def insertAnalysisFromTable(filename, projectID, **kw):
resp = {'ok': False, 'msg': [], 'rows': 0}
# Test for needed arguments
try:
dataset = kw['dataset']
method = kw['method']
category = kw['category']
except:
resp['msg'].append(
'Missing arguments. The "dataset", "method", and "category" information is required.'
)
return resp
# Test File input
try:
fh = open(filename, 'U')
except:
resp['msg'].append('Error in accessing file "%s"' % (filename))
return resp
# Test project ID input
try:
project = Project.objects.get(
pk=int(projectID)) # attempt to find the project for the given ID
except Project.DoesNotExist:
resp["msg"].append(
'Project "%s" does not exist' %
projectID) # report if the project ID does not exist
return resp
header = fh.next().strip().split('\t')
ctr = 0
regex = re.compile(
'[^0-9a-zA-Z_.]') # this is a list of the valid characters
for row in fh:
ctr += 1
line = row.strip().split('\t')
s = line[0]
if len(line) != len(header):
resp["msg"].append(
'Data row "%d" has a length "%d", which does not match the header length "%d"'
% (ctr, len(line), len(header)))
if bool(regex.findall(s)):
resp["msg"].append(
'Sample syntax is invalid at sample "%s", only alpha, numeric, underscore, and dot characters are allowed'
% s)
if s[0].isdigit() or s[0] == "_":
resp["msg"].append(
'Sample syntax is invalid at sample "%s", the first character must be a letter or a dot'
% s)
if s[0] == "." and s[1].isdigit():
resp["msg"].append(
'Sample syntax is invalid at sample "%s", the first character is a dot, so the second must be a letter'
% s)
for i in range(1, len(line)):
try:
float(line[i])
except:
resp["msg"].append(
'File "%s" fails validation. Field "%s" contains non-numeric data. Line: %d. Column %d'
% (filename, field, ctr, i + 1))
if len(resp['msg']) != 0:
return resp
else:
resp['ok'] = True
resp['rows'] = ctr
fh.seek(0)
fh.next()
for row in fh:
line = row.strip().split('\t')
s = line[0]
for i in range(1, len(line)):
entity = header[i]
value = float(line[i])
analysis = Analysis(
project=project,
sample=s,
dataset=dataset,
method=method,
category=category,
entity=entity,
numreads=int(round(value)),
profile=value,
avgscore=1.0,
)
analysis.save()
ok = updateSampleCounts(projectID)
return resp
def insertMetadataFromFile(filename, projectID):
resp = {'ok': False, 'msg': [], 'rows': 0}
# Test File input
try:
#inputdata = read_input(filename)
fh = open(filename)
except:
resp['msg'].append('Error in accessing file "%s"' % (filename))
return resp
# Test project ID input
try:
project = Project.objects.get(
pk=int(projectID)) # attempt to find the project for the given ID
except Project.DoesNotExist:
resp["msg"].append(
'Project "%s" does not exist' %
projectID) # report if the project ID does not exist
return resp
# Once usable data was input check validity of file contents
#header = inputdata.pop(0)
header = fh.next().strip().split('\t')
count = 0
typecheck = {}
validsamples = {}
#for idx, line in enumerate(inputdata):
for line_str in fh:
count += 1
line = line_str.strip().split('\t')
s = line[0]
# attribute data has 4 columns
if len(line) != 4:
resp["msg"].append(
'File "%s" fails validation. Must be 4 columns. Line: %d. Content: %s'
% (filename, count, line_str))
return resp
#if line[3] not in typecheck:
# typecheck[line[2]] = getfieldtype(line[3])
#elif getfieldtype(line[3]) != typecheck[line[2]]:
# resp["msg"].append('File "%s" fails validation. Field "%s" has inconsistent type. Expected a "%s" but found a "%s". Line: %d. Content: %s' % (filename, line[2], typecheck[line[2]], getfieldtype(line[3]), count, line))
# Check if entries for these samples already exist and sample formatting
#samples = list(set(zip(*inputdata)[0]))
regex = re.compile(
'[^0-9a-zA-Z_.]') # this is a list of the valid characters
#for s in samples:
if s in validsamples: continue
if bool(regex.findall(s)):
resp["msg"].append(
'Sample syntax is invalid at sample "%s", only alpha, numeric, underscore, and dot characters are allowed'
% s)
if s[0].isdigit() or s[0] == "_":
resp["msg"].append(
'Sample syntax is invalid at sample "%s", the first character must be a letter or a dot'
% s)
if s[0] == "." and s[1].isdigit():
resp["msg"].append(
'Sample syntax is invalid at sample "%s", the first character is a dot, so the second must be a letter'
% s)
if Attributes.objects.filter(project=project, sample=s,
field=line[3]).exists():
resp["msg"].append(
'Data exists for field %s for sample %s in project %s (%s). Aborting load...'
