def _load_from_stream(self):
"""Load content from memory
:params stream file_content: the actual file content in memory
:returns: a book
"""
self.__load_from_memory_flag = True
self.__line_terminator = self._keywords.get(
constants.KEYWORD_LINE_TERMINATOR, self.__line_terminator
)
separator = DEFAULT_SHEET_SEPARATOR_FORMATTER % self.__line_terminator
if self.__multiple_sheets:
# will be slow for large files
self._file_stream.seek(0)
content = self._file_stream.read()
sheets = content.split(separator)
named_contents = []
for sheet in sheets:
if sheet == "": # skip empty named sheet
continue
lines = sheet.split(self.__line_terminator)
result = re.match(constants.SEPARATOR_MATCHER, lines[0])
new_content = "\n".join(lines[1:])
new_sheet = NamedContent(
result.group(1), compact.StringIO(new_content)
)
named_contents.append(new_sheet)
return named_contents
else:
if hasattr(self._file_stream, "seek"):
self._file_stream.seek(0)
return [NamedContent(self._file_type, self._file_stream)]
def get_file_handle(self):
if compact.PY2:
f = UTF8Recorder(self._native_sheet.payload, self._encoding)
else:
if isinstance(self._native_sheet.payload, compact.BytesIO):
content = self._native_sheet.payload.read()
f = compact.StringIO(content.decode(self._encoding))
else:
f = self._native_sheet.payload
return f
def get_file_handle(self):
if isinstance(self._native_sheet.payload, compact.BytesIO):
# please note that
# if the end developer feed us bytesio in python3
# we will do the conversion to StriongIO but that
# comes at a cost.
content = self._native_sheet.payload.read()
unicode_reader = compact.StringIO(content.decode(self._encoding))
else:
unicode_reader = self._native_sheet.payload
return unicode_reader
def get_file_handle(self):
if compact.PY2:
if hasattr(self._native_sheet.payload, 'read'):
f = UTF8Recorder(self._native_sheet.payload, self._encoding)
else:
f = self._native_sheet.payload
else:
if isinstance(self._native_sheet.payload, compact.BytesIO):
# please note that
# if the end developer feed us bytesio in python3
# we will do the conversion to StriongIO but that
# comes at a cost.
content = self._native_sheet.payload.read()
f = compact.StringIO(content.decode(self._encoding))
else:
f = self._native_sheet.payload
return f
def __init__(self,
file_stream,
file_type,
multiple_sheets=False,
**keywords):
"""Load content from memory
:params stream file_content: the actual file content in memory
:returns: a book
"""
self.handles = []
self.keywords = keywords
if file_type == constants.FILE_FORMAT_TSV:
self.keywords["dialect"] = constants.KEYWORD_TSV_DIALECT
self.file_type = file_type
self.__load_from_memory_flag = True
self.__line_terminator = keywords.get(
constants.KEYWORD_LINE_TERMINATOR, constants.DEFAULT_CSV_NEWLINE)
separator = DEFAULT_SHEET_SEPARATOR_FORMATTER % self.__line_terminator
if multiple_sheets:
# will be slow for large files
file_stream.seek(0)
content = file_stream.read()
sheets = content.split(separator)
named_contents = []
for sheet in sheets:
if sheet == "": # skip empty named sheet
continue
lines = sheet.split(self.__line_terminator)
result = re.match(constants.SEPARATOR_MATCHER, lines[0])
new_content = "\n".join(lines[1:])
new_sheet = NamedContent(result.group(1),
compact.StringIO(new_content))
named_contents.append(new_sheet)
self.content_array = named_contents
else:
if hasattr(file_stream, "seek"):
file_stream.seek(0)
self.content_array = [NamedContent(self.file_type, file_stream)]
def __init__(self, f, encoding="utf-8", **kwds):
# Redirect output to a queue
self.queue = compact.StringIO()
self.writer = csv.writer(self.queue, **kwds)
self.stream = f
self.encoder = codecs.getincrementalencoder(encoding)()
