crbanim.helpers.sanitize_url() - Code Metrics - Inspection of ":arrow_up: upgrade image_validation library" - cnr-ibba/IMAGE-InjectTool - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Push — master ( 8738db...eecf08 )

by Paolo

created 2019-10-03 09:30 UTC

crbanim.helpers.sanitize_url() A

↳ Parent: crbanim.helpers

Complexity

Conditions

Size

Total Lines	4
Code Lines	2

Duplication

Lines	0
Ratio	0 %

Importance

Changes

Metric	Value
eloc	2
dl	0
loc	4
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 urllib
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 sanitize_url(url):
    """Quote URLs for accession"""

    return urllib.parse.quote(url, ':/#?=')


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': sanitize_url(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 urllib
12			import logging
13			import pycountry
14
15			from collections import defaultdict, namedtuple
16
17			from django.utils.dateparse import parse_date
18
19			from common.constants import LOADED, ERROR, MISSING, SAMPLE_STORAGE
20			from common.helpers import image_timedelta
21			from image_app.helpers import (
22			FileDataSourceMixin, get_or_create_obj, update_or_create_obj)
23			from image_app.models import (
24			DictSpecie, DictSex, DictCountry, DictBreed, Name, Animal, Sample,
25			DictUberon, Publication)
26			from submissions.helpers import send_message
27			from validation.helpers import construct_validation_message
28			from validation.models import ValidationSummary
29
30			# Get an instance of a logger
31			logger = logging.getLogger(__name__)
32
33
34			# A class to deal with cryoweb import errors
35			class CRBAnimImportError(Exception):
36			pass
37
38
39			class CRBAnimReader(FileDataSourceMixin):
40			mandatory_columns = [
41			'sex',
42			'species_latin_name',
43			'country_of_origin',
44			'breed_name',
45			'animal_ID',
46			'sample_bibliographic_references',
47			'sample_identifier',
48			'animal_birth_date',
49			'sample_storage_temperature',
50			'sample_type_name',
51			'body_part_name',
52			'sampling_date',
53			'sampling_protocol_url',
54			'sample_availability',
55			'EBI_Biosample_identifier',
56			]
57
58			def __init__(self):
59			self.data = None
60			self.header = None
61			self.dialect = None
62			self.items = None
63			self.filename = None
64
65			@classmethod
66			def get_dialect(cls, chunk):
67			"""Determine dialect of a CSV from a chunk"""
68
69			return csv.Sniffer().sniff(chunk)
70
71			@classmethod
72			def is_valid(cls, chunk):
73			"""Try to determine if CRBanim has at least the required columns
74			or not"""
75
76			dialect = cls.get_dialect(chunk)
77
78			# get a handle from a string
79			handle = io.StringIO(chunk)
80
81			# read chunk
82			reader = csv.reader(handle, dialect)
83			header = next(reader)
84
85			not_found = []
86
87			for column in cls.mandatory_columns:
88			if column not in header:
89			not_found.append(column)
90
91			if len(not_found) == 0:
92			logger.debug("This seems to be a valid CRBanim file")
93			return True, []
94
95			else:
96			logger.error("Couldn't not find mandatory CRBanim columns %s" % (
97			not_found))
98			return False, not_found
99
100			def read_file(self, filename):
101			"""Read crb anim files and set tit to class attribute"""
102
103			with open(filename, newline='') as handle:
104			# initialize data
105			self.filename = filename
106			self.data = []
107
108			# get dialect
109			chunk = handle.read(2048)
110			self.dialect = self.get_dialect(chunk)
111
112			# restart filename from the beginning
113			handle.seek(0)
114
115			# read csv file
116			reader = csv.reader(handle, self.dialect)
117			self.header = next(reader)
118
119			# find sex index column
120			sex_idx = self.header.index('sex')
121
122			# create a namedtuple object
123			Data = namedtuple("Data", self.header)
124
125			# add records to data
126			for record in reader:
127			# replace all "\\N" occurences in a list
128			record = [None if col in ["\\N", ""]
129			else col for col in record]
130
131			# 'unknown' sex should be replaced with 'record of unknown sex'