% (line[3], s, projectID, project.name))
if len(resp['msg']) != 0:
return resp
else:
resp['ok'] = True
resp['rows'] = count
fh.seek(0)
fh.next()
# Insert Validated Input into the database for the project
for line in fh:
data = line.strip().split('\t')
data[3] = dateparse(data[3])
if isinstance(data[3], datetime.datetime):
data[3] = data[3].strftime('%Y-%m-%d')
attributes = Attributes(
project=project,
sample=data[0],
category=data[1],
field=data[2],
value=data[3],
)
if not AttributeInfo.objects.filter(project=project,
name=attributes.field).exists():
newinfo = AttributeInfo(
project=project,
name=attributes.field,
fieldtype=getfieldtype(attributes.value),
values="",
)
newinfo.save()
attributes.save()
fh.close()
ok = updateSampleCounts(projectID)
return resp
def insertMetadataFromTable(filename, projectID):
resp = {'ok': False, 'msg': [], 'rows': 0}
# Test File input
try:
fh = open(filename, 'U')
except:
resp['msg'].append('Error in accessing file "%s"' % (filename))
return resp
# Test project ID input
try:
project = Project.objects.get(
pk=int(projectID)) # attempt to find the project for the given ID
except Project.DoesNotExist:
resp["msg"].append(
'Project "%s" does not exist' %
projectID) # report if the project ID does not exist
return resp
# Once usable data was input check validity of file contents
header = fh.next().strip().split('\t')
ctr = 0
typecheck = dict()
regex = re.compile(
'[^0-9a-zA-Z_.]') # this is a list of the valid characters
for row in fh:
ctr += 1
line = row.strip().split('\t')
s = line[0]
if len(line) != len(header):
resp["msg"].append(
'Data row "%d" has a length "%d", which does not match the header length "%d"'
% (ctr, len(line), len(header)))
if bool(regex.findall(s)):
resp["msg"].append(
'Sample syntax is invalid at sample "%s", only alpha, numeric, underscore, and dot characters are allowed'
% s)
if s[0].isdigit() or s[0] == "_":
resp["msg"].append(
'Sample syntax is invalid at sample "%s", the first character must be a letter or a dot'
% s)
if s[0] == "." and s[1].isdigit():
resp["msg"].append(
'Sample syntax is invalid at sample "%s", the first character is a dot, so the second must be a letter'
% s)
if Attributes.objects.filter(project=project,
sample=s,
field__in=line[1:]).exists():
resp["msg"].append(
'Field data exists for sample %s in project %s (%s). Aborting load...'
