def parse_clause(self, clause, base_clause_parser, table):
clause = clause.strip()
depth = 0
min_depth = 100000 #Arbitrary bound on nr of nested clauses.
in_wildcard_clause = False
for i in range(0, len(clause)):
if clause[i] == '[':
in_wildcard_clause = True
elif clause[i] == ']':
in_wildcard_clause = False
elif in_wildcard_clause: #currently in wildcard thingy, so doesn't mean anything. Move on.
continue
elif clause[i] == '(':
depth += 1
elif clause[i] == ')':
depth -= 1
elif i < len(clause) - 2:
if clause[i:i + 2] == "||":
if depth == 0:
left = self.parse_clause(clause[:i].strip(),
base_clause_parser, table)
right = self.parse_clause(clause[i + 2:].strip(),
base_clause_parser, table)
return OR_expression(left, right)
else:
min_depth = min(min_depth, depth)
elif clause[i:i + 2] == "&&":
if depth == 0:
left = self.parse_clause(clause[:i].strip(),
base_clause_parser, table)
right = self.parse_clause(clause[i + 2:].strip(),
base_clause_parser, table)
return AND_expression(left, right)
else:
min_depth = min(min_depth, depth)
elif i < len(clause) - 3:
if clause[i:i + 3] == "NOT":
if depth == 0:
body = self.parse_clause(clause[i + 3:].strip(),
base_clause_parser, table)
return NOT_expression(
body, table, self.get_partition_key(table),
self.n_variants)
else:
min_depth = min(min_depth, depth)
if depth == 0:
if min_depth < 100000:
#Strip away all brackets to expose uppermost boolean operator
return self.parse_clause(
clause[min_depth:len(clause) - min_depth],
base_clause_parser, table)
else:
#No more boolean operators, strip all remaining brackets
token = clause.strip('(').strip(')')
return base_clause_parser(token)
else:
sys.exit("ERROR in %s. Brackets don't match" % clause)
def run(self,
query,
gt_filter=None,
show_variant_samples=False,
variant_samples_delim=',',
predicates=None,
needs_genotypes=False,
needs_genes=False,
show_families=False,
test_mode=False,
needs_sample_names=False,
nr_cores=1,
start_time=-42,
use_header=False,
exp_id="Oink",
timeout=10.0,
batch_size=100):
"""
Execute a query against a Gemini database. The user may
specify:
1. (reqd.) an SQL `query`.
2. (opt.) a genotype filter.
"""
self.query = self.formatter.format_query(query).replace('==', '=')
self.gt_filter = gt_filter
#print self.query + '; gt-filter = %s \n' % gt_filter
self.nr_cores = nr_cores
self.start_time = start_time
self.use_header = use_header
self.exp_id = exp_id
self.timeout = timeout
self.batch_size = batch_size
if self._is_gt_filter_safe() is False:
sys.exit("ERROR: unsafe --gt-filter command.")
self.show_variant_samples = show_variant_samples
self.variant_samples_delim = variant_samples_delim
self.test_mode = test_mode
self.needs_genotypes = needs_genotypes
self.needs_vcf_columns = False
if self.formatter.name == 'vcf':
self.needs_vcf_columns = True
self.needs_sample_names = needs_sample_names
self.needs_genes = needs_genes
self.show_families = show_families
if predicates:
self.predicates += predicates
# make sure the SELECT columns are separated by a
# comma and a space. then tokenize by spaces.
self.query = self.query.replace(',', ', ')
self.query_pieces = self.query.split()
if not any(s.startswith("gt") for s in self.query_pieces) and \
not any(s.startswith("(gt") for s in self.query_pieces) and \
not any(".gt" in s for s in self.query_pieces):
if self.gt_filter is None:
self.query_type = "no-genotypes"
else:
self.gt_filter_exp = self._correct_genotype_filter()
self.query_type = "filter-genotypes"
else:
if self.gt_filter is None:
self.query_type = "select-genotypes"
else:
self.gt_filter_exp = self._correct_genotype_filter()
self.query_type = "filter-genotypes"
(self.requested_columns, self.from_table, where_clause,
self.rest_of_query) = get_query_parts(self.query)
self.extra_columns = []
if where_clause != '':
self.where_exp = self.parse_where_clause(where_clause,
self.from_table)
if not self.gt_filter is None:
self.where_exp = AND_expression(self.where_exp,
self.gt_filter_exp)
else:
if not self.gt_filter is None:
self.where_exp = self.gt_filter_exp
else:
self.where_exp = None
self._apply_query()
self.query_executed = True
def run(self, query, gt_filter=None, show_variant_samples=False,
variant_samples_delim=',', predicates=None,
needs_genotypes=False, needs_genes=False,
show_families=False, test_mode=False,
needs_sample_names=False, nr_cores = 1,
start_time = -42, use_header = False,
exp_id="Oink", timeout=10.0, batch_size = 100):
"""
Execute a query against a Gemini database. The user may
specify:
1. (reqd.) an SQL `query`.
2. (opt.) a genotype filter.
"""
self.query = self.formatter.format_query(query).replace('==','=')
self.gt_filter = gt_filter
#print self.query + '; gt-filter = %s \n' % gt_filter
self.nr_cores = nr_cores
self.start_time = start_time
self.use_header = use_header
self.exp_id = exp_id
self.timeout = timeout
self.batch_size = batch_size
if self._is_gt_filter_safe() is False:
sys.exit("ERROR: unsafe --gt-filter command.")
self.show_variant_samples = show_variant_samples
self.variant_samples_delim = variant_samples_delim
self.test_mode = test_mode
self.needs_genotypes = needs_genotypes
self.needs_vcf_columns = False
if self.formatter.name == 'vcf':
self.needs_vcf_columns = True
self.needs_sample_names = needs_sample_names
self.needs_genes = needs_genes
self.show_families = show_families
if predicates:
self.predicates += predicates
# make sure the SELECT columns are separated by a
# comma and a space. then tokenize by spaces.
self.query = self.query.replace(',', ', ')
self.query_pieces = self.query.split()
if not any(s.startswith("gt") for s in self.query_pieces) and \
not any(s.startswith("(gt") for s in self.query_pieces) and \
not any(".gt" in s for s in self.query_pieces):
if self.gt_filter is None:
self.query_type = "no-genotypes"
else:
self.gt_filter_exp = self._correct_genotype_filter()
self.query_type = "filter-genotypes"
else:
if self.gt_filter is None:
self.query_type = "select-genotypes"
else:
self.gt_filter_exp = self._correct_genotype_filter()
self.query_type = "filter-genotypes"
(self.requested_columns, self.from_table, where_clause, self.rest_of_query) = get_query_parts(self.query)
self.extra_columns = []
if where_clause != '':
self.where_exp = self.parse_where_clause(where_clause, self.from_table)
if not self.gt_filter is None:
self.where_exp = AND_expression(self.where_exp, self.gt_filter_exp)
else:
if not self.gt_filter is None:
self.where_exp = self.gt_filter_exp
else:
self.where_exp = None
self._apply_query()
self.query_executed = True
class GeminiQuery(object):
"""
An interface to submit queries to an existing Gemini database
and iterate over the results of the query.
