crbanim.helpers.CRBAnimReader.check_sex() - Code Metrics - Inspection of ":sparkles: import from Template file" - cnr-ibba/IMAGE-InjectTool - Measure and Improve Code Quality continuously with Scrutinizer

Passed

Pull Request — master (#40)

by Paolo

created 2019-07-09 09:48 UTC

crbanim.helpers.CRBAnimReader.check_sex() A

↳ Parent: crbanim.helpers

Complexity

Conditions

Size

Total Lines	9
Code Lines	4

Duplication

Lines	0
Ratio	0 %

Importance

Changes

Metric	Value
eloc	4
dl	0
loc	9
rs	10
c	0
b	0
f	0
cc	1
nop	1

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Thu Feb 21 15:37:16 2019

@author: Paolo Cozzi <[email protected]>
"""

import io
import csv
import logging
import pycountry

from collections import defaultdict, namedtuple

from django.utils.dateparse import parse_date

from common.constants import LOADED, ERROR, MISSING, SAMPLE_STORAGE
from common.helpers import image_timedelta
from image_app.helpers import (
    FileDataSourceMixin, get_or_create_obj, update_or_create_obj)
from image_app.models import (
    DictSpecie, DictSex, DictCountry, DictBreed, Name, Animal, Sample,
    DictUberon, Publication)
from submissions.helpers import send_message
from validation.helpers import construct_validation_message
from validation.models import ValidationSummary

# Get an instance of a logger
logger = logging.getLogger(__name__)


# A class to deal with cryoweb import errors
class CRBAnimImportError(Exception):
    pass


class CRBAnimReader(FileDataSourceMixin):
    mandatory_columns = [
            'sex',
            'species_latin_name',
            'country_of_origin',
            'breed_name',
            'animal_ID',
            'sample_bibliographic_references',
            'sample_identifier',
            'animal_birth_date',
            'sample_storage_temperature',
            'sample_type_name',
            'body_part_name',
            'sampling_date',
            'sampling_protocol_url',
            'sample_availability',
            'EBI_Biosample_identifier',
        ]

    def __init__(self):
        self.data = None
        self.header = None
        self.dialect = None
        self.items = None
        self.filename = None

    @classmethod
    def get_dialect(cls, chunk):
        """Determine dialect of a CSV from a chunk"""

        return csv.Sniffer().sniff(chunk)

    @classmethod
    def is_valid(cls, chunk):
        """Try to determine if CRBanim has at least the required columns
        or not"""

        dialect = cls.get_dialect(chunk)

        # get a handle from a string
        handle = io.StringIO(chunk)

        # read chunk
        reader = csv.reader(handle, dialect)
        header = next(reader)

        not_found = []

        for column in cls.mandatory_columns:
            if column not in header:
                not_found.append(column)

        if len(not_found) == 0:
            logger.debug("This seems to be a valid CRBanim file")
            return True, []

        else:
            logger.error("Couldn't not find mandatory CRBanim columns %s" % (
                not_found))
            return False, not_found

    def read_file(self, filename):
        """Read crb anim files and set tit to class attribute"""

        with open(filename, newline='') as handle:
            # initialize data
            self.filename = filename
            self.data = []

            # get dialect
            chunk = handle.read(2048)
            self.dialect = self.get_dialect(chunk)

            # restart filename from the beginning
            handle.seek(0)

            # read csv file
            reader = csv.reader(handle, self.dialect)
            self.header = next(reader)

            # find sex index column
            sex_idx = self.header.index('sex')

            # create a namedtuple object
            Data = namedtuple("Data", self.header)

            # add records to data
            for record in reader:
                # replace all "\\N" occurences in a list
                record = [None if col in ["\\N", ""]
                          else col for col in record]

                # 'unknown' sex should be replaced with 'record of unknown sex'
                if record[sex_idx].lower() == 'unknown':
                    logger.debug(
                        "Changing '%s' with '%s'" % (
                            record[sex_idx], 'record of unknown sex'))
                    record[sex_idx] = 'record of unknown sex'

                record = Data._make(record)
                self.data.append(record)

        self.items = self.eval_columns()

    def eval_columns(self):
        """define a set from column data"""