% (line[3], s, projectID, project.name))
for i in range(1, len(line)):
field = header[i]
value = line[i]
if field not in typecheck:
typecheck[field] = getfieldtype(value)
elif getfieldtype(value) != typecheck[field]:
resp["msg"].append(
'File "%s" fails validation. Field "%s" has inconsistent type. Expected a "%s" but found a "%s". Line: %d. Column %d'
% (filename, field, typecheck[field], getfieldtype(value),
ctr, i + 1))
if len(resp['msg']) != 0:
return resp
else:
resp['ok'] = True
resp['rows'] = ctr
fh.seek(0)
fh.next()
for row in fh:
line = row.strip().split('\t')
s = line[0]
for i in range(1, len(line)):
field = header[i]
value = line[i]
value = dateparse(value)
if isinstance(value, datetime.datetime):
value = value.strftime('%Y-%m-%d')
attributes = Attributes(
project=project,
sample=s,
category='metadata',
field=field,
value=value,
)
if not AttributeInfo.objects.filter(
project=project, name=attributes.field).exists():
newinfo = AttributeInfo(
project=project,
name=attributes.field,
fieldtype=getfieldtype(attributes.value),
values="",
)
newinfo.save()
attributes.save()
fh.close()
ok = updateSampleCounts(projectID)
return resp
def insertAnalysisFromFile(filename, projectID, taxonomy=None):
resp = {'ok': False, 'msg': [], 'rows': 0}
samples = dict()
# Test File input
try:
fh = fh = open(filename, 'rU')
num_lines = sum(1 for line in fh)
fh.seek(0)
except:
resp['msg'].append('Error in accessing file "%s"' % (filename))
return resp
print >> sys.stderr, "file contains %d lines" % num_lines
# Test project ID input
try:
project = Project.objects.get(
pk=int(projectID)) # attempt to find the project for the given ID
except Project.DoesNotExist:
resp['msg'].append(
'Project "%s" does not exist' %
projectID) # report if the project ID does not exist
return resp
# test taxonomy input
if taxonomy:
try:
taxonomy = Taxonomy.objects.get(pk=int(taxonomy))
except:
resp['msg'].append(
'Unable to determine the taxonomy information (%s)' % taxonomy)
return resp
# Once usable data was input check validity of file contents
header = fh.next()
count = 0
taxas = dict()
print >> sys.stderr, "starting line validation"
start = datetime.datetime.now()
regex = re.compile(
'[^0-9a-zA-Z_.]') # this is a list of the valid characters
for idx, line in enumerate(fh):
count += 1
line = prepareAnalysisLine(line)
samples[line[0]] = samples.get(line[0], 0) + 1
acceptedSamples = dict()
# check for previous submission
if line[0] not in acceptedSamples:
s = line[0]
if bool(regex.findall(s)):
resp['msg'].append(
'Sample syntax is invalid at sample "%s", only alpha, numeric, underscore, and dot characters are allowed'
% s)
return resp
if s[0].isdigit() or s[0] == "_":
resp['msg'].append(
'Sample syntax is invalid at sample "%s", the first character must be a letter or a dot'
% s)
return resp
if s[0] == "." and s[1].isdigit():
resp['msg'].append(
'Sample syntax is invalid at sample "%s", the first character is a dot, so the second must be a letter'
% s)
return resp
if Analysis.objects.filter(project=project,
sample=s,
dataset=line[1],
method=line[2],
category=line[3],
entity=line[4]).exists():
resp['msg'].append(
'Data exists for sample %s in project %s (%s). Aborting load...'
% (s, projectID, project.name))
return resp
acceptedSamples[line[0]] = True
# analysis data has 9 columns
if len(line) != 9:
resp['msg'].append(
'File "%s" fails validation. Must be 9 columns. Line: %d. Content: %s'
% (filename, count, line))
return resp
# Num of reads must be integer
#try: inputdata[idx][6] = int(line[6])
try:
int(line[6])
except:
resp['msg'].append(
'File "%s" fails validation. Column 7 must be an integer. Line: %d. Content: %s'
% (filename, count, line))
return resp
# profile must be a float
#try: inputdata[idx][7] = float(line[7])
try:
float(line[7])
except:
resp['msg'].append(
'File "%s" fails validation. Column 8 must be a float. Line: %d. Content: %s'
% (filename, count, line))
return resp
# average score must be a float
#try: inputdata[idx][8] =float(line[8])
try:
float(line[8])
except:
resp['msg'].append(