We create a GeminiQuery object by specifying database to which to
connect::
from geminicassandra import GeminiQuery
gq = GeminiQuery("my.db")
We can then issue a query against the database and iterate through
the results by using the ``run()`` method::
for row in gq:
print row
Instead of printing the entire row, one access print specific columns::
gq.run("select chrom, start, end from variants")
for row in gq:
print row['chrom']
Also, all of the underlying numpy genotype arrays are
always available::
gq.run("select chrom, start, end from variants")
for row in gq:
gts = row.gts
print row['chrom'], gts
# yields "chr1" ['A/G' 'G/G' ... 'A/G']
The ``run()`` methods also accepts genotype filter::
query = "select chrom, start, end" from variants"
gt_filter = "gt_types.NA20814 == HET"
gq.run(query)
for row in gq:
print row
Lastly, one can use the ``sample_to_idx`` and ``idx_to_sample``
dictionaries to gain access to sample-level genotype information
either by sample name or by sample index::
# grab dict mapping sample to genotype array indices
smp2idx = gq.sample_to_idx
query = "select chrom, start, end from variants"
gt_filter = "gt_types.NA20814 == HET"
gq.run(query, gt_filter)
# print a header listing the selected columns
print gq.header
for row in gq:
# access a NUMPY array of the sample genotypes.
gts = row['gts']
# use the smp2idx dict to access sample genotypes
idx = smp2idx['NA20814']
print row, gts[idx]
"""
def __init__(self,
db_contact_points,
keyspace,
include_gt_cols=False,
out_format=DefaultRowFormat(None)):
self.db_contact_points = map(strip, db_contact_points.split(','))
self.keyspace = keyspace
self.query_executed = False
self.for_browser = False
self.include_gt_cols = include_gt_cols
# try to connect to the provided database
self._connect_to_database()
self.n_variants = self.get_n_variants()
# list of samples ids for each clause in the --gt-filter
self.sample_info = collections.defaultdict(list)
'''# map sample names to indices. e.g. self.sample_to_idx[NA20814] -> 323
self.sample_to_idx = util.map_samples_to_indices(self.session)
# and vice versa. e.g., self.idx_to_sample[323] -> NA20814
self.idx_to_sample = util.map_indices_to_samples(self.session)
self.idx_to_sample_object = util.map_indices_to_sample_objects(self.session)
self.sample_to_sample_object = util.map_samples_to_sample_objects(self.session)'''
self.formatter = out_format
self.predicates = [self.formatter.predicate]
def get_n_variants(self):
res = self.session.execute(
"select n_rows from row_counts where table_name = 'variants'")
return res[0].n_rows
def run(self,
query,
gt_filter=None,
show_variant_samples=False,
variant_samples_delim=',',
predicates=None,
needs_genotypes=False,
needs_genes=False,
show_families=False,
test_mode=False,
needs_sample_names=False,
nr_cores=1,
start_time=-42,
use_header=False,
exp_id="Oink",
timeout=10.0,
batch_size=100):
"""
Execute a query against a Gemini database. The user may
specify:
1. (reqd.) an SQL `query`.
2. (opt.) a genotype filter.
"""
self.query = self.formatter.format_query(query).replace('==', '=')
self.gt_filter = gt_filter
#print self.query + '; gt-filter = %s \n' % gt_filter
self.nr_cores = nr_cores
self.start_time = start_time
self.use_header = use_header
self.exp_id = exp_id
self.timeout = timeout
self.batch_size = batch_size
if self._is_gt_filter_safe() is False:
sys.exit("ERROR: unsafe --gt-filter command.")
self.show_variant_samples = show_variant_samples
self.variant_samples_delim = variant_samples_delim
self.test_mode = test_mode
self.needs_genotypes = needs_genotypes
self.needs_vcf_columns = False
if self.formatter.name == 'vcf':
self.needs_vcf_columns = True
self.needs_sample_names = needs_sample_names
self.needs_genes = needs_genes
self.show_families = show_families
if predicates:
self.predicates += predicates
# make sure the SELECT columns are separated by a
# comma and a space. then tokenize by spaces.
self.query = self.query.replace(',', ', ')
self.query_pieces = self.query.split()
if not any(s.startswith("gt") for s in self.query_pieces) and \
not any(s.startswith("(gt") for s in self.query_pieces) and \
not any(".gt" in s for s in self.query_pieces):
if self.gt_filter is None:
self.query_type = "no-genotypes"
else:
self.gt_filter_exp = self._correct_genotype_filter()
self.query_type = "filter-genotypes"
else:
if self.gt_filter is None:
self.query_type = "select-genotypes"
else:
self.gt_filter_exp = self._correct_genotype_filter()
self.query_type = "filter-genotypes"
(self.requested_columns, self.from_table, where_clause,
self.rest_of_query) = get_query_parts(self.query)
self.extra_columns = []
if where_clause != '':
self.where_exp = self.parse_where_clause(where_clause,
self.from_table)
if not self.gt_filter is None:
self.where_exp = AND_expression(self.where_exp,
self.gt_filter_exp)
else:
if not self.gt_filter is None:
self.where_exp = self.gt_filter_exp
else:
self.where_exp = None
self._apply_query()
self.query_executed = True
def run_simple_query(self, query):
(requested_columns, from_table, where_clause,
rest_of_query) = get_query_parts(query)
if where_clause != '':
where_exp = self.parse_where_clause(where_clause, from_table)
if not where_exp is None:
matches = where_exp.evaluate(self.session, "*")
if len(matches) == 0:
return OrderedDict([])
else:
try:
dink_query = "SELECT %s FROM %s" % (
','.join(requested_columns), from_table)
if matches != "*":
if from_table.startswith('samples'):
in_clause = "','".join(matches)
dink_query += " WHERE %s IN ('%s')" % (
self.get_partition_key(from_table), in_clause)
else:
in_clause = ",".join(map(str, matches))
dink_query += " WHERE %s IN (%s)" % (
self.get_partition_key(from_table), in_clause)
dink_query += " " + rest_of_query
self.session.row_factory = ordered_dict_factory
return self.session.execute(dink_query)
except cassandra.protocol.SyntaxException as e:
print "Cassandra error: {0}".format(e)
sys.exit("The query issued (%s) has a syntax error." % query)
def __iter__(self):
return self
@property
def header(self):
"""
Return a header describing the columns that
were selected in the query issued to a GeminiQuery object.
"""
h = [col for col in self.report_cols]
if self.show_variant_samples:
h += ["variant_samples", "HET_samples", "HOM_ALT_samples"]
if self.show_families:
h += ["families"]
return self.formatter.header(h)
def next(self):
"""
Return the GeminiRow object for the next query result.