        # target_columns = ['sex', 'species_latin_name', 'breed_name']
        target_columns = self.header

        items = defaultdict(list)

        for line in self.data:
            for column in target_columns:
                idx = self.header.index(column)
                items[column].append(line[idx])

        # now get a set of object
        for column in target_columns:
            items[column] = set(items[column])

        return items

    def print_line(self, num):
        """print a record with its column names"""

        for i, column in enumerate(self.header):
            logger.debug("%s: %s" % (column, self.data[num][i]))

    def filter_by_column_values(self, column, values, ignorecase=False):
        if ignorecase is True:
            # lower values
            values = [value.lower() for value in values]

        for line in self.data:
            # search for case insensitive value (lower attrib in lower values)
            if ignorecase is True:
                if getattr(line, column).lower() in values:
                    yield line

                else:
                    logger.debug("Filtering: %s" % (str(line)))

            else:
                if getattr(line, column) in values:
                    yield line

                else:
                    logger.debug("Filtering: %s" % (str(line)))

            # ignore case or not

        # cicle for line

    # a function to detect if crbanim species are in UID database or not
    def check_species(self, country):
        """Check if all species are defined in UID DictSpecies"""

        # CRBAnim usually have species in the form required for UID
        # However sometimes there could be a common name, not a DictSpecie one
        column = 'species_latin_name'
        item_set = self.items[column]

        # call FileDataSourceMixin.check_species
        return super().check_species(column, item_set, country)

    # check that dict sex table contains data
    def check_sex(self):
        """check that dict sex table contains data"""

        # item.sex are in uppercase
        column = 'sex'
        item_set = [item.lower() for item in self.items[column]]

        # call FileDataSourceMixin.check_items
        return self.check_items(item_set, DictSex, column)


def fill_uid_breed(record, language):
    """Fill DictBreed from a crbanim record"""

    # get a DictSpecie object. Species are in latin names, but I can
    # find also a common name in translation tables
    specie = DictSpecie.get_specie_check_synonyms(
            species_label=record.species_latin_name,
            language=language)

    # get country name using pycountries
    country_name = pycountry.countries.get(
        alpha_2=record.country_of_origin).name

    # get country for breeds. Ideally will be the same of submission,
    # however, it could be possible to store data from other contries
    country = get_or_create_obj(
        DictCountry,
        label=country_name)

    breed = get_or_create_obj(
        DictBreed,
        supplied_breed=record.breed_name,
        specie=specie,
        country=country)

    # return a DictBreed object
    return breed


def fill_uid_names(record, submission):
    """fill Names table from crbanim record"""

    # in the same record I have the sample identifier and animal identifier
    # a name record for animal
    animal_name = get_or_create_obj(
        Name,
        name=record.animal_ID,
        submission=submission,
        owner=submission.owner)

    # get a publication (if present)
    publication = None

    if record.sample_bibliographic_references:
        publication = get_or_create_obj(
            Publication,
            doi=record.sample_bibliographic_references)

    # name record for sample
    sample_name = get_or_create_obj(
        Name,
        name=record.sample_identifier,
        submission=submission,
        owner=submission.owner,
        publication=publication)

    # returning 2 Name instances
    return animal_name, sample_name


def fill_uid_animal(record, animal_name, breed, submission, animals):
    """Helper function to fill animal data in UID animal table"""

    # HINT: does CRBAnim models mother and father?

    # check if such animal is already beed updated
    if animal_name.name in animals:
        logger.debug(
            "Ignoring %s: already created or updated" % (animal_name))

        # return an animal object
        animal = animals[animal_name.name]

    else:
        # determine sex. Check for values
        sex = DictSex.objects.get(label__iexact=record.sex)

        # there's no birth_location for animal in CRBAnim
        accuracy = MISSING

        # create a new object. Using defaults to avoid collisions when
        # updating data
        # HINT: CRBanim has less attribute than cryoweb
        defaults = {
            # HINT: is a duplication of name. Can this be non-mandatory?
            'alternative_id': animal_name.name,
            'breed': breed,
            'sex': sex,
            'birth_date': record.animal_birth_date,
            'birth_location_accuracy': accuracy,
            'owner': submission.owner
        }