'File "%s" fails validation. Column 9 must be a float. Line: %d. Content: %s'
% (filename, count, line))
return resp
# try to find taxa if applicable
if taxonomy:
try:
#tax_id = int(line[5])
tax_id = line[5]
if tax_id not in taxas:
taxas[tax_id] = TaxaTree.objects.get(taxonomy=taxonomy,
tax_id=int(line[5]))
except:
resp['msg'].append(
'File "%s" fails validation. Unable to find taxa. Line %d. Content: %s'
% (filename, count, line))
return resp
else:
#taxas[int(line[5])] = None
taxas[line[5]] = None
if (count % 10000 == 0):
elapsed = (datetime.datetime.now() - start).seconds
if elapsed == 0: elapsed = 1
print >> sys.stderr, "\t%d lines checked of %d [about %d seconds remaining]" % (
count, num_lines, (num_lines - count) / (count / elapsed))
print >> sys.stderr, "starting sample validation"
# Check if entries for these samples already exist
#regex = re.compile('[^0-9a-zA-Z_.]') # this is a list of the valid characters
#for s in samples:
# if bool(regex.findall(s)):
# resp['msg'].append('Sample syntax is invalid at sample "%s", only alpha, numeric, underscore, and dot characters are allowed' %s)
# return resp
# if s[0].isdigit() or s[0] == "_":
# resp['msg'].append('Sample syntax is invalid at sample "%s", the first character must be a letter or a dot' %s)
# return resp
# if s[0] == "." and s[1].isdigit():
# resp['msg'].append('Sample syntax is invalid at sample "%s", the first character is a dot, so the second must be a letter' %s)
# return resp
# Insert Validated Input into the database for the project
fh.seek(0)
fh.next() # skip header
print >> sys.stderr, "inserting data"
count = 0
batch = []
start = datetime.datetime.now()
for line in fh:
count += 1
data = prepareAnalysisLine(line)
analysis = Analysis(
project=project,
sample=data[0],
dataset=data[1],
method=data[2],
category=data[3],
entity=data[4],
#taxatree = taxas[int(data[5])],
taxatree=taxas[data[5]],
numreads=int(data[6]),
profile=float(data[7]),
avgscore=float(data[8]),
)
batch.append(analysis)
if (count % 10000 == 0):
Analysis.objects.bulk_create(batch)
elapsed = (datetime.datetime.now() - start).seconds
if elapsed == 0: elapsed = 1
print >> sys.stderr, "\t%d rows added of %d [about %d seconds remaining]" % (
count, num_lines, (num_lines - count) / (count / elapsed))
batch = []
if len(batch) > 0:
Analysis.objects.bulk_create(batch)
ok = updateSampleCounts(projectID)
if len(resp['msg']) == 0: resp['ok'] = True
resp['rows'] = count
fh.close()
return resp
def insertAnalysisFromFile(filename, projectID, taxonomy=None):
resp = {'ok': False, 'msg': [], 'rows': 0}
samples = dict()
# Test File input
try:
fh = fh = open(filename, 'rU')
num_lines = sum(1 for line in fh)
fh.seek(0)
except:
resp['msg'].append('Error in accessing file "%s"' % (filename))
return resp
print >> sys.stderr, "file contains %d lines" % num_lines
# Test project ID input
try: project = Project.objects.get(pk=int(projectID)) # attempt to find the project for the given ID
except Project.DoesNotExist:
resp['msg'].append('Project "%s" does not exist' % projectID) # report if the project ID does not exist
return resp
# test taxonomy input
if taxonomy:
try:
taxonomy = Taxonomy.objects.get(pk=int(taxonomy))
except:
resp['msg'].append('Unable to determine the taxonomy information (%s)' % taxonomy)
return resp
# Once usable data was input check validity of file contents
header = fh.next()
count = 0
taxas = dict()
print >> sys.stderr, "starting line validation"
start = datetime.datetime.now()
regex = re.compile('[^0-9a-zA-Z_.]') # this is a list of the valid characters
for idx, line in enumerate(fh):
count+=1
line = prepareAnalysisLine(line)
samples[line[0]] = samples.get(line[0], 0) + 1
acceptedSamples = dict()
# check for previous submission
if line[0] not in acceptedSamples:
s = line[0]
if bool(regex.findall(s)):
resp['msg'].append('Sample syntax is invalid at sample "%s", only alpha, numeric, underscore, and dot characters are allowed' %s)
return resp
if s[0].isdigit() or s[0] == "_":
resp['msg'].append('Sample syntax is invalid at sample "%s", the first character must be a letter or a dot' %s)