132			if record[sex_idx].lower() == 'unknown':
133			logger.debug(
134			"Changing '%s' with '%s'" % (
135			record[sex_idx], 'record of unknown sex'))
136			record[sex_idx] = 'record of unknown sex'
137
138			record = Data._make(record)
139			self.data.append(record)
140
141			self.items = self.eval_columns()
142
143			def eval_columns(self):
144			"""define a set from column data"""
145
146			# target_columns = ['sex', 'species_latin_name', 'breed_name']
147			target_columns = self.header
148
149			items = defaultdict(list)
150
151			for line in self.data:
152			for column in target_columns:
153			idx = self.header.index(column)
154			items[column].append(line[idx])
155
156			# now get a set of object
157			for column in target_columns:
158			items[column] = set(items[column])
159
160			return items
161
162			def print_line(self, num):
163			"""print a record with its column names"""
164
165			for i, column in enumerate(self.header):
166			logger.debug("%s: %s" % (column, self.data[num][i]))
167
168			def filter_by_column_values(self, column, values, ignorecase=False):
169			if ignorecase is True:
170			# lower values
171			values = [value.lower() for value in values]
172
173			for line in self.data:
174			# search for case insensitive value (lower attrib in lower values)
175			if ignorecase is True:
176			if getattr(line, column).lower() in values:
177			yield line
178
179			else:
180			logger.debug("Filtering: %s" % (str(line)))
181
182			else:
183			if getattr(line, column) in values:
184			yield line
185
186			else:
187			logger.debug("Filtering: %s" % (str(line)))
188
189			# ignore case or not
190
191			# cicle for line
192
193			# a function to detect if crbanim species are in UID database or not
194			def check_species(self, country):
195			"""Check if all species are defined in UID DictSpecies"""
196
197			# CRBAnim usually have species in the form required for UID
198			# However sometimes there could be a common name, not a DictSpecie one
199			column = 'species_latin_name'
200			item_set = self.items[column]
201
202			# call FileDataSourceMixin.check_species
203			return super().check_species(column, item_set, country)
204
205			# check that dict sex table contains data
206			def check_sex(self):
207			"""check that dict sex table contains data"""
208
209			# item.sex are in uppercase
210			column = 'sex'
211			item_set = [item.lower() for item in self.items[column]]
212
213			# call FileDataSourceMixin.check_items
214			return self.check_items(item_set, DictSex, column)
215
216
217			def fill_uid_breed(record, language):
218			"""Fill DictBreed from a crbanim record"""
219
220			# get a DictSpecie object. Species are in latin names, but I can
221			# find also a common name in translation tables
222			specie = DictSpecie.get_specie_check_synonyms(
223			species_label=record.species_latin_name,
224			language=language)
225
226			# get country name using pycountries
227			country_name = pycountry.countries.get(
228			alpha_2=record.country_of_origin).name
229
230			# get country for breeds. Ideally will be the same of submission,
231			# however, it could be possible to store data from other contries
232			country = get_or_create_obj(
233			DictCountry,
234			label=country_name)
235
236			breed = get_or_create_obj(
237			DictBreed,
238			supplied_breed=record.breed_name,
239			specie=specie,
240			country=country)
241
242			# return a DictBreed object
243			return breed
244
245
246			def fill_uid_names(record, submission):
247			"""fill Names table from crbanim record"""
248
249			# in the same record I have the sample identifier and animal identifier
250			# a name record for animal
251			animal_name = get_or_create_obj(
252			Name,
253			name=record.animal_ID,
254			submission=submission,
255			owner=submission.owner)
256
257			# get a publication (if present)
258			publication = None
259
260			if record.sample_bibliographic_references:
261			publication = get_or_create_obj(
262			Publication,
263			doi=record.sample_bibliographic_references)
264
265			# name record for sample
