def export_to_html(table, filename_or_fobj=None, encoding='utf-8'):
fields = table.fields.keys()
result = ['<table>\n\n', ' <thead>\n', ' <tr>\n']
header = [' <th> {} </th>\n'.format(field) for field in fields]
result.extend(header)
result.extend([' </tr>\n', ' </thead>\n', '\n', ' <tbody>\n', '\n'])
for index, row in enumerate(serialize(table, encoding=encoding), start=1):
css_class = 'odd' if index % 2 == 1 else 'even'
result.append(' <tr class="{}">\n'.format(css_class))
for value in row:
result.extend([' <td> ', value, ' </td>\n'])
result.append(' </tr>\n\n')
result.append(' </tbody>\n\n</table>\n')
new_result = [
value.encode(encoding) if isinstance(value, unicode) else value
for value in result
]
html = ''.encode(encoding).join(new_result)
if filename_or_fobj is not None:
filename, fobj = get_filename_and_fobj(filename_or_fobj, mode='w')
fobj.write(html)
fobj.flush()
return fobj
else:
return html
def export_to_txt(table, filename_or_fobj, encoding='utf-8', *args, **kwargs):
# TODO: will work only if table.fields is OrderedDict
# TODO: should use fobj? What about creating a method like json.dumps?
filename, fobj = get_filename_and_fobj(filename_or_fobj, mode='w')
max_sizes = _max_column_sizes(table, encoding, *args, **kwargs)
fields = table.fields.keys()
dashes = [DASH * (max_sizes[field] + 2) for field in fields]
header = [field.center(max_sizes[field]) for field in fields]
header = '{} {} {}'.format(PIPE, ' {} '.format(PIPE).join(header), PIPE)
split_line = PLUS + PLUS.join(dashes) + PLUS
result = [split_line, header, split_line]
for row in serialize(table):
values = [
value.rjust(max_sizes[field_name])
for field_name, value in zip(fields, row)
]
row_data = ' {} '.format(PIPE).join(values)
result.append('{} {} {}'.format(PIPE, row_data, PIPE))
result.extend([split_line, '\n'])
data = '\n'.join(result).encode(encoding)
fobj.write(data)
fobj.flush()
return fobj
def export_to_html(table, filename_or_fobj=None, encoding='utf-8'):
fields = table.fields.keys()
result = ['<table>\n\n', ' <thead>\n', ' <tr>\n']
header = [' <th> {} </th>\n'.format(field) for field in fields]
result.extend(header)
result.extend([' </tr>\n', ' </thead>\n', '\n', ' <tbody>\n', '\n'])
for index, row in enumerate(serialize(table, encoding=encoding), start=1):
css_class = 'odd' if index % 2 == 1 else 'even'
result.append(' <tr class="{}">\n'.format(css_class))
for value in row:
result.extend([' <td> ', value, ' </td>\n'])
result.append(' </tr>\n\n')
result.append(' </tbody>\n\n</table>\n')
new_result = [value.encode(encoding) if isinstance(value, unicode)
else value
for value in result]
html = ''.encode(encoding).join(new_result)
if filename_or_fobj is not None:
filename, fobj = get_filename_and_fobj(filename_or_fobj, mode='w')
fobj.write(html)
fobj.flush()
return fobj
else:
return html
def export_to_txt(table, filename_or_fobj, encoding='utf-8', *args, **kwargs):
# TODO: will work only if table.fields is OrderedDict
# TODO: should use fobj? What about creating a method like json.dumps?
filename, fobj = get_filename_and_fobj(filename_or_fobj, mode='w')
max_sizes = _max_column_sizes(table, encoding, *args, **kwargs)
fields = table.fields.keys()
dashes = [DASH * (max_sizes[field] + 2) for field in fields]
header = [field.center(max_sizes[field]) for field in fields]
header = '{} {} {}'.format(PIPE, ' {} '.format(PIPE).join(header), PIPE)
split_line = PLUS + PLUS.join(dashes) + PLUS
result = [split_line, header, split_line]
for row in serialize(table):
values = [value.rjust(max_sizes[field_name])
for field_name, value in zip(fields, row)]
row_data = ' {} '.format(PIPE).join(values)
result.append('{} {} {}'.format(PIPE, row_data, PIPE))
result.extend([split_line, '\n'])
data = '\n'.join(result).encode(encoding)
fobj.write(data)
fobj.flush()
return fobj
def import_from_html(filename_or_fobj, encoding='utf-8', index=0,
ignore_colspan=True, preserve_html=False, row_tag='tr',
column_tag='td|th', *args, **kwargs):
# TODO: unescape before returning: html_parser.unescape(html)
# TODO: lxml -> unicode?
filename, fobj = get_filename_and_fobj(filename_or_fobj)
kwargs['encoding'] = encoding
html = fobj.read().decode(encoding)
html_tree = document_fromstring(html)
tables = html_tree.xpath('//table')
table = tables[index]
strip_tags(table, 'thead')
strip_tags(table, 'tbody')
row_elements = table.xpath(row_tag)
if not preserve_html:
table_rows = [[value_element.text_content().strip()
for value_element in row.xpath(column_tag)]
for row in row_elements]
else:
table_rows = [[_get_content(value_element)
for value_element in row.xpath(column_tag)]
for row in row_elements]
max_columns = max(len(row) for row in table_rows)
if ignore_colspan:
table_rows = filter(lambda row: len(row) == max_columns, table_rows)
meta = {'imported_from': 'html', 'filename': filename,}
return create_table(table_rows, meta=meta, *args, **kwargs)
def import_from_csv(filename_or_fobj, encoding='utf-8', delimiter=',',
quotechar='"', *args, **kwargs):
'Import data from a CSV file'
filename, fobj = get_filename_and_fobj(filename_or_fobj)
kwargs['encoding'] = encoding
csv_reader = unicodecsv.reader(fobj, encoding=encoding,
delimiter=str(delimiter),
quotechar=str(quotechar))
meta = {'imported_from': 'csv', 'filename': filename,}
return create_table(csv_reader, meta=meta, *args, **kwargs)
def export_to_csv(table, filename_or_fobj, encoding='utf-8'):
# TODO: will work only if table.fields is OrderedDict
# TODO: should use fobj? What about creating a method like json.dumps?
filename, fobj = get_filename_and_fobj(filename_or_fobj, mode='w')
csv_writer = unicodecsv.writer(fobj, encoding=encoding)
csv_writer.writerow(table.fields.keys())
for row in serialize(table, encoding=encoding):
csv_writer.writerow(row)
fobj.flush()
return fobj
def import_from_xls(filename_or_fobj,
sheet_name=None,
sheet_index=0,
start_row=0,
start_column=0,
*args,
**kwargs):
filename, _ = get_filename_and_fobj(filename_or_fobj)
book = xlrd.open_workbook(filename, formatting_info=True)
if sheet_name is not None:
sheet = book.sheet_by_name(sheet_name)
else:
sheet = book.sheet_by_index(sheet_index)
# TODO: may re-use Excel data types
# Get field names
# TODO: may use sheet.col_values or even sheet.ncols
column_count = 0
header = []
column_value = cell_value(sheet, start_row, start_column + column_count)
while column_value:
header.append(column_value)
column_count += 1
column_value = cell_value(sheet, start_row,
start_column + column_count)
# Get sheet rows
# TODO: may use sheel.col_slice or even sheet.nrows
table_rows = []
row_count = 0
start_row += 1
cell_is_empty = False
while not cell_is_empty:
row = [
cell_value(sheet, start_row + row_count,
start_column + column_index)
for column_index in range(column_count)
]
cell_is_empty = not any(row)
if not cell_is_empty:
table_rows.append(row)
row_count += 1
meta = {
'imported_from': 'xls',
'filename': filename,
}
return create_table([header] + table_rows, meta=meta, *args, **kwargs)