return resp
if s[0] == "." and s[1].isdigit():
resp['msg'].append('Sample syntax is invalid at sample "%s", the first character is a dot, so the second must be a letter' %s)
return resp
if Analysis.objects.filter(project = project, sample = s, dataset=line[1], method=line[2], category=line[3], entity=line[4]).exists():
resp['msg'].append('Data exists for sample %s in project %s (%s). Aborting load...' %(s, projectID, project.name) )
return resp
acceptedSamples[line[0]] = True
# analysis data has 9 columns
if len(line) != 9:
resp['msg'].append('File "%s" fails validation. Must be 9 columns. Line: %d. Content: %s' % (filename, count, line))
return resp
# Num of reads must be integer
#try: inputdata[idx][6] = int(line[6])
try: int(line[6])
except:
resp['msg'].append('File "%s" fails validation. Column 7 must be an integer. Line: %d. Content: %s' % (filename, count, line))
return resp
# profile must be a float
#try: inputdata[idx][7] = float(line[7])
try: float(line[7])
except:
resp['msg'].append('File "%s" fails validation. Column 8 must be a float. Line: %d. Content: %s' % (filename, count, line))
return resp
# average score must be a float
#try: inputdata[idx][8] =float(line[8])
try: float(line[8])
except:
resp['msg'].append('File "%s" fails validation. Column 9 must be a float. Line: %d. Content: %s' % (filename, count, line))
return resp
# try to find taxa if applicable
if taxonomy:
try:
#tax_id = int(line[5])
tax_id = line[5]
if tax_id not in taxas:
taxas[tax_id] = TaxaTree.objects.get(taxonomy=taxonomy, tax_id=int(line[5]))
except:
resp['msg'].append('File "%s" fails validation. Unable to find taxa. Line %d. Content: %s' % (filename, count, line))
return resp
else:
#taxas[int(line[5])] = None
taxas[line[5]] = None
if (count % 10000 == 0):
elapsed = (datetime.datetime.now() - start).seconds
if elapsed == 0: elapsed = 1
print >> sys.stderr, "\t%d lines checked of %d [about %d seconds remaining]" % (count, num_lines, (num_lines - count)/(count/elapsed))
print >> sys.stderr, "starting sample validation"
# Check if entries for these samples already exist
#regex = re.compile('[^0-9a-zA-Z_.]') # this is a list of the valid characters
#for s in samples:
# if bool(regex.findall(s)):
# resp['msg'].append('Sample syntax is invalid at sample "%s", only alpha, numeric, underscore, and dot characters are allowed' %s)
# return resp
# if s[0].isdigit() or s[0] == "_":
# resp['msg'].append('Sample syntax is invalid at sample "%s", the first character must be a letter or a dot' %s)
# return resp
# if s[0] == "." and s[1].isdigit():
# resp['msg'].append('Sample syntax is invalid at sample "%s", the first character is a dot, so the second must be a letter' %s)
# return resp
# Insert Validated Input into the database for the project
fh.seek(0)
fh.next() # skip header
print >> sys.stderr, "inserting data"
count= 0
batch = []
start = datetime.datetime.now()
for line in fh:
count+=1
data = prepareAnalysisLine(line)
analysis = Analysis(project = project,
sample = data[0],
dataset = data[1],
method = data[2],
category = data[3],
entity = data[4],
#taxatree = taxas[int(data[5])],
taxatree = taxas[data[5]],
numreads = int(data[6]),
profile = float(data[7]),
avgscore = float(data[8]),
)
batch.append(analysis)
if (count % 10000 == 0):
Analysis.objects.bulk_create(batch)
elapsed = (datetime.datetime.now() - start).seconds
if elapsed == 0: elapsed = 1
print >> sys.stderr, "\t%d rows added of %d [about %d seconds remaining]" % (count, num_lines, (num_lines - count)/(count/elapsed))
batch = []
if len(batch) > 0:
Analysis.objects.bulk_create(batch)
ok = updateSampleCounts(projectID)
if len(resp['msg']) == 0: resp['ok']=True
resp['rows'] = count
fh.close()
return resp
def insertAnalysisFromTable(filename,projectID,**kw):
resp = {'ok': False, 'msg': [], 'rows': 0}
# Test for needed arguments
try:
dataset = kw['dataset']
method = kw['method']
category= kw['category']
except:
resp['msg'].append('Missing arguments. The "dataset", "method", and "category" information is required.')
