h"
typedef struct {
int32_t taxid;
int32_t rank;
int32_t parent;
int32_t name_length;
char name[1];
} ecotxformat_t;
typedef struct ecotxnode {
int32_t taxid;
int32_t rank;
int32_t farest;
struct ecotxnode* parent;
char* name;
} ecotx_t;
typedef struct {
int32_t count;
int32_t max_taxid;
int32_t buffer_size;
ecotx_t taxon[1];
} ecotxidx_t;
typedef struct {
int32_t count;
char* label[1];
} ecorankidx_t;
typedef struct {
int32_t is_scientific_name;
int32_t name_length;
int32_t class_length;
int32_t taxid;
char names[1];
} econameformat_t;
typedef struct {
char* name;
char* class_name;
int32_t is_scientific_name;
struct ecotxnode* taxon;
} econame_t;
typedef struct {
int32_t count;
econame_t names[1];
} econameidx_t;
typedef struct OBIDMS_taxonomy_t {
ecorankidx_t* ranks;
econameidx_t* names;
ecotxidx_t* taxa;
} OBIDMS_taxonomy_t, *OBIDMS_taxonomy_p;
OBIDMS_taxonomy_p obi_read_taxonomy(OBIDMS_p dms, const char* taxonomy_name, bool read_alternative_names);
int obi_close_taxonomy(OBIDMS_taxonomy_p taxonomy);
ecotx_t* obi_taxo_get_parent_at_rank(ecotx_t* taxon, int32_t rankidx);
ecotx_t* obi_taxo_get_taxon_with_taxid(OBIDMS_taxonomy_p taxonomy, int32_t taxid);
bool obi_taxo_is_taxon_under_taxid(ecotx_t* taxon, int32_t other_taxid);
ecotx_t* obi_taxo_get_species(ecotx_t* taxon, OBIDMS_taxonomy_p taxonomy);
ecotx_t* obi_taxo_get_genus(ecotx_t* taxon, OBIDMS_taxonomy_p taxonomy);
ecotx_t* obi_taxo_get_family(ecotx_t* taxon, OBIDMS_taxonomy_p taxonomy);
ecotx_t* obi_taxo_get_kingdom(ecotx_t* taxon, OBIDMS_taxonomy_p taxonomy);
ecotx_t* obi_taxo_get_superkingdom(ecotx_t* taxon, OBIDMS_taxonomy_p taxonomy);
......@@ -102,6 +102,8 @@ extern int obi_errno;
*/
#define OBIVIEW_ERROR (21) /** Error while handling an OBIView
*/
#define OBI_TAXONOMY_ERROR (22) /** Error while handling binary taxonomy files
*/
/**@}*/
#endif /* OBIERRNO_H_ */
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment