def createReader(readable, **keywords):
"""
A reader fitting the contents of ``readable``. Supported formats are
delimited data (such as CSV), ODS and Excel. When iterating the reader,
it returns a Python array for each row of data.
"""
# TODO: Get rid of circular import.
import _parsers
assert readable is not None
encoding = keywords.get("encoding", DEFAULT_ENCODING)
assert encoding is not None
result = None
icdHeader = readable.read(4)
_log.debug(u"header=%r", icdHeader)
if icdHeader == _ODS_HEADER:
# Consider ICD to be ODS.
readable.seek(0)
result = _parsers.odsReader(readable)
else:
icdHeader += readable.read(4)
readable.seek(0)
if _tools.isEqualBytes(icdHeader, _EXCEL_HEADER):
# Consider ICD to be Excel.
result = _parsers.excelReader(readable)
else:
# Consider ICD to be CSV.
dialect = _parsers.DelimitedDialect()
dialect.lineDelimiter = _parsers.AUTO
dialect.itemDelimiter = _parsers.AUTO
dialect.quoteChar = "\""
dialect.escapeChar = "\""
result = _parsers.delimitedReader(readable, dialect, encoding)
return result
def testReader(self):
dialect = self._createDefaultDialect()
dataStream = StringIO.StringIO("hugo,was" + _parsers.LF + "here,again")
csvReader = _parsers.delimitedReader(dataStream, dialect)
rowCount = 0
for row in csvReader:
rowCount += 1
self.assertEqual(2, len(row))
self.assertEqual(2, rowCount)
def testReader(self):
dialect = self._createDefaultDialect()
dataStream = StringIO.StringIO("hugo,was" + _parsers.LF + "here,again")
csvReader = _parsers.delimitedReader(dataStream, dialect)
rowCount = 0
for row in csvReader:
rowCount += 1
self.assertEqual(2, len(row))
self.assertEqual(2, rowCount)
def testAutoItemDelimiter(self):
dialect = self._createDefaultDialect()
dialect.itemDelimiter = _parsers.AUTO
dataStream = StringIO.StringIO("some;items;using;a;semicolon;as;separator")
csvReader = _parsers.delimitedReader(dataStream, dialect)
rowCount = 0
for row in csvReader:
rowCount += 1
self.assertEqual(7, len(row))
self.assertEqual(1, rowCount)
def testAutoItemDelimiter(self):
dialect = self._createDefaultDialect()
dialect.itemDelimiter = _parsers.AUTO
dataStream = StringIO.StringIO(
"some;items;using;a;semicolon;as;separator")
csvReader = _parsers.delimitedReader(dataStream, dialect)
rowCount = 0
for row in csvReader:
rowCount += 1
self.assertEqual(7, len(row))
self.assertEqual(1, rowCount)
def _assertRowsEqual(self, expectedRows, readable, dialect=None):
"""
Simply parse all items of `readable` using `dialect` and assert that the number of items read matches `expectedItem`.
"""
assert expectedRows is not None
assert readable is not None
actualReadable = self.possiblyStringIoedReadable(readable)
if dialect is None:
actualDialect = self._createDefaultDialect()
else:
actualDialect = dialect
reader = _parsers.delimitedReader(actualReadable, actualDialect)
self.readAndAssertEquals(expectedRows, reader)
def _assertRowsEqual(self, expectedRows, readable, dialect=None):
"""
Simply parse all items of `readable` using `dialect` and assert that the number of items read matches `expectedItem`.
"""
assert expectedRows is not None
assert readable is not None
actualReadable = self.possiblyStringIoedReadable(readable)
if dialect is None:
actualDialect = self._createDefaultDialect()
else:
actualDialect = dialect
reader = _parsers.delimitedReader(actualReadable, actualDialect)
self.readAndAssertEquals(expectedRows, reader)
def _assertRaisesParserSyntaxError(self, readable, dialect=None):
"""
Attempt to parse all items of `readable` using `dialect` and assert that this raises _`_parsers.ParserSyntaxError`.
"""
assert readable is not None
actualReadable = self.possiblyStringIoedReadable(readable)
if dialect is None:
actualDialect = self._createDefaultDialect()
else:
actualDialect = dialect
try:
reader = _parsers.delimitedReader(actualReadable, actualDialect)
for dummy in reader:
pass
# FIXME: self.fail(u"readable must raise %s" % _parsers.ParserSyntaxError.__name__)
except _parsers.ParserSyntaxError:
# Ignore expected error.
pass
def _assertRaisesParserSyntaxError(self, readable, dialect=None):
"""
Attempt to parse all items of `readable` using `dialect` and assert that this raises _`_parsers.ParserSyntaxError`.
"""
assert readable is not None
actualReadable = self.possiblyStringIoedReadable(readable)
if dialect is None:
actualDialect = self._createDefaultDialect()
else:
actualDialect = dialect
try:
reader = _parsers.delimitedReader(actualReadable, actualDialect)
for dummy in reader:
pass
# FIXME: self.fail(u"readable must raise %s" % _parsers.ParserSyntaxError.__name__)
except _parsers.ParserSyntaxError:
# Ignore expected error.
pass