        # HINT: I could have the same animal again and again. Should I update
        # every times?
        animal = update_or_create_obj(
            Animal,
            name=animal_name,
            defaults=defaults)

        # track this animal in dictionary
        animals[animal_name.name] = animal

    # I need to track animal to relate the sample
    return animal


def find_storage_type(record):
    """Determine a sample storage relying on a dictionary"""

    mapping = {
        '-196°C': 'frozen, liquid nitrogen',
        '-20°C': 'frozen, -20 degrees Celsius freezer',
        '-30°C': 'frozen, -20 degrees Celsius freezer',
        '-80°C': 'frozen, -80 degrees Celsius freezer'}

    if record.sample_storage_temperature in mapping:
        # get ENUM conversion
        storage = SAMPLE_STORAGE.get_value_by_desc(
            mapping[record.sample_storage_temperature])

        return storage

    else:
        logging.warning("Couldn't find %s in storage types mapping" % (
            record.sample_storage_temperature))

        return None


def fill_uid_sample(record, sample_name, animal, submission):
    """Helper function to fill animal data in UID sample table"""

    # name and animal name come from parameters
    organism_part_label = None
    sample_type_name = record.sample_type_name.lower()
    body_part_name = record.body_part_name.lower()

    # sylvain has proposed to apply the following decision rule:
    if body_part_name != "unknown" and body_part_name != "not relevant":
        organism_part_label = body_part_name

    else:
        organism_part_label = sample_type_name

    # get a organism part. Organism parts need to be in lowercases
    organism_part = get_or_create_obj(
        DictUberon,
        label=organism_part_label
    )

    # calculate animal age at collection
    animal_birth_date = parse_date(record.animal_birth_date)
    sampling_date = parse_date(record.sampling_date)
    animal_age_at_collection, time_units = image_timedelta(
        sampling_date, animal_birth_date)

    # create a new object. Using defaults to avoid collisions when
    # updating data
    defaults = {
        # HINT: is a duplication of name. Can this be non-mandatory?
        'alternative_id': sample_name.name,
        'collection_date': record.sampling_date,
        'protocol': record.sampling_protocol_url,
        'organism_part': organism_part,
        'animal': animal,
        # 'description': v_vessel.comment,
        'owner': submission.owner,
        'storage': find_storage_type(record),
        'availability': record.sample_availability,
        'animal_age_at_collection': animal_age_at_collection,
        'animal_age_at_collection_units': time_units
    }

    sample = update_or_create_obj(
        Sample,
        name=sample_name,
        defaults=defaults)

    return sample


def process_record(record, submission, animals, language):
    # Peter mail 26/02/19 18:30: I agree that it sounds like we will
    # need to create sameAs BioSamples for the IMAGE project, and it makes
    # sense that the inject tool is able to do this.  It may be that we
    # tackle these cases after getting the main part of the inject tool
    # functioning and hold or ignore these existing BioSamples for now.
    # HINT: record with a biosample id should be ignored, for the moment
    if record.EBI_Biosample_identifier is not None:
        logger.warning("Ignoring %s: already in biosample!" % str(record))
        return

    # filling breeds
    breed = fill_uid_breed(record, language)

    # filling name tables
    animal_name, sample_name = fill_uid_names(record, submission)

    # fill animal
    animal = fill_uid_animal(record, animal_name, breed, submission, animals)

    # fill sample
    fill_uid_sample(record, sample_name, animal, submission)


def upload_crbanim(submission):
    # debug
    logger.info("Importing from CRB-Anim file")

    # this is the full path in docker container
    fullpath = submission.get_uploaded_file_path()

    # read submission data
    reader = CRBAnimReader()
    reader.read_file(fullpath)

    # start data loading
    try:
        # check for species and sex in a similar way as cryoweb does
        check, not_found = reader.check_sex()

        if not check:
            message = (
                "Not all Sex terms are loaded into database: "
                "check for %s in your dataset" % (not_found))

            raise CRBAnimImportError(message)

        check, not_found = reader.check_species(submission.gene_bank_country)

        if not check:
            raise CRBAnimImportError(
                "Some species are not loaded in UID database: "
                "%s" % (not_found))

        # ok get languages from submission (useful for translation)
        # HINT: no traslations implemented, at the moment
        language = submission.gene_bank_country.label

        # a dictionary in which store animal data
        animals = {}

        for record in reader.data:
            process_record(record, submission, animals, language)

        # after processing records, initilize validationsummary objects
        # create a validation summary object and set all_count
        vs_animal = get_or_create_obj(
            ValidationSummary,
            submission=submission,
            type="animal")

        # reset counts
        vs_animal.reset_all_count()

        vs_sample = get_or_create_obj(
            ValidationSummary,
            submission=submission,
            type="sample")

        # reset counts
        vs_sample.reset_all_count()

    except Exception as exc:
        # set message:
        message = "Error in importing data: %s" % (str(exc))

        # save a message in database
        submission.status = ERROR
        submission.message = message
        submission.save()

        # send async message
        send_message(submission)

        # debug
        logger.error("error in importing from crbanim: %s" % (exc))
        logger.exception(exc)

        return False

    else:
        message = "CRBAnim import completed for submission: %s" % (
            submission.id)

        submission.message = message
        submission.status = LOADED
        submission.save()

        # send async message
        send_message(
            submission,
            validation_message=construct_validation_message(submission))

    logger.info("Import from CRBAnim is complete")

    return True


1			#!/usr/bin/env python3
2			# -- coding: utf-8 --
3			"""
4			Created on Thu Feb 21 15:37:16 2019
5
6			@author: Paolo Cozzi <[email protected]>
7			"""
8
9			import io
10			import csv
11			import logging
12			import pycountry
13
14			from collections import defaultdict, namedtuple
15
16			from django.utils.dateparse import parse_date
17
18			from common.constants import LOADED, ERROR, MISSING, SAMPLE_STORAGE
19			from common.helpers import image_timedelta
20			from image_app.helpers import (
21			FileDataSourceMixin, get_or_create_obj, update_or_create_obj)
22			from image_app.models import (
23			DictSpecie, DictSex, DictCountry, DictBreed, Name, Animal, Sample,
24			DictUberon, Publication)
25			from submissions.helpers import send_message
26			from validation.helpers import construct_validation_message
27			from validation.models import ValidationSummary
28
29			# Get an instance of a logger
30			logger = logging.getLogger(__name__)
31
32
33			# A class to deal with cryoweb import errors
34			class CRBAnimImportError(Exception):
35			pass
36
37
38			class CRBAnimReader(FileDataSourceMixin):
39			mandatory_columns = [
40			'sex',
41			'species_latin_name',
42			'country_of_origin',
43			'breed_name',
44			'animal_ID',
45			'sample_bibliographic_references',
46			'sample_identifier',
47			'animal_birth_date',
48			'sample_storage_temperature',
49			'sample_type_name',
50			'body_part_name',
51			'sampling_date',
52			'sampling_protocol_url',
53			'sample_availability',
54			'EBI_Biosample_identifier',
55			]
56
57			def __init__(self):
58			self.data = None
59			self.header = None
60			self.dialect = None
61			self.items = None
62			self.filename = None
63
64			@classmethod
65			def get_dialect(cls, chunk):
66			"""Determine dialect of a CSV from a chunk"""
67
68			return csv.Sniffer().sniff(chunk)
69
70			@classmethod
71			def is_valid(cls, chunk):
72			"""Try to determine if CRBanim has at least the required columns
73			or not"""
74
75			dialect = cls.get_dialect(chunk)
76
77			# get a handle from a string
78			handle = io.StringIO(chunk)
79
80			# read chunk
81			reader = csv.reader(handle, dialect)
82			header = next(reader)
83
84			not_found = []
85
86			for column in cls.mandatory_columns:
87			if column not in header:
88			not_found.append(column)
89
90			if len(not_found) == 0:
91			logger.debug("This seems to be a valid CRBanim file")
92			return True, []
93
94			else:
95			logger.error("Couldn't not find mandatory CRBanim columns %s" % (
96			not_found))
97			return False, not_found
98
99			def read_file(self, filename):