"""
while (1):
try:
row = self.result.next()
except Exception as e:
sys.__stderr__.write(str(e) + "\n")
self.cluster.shutdown()
raise StopIteration
return self.row_2_GeminiRow(row)
def row_2_GeminiRow(self, row):
variant_names = []
het_names = []
hom_alt_names = []
hom_ref_names = []
unknown_names = []
info = None
if 'info' in self.report_cols:
info = compression.unpack_ordereddict_blob(row['info'])
fields = OrderedDict()
for col in self.report_cols:
if col == "*":
continue
if not col == "info":
fields[col] = row[col]
elif col == "info":
fields[col] = _info_dict_to_string(info)
if self.show_variant_samples or self.needs_sample_names:
het_names = self._get_variant_samples(row['variant_id'], HET)
hom_alt_names = self._get_variant_samples(row['variant_id'],
HOM_ALT)
hom_ref_names = self._get_variant_samples(row['variant_id'],
HOM_REF)
unknown_names = self._get_variant_samples(row['variant_id'],
UNKNOWN)
variant_names = het_names | hom_alt_names
if self.show_variant_samples:
fields["variant_samples"] = \
self.variant_samples_delim.join(variant_names)
fields["HET_samples"] = \
self.variant_samples_delim.join(het_names)
fields["HOM_ALT_samples"] = \
self.variant_samples_delim.join(hom_alt_names)
if self.show_families:
families = map(
str,
list(
set([
self.sample_to_sample_object[x].family_id
for x in variant_names
])))
fields["families"] = self.variant_samples_delim.join(families)
gemini_row = GeminiRow(fields,
variant_names,
het_names,
hom_alt_names,
hom_ref_names,
unknown_names,
info,
formatter=self.formatter)
if not all([predicate(gemini_row) for predicate in self.predicates]):
return None
if not self.for_browser:
return gemini_row
else:
return fields
def _get_variant_samples(self, variant_id, gt_type):
query = "SELECT sample_name from samples_by_variants_gt_type WHERE variant_id = %s AND gt_type = %s"
self.session.row_factory = tuple_factory
return async_rows_as_set(self.session, query % (variant_id, gt_type))
def _group_samples_by_genotype(self, gt_types):
"""
make dictionary keyed by genotype of list of samples with that genotype
"""
key_fn = lambda x: x[1]
val_fn = lambda x: self.idx_to_sample[x[0]]
return partition_by_fn(enumerate(gt_types),
key_fn=key_fn,
val_fn=val_fn)
def _connect_to_database(self):
"""
Establish a connection to the requested Gemini database.
"""
# open up a new database
self.cluster = Cluster(self.db_contact_points)
self.session = self.cluster.connect(self.keyspace)
def _is_gt_filter_safe(self):
"""
Test to see if the gt_filter string is potentially malicious.
A future improvement would be to use pyparsing to
traverse and directly validate the string.
"""
if self.gt_filter is None or len(self.gt_filter.strip()) == 0:
return True
# avoid builtins
# http://nedbatchelder.com/blog/201206/eval_really_is_dangerous.html
if "__" in self.gt_filter:
return False
# avoid malicious commands
evil = [" rm ", "os.system"]
if any(s in self.gt_filter for s in evil):
return False
# make sure a "gt" col is in the string
valid_cols = [
"gts.", "gt_types.", "gt_phases.", "gt_quals.", "gt_depths.",
"gt_ref_depths.", "gt_alt_depths.", "gt_copy_numbers.", "[gts].",
"[gt_types].", "[gt_phases].", "[gt_quals].", "[gt_copy_numbers].",
"[gt_depths].", "[gt_ref_depths].", "[gt_alt_depths]."
]
if any(s in self.gt_filter for s in valid_cols):
return True
# assume the worst
return False
def _execute_query(self):
n_matches = len(self.matches)
self.session.row_factory = ordered_dict_factory
query = "SELECT %s FROM %s" % (','.join(self.requested_columns +
self.extra_columns),
self.from_table)
error_count = 0
if n_matches == 0:
print "No results!"
time_taken = time.time() - self.start_time
elif not self.test_mode:
output_folder = self.exp_id + "_results"
if not os.path.exists(output_folder):
os.makedirs(output_folder)
output_path = output_folder + "/%s"
if self.matches == "*":
print "All rows match query."
query += " " + self.rest_of_query
error_count += execute_async_blocking(self.session, query,
output_path % 0,
self.extra_columns, (),
self.timeout)
else:
print "%d rows match query." % len(self.matches)
step = len(self.matches) / self.nr_cores
procs = []
conns = []
for i in range(self.nr_cores):
parent_conn, child_conn = Pipe()
conns.append(parent_conn)
p = Process(target=fetch_matches,
args=(child_conn, i, output_path % i, query, self.from_table,\
self.get_partition_key(self.from_table), self.extra_columns,\
self.db_contact_points, self.keyspace, self.batch_size))
procs.append(p)
p.start()
for i in range(self.nr_cores):
n = len(self.matches)
begin = i * step + min(i, n % self.nr_cores)
end = begin + step
if i < n % self.nr_cores:
end += 1
conns[i].send(self.matches[begin:end])
for i in range(self.nr_cores):
errs = conns[i].recv()
error_count += errs
conns[i].close()
procs[i].join()
time_taken = time.time() - self.start_time
print "Query %s completed in %s s." % (self.exp_id, time_taken)
print "Query %s encountered %d errors" % (self.exp_id, error_count)
else:
signal(SIGPIPE, SIG_DFL)
'''
Test mode stuff: no extra prints, synchronous execution, sort results
'''
try:
query = "SELECT %s FROM %s" % (
','.join(self.requested_columns + self.extra_columns),
self.from_table)
if self.matches != "*":
if self.from_table != 'samples':
in_clause = ",".join(map(str, self.matches))
query += " WHERE %s IN (%s)" % (self.get_partition_key(
self.from_table), in_clause)
else:
in_clause = "','".join(self.matches)
query += " WHERE %s IN ('%s')" % (
self.get_partition_key(self.from_table), in_clause)
query += " " + self.rest_of_query
res = self.session.execute(query)
if self.from_table == 'variants':
res = sorted(res, key=lambda x: x['start'])
elif self.from_table == 'samples':
res = sorted(res, key=lambda x: x['sample_id'])
self.report_cols = filter(
lambda x: not x in self.extra_columns, res[0].keys())
if self.use_header and self.header:
print self.header
for row in res:
print self.row_2_GeminiRow(row)
except Exception as e:
sys.stderr.write(str(e))
finally:
time_taken = time.time() - self.start_time
with open("querylog", 'a') as log:
log.write("2::%s;%s;%d\n" % (self.exp_id, time_taken, error_count))
def set_report_cols(self, rep_cols):
self.report_cols = rep_cols
def _apply_query(self):
"""
Execute a query. Intercept gt* columns and
replace sample names with indices where necessary.