266			sample_name = get_or_create_obj(
267			Name,
268			name=record.sample_identifier,
269			submission=submission,
270			owner=submission.owner,
271			publication=publication)
272
273			# returning 2 Name instances
274			return animal_name, sample_name
275
276
277			def fill_uid_animal(record, animal_name, breed, submission, animals):
278			"""Helper function to fill animal data in UID animal table"""
279
280			# HINT: does CRBAnim models mother and father?
281
282			# check if such animal is already beed updated
283			if animal_name.name in animals:
284			logger.debug(
285			"Ignoring %s: already created or updated" % (animal_name))
286
287			# return an animal object
288			animal = animals[animal_name.name]
289
290			else:
291			# determine sex. Check for values
292			sex = DictSex.objects.get(label__iexact=record.sex)
293
294			# there's no birth_location for animal in CRBAnim
295			accuracy = MISSING
296
297			# create a new object. Using defaults to avoid collisions when
298			# updating data
299			# HINT: CRBanim has less attribute than cryoweb
300			defaults = {
301			# HINT: is a duplication of name. Can this be non-mandatory?
302			'alternative_id': animal_name.name,
303			'breed': breed,
304			'sex': sex,
305			'birth_date': record.animal_birth_date,
306			'birth_location_accuracy': accuracy,
307			'owner': submission.owner
308			}
309
310			# HINT: I could have the same animal again and again. Should I update
311			# every times?
312			animal = update_or_create_obj(
313			Animal,
314			name=animal_name,
315			defaults=defaults)
316
317			# track this animal in dictionary
318			animals[animal_name.name] = animal
319
320			# I need to track animal to relate the sample
321			return animal
322
323
324			def find_storage_type(record):
325			"""Determine a sample storage relying on a dictionary"""
326
327			mapping = {
328			'-196°C': 'frozen, liquid nitrogen',
329			'-20°C': 'frozen, -20 degrees Celsius freezer',
330			'-30°C': 'frozen, -20 degrees Celsius freezer',
331			'-80°C': 'frozen, -80 degrees Celsius freezer'}
332
333			if record.sample_storage_temperature in mapping:
334			# get ENUM conversion
335			storage = SAMPLE_STORAGE.get_value_by_desc(
336			mapping[record.sample_storage_temperature])
337
338			return storage
339
340			else:
341			logging.warning("Couldn't find %s in storage types mapping" % (
342			record.sample_storage_temperature))
343
344			return None
345
346
347			def sanitize_url(url):
348			"""Quote URLs for accession"""
349
350			return urllib.parse.quote(url, ':/#?=')
351
352
353			def fill_uid_sample(record, sample_name, animal, submission):
354			"""Helper function to fill animal data in UID sample table"""
355
356			# name and animal name come from parameters
357			organism_part_label = None
358			sample_type_name = record.sample_type_name.lower()
359			body_part_name = record.body_part_name.lower()
360
361			# sylvain has proposed to apply the following decision rule:
362			if body_part_name != "unknown" and body_part_name != "not relevant":
363			organism_part_label = body_part_name
364
365			else:
366			organism_part_label = sample_type_name
367
368			# get a organism part. Organism parts need to be in lowercases
369			organism_part = get_or_create_obj(
370			DictUberon,
371			label=organism_part_label
372			)
373
374			# calculate animal age at collection
375			animal_birth_date = parse_date(record.animal_birth_date)
376			sampling_date = parse_date(record.sampling_date)
377			animal_age_at_collection, time_units = image_timedelta(
378			sampling_date, animal_birth_date)
379
380			# create a new object. Using defaults to avoid collisions when
381			# updating data
382			defaults = {
383			# HINT: is a duplication of name. Can this be non-mandatory?
384			'alternative_id': sample_name.name,
385			'collection_date': record.sampling_date,
386			'protocol': record.sampling_protocol_url,
387			'organism_part': organism_part,
388			'animal': animal,
389			# 'description': v_vessel.comment,
390			'owner': submission.owner,
391			'storage': find_storage_type(record),