100			"""Read crb anim files and set tit to class attribute"""
101
102			with open(filename, newline='') as handle:
103			# initialize data
104			self.filename = filename
105			self.data = []
106
107			# get dialect
108			chunk = handle.read(2048)
109			self.dialect = self.get_dialect(chunk)
110
111			# restart filename from the beginning
112			handle.seek(0)
113
114			# read csv file
115			reader = csv.reader(handle, self.dialect)
116			self.header = next(reader)
117
118			# find sex index column
119			sex_idx = self.header.index('sex')
120
121			# create a namedtuple object
122			Data = namedtuple("Data", self.header)
123
124			# add records to data
125			for record in reader:
126			# replace all "\\N" occurences in a list
127			record = [None if col in ["\\N", ""]
128			else col for col in record]
129
130			# 'unknown' sex should be replaced with 'record of unknown sex'
131			if record[sex_idx].lower() == 'unknown':
132			logger.debug(
133			"Changing '%s' with '%s'" % (
134			record[sex_idx], 'record of unknown sex'))
135			record[sex_idx] = 'record of unknown sex'
136
137			record = Data._make(record)
138			self.data.append(record)
139
140			self.items = self.eval_columns()
141
142			def eval_columns(self):
143			"""define a set from column data"""
144
145			# target_columns = ['sex', 'species_latin_name', 'breed_name']
146			target_columns = self.header
147
148			items = defaultdict(list)
149
150			for line in self.data:
151			for column in target_columns:
152			idx = self.header.index(column)
153			items[column].append(line[idx])
154
155			# now get a set of object
156			for column in target_columns:
157			items[column] = set(items[column])
158
159			return items
160
161			def print_line(self, num):
162			"""print a record with its column names"""
163
164			for i, column in enumerate(self.header):
165			logger.debug("%s: %s" % (column, self.data[num][i]))
166
167			def filter_by_column_values(self, column, values, ignorecase=False):
168			if ignorecase is True:
169			# lower values
170			values = [value.lower() for value in values]
171
172			for line in self.data:
173			# search for case insensitive value (lower attrib in lower values)
174			if ignorecase is True:
175			if getattr(line, column).lower() in values:
176			yield line
177
178			else:
179			logger.debug("Filtering: %s" % (str(line)))
180
181			else:
182			if getattr(line, column) in values:
183			yield line
184
185			else:
186			logger.debug("Filtering: %s" % (str(line)))
187
188			# ignore case or not
189
190			# cicle for line
191
192			# a function to detect if crbanim species are in UID database or not
193			def check_species(self, country):
194			"""Check if all species are defined in UID DictSpecies"""
195
196			# CRBAnim usually have species in the form required for UID
197			# However sometimes there could be a common name, not a DictSpecie one
198			column = 'species_latin_name'
199			item_set = self.items[column]
200
201			# call FileDataSourceMixin.check_species
202			return super().check_species(column, item_set, country)
203
204			# check that dict sex table contains data
205			def check_sex(self):
206			"""check that dict sex table contains data"""
207
208			# item.sex are in uppercase
209			column = 'sex'
210			item_set = [item.lower() for item in self.items[column]]
211
212			# call FileDataSourceMixin.check_items
213			return self.check_items(item_set, DictSex, column)
214
215
216			def fill_uid_breed(record, language):
217			"""Fill DictBreed from a crbanim record"""
218
219			# get a DictSpecie object. Species are in latin names, but I can
220			# find also a common name in translation tables
221			specie = DictSpecie.get_specie_check_synonyms(
222			species_label=record.species_latin_name,
223			language=language)
224
225			# get country name using pycountries
226			country_name = pycountry.countries.get(
227			alpha_2=record.country_of_origin).name
228
229			# get country for breeds. Ideally will be the same of submission,
230			# however, it could be possible to store data from other contries
231			country = get_or_create_obj(
232			DictCountry,
233			label=country_name)
234
235			breed = get_or_create_obj(
236			DictBreed,
237			supplied_breed=record.breed_name,
238			specie=specie,
239			country=country)
240
241			# return a DictBreed object