"""
if self.needs_genes:
self.requested_columns.append("gene")
'''if self.needs_vcf_columns:
self.query = self._add_vcf_cols_to_query()'''
self.matches = "*"
if not self.where_exp is None:
self.matches = list(self.where_exp.evaluate(self.session, "*"))
mid_time = time.time()
log = open("querylog", 'a')
log.write("1::%s;%s\n" % (self.exp_id, mid_time - self.start_time))
log.close()
if self._query_needs_genotype_info():
# break up the select statement into individual
# pieces and replace genotype columns using sample
# names with sample indices
self._split_select()
if self.show_families or self.show_variant_samples or self.needs_sample_names:
if (not 'variant_id' in self.requested_columns) and (
not "*" in self.requested_columns):
self.extra_columns.append('variant_id')
if self.test_mode and self.from_table == 'variants' and not 'start' in self.requested_columns:
self.extra_columns.append('start')
self._execute_query()
def shutdown(self):
self.session.shutdown()
self.cluster.shutdown()
def _get_matching_sample_ids(self, wildcard):
if wildcard.strip() != "*":
query = self.parse_where_clause(wildcard, 'samples')
else:
query = Basic_expression('samples', 'name', "")
return list(query.evaluate(self.session, '*'))
def get_partition_key(self, table):
key = self.cluster.metadata.keyspaces[
self.keyspace].tables[table].partition_key[0].name
return key
def _correct_genotype_filter(self):
"""
This converts a raw genotype filter that contains
'wildcard' statements into a filter that can be eval()'ed.
Specifically, we must convert a _named_ genotype index
to a _numerical_ genotype index so that the appropriate
value can be extracted for the sample from the genotype
numpy arrays.
For example, without WILDCARDS, this converts:
--gt-filter "(gt_types.1478PC0011 == 1)"
to:
(gt_types[11] == 1)
With WILDCARDS, this converts things like:
"(gt_types).(phenotype==1).(==HET)"
to:
"gt_types[2] == HET and gt_types[5] == HET"
"""
return self.parse_clause(self.gt_filter, self.gt_base_parser,
'variants')
def parse_where_clause(self, where_clause, table):
return self.parse_clause(
where_clause, lambda x: self.where_clause_to_exp(
table, self.get_partition_key(table), x), table)
def parse_clause(self, clause, base_clause_parser, table):
clause = clause.strip()
depth = 0
min_depth = 100000 #Arbitrary bound on nr of nested clauses.
in_wildcard_clause = False
for i in range(0, len(clause)):
if clause[i] == '[':
in_wildcard_clause = True
elif clause[i] == ']':
in_wildcard_clause = False
elif in_wildcard_clause: #currently in wildcard thingy, so doesn't mean anything. Move on.
continue
elif clause[i] == '(':
depth += 1
elif clause[i] == ')':
depth -= 1
elif i < len(clause) - 2:
if clause[i:i + 2] == "||":
if depth == 0:
left = self.parse_clause(clause[:i].strip(),
base_clause_parser, table)
right = self.parse_clause(clause[i + 2:].strip(),
base_clause_parser, table)
return OR_expression(left, right)
else:
min_depth = min(min_depth, depth)
elif clause[i:i + 2] == "&&":
if depth == 0:
left = self.parse_clause(clause[:i].strip(),
base_clause_parser, table)
right = self.parse_clause(clause[i + 2:].strip(),
base_clause_parser, table)
return AND_expression(left, right)
else:
min_depth = min(min_depth, depth)
elif i < len(clause) - 3:
if clause[i:i + 3] == "NOT":
if depth == 0:
body = self.parse_clause(clause[i + 3:].strip(),
base_clause_parser, table)
return NOT_expression(
body, table, self.get_partition_key(table),
self.n_variants)
else:
min_depth = min(min_depth, depth)
if depth == 0:
if min_depth < 100000:
#Strip away all brackets to expose uppermost boolean operator
return self.parse_clause(
clause[min_depth:len(clause) - min_depth],
base_clause_parser, table)
else:
#No more boolean operators, strip all remaining brackets
token = clause.strip('(').strip(')')
return base_clause_parser(token)
else:
sys.exit("ERROR in %s. Brackets don't match" % clause)
def where_clause_to_exp(self, table, cols, clause):
target_table = self.get_table_from_where_clause(table, clause)
exp = Basic_expression(target_table, cols, clause)
return exp
def get_table_from_where_clause(self, table, where_clause):
where_clause = where_clause.replace('==', '=')
clauses = where_clause.split("and")
range_clauses = filter(lambda x: '<' in x or '>' in x, clauses)
range_col = None
for range_clause in range_clauses:
i = 0
for op in ["<", ">"]:
temp = range_clause.find(op)
if temp > -1:
i = temp
break
col = range_clause[0:i].strip()
if range_col and (col != range_col):
sys.exit(
"ERROR: range clauses only possible on at most one column")
else:
range_col = col
eq_clauses = filter(lambda x: not ('<' in x or '>' in x), clauses)
eq_clauses = map(lambda x: x.split('='), eq_clauses)
eq_columns = map(lambda x: x[0].strip(), eq_clauses)
relevant_tables = self.get_relevant_tables(table)
for t in relevant_tables:
if all(x in eq_columns for x in t.partition_key):
other_cols = filter(lambda x: not x in t.partition_key,
eq_columns)
if all(x in t.clustering_key[:len(other_cols)]
for x in other_cols):
if range_col:
if range_col == t.clustering_key[len(other_cols)]:
return t.name
else:
return t.name
sys.exit("ERROR: No suitable table found for query: %s" % where_clause)
def get_relevant_tables(self, table):
Row = namedtuple('Row', 'name partition_key clustering_key')
tables = self.cluster.metadata.keyspaces[self.keyspace].tables
interesting_tables = filter(lambda x: x.startswith(table),
tables.keys())
res = []
for table in interesting_tables:
res.append(Row(table, map(lambda y: y.name, tables[table].partition_key),\
map(lambda y: y.name, tables[table].clustering_key)))
return res
def gt_base_parser(self, clause):
if (clause.find("gt") >= 0
or clause.find("GT") >= 0) and not '[' in clause:
return self.gt_filter_to_query_exp(clause.replace('==', '='))
elif (clause.find("gt") >= 0
or clause.find("GT") >= 0) and '[' in clause:
dink = self.parse_gt_wildcard(clause)
#print dink.to_string()
return dink
else:
sys.exit("ERROR: invalid --gt-filter command")
def gt_filter_to_query_exp(self, gt_filter):
i = -1
operators = ['!=', '<=', '>=', '=', '<', '>']
for op in operators:
temp = gt_filter.find(op)
if temp > -1:
i = temp
break
if i > -1:
left = gt_filter[0:i].strip()
clause = self._swap_genotype_for_number(gt_filter[i:].strip())
else:
sys.exit("ERROR: invalid --gt-filter command 858.")
not_exp = False
if clause.startswith('!'):
not_exp = True
clause = clause[1:]
(column, sample) = left.split('.', 1)
exp = Basic_expression(
'variants_by_samples_' + column, 'variant_id',
"sample_name = '" + sample + "' AND " + column + clause)
if not_exp:
return NOT_expression(exp, 'variants', 'variant_id',
self.n_variants)
else:
return exp
def parse_gt_wildcard(self, token):
if token.count('.') != 3 or \
token.count('[') != 4 or \
token.count(']') != 4:
sys.exit("Wildcard filter should consist of 4 elements. Exiting.")