return resp
# Test File input
try:
fh = open(filename, 'U')
except:
resp['msg'].append('Error in accessing file "%s"' % (filename))
return resp
# Test project ID input
try:
project = Project.objects.get(pk=int(projectID)) # attempt to find the project for the given ID
except Project.DoesNotExist:
resp["msg"].append('Project "%s" does not exist' % projectID) # report if the project ID does not exist
return resp
header = fh.next().strip().split('\t')
ctr = 0
regex = re.compile('[^0-9a-zA-Z_.]') # this is a list of the valid characters
for row in fh:
ctr+=1
line = row.strip().split('\t')
s = line[0]
if len(line) != len(header):
resp["msg"].append('Data row "%d" has a length "%d", which does not match the header length "%d"' % ( ctr, len(line), len(header) ))
if bool(regex.findall(s)):
resp["msg"].append('Sample syntax is invalid at sample "%s", only alpha, numeric, underscore, and dot characters are allowed' %s)
if s[0].isdigit() or s[0] == "_":
resp["msg"].append('Sample syntax is invalid at sample "%s", the first character must be a letter or a dot' %s)
if s[0] == "." and s[1].isdigit():
resp["msg"].append('Sample syntax is invalid at sample "%s", the first character is a dot, so the second must be a letter' %s)
for i in range(1, len(line)):
try:
float(line[i])
except:
resp["msg"].append('File "%s" fails validation. Field "%s" contains non-numeric data. Line: %d. Column %d' % (filename, field, ctr, i+1))
if len(resp['msg']) != 0:
return resp
else:
resp['ok'] = True
resp['rows'] = ctr
fh.seek(0)
fh.next()
for row in fh:
line = row.strip().split('\t')
s = line[0]
for i in range(1, len(line)):
entity = header[i]
value = float(line[i])
analysis = Analysis(project = project,
sample = s,
dataset = dataset,
method = method,
category = category,
entity = entity,
numreads = int(round(value)),
profile = value,
avgscore = 1.0,
)
analysis.save()
ok = updateSampleCounts(projectID)
return resp
def insertMetadataFromTable(filename, projectID):
resp = {'ok': False, 'msg': [], 'rows': 0}
# Test File input
try:
fh = open(filename, 'U')
except:
resp['msg'].append('Error in accessing file "%s"' % (filename))
return resp
# Test project ID input
try:
project = Project.objects.get(pk=int(projectID)) # attempt to find the project for the given ID
except Project.DoesNotExist:
resp["msg"].append('Project "%s" does not exist' % projectID) # report if the project ID does not exist
return resp
# Once usable data was input check validity of file contents
header = fh.next().strip().split('\t')
ctr = 0
typecheck = dict()
regex = re.compile('[^0-9a-zA-Z_.]') # this is a list of the valid characters
for row in fh:
ctr+=1
line = row.strip().split('\t')
s = line[0]
if len(line) != len(header):
resp["msg"].append('Data row "%d" has a length "%d", which does not match the header length "%d"' % ( ctr, len(line), len(header) ))
if bool(regex.findall(s)):
resp["msg"].append('Sample syntax is invalid at sample "%s", only alpha, numeric, underscore, and dot characters are allowed' %s)
if s[0].isdigit() or s[0] == "_":
resp["msg"].append('Sample syntax is invalid at sample "%s", the first character must be a letter or a dot' %s)
if s[0] == "." and s[1].isdigit():
resp["msg"].append('Sample syntax is invalid at sample "%s", the first character is a dot, so the second must be a letter' %s)
if Attributes.objects.filter(project = project, sample = s, field__in=line[1:]).exists():
resp["msg"].append('Field data exists for sample %s in project %s (%s). Aborting load...' %(line[3], s, projectID, project.name) )
for i in range(1, len(line)):
field = header[i]
value = line[i]
if field not in typecheck:
typecheck[field] = getfieldtype(value)
elif getfieldtype(value) != typecheck[field]:
resp["msg"].append('File "%s" fails validation. Field "%s" has inconsistent type. Expected a "%s" but found a "%s". Line: %d. Column %d' % (filename, field, typecheck[field], getfieldtype(value), ctr, i+1))