242			return breed
243
244
245			def fill_uid_names(record, submission):
246			"""fill Names table from crbanim record"""
247
248			# in the same record I have the sample identifier and animal identifier
249			# a name record for animal
250			animal_name = get_or_create_obj(
251			Name,
252			name=record.animal_ID,
253			submission=submission,
254			owner=submission.owner)
255
256			# get a publication (if present)
257			publication = None
258
259			if record.sample_bibliographic_references:
260			publication = get_or_create_obj(
261			Publication,
262			doi=record.sample_bibliographic_references)
263
264			# name record for sample
265			sample_name = get_or_create_obj(
266			Name,
267			name=record.sample_identifier,
268			submission=submission,
269			owner=submission.owner,
270			publication=publication)
271
272			# returning 2 Name instances
273			return animal_name, sample_name
274
275
276			def fill_uid_animal(record, animal_name, breed, submission, animals):
277			"""Helper function to fill animal data in UID animal table"""
278
279			# HINT: does CRBAnim models mother and father?
280
281			# check if such animal is already beed updated
282			if animal_name.name in animals:
283			logger.debug(
284			"Ignoring %s: already created or updated" % (animal_name))
285
286			# return an animal object
287			animal = animals[animal_name.name]
288
289			else:
290			# determine sex. Check for values
291			sex = DictSex.objects.get(label__iexact=record.sex)
292
293			# there's no birth_location for animal in CRBAnim
294			accuracy = MISSING
295
296			# create a new object. Using defaults to avoid collisions when
297			# updating data
298			# HINT: CRBanim has less attribute than cryoweb
299			defaults = {
300			# HINT: is a duplication of name. Can this be non-mandatory?
301			'alternative_id': animal_name.name,
302			'breed': breed,
303			'sex': sex,
304			'birth_date': record.animal_birth_date,
305			'birth_location_accuracy': accuracy,
306			'owner': submission.owner
307			}
308
309			# HINT: I could have the same animal again and again. Should I update
310			# every times?
311			animal = update_or_create_obj(
312			Animal,
313			name=animal_name,
314			defaults=defaults)
315
316			# track this animal in dictionary
317			animals[animal_name.name] = animal
318
319			# I need to track animal to relate the sample
320			return animal
321
322
323			def find_storage_type(record):
324			"""Determine a sample storage relying on a dictionary"""
325
326			mapping = {
327			'-196°C': 'frozen, liquid nitrogen',
328			'-20°C': 'frozen, -20 degrees Celsius freezer',
329			'-30°C': 'frozen, -20 degrees Celsius freezer',
330			'-80°C': 'frozen, -80 degrees Celsius freezer'}
331
332			if record.sample_storage_temperature in mapping:
333			# get ENUM conversion
334			storage = SAMPLE_STORAGE.get_value_by_desc(
335			mapping[record.sample_storage_temperature])
336
337			return storage
338
339			else:
340			logging.warning("Couldn't find %s in storage types mapping" % (
341			record.sample_storage_temperature))
342
343			return None
344
345
346			def fill_uid_sample(record, sample_name, animal, submission):
347			"""Helper function to fill animal data in UID sample table"""
348
349			# name and animal name come from parameters
350			organism_part_label = None
351			sample_type_name = record.sample_type_name.lower()
352			body_part_name = record.body_part_name.lower()
353
354			# sylvain has proposed to apply the following decision rule:
355			if body_part_name != "unknown" and body_part_name != "not relevant":
356			organism_part_label = body_part_name
357
358			else:
359			organism_part_label = sample_type_name
360
361			# get a organism part. Organism parts need to be in lowercases
362			organism_part = get_or_create_obj(
363			DictUberon,
364			label=organism_part_label
365			)
366
367			# calculate animal age at collection
368			animal_birth_date = parse_date(record.animal_birth_date)
369			sampling_date = parse_date(record.sampling_date)
370			animal_age_at_collection, time_units = image_timedelta(
371			sampling_date, animal_birth_date)
372
373			# create a new object. Using defaults to avoid collisions when
374			# updating data
375			defaults = {
376			# HINT: is a duplication of name. Can this be non-mandatory?
377			'alternative_id': sample_name.name,