def import_from_html(filename_or_fobj,
encoding='utf-8',
index=0,
ignore_colspan=True,
preserve_html=False,
row_tag='tr',
column_tag='td|th',
*args,
**kwargs):
# TODO: unescape before returning: html_parser.unescape(html)
# TODO: lxml -> unicode?
filename, fobj = get_filename_and_fobj(filename_or_fobj)
kwargs['encoding'] = encoding
html = fobj.read().decode(encoding)
html_tree = document_fromstring(html)
tables = html_tree.xpath('//table')
table = tables[index]
strip_tags(table, 'thead')
strip_tags(table, 'tbody')
row_elements = table.xpath(row_tag)
if not preserve_html:
table_rows = [[
value_element.text_content().strip()
for value_element in row.xpath(column_tag)
] for row in row_elements]
else:
table_rows = [[
_get_content(value_element)
for value_element in row.xpath(column_tag)
] for row in row_elements]
max_columns = max(len(row) for row in table_rows)
if ignore_colspan:
table_rows = filter(lambda row: len(row) == max_columns, table_rows)
meta = {
'imported_from': 'html',
'filename': filename,
}
return create_table(table_rows, meta=meta, *args, **kwargs)
def import_from_xls(filename_or_fobj, sheet_name=None, sheet_index=0,
start_row=0, start_column=0, *args, **kwargs):
filename, _ = get_filename_and_fobj(filename_or_fobj)
book = xlrd.open_workbook(filename, formatting_info=True)
if sheet_name is not None:
sheet = book.sheet_by_name(sheet_name)
else:
sheet = book.sheet_by_index(sheet_index)
# TODO: may re-use Excel data types
# Get field names
# TODO: may use sheet.col_values or even sheet.ncols
column_count = 0
header = []
column_value = cell_value(sheet, start_row, start_column + column_count)
while column_value:
header.append(column_value)
column_count += 1
column_value = cell_value(sheet, start_row,
start_column + column_count)
# Get sheet rows
# TODO: may use sheel.col_slice or even sheet.nrows
table_rows = []
row_count = 0
start_row += 1
cell_is_empty = False
while not cell_is_empty:
row = [cell_value(sheet, start_row + row_count,
start_column + column_index)
for column_index in range(column_count)]
cell_is_empty = not any(row)
if not cell_is_empty:
table_rows.append(row)
row_count += 1
meta = {'imported_from': 'xls', 'filename': filename,}
return create_table([header] + table_rows, meta=meta, *args, **kwargs)
def export_to_xls(table, filename_or_fobj, sheet_name='Sheet1'):
filename, fobj = get_filename_and_fobj(filename_or_fobj, mode='wb')
work_book = xlwt.Workbook()
sheet = work_book.add_sheet(sheet_name)
fields = [(index, field_name)
for index, field_name in enumerate(table.fields)]
for index, field_name in fields:
sheet.write(0, index, field_name)
for row_index, row in enumerate(table, start=1):
for column_index, field_name in fields:
value = getattr(row, field_name)
field_type = table.fields[field_name]
data = {}
if field_type in FORMATTING_STYLES:
data['style'] = FORMATTING_STYLES[field_type]
sheet.write(row_index, column_index, value, **data)
work_book.save(fobj)
fobj.flush()
return fobj
def export_to_xls(table, filename_or_fobj, sheet_name='Sheet1'):
filename, fobj = get_filename_and_fobj(filename_or_fobj, mode='wb')
work_book = xlwt.Workbook()
sheet = work_book.add_sheet(sheet_name)
fields = [(index, field_name)
for index, field_name in enumerate(table.fields)]
for index, field_name in fields:
sheet.write(0, index, field_name)
for row_index, row in enumerate(table, start=1):
for column_index, field_name in fields:
value = getattr(row, field_name)
field_type = table.fields[field_name]
data = {}
if field_type in FORMATTING_STYLES:
data['style'] = FORMATTING_STYLES[field_type]
sheet.write(row_index, column_index, value, **data)
work_book.save(fobj)
fobj.flush()
return fobj
