opensource/galaxy
3
0
Fork 0
mirror of https://github.com/galaxyproject/galaxy.git synced 2026-09-24 16:30:27 +08:00
Code Issues Packages Projects Releases Wiki Activity

Implement Nicola Sorenzo's proposal for set_meta() method.

Browse Source
This commit is contained in:
Pierrick Roger Mele
2017-07-06 08:41:49 +02:00
parent edbe273a95
commit 8b01873175
1 changed files with 28 additions and 29 deletions
Download Patch File Download Diff File Expand all files Collapse all files
lib/galaxy/datatypes/tabular.py
+28 -29
View File
@@ -968,38 +968,37 @@ class BaseCSV( TabularData ):
return False
def set_meta( self, dataset, **kwd ):
with open(dataset.file_name, 'r') as csvfile:
# Parse file with the correct dialect
reader = csv.reader(csvfile, self.dialect)
data_row = None
header_row = None
try:
header_row = next(reader)
data_row = next(reader)
for row in reader:
column_types = []
header_row = []
data_row = []
data_lines = 0
if dataset.has_data():
with open(dataset.file_name, 'r') as csvfile:
# Parse file with the correct dialect
reader = csv.reader(csvfile, self.dialect)
try:
header_row = next(reader)
data_row = next(reader)
for row in reader:
pass
except StopIteration:
pass
except StopIteration:
pass
except csv.Error as e:
raise Exception('CSV reader error - line %d: %s' % (reader.line_num, e))
except csv.Error as e:
raise Exception('CSV reader error - line %d: %s' % (reader.line_num, e))
else:
data_lines = reader.line_num - 1
# Guess column types
column_types = []
if data_row is not None:
for cell in data_row:
column_types.append(self.guess_type(cell))
# Set metadata
dataset.metadata.data_lines = reader.line_num - 1
dataset.metadata.comment_lines = 1
dataset.metadata.column_types = column_types
if data_row is None:
dataset.metadata.columns = len( header_row )
else:
dataset.metadata.columns = max( len( header_row ), len( data_row ) )
dataset.metadata.column_names = header_row
dataset.metadata.delimiter = reader.dialect.delimiter
# Guess column types
for cell in data_row:
column_types.append(self.guess_type(cell))
# Set metadata
dataset.metadata.data_lines = data_lines
dataset.metadata.comment_lines = int(bool(header_row))
dataset.metadata.column_types = column_types
dataset.metadata.columns = max( len( header_row ), len( data_row ) )
dataset.metadata.column_names = header_row
dataset.metadata.delimiter = self.dialect.delimiter
@dataproviders.decorators.has_dataproviders
class CSV( BaseCSV ):