(column, wildcard, wildcard_rule, wildcard_op) = token.split('.')
column = column.strip('[').strip(']').strip()
wildcard = wildcard.strip('[').strip(']').strip().replace('==', '=')
wildcard_rule = wildcard_rule.strip('[').strip(']').strip()
wildcard_op = wildcard_op.strip('[').strip(']').strip()
if not (wildcard_op.lower().strip() in ['any', 'all', 'none']
or wildcard_op.lower().strip().startswith('count')):
sys.exit("Unsupported wildcard operation: (%s). Exiting." %
wildcard_op)
sample_names = self._get_matching_sample_ids(wildcard)
if not self.test_mode:
print "%d samples matching sample wildcard" % len(sample_names)
# Replace HET, etc. with 1, et.session to avoid eval() issues.
wildcard_rule = self._swap_genotype_for_number(wildcard_rule)
wildcard_rule = wildcard_rule.replace('==', '=')
actual_nr_cores = min(len(sample_names), self.nr_cores)
return GT_wildcard_expression(column, wildcard_rule, wildcard_op, sample_names, \
self.db_contact_points, self.keyspace, self.n_variants, actual_nr_cores)
def _swap_genotype_for_number(self, token):
if any(g in token for g in ['HET', 'HOM_ALT', 'HOM_REF', 'UNKNOWN']):
token = token.replace('HET', str(HET))
token = token.replace('HOM_ALT', str(HOM_ALT))
token = token.replace('HOM_REF', str(HOM_REF))
token = token.replace('UNKNOWN', str(UNKNOWN))
return token
def _split_select(self):
"""
Build a list of _all_ columns in the SELECT statement
and segregated the non-genotype specific SELECT columns.
This is used to control how to report the results, as the
genotype-specific columns need to be eval()'ed whereas others
do not.
For example: "SELECT chrom, start, end, gt_types.1478PC0011"
will populate the lists as follows:
select_columns = ['chrom', 'start', 'end']
all_columns = ['chrom', 'start', 'end', 'gt_types[11]']
"""
self.select_columns = []
self.all_columns_new = []
self.all_columns_orig = []
# iterate through all of the select columns and clear
# distinguish the genotype-specific columns from the base columns
if "from" not in self.query.lower():
sys.exit("Malformed query: expected a FROM keyword.")
tokens_to_be_removed = set()
for token in self.requested_columns:
# it is a WILDCARD
if (token.find("gt") >= 0 or token.find("GT") >= 0) \
and '.(' in token and ').' in token:
# break the wildcard into its pieces. That is:
# (COLUMN).(WILDCARD)
(column, wildcard) = token.split('.')
# remove the syntactic parentheses
wildcard = wildcard.strip('(').strip(')')
column = column.strip('(').strip(')')
# convert "gt_types.(affected==1)"
# to: gt_types[3] == HET and gt_types[9] == HET
sample_info = self._get_matching_sample_ids(wildcard)
# maintain a list of the sample indices that should
# be displayed as a result of the SELECT'ed wildcard
for sample in sample_info:
wildcard_col = column + '_' + sample
self.requested_columns.append(wildcard_col)
tokens_to_be_removed.add(token)
for token in tokens_to_be_removed:
self.requested_columns.remove(token)
def _tokenize_query(self):
tokens = list(flatten([x.split(",") for x in self.query.split(" ")]))
return tokens
def _query_needs_genotype_info(self):
tokens = self._tokenize_query()
requested_genotype = "variants" in tokens and \
(any([x.startswith("gt") for x in tokens]) or \
any([x.startswith("(gt") for x in tokens]) or \
any(".gt" in x for x in tokens))
return requested_genotype or \
self.include_gt_cols or \
self.show_variant_samples or \
self.needs_genotypes
class GeminiQuery(object):
"""
An interface to submit queries to an existing Gemini database
and iterate over the results of the query.
We create a GeminiQuery object by specifying database to which to
connect::
from geminicassandra import GeminiQuery
gq = GeminiQuery("my.db")
We can then issue a query against the database and iterate through
the results by using the ``run()`` method::
for row in gq:
print row
Instead of printing the entire row, one access print specific columns::
gq.run("select chrom, start, end from variants")
for row in gq:
print row['chrom']
Also, all of the underlying numpy genotype arrays are
always available::
gq.run("select chrom, start, end from variants")
for row in gq:
gts = row.gts
print row['chrom'], gts
# yields "chr1" ['A/G' 'G/G' ... 'A/G']
The ``run()`` methods also accepts genotype filter::
query = "select chrom, start, end" from variants"
gt_filter = "gt_types.NA20814 == HET"
gq.run(query)
for row in gq:
print row
Lastly, one can use the ``sample_to_idx`` and ``idx_to_sample``
dictionaries to gain access to sample-level genotype information
either by sample name or by sample index::
# grab dict mapping sample to genotype array indices
smp2idx = gq.sample_to_idx
query = "select chrom, start, end from variants"
gt_filter = "gt_types.NA20814 == HET"
gq.run(query, gt_filter)
# print a header listing the selected columns
print gq.header
for row in gq:
# access a NUMPY array of the sample genotypes.
gts = row['gts']
# use the smp2idx dict to access sample genotypes
idx = smp2idx['NA20814']
print row, gts[idx]
"""
def __init__(self, db_contact_points, keyspace, include_gt_cols=False,
out_format=DefaultRowFormat(None)):
self.db_contact_points = map(strip, db_contact_points.split(','))
self.keyspace = keyspace
self.query_executed = False
self.for_browser = False
self.include_gt_cols = include_gt_cols
# try to connect to the provided database
self._connect_to_database()
self.n_variants = self.get_n_variants()
# list of samples ids for each clause in the --gt-filter
self.sample_info = collections.defaultdict(list)
'''# map sample names to indices. e.g. self.sample_to_idx[NA20814] -> 323
self.sample_to_idx = util.map_samples_to_indices(self.session)
# and vice versa. e.g., self.idx_to_sample[323] -> NA20814
self.idx_to_sample = util.map_indices_to_samples(self.session)
self.idx_to_sample_object = util.map_indices_to_sample_objects(self.session)
self.sample_to_sample_object = util.map_samples_to_sample_objects(self.session)'''
self.formatter = out_format
self.predicates = [self.formatter.predicate]
def get_n_variants(self):
res = self.session.execute("select n_rows from row_counts where table_name = 'variants'")
return res[0].n_rows
def run(self, query, gt_filter=None, show_variant_samples=False,
variant_samples_delim=',', predicates=None,
needs_genotypes=False, needs_genes=False,
show_families=False, test_mode=False,
needs_sample_names=False, nr_cores = 1,
start_time = -42, use_header = False,
exp_id="Oink", timeout=10.0, batch_size = 100):
"""
Execute a query against a Gemini database. The user may
specify:
1. (reqd.) an SQL `query`.
2. (opt.) a genotype filter.