if len(resp['msg']) != 0:
return resp
else:
resp['ok'] = True
resp['rows'] = ctr
fh.seek(0)
fh.next()
for row in fh:
line = row.strip().split('\t')
s = line[0]
for i in range(1, len(line)):
field = header[i]
value = line[i]
value = dateparse(value)
if isinstance(value, datetime.datetime): value = value.strftime('%Y-%m-%d')
attributes = Attributes(project = project,
sample = s,
category = 'metadata',
field = field,
value = value,
)
if not AttributeInfo.objects.filter(project=project, name=attributes.field).exists():
newinfo = AttributeInfo(project = project,
name = attributes.field,
fieldtype = getfieldtype(attributes.value),
values = "",
)
newinfo.save()
attributes.save()
fh.close()
ok = updateSampleCounts(projectID)
return resp
def insertMetadataFromFile(filename, projectID):
resp = {'ok': False, 'msg': [], 'rows': 0}
# Test File input
try:
#inputdata = read_input(filename)
fh = open(filename)
except:
resp['msg'].append('Error in accessing file "%s"' % (filename))
return resp
# Test project ID input
try:
project = Project.objects.get(pk=int(projectID)) # attempt to find the project for the given ID
except Project.DoesNotExist:
resp["msg"].append('Project "%s" does not exist' % projectID) # report if the project ID does not exist
return resp
# Once usable data was input check validity of file contents
#header = inputdata.pop(0)
header = fh.next().strip().split('\t')
count = 0
typecheck = {}
validsamples = {}
#for idx, line in enumerate(inputdata):
for line_str in fh:
count+=1
line = line_str.strip().split('\t')
s = line[0]
# attribute data has 4 columns
if len(line) != 4:
resp["msg"].append('File "%s" fails validation. Must be 4 columns. Line: %d. Content: %s' % (filename, count, line_str))
return resp
#if line[3] not in typecheck:
# typecheck[line[2]] = getfieldtype(line[3])
#elif getfieldtype(line[3]) != typecheck[line[2]]:
# resp["msg"].append('File "%s" fails validation. Field "%s" has inconsistent type. Expected a "%s" but found a "%s". Line: %d. Content: %s' % (filename, line[2], typecheck[line[2]], getfieldtype(line[3]), count, line))
# Check if entries for these samples already exist and sample formatting
#samples = list(set(zip(*inputdata)[0]))
regex = re.compile('[^0-9a-zA-Z_.]') # this is a list of the valid characters
#for s in samples:
if s in validsamples: continue
if bool(regex.findall(s)):
resp["msg"].append('Sample syntax is invalid at sample "%s", only alpha, numeric, underscore, and dot characters are allowed' %s)
if s[0].isdigit() or s[0] == "_":
resp["msg"].append('Sample syntax is invalid at sample "%s", the first character must be a letter or a dot' %s)
if s[0] == "." and s[1].isdigit():
resp["msg"].append('Sample syntax is invalid at sample "%s", the first character is a dot, so the second must be a letter' %s)
if Attributes.objects.filter(project = project, sample = s, field=line[3]).exists():
resp["msg"].append('Data exists for field %s for sample %s in project %s (%s). Aborting load...' %(line[3], s, projectID, project.name) )
if len(resp['msg']) != 0:
return resp
else:
resp['ok'] = True
resp['rows'] = count
fh.seek(0)
fh.next()
# Insert Validated Input into the database for the project
for line in fh:
data = line.strip().split('\t')
data[3] = dateparse(data[3])
if isinstance(data[3], datetime.datetime): data[3] = data[3].strftime('%Y-%m-%d')
attributes = Attributes(project = project,
sample = data[0],
category = data[1],
field = data[2],
value = data[3],
)
if not AttributeInfo.objects.filter(project=project, name=attributes.field).exists():
newinfo = AttributeInfo(project = project,
name = attributes.field,
fieldtype = getfieldtype(attributes.value),
values = "",
)
newinfo.save()
attributes.save()
fh.close()
ok = updateSampleCounts(projectID)
return resp