392			'availability': sanitize_url(record.sample_availability),
393			'animal_age_at_collection': animal_age_at_collection,
394			'animal_age_at_collection_units': time_units
395			}
396
397			sample = update_or_create_obj(
398			Sample,
399			name=sample_name,
400			defaults=defaults)
401
402			return sample
403
404
405			def process_record(record, submission, animals, language):
406			# Peter mail 26/02/19 18:30: I agree that it sounds like we will
407			# need to create sameAs BioSamples for the IMAGE project, and it makes
408			# sense that the inject tool is able to do this. It may be that we
409			# tackle these cases after getting the main part of the inject tool
410			# functioning and hold or ignore these existing BioSamples for now.
411			# HINT: record with a biosample id should be ignored, for the moment
412			if record.EBI_Biosample_identifier is not None:
413			logger.warning("Ignoring %s: already in biosample!" % str(record))
414			return
415
416			# filling breeds
417			breed = fill_uid_breed(record, language)
418
419			# filling name tables
420			animal_name, sample_name = fill_uid_names(record, submission)
421
422			# fill animal
423			animal = fill_uid_animal(record, animal_name, breed, submission, animals)
424
425			# fill sample
426			fill_uid_sample(record, sample_name, animal, submission)
427
428
429			def upload_crbanim(submission):
430			# debug
431			logger.info("Importing from CRB-Anim file")
432
433			# this is the full path in docker container
434			fullpath = submission.get_uploaded_file_path()
435
436			# read submission data
437			reader = CRBAnimReader()
438			reader.read_file(fullpath)
439
440			# start data loading
441			try:
442			# check for species and sex in a similar way as cryoweb does
443			check, not_found = reader.check_sex()
444
445			if not check:
446			message = (
447			"Not all Sex terms are loaded into database: "
448			"check for %s in your dataset" % (not_found))
449
450			raise CRBAnimImportError(message)
451
452			check, not_found = reader.check_species(submission.gene_bank_country)
453
454			if not check:
455			raise CRBAnimImportError(
456			"Some species are not loaded in UID database: "
457			"%s" % (not_found))
458
459			# ok get languages from submission (useful for translation)
460			# HINT: no traslations implemented, at the moment
461			language = submission.gene_bank_country.label
462
463			# a dictionary in which store animal data
464			animals = {}
465
466			for record in reader.data:
467			process_record(record, submission, animals, language)
468
469			# after processing records, initilize validationsummary objects
470			# create a validation summary object and set all_count
471			vs_animal = get_or_create_obj(
472			ValidationSummary,
473			submission=submission,
474			type="animal")
475
476			# reset counts
477			vs_animal.reset_all_count()
478
479			vs_sample = get_or_create_obj(
480			ValidationSummary,
481			submission=submission,
482			type="sample")
483
484			# reset counts
485			vs_sample.reset_all_count()
486
487			except Exception as exc:
488			# set message:
489			message = "Error in importing data: %s" % (str(exc))
490
491			# save a message in database
492			submission.status = ERROR
493			submission.message = message
494			submission.save()
495
496			# send async message
497			send_message(submission)
498
499			# debug
500			logger.error("error in importing from crbanim: %s" % (exc))
501			logger.exception(exc)
502
503			return False
504
505			else:
506			message = "CRBAnim import completed for submission: %s" % (
507			submission.id)
508
509			submission.message = message
510			submission.status = LOADED
511			submission.save()
512
513			# send async message
514			send_message(
515			submission,
516			validation_message=construct_validation_message(submission))
517
518			logger.info("Import from CRBAnim is complete")
519
520			return True
521

cnr-ibba / IMAGE-InjectTool

Push — master ( 8738db...eecf08 )

crbanim.helpers.sanitize_url() A

Complexity

Size

Duplication

Importance

Duplication Side-by-Side

Filter issues like