"""
self.query = self.formatter.format_query(query).replace('==','=')
self.gt_filter = gt_filter
#print self.query + '; gt-filter = %s \n' % gt_filter
self.nr_cores = nr_cores
self.start_time = start_time
self.use_header = use_header
self.exp_id = exp_id
self.timeout = timeout
self.batch_size = batch_size
if self._is_gt_filter_safe() is False:
sys.exit("ERROR: unsafe --gt-filter command.")
self.show_variant_samples = show_variant_samples
self.variant_samples_delim = variant_samples_delim
self.test_mode = test_mode
self.needs_genotypes = needs_genotypes
self.needs_vcf_columns = False
if self.formatter.name == 'vcf':
self.needs_vcf_columns = True
self.needs_sample_names = needs_sample_names
self.needs_genes = needs_genes
self.show_families = show_families
if predicates:
self.predicates += predicates
# make sure the SELECT columns are separated by a
# comma and a space. then tokenize by spaces.
self.query = self.query.replace(',', ', ')
self.query_pieces = self.query.split()
if not any(s.startswith("gt") for s in self.query_pieces) and \
not any(s.startswith("(gt") for s in self.query_pieces) and \
not any(".gt" in s for s in self.query_pieces):
if self.gt_filter is None:
self.query_type = "no-genotypes"
else:
self.gt_filter_exp = self._correct_genotype_filter()
self.query_type = "filter-genotypes"
else:
if self.gt_filter is None:
self.query_type = "select-genotypes"
else:
self.gt_filter_exp = self._correct_genotype_filter()
self.query_type = "filter-genotypes"
(self.requested_columns, self.from_table, where_clause, self.rest_of_query) = get_query_parts(self.query)
self.extra_columns = []
if where_clause != '':
self.where_exp = self.parse_where_clause(where_clause, self.from_table)
if not self.gt_filter is None:
self.where_exp = AND_expression(self.where_exp, self.gt_filter_exp)
else:
if not self.gt_filter is None:
self.where_exp = self.gt_filter_exp
else:
self.where_exp = None
self._apply_query()
self.query_executed = True
def run_simple_query(self, query):
(requested_columns, from_table, where_clause, rest_of_query) = get_query_parts(query)
if where_clause != '':
where_exp = self.parse_where_clause(where_clause, from_table)
if not where_exp is None:
matches = where_exp.evaluate(self.session, "*")
if len(matches) == 0:
return OrderedDict([])
else:
try:
dink_query = "SELECT %s FROM %s" % (','.join(requested_columns), from_table)
if matches != "*":
if from_table.startswith('samples'):
in_clause = "','".join(matches)
dink_query += " WHERE %s IN ('%s')" % (self.get_partition_key(from_table), in_clause)
else:
in_clause = ",".join(map(str, matches))
dink_query += " WHERE %s IN (%s)" % (self.get_partition_key(from_table), in_clause)
dink_query += " " + rest_of_query
self.session.row_factory = ordered_dict_factory
return self.session.execute(dink_query)
except cassandra.protocol.SyntaxException as e:
print "Cassandra error: {0}".format(e)
sys.exit("The query issued (%s) has a syntax error." % query)
def __iter__(self):
return self
@property
def header(self):
"""
Return a header describing the columns that
were selected in the query issued to a GeminiQuery object.
"""
h = [col for col in self.report_cols]
if self.show_variant_samples:
h += ["variant_samples", "HET_samples", "HOM_ALT_samples"]
if self.show_families:
h += ["families"]
return self.formatter.header(h)
def next(self):
"""
Return the GeminiRow object for the next query result.
"""
while (1):
try:
row = self.result.next()
except Exception as e:
sys.__stderr__.write(str(e) + "\n")
self.cluster.shutdown()
raise StopIteration
return self.row_2_GeminiRow(row)
def row_2_GeminiRow(self, row):
variant_names = []
het_names = []
hom_alt_names = []
hom_ref_names = []
unknown_names = []
info = None
if 'info' in self.report_cols:
info = compression.unpack_ordereddict_blob(row['info'])
fields = OrderedDict()
for col in self.report_cols:
if col == "*":
continue
if not col == "info":
fields[col] = row[col]
elif col == "info":
fields[col] = _info_dict_to_string(info)
if self.show_variant_samples or self.needs_sample_names:
het_names = self._get_variant_samples(row['variant_id'], HET)
hom_alt_names = self._get_variant_samples(row['variant_id'], HOM_ALT)
hom_ref_names = self._get_variant_samples(row['variant_id'], HOM_REF)
unknown_names = self._get_variant_samples(row['variant_id'], UNKNOWN)
variant_names = het_names | hom_alt_names
if self.show_variant_samples:
fields["variant_samples"] = \
self.variant_samples_delim.join(variant_names)
fields["HET_samples"] = \
self.variant_samples_delim.join(het_names)
fields["HOM_ALT_samples"] = \
self.variant_samples_delim.join(hom_alt_names)
if self.show_families:
families = map(str, list(set([self.sample_to_sample_object[x].family_id
for x in variant_names])))
fields["families"] = self.variant_samples_delim.join(families)
gemini_row = GeminiRow(fields, variant_names, het_names, hom_alt_names,
hom_ref_names, unknown_names, info,
formatter=self.formatter)
if not all([predicate(gemini_row) for predicate in self.predicates]):
return None
if not self.for_browser:
return gemini_row
else:
return fields
def _get_variant_samples(self, variant_id, gt_type):
query = "SELECT sample_name from samples_by_variants_gt_type WHERE variant_id = %s AND gt_type = %s"
self.session.row_factory = tuple_factory
return async_rows_as_set(self.session, query % (variant_id, gt_type))
def _group_samples_by_genotype(self, gt_types):
"""
make dictionary keyed by genotype of list of samples with that genotype
"""
key_fn = lambda x: x[1]
val_fn = lambda x: self.idx_to_sample[x[0]]
return partition_by_fn(enumerate(gt_types), key_fn=key_fn, val_fn=val_fn)
def _connect_to_database(self):
"""
Establish a connection to the requested Gemini database.
"""
# open up a new database
self.cluster = Cluster(self.db_contact_points)
self.session = self.cluster.connect(self.keyspace)
def _is_gt_filter_safe(self):
"""
Test to see if the gt_filter string is potentially malicious.
A future improvement would be to use pyparsing to
traverse and directly validate the string.