378			'collection_date': record.sampling_date,
379			'protocol': record.sampling_protocol_url,
380			'organism_part': organism_part,
381			'animal': animal,
382			# 'description': v_vessel.comment,
383			'owner': submission.owner,
384			'storage': find_storage_type(record),
385			'availability': record.sample_availability,
386			'animal_age_at_collection': animal_age_at_collection,
387			'animal_age_at_collection_units': time_units
388			}
389
390			sample = update_or_create_obj(
391			Sample,
392			name=sample_name,
393			defaults=defaults)
394
395			return sample
396
397
398			def process_record(record, submission, animals, language):
399			# Peter mail 26/02/19 18:30: I agree that it sounds like we will
400			# need to create sameAs BioSamples for the IMAGE project, and it makes
401			# sense that the inject tool is able to do this. It may be that we
402			# tackle these cases after getting the main part of the inject tool
403			# functioning and hold or ignore these existing BioSamples for now.
404			# HINT: record with a biosample id should be ignored, for the moment
405			if record.EBI_Biosample_identifier is not None:
406			logger.warning("Ignoring %s: already in biosample!" % str(record))
407			return
408
409			# filling breeds
410			breed = fill_uid_breed(record, language)
411
412			# filling name tables
413			animal_name, sample_name = fill_uid_names(record, submission)
414
415			# fill animal
416			animal = fill_uid_animal(record, animal_name, breed, submission, animals)
417
418			# fill sample
419			fill_uid_sample(record, sample_name, animal, submission)
420
421
422			def upload_crbanim(submission):
423			# debug
424			logger.info("Importing from CRB-Anim file")
425
426			# this is the full path in docker container
427			fullpath = submission.get_uploaded_file_path()
428
429			# read submission data
430			reader = CRBAnimReader()
431			reader.read_file(fullpath)
432
433			# start data loading
434			try:
435			# check for species and sex in a similar way as cryoweb does
436			check, not_found = reader.check_sex()
437
438			if not check:
439			message = (
440			"Not all Sex terms are loaded into database: "
441			"check for %s in your dataset" % (not_found))
442
443			raise CRBAnimImportError(message)
444
445			check, not_found = reader.check_species(submission.gene_bank_country)
446
447			if not check:
448			raise CRBAnimImportError(
449			"Some species are not loaded in UID database: "
450			"%s" % (not_found))
451
452			# ok get languages from submission (useful for translation)
453			# HINT: no traslations implemented, at the moment
454			language = submission.gene_bank_country.label
455
456			# a dictionary in which store animal data
457			animals = {}
458
459			for record in reader.data:
460			process_record(record, submission, animals, language)
461
462			# after processing records, initilize validationsummary objects
463			# create a validation summary object and set all_count
464			vs_animal = get_or_create_obj(
465			ValidationSummary,
466			submission=submission,
467			type="animal")
468
469			# reset counts
470			vs_animal.reset_all_count()
471
472			vs_sample = get_or_create_obj(
473			ValidationSummary,
474			submission=submission,
475			type="sample")
476
477			# reset counts
478			vs_sample.reset_all_count()
479
480			except Exception as exc:
481			# set message:
482			message = "Error in importing data: %s" % (str(exc))
483
484			# save a message in database
485			submission.status = ERROR
486			submission.message = message
487			submission.save()
488
489			# send async message
490			send_message(submission)
491
492			# debug
493			logger.error("error in importing from crbanim: %s" % (exc))
494			logger.exception(exc)
495
496			return False
497
498			else:
499			message = "CRBAnim import completed for submission: %s" % (
500			submission.id)
501
502			submission.message = message
503			submission.status = LOADED
504			submission.save()
505
506			# send async message
507			send_message(
508			submission,
509			validation_message=construct_validation_message(submission))
510
511			logger.info("Import from CRBAnim is complete")
512
513			return True
514

cnr-ibba / IMAGE-InjectTool

Pull Request — master (#40)

crbanim.helpers.CRBAnimReader.check_sex() A

Complexity

Size

Duplication

Importance

Duplication Side-by-Side

Filter issues like