def align_from_files(self, reffile, targetfile,
ref_indexes=None, target_indexes=None,
ref_encoding=None, target_encoding=None,
ref_separator='\t', target_separator='\t',
get_matrix=True):
""" Align data from files
Parameters
----------
reffile: name of the reference file
targetfile: name of the target file
ref_encoding: if given (e.g. 'utf-8' or 'latin-1'), it will
be used to read the files.
target_encoding: if given (e.g. 'utf-8' or 'latin-1'), it will
be used to read the files.
ref_separator: separator of the reference file
target_separator: separator of the target file
"""
refset = parsefile(reffile, indexes=ref_indexes,
encoding=ref_encoding, delimiter=ref_separator)
targetset = parsefile(targetfile, indexes=target_indexes,
encoding=target_encoding, delimiter=target_separator)
return self.align(refset, targetset, get_matrix=get_matrix)
def get_aligned_pairs_from_files(self, reffile, targetfile,
ref_indexes=None, target_indexes=None,
ref_encoding=None, target_encoding=None,
ref_separator='\t', target_separator='\t',
unique=True):
""" Get the pairs of aligned elements
"""
refset = parsefile(reffile, indexes=ref_indexes,
encoding=ref_encoding, delimiter=ref_separator)
targetset = parsefile(targetfile, indexes=target_indexes,
encoding=target_encoding, delimiter=target_separator)
global_mat, global_matched = self.align(refset, targetset, get_matrix=False)
for pair in iter_aligned_pairs(refset, targetset, global_mat, global_matched, unique):
yield pair
def test_parser(self):
data = parsefile(path.join(TESTDIR, 'data', 'file2parse'),
[0, (2, 3), 4, 1], delimiter=',')
self.assertEqual([[1, (12, 19), u'apple', u'house'],
[2, (21.9, 19), u'stramberry', u'horse'],
[3, (23, 2.17), u'cherry', u'flower']], data)
data = parsefile(path.join(TESTDIR, 'data', 'file2parse'),
[0, (2, 3), 4, 1], delimiter=',', formatopt={2:str})
self.assertEqual([[1, ('12', 19), u'apple', u'house'],
[2, ('21.9', 19), u'stramberry', u'horse'],
[3, ('23', 2.17), u'cherry', u'flower']], data)
data = parsefile(path.join(TESTDIR, 'data', 'file2parse'),
[0, (2, 3), 4, 1], delimiter=',', autocast_data=False)
self.assertEqual([['1', ('12', '19'), 'apple', 'house'],
['2', ('21.9', '19'), 'stramberry', 'horse'],
['3', ('23', '2.17'), 'cherry', 'flower']], data)