"""
if self.gt_filter is None or len(self.gt_filter.strip()) == 0:
return True
# avoid builtins
# http://nedbatchelder.com/blog/201206/eval_really_is_dangerous.html
if "__" in self.gt_filter:
return False
# avoid malicious commands
evil = [" rm ", "os.system"]
if any(s in self.gt_filter for s in evil):
return False
# make sure a "gt" col is in the string
valid_cols = ["gts.", "gt_types.", "gt_phases.", "gt_quals.",
"gt_depths.", "gt_ref_depths.", "gt_alt_depths.", "gt_copy_numbers.",
"[gts].", "[gt_types].", "[gt_phases].", "[gt_quals].", "[gt_copy_numbers].",
"[gt_depths].", "[gt_ref_depths].", "[gt_alt_depths]."]
if any(s in self.gt_filter for s in valid_cols):
return True
# assume the worst
return False
def _execute_query(self):
n_matches = len(self.matches)
self.session.row_factory = ordered_dict_factory
query = "SELECT %s FROM %s" % (','.join(self.requested_columns + self.extra_columns), self.from_table)
error_count = 0
if n_matches == 0:
print "No results!"
time_taken = time.time() - self.start_time
elif not self.test_mode:
output_folder = self.exp_id + "_results"
if not os.path.exists(output_folder):
os.makedirs(output_folder)
output_path = output_folder + "/%s"
if self.matches == "*":
print "All rows match query."
query += " " + self.rest_of_query
error_count += execute_async_blocking(self.session, query, output_path % 0, self.extra_columns, (), self.timeout)
else:
print "%d rows match query." % len(self.matches)
step = len(self.matches) / self.nr_cores
procs = []
conns = []
for i in range(self.nr_cores):
parent_conn, child_conn = Pipe()
conns.append(parent_conn)
p = Process(target=fetch_matches,
args=(child_conn, i, output_path % i, query, self.from_table,\
self.get_partition_key(self.from_table), self.extra_columns,\
self.db_contact_points, self.keyspace, self.batch_size))
procs.append(p)
p.start()
for i in range(self.nr_cores):
n = len(self.matches)
begin = i*step + min(i, n % self.nr_cores)
end = begin + step
if i < n % self.nr_cores:
end += 1
conns[i].send(self.matches[begin:end])
for i in range(self.nr_cores):
errs = conns[i].recv()
error_count += errs
conns[i].close()
procs[i].join()
time_taken = time.time() - self.start_time
print "Query %s completed in %s s." % (self.exp_id, time_taken)
print "Query %s encountered %d errors" % (self.exp_id, error_count)
else:
signal(SIGPIPE,SIG_DFL)
'''
Test mode stuff: no extra prints, synchronous execution, sort results
'''
try:
query = "SELECT %s FROM %s" % (','.join(self.requested_columns + self.extra_columns), self.from_table)
if self.matches != "*":
if self.from_table != 'samples':
in_clause = ",".join(map(str, self.matches))
query += " WHERE %s IN (%s)" % (self.get_partition_key(self.from_table), in_clause)
else:
in_clause = "','".join(self.matches)
query += " WHERE %s IN ('%s')" % (self.get_partition_key(self.from_table), in_clause)
query += " " + self.rest_of_query
res = self.session.execute(query)
if self.from_table == 'variants':
res = sorted(res, key = lambda x: x['start'])
elif self.from_table == 'samples':
res = sorted(res, key = lambda x: x['sample_id'])
self.report_cols = filter(lambda x: not x in self.extra_columns, res[0].keys())
if self.use_header and self.header:
print self.header
for row in res:
print self.row_2_GeminiRow(row)
except Exception as e:
sys.stderr.write(str(e))
finally:
time_taken = time.time() - self.start_time
with open("querylog", 'a') as log:
log.write("2::%s;%s;%d\n" % (self.exp_id, time_taken, error_count))
def set_report_cols(self, rep_cols):
self.report_cols = rep_cols
def _apply_query(self):
"""
Execute a query. Intercept gt* columns and
replace sample names with indices where necessary.
"""
if self.needs_genes:
self.requested_columns.append("gene")
'''if self.needs_vcf_columns:
self.query = self._add_vcf_cols_to_query()'''
self.matches = "*"
if not self.where_exp is None:
self.matches = list(self.where_exp.evaluate(self.session, "*"))
mid_time = time.time()
log = open("querylog", 'a')
log.write("1::%s;%s\n" % (self.exp_id, mid_time - self.start_time))
log.close()
if self._query_needs_genotype_info():
# break up the select statement into individual
# pieces and replace genotype columns using sample
# names with sample indices
self._split_select()
if self.show_families or self.show_variant_samples or self.needs_sample_names:
if (not 'variant_id' in self.requested_columns) and (not "*" in self.requested_columns):
self.extra_columns.append('variant_id')
if self.test_mode and self.from_table == 'variants' and not 'start' in self.requested_columns:
self.extra_columns.append('start')
self._execute_query()
def shutdown(self):
self.session.shutdown()
self.cluster.shutdown()
def _get_matching_sample_ids(self, wildcard):
if wildcard.strip() != "*":
query = self.parse_where_clause(wildcard, 'samples')
else:
query = Basic_expression('samples', 'name', "")
return list(query.evaluate(self.session, '*'))
def get_partition_key(self, table):
key = self.cluster.metadata.keyspaces[self.keyspace].tables[table].partition_key[0].name
return key
def _correct_genotype_filter(self):
"""
This converts a raw genotype filter that contains
'wildcard' statements into a filter that can be eval()'ed.
Specifically, we must convert a _named_ genotype index
to a _numerical_ genotype index so that the appropriate
value can be extracted for the sample from the genotype
numpy arrays.
For example, without WILDCARDS, this converts:
--gt-filter "(gt_types.1478PC0011 == 1)"
to:
(gt_types[11] == 1)
With WILDCARDS, this converts things like:
"(gt_types).(phenotype==1).(==HET)"
to:
"gt_types[2] == HET and gt_types[5] == HET"
"""
return self.parse_clause(self.gt_filter, self.gt_base_parser, 'variants')
def parse_where_clause(self, where_clause, table):
return self.parse_clause(where_clause, lambda x: self.where_clause_to_exp(table, self.get_partition_key(table), x), table)
def parse_clause(self, clause, base_clause_parser, table):
clause = clause.strip()
depth = 0
min_depth = 100000 #Arbitrary bound on nr of nested clauses.
in_wildcard_clause = False
for i in range(0,len(clause)):
if clause[i] == '[':
in_wildcard_clause = True
elif clause[i] == ']':
in_wildcard_clause = False
elif in_wildcard_clause: #currently in wildcard thingy, so doesn't mean anything. Move on.
continue
elif clause[i] == '(':
depth += 1
elif clause[i] == ')':
depth -= 1
elif i < len(clause) - 2:
if clause[i:i+2] == "||":
if depth == 0:
left = self.parse_clause(clause[:i].strip(), base_clause_parser, table)
right = self.parse_clause(clause[i+2:].strip(), base_clause_parser, table)
return OR_expression(left, right)
else:
min_depth = min(min_depth, depth)
elif clause[i:i+2] == "&&":
if depth == 0:
left = self.parse_clause(clause[:i].strip(), base_clause_parser, table)
right = self.parse_clause(clause[i+2:].strip(), base_clause_parser, table)
return AND_expression(left, right)
else:
min_depth = min(min_depth, depth)
elif i < len(clause) - 3:
if clause[i:i+3] == "NOT":
if depth == 0:
body = self.parse_clause(clause[i+3:].strip(), base_clause_parser, table)
return NOT_expression(body, table, self.get_partition_key(table), self.n_variants)
else:
min_depth = min(min_depth, depth)
if depth == 0:
if min_depth < 100000:
#Strip away all brackets to expose uppermost boolean operator
return self.parse_clause(clause[min_depth:len(clause)-min_depth], base_clause_parser, table)
else:
#No more boolean operators, strip all remaining brackets
token = clause.strip('(').strip(')')
return base_clause_parser(token)
else:
sys.exit("ERROR in %s. Brackets don't match" % clause)
def where_clause_to_exp(self, table, cols, clause):
target_table = self.get_table_from_where_clause(table, clause)
exp = Basic_expression(target_table, cols, clause)
return exp
def get_table_from_where_clause(self, table, where_clause):
where_clause = where_clause.replace('==','=')
clauses = where_clause.split("and")
range_clauses = filter(lambda x: '<' in x or '>' in x, clauses)
range_col = None
for range_clause in range_clauses:
i = 0
for op in ["<", ">"]:
temp = range_clause.find(op)
if temp > -1:
i = temp
break
col = range_clause[0:i].strip()
if range_col and (col != range_col):
sys.exit("ERROR: range clauses only possible on at most one column")
else:
range_col = col
eq_clauses = filter(lambda x: not ('<' in x or '>' in x), clauses)
eq_clauses = map(lambda x: x.split('='), eq_clauses)
eq_columns = map(lambda x: x[0].strip(), eq_clauses)
relevant_tables = self.get_relevant_tables(table)
for t in relevant_tables:
if all(x in eq_columns for x in t.partition_key):
other_cols = filter(lambda x: not x in t.partition_key, eq_columns)
if all (x in t.clustering_key[:len(other_cols)] for x in other_cols):
if range_col:
if range_col == t.clustering_key[len(other_cols)]:
return t.name
else:
return t.name
sys.exit("ERROR: No suitable table found for query: %s" % where_clause)
def get_relevant_tables(self, table):
Row = namedtuple('Row', 'name partition_key clustering_key')
tables = self.cluster.metadata.keyspaces[self.keyspace].tables
interesting_tables = filter(lambda x: x.startswith(table), tables.keys())
res = []
for table in interesting_tables:
res.append(Row(table, map(lambda y: y.name, tables[table].partition_key),\
map(lambda y: y.name, tables[table].clustering_key)))
return res
def gt_base_parser(self, clause):
if (clause.find("gt") >= 0 or clause.find("GT") >= 0) and not '[' in clause:
return self.gt_filter_to_query_exp(clause.replace('==', '='))
elif (clause.find("gt") >= 0 or clause.find("GT") >= 0) and '[' in clause:
dink = self.parse_gt_wildcard(clause)
#print dink.to_string()
return dink
else:
sys.exit("ERROR: invalid --gt-filter command")
def gt_filter_to_query_exp(self, gt_filter):
i = -1
operators = ['!=', '<=', '>=', '=', '<', '>']
for op in operators:
temp = gt_filter.find(op)
if temp > -1:
i = temp
break
if i > -1:
left = gt_filter[0:i].strip()
clause = self._swap_genotype_for_number(gt_filter[i:].strip())
else:
sys.exit("ERROR: invalid --gt-filter command 858.")
not_exp = False
if clause.startswith('!'):
not_exp = True
clause = clause[1:]
(column, sample) = left.split('.', 1)
exp = Basic_expression('variants_by_samples_' + column, 'variant_id' , "sample_name = '" + sample + "' AND " + column + clause)
if not_exp:
return NOT_expression(exp, 'variants', 'variant_id', self.n_variants)
else:
return exp
def parse_gt_wildcard(self, token):
if token.count('.') != 3 or \
token.count('[') != 4 or \
token.count(']') != 4:
sys.exit("Wildcard filter should consist of 4 elements. Exiting.")
(column, wildcard, wildcard_rule, wildcard_op) = token.split('.')
column = column.strip('[').strip(']').strip()
wildcard = wildcard.strip('[').strip(']').strip().replace('==', '=')
wildcard_rule = wildcard_rule.strip('[').strip(']').strip()
wildcard_op = wildcard_op.strip('[').strip(']').strip()
if not (wildcard_op.lower().strip() in ['any', 'all', 'none'] or wildcard_op.lower().strip().startswith('count')):
sys.exit("Unsupported wildcard operation: (%s). Exiting." % wildcard_op)
sample_names = self._get_matching_sample_ids(wildcard)
if not self.test_mode:
print "%d samples matching sample wildcard" % len(sample_names)
# Replace HET, etc. with 1, et.session to avoid eval() issues.
wildcard_rule = self._swap_genotype_for_number(wildcard_rule)
wildcard_rule = wildcard_rule.replace('==', '=')
actual_nr_cores = min(len(sample_names), self.nr_cores)
return GT_wildcard_expression(column, wildcard_rule, wildcard_op, sample_names, \
self.db_contact_points, self.keyspace, self.n_variants, actual_nr_cores)
def _swap_genotype_for_number(self, token):
if any(g in token for g in ['HET', 'HOM_ALT', 'HOM_REF', 'UNKNOWN']):
token = token.replace('HET', str(HET))
token = token.replace('HOM_ALT', str(HOM_ALT))
token = token.replace('HOM_REF', str(HOM_REF))
token = token.replace('UNKNOWN', str(UNKNOWN))
return token
def _split_select(self):
"""
Build a list of _all_ columns in the SELECT statement
and segregated the non-genotype specific SELECT columns.
This is used to control how to report the results, as the
genotype-specific columns need to be eval()'ed whereas others
do not.
For example: "SELECT chrom, start, end, gt_types.1478PC0011"
will populate the lists as follows:
select_columns = ['chrom', 'start', 'end']
all_columns = ['chrom', 'start', 'end', 'gt_types[11]']
"""
self.select_columns = []
self.all_columns_new = []
self.all_columns_orig = []
# iterate through all of the select columns and clear
# distinguish the genotype-specific columns from the base columns
if "from" not in self.query.lower():
sys.exit("Malformed query: expected a FROM keyword.")
tokens_to_be_removed = set()
for token in self.requested_columns:
# it is a WILDCARD
if (token.find("gt") >= 0 or token.find("GT") >= 0) \
and '.(' in token and ').' in token:
# break the wildcard into its pieces. That is:
# (COLUMN).(WILDCARD)
(column, wildcard) = token.split('.')
# remove the syntactic parentheses
wildcard = wildcard.strip('(').strip(')')
column = column.strip('(').strip(')')
# convert "gt_types.(affected==1)"
# to: gt_types[3] == HET and gt_types[9] == HET
sample_info = self._get_matching_sample_ids(wildcard)
# maintain a list of the sample indices that should
# be displayed as a result of the SELECT'ed wildcard
for sample in sample_info:
wildcard_col = column + '_' + sample
self.requested_columns.append(wildcard_col)
tokens_to_be_removed.add(token)
for token in tokens_to_be_removed:
self.requested_columns.remove(token)
def _tokenize_query(self):
tokens = list(flatten([x.split(",") for x in self.query.split(" ")]))
return tokens
def _query_needs_genotype_info(self):
tokens = self._tokenize_query()
requested_genotype = "variants" in tokens and \
(any([x.startswith("gt") for x in tokens]) or \
any([x.startswith("(gt") for x in tokens]) or \
any(".gt" in x for x in tokens))
return requested_genotype or \
self.include_gt_cols or \
self.show_variant_samples or \
self.needs_genotypes
