` element of the html table.
+ """
+ def __init__(self, tr, parent=None):
+ super(TableRow, self).__init__(tr, parent=parent)
+ self.cells = [TableCell(cell, parent=self) for cell in tr.findall('th') + tr.findall('td')]
+
+
+def element_to_string(el):
+ return _element_to_string(el).strip()
+
+
+def _element_to_string(el):
+ string = ''
+
+ for x in el.iterchildren():
+ string += '\n' + _element_to_string(x)
+
+ text = el.text.strip() if el.text else ''
+ tail = el.tail.strip() if el.tail else ''
+
+ return text + string + '\n' + tail
+
+
+class TableCell(Element):
+ """
+ This class maps to the `| ` element of the html table.
+ """
+ CELL_TYPES = {'TYPE_STRING', 'TYPE_FORMULA', 'TYPE_NUMERIC', 'TYPE_BOOL', 'TYPE_CURRENCY', 'TYPE_PERCENTAGE',
+ 'TYPE_NULL', 'TYPE_INLINE', 'TYPE_ERROR', 'TYPE_FORMULA_CACHE_STRING', 'TYPE_INTEGER'}
+
+ def __init__(self, cell, parent=None):
+ super(TableCell, self).__init__(cell, parent=parent)
+ self.value = element_to_string(cell)
+ self.number_format = self.get_number_format()
+
+ def data_type(self):
+ cell_types = self.CELL_TYPES & set(self.element.get('class', '').split())
+ if cell_types:
+ if 'TYPE_FORMULA' in cell_types:
+ # Make sure TYPE_FORMULA takes precedence over the other classes in the set.
+ cell_type = 'TYPE_FORMULA'
+ elif cell_types & {'TYPE_CURRENCY', 'TYPE_INTEGER', 'TYPE_PERCENTAGE'}:
+ cell_type = 'TYPE_NUMERIC'
+ else:
+ cell_type = cell_types.pop()
+ else:
+ cell_type = 'TYPE_STRING'
+ return getattr(cell, cell_type)
+
+ def get_number_format(self):
+ if 'TYPE_CURRENCY' in self.element.get('class', '').split():
+ return FORMAT_CURRENCY_USD_SIMPLE
+ if 'TYPE_INTEGER' in self.element.get('class', '').split():
+ return '#,##0'
+ if 'TYPE_PERCENTAGE' in self.element.get('class', '').split():
+ return FORMAT_PERCENTAGE
+ if 'TYPE_DATE' in self.element.get('class', '').split():
+ return FORMAT_DATE_MMDDYYYY
+ if self.data_type() == cell.TYPE_NUMERIC:
+ try:
+ int(self.value)
+ except ValueError:
+ return '#,##0.##'
+ else:
+ return '#,##0'
+
+ def format(self, cell):
+ cell.style = self.style()
+ data_type = self.data_type()
+ if data_type:
+ cell.data_type = data_type
\ No newline at end of file
diff --git a/ppstructure/table/tablepyxl/tablepyxl.py b/ppstructure/table/tablepyxl/tablepyxl.py
new file mode 100644
index 0000000000000000000000000000000000000000..ba3cc0fc499fccd93ffe3993a99296bc6603ed8a
--- /dev/null
+++ b/ppstructure/table/tablepyxl/tablepyxl.py
@@ -0,0 +1,118 @@
+# Do imports like python3 so our package works for 2 and 3
+from __future__ import absolute_import
+
+from lxml import html
+from openpyxl import Workbook
+from openpyxl.utils import get_column_letter
+from premailer import Premailer
+from tablepyxl.style import Table
+
+
+def string_to_int(s):
+ if s.isdigit():
+ return int(s)
+ return 0
+
+
+def get_Tables(doc):
+ tree = html.fromstring(doc)
+ comments = tree.xpath('//comment()')
+ for comment in comments:
+ comment.drop_tag()
+ return [Table(table) for table in tree.xpath('//table')]
+
+
+def write_rows(worksheet, elem, row, column=1):
+ """
+ Writes every tr child element of elem to a row in the worksheet
+ returns the next row after all rows are written
+ """
+ from openpyxl.cell.cell import MergedCell
+
+ initial_column = column
+ for table_row in elem.rows:
+ for table_cell in table_row.cells:
+ cell = worksheet.cell(row=row, column=column)
+ while isinstance(cell, MergedCell):
+ column += 1
+ cell = worksheet.cell(row=row, column=column)
+
+ colspan = string_to_int(table_cell.element.get("colspan", "1"))
+ rowspan = string_to_int(table_cell.element.get("rowspan", "1"))
+ if rowspan > 1 or colspan > 1:
+ worksheet.merge_cells(start_row=row, start_column=column,
+ end_row=row + rowspan - 1, end_column=column + colspan - 1)
+
+ cell.value = table_cell.value
+ table_cell.format(cell)
+ min_width = table_cell.get_dimension('min-width')
+ max_width = table_cell.get_dimension('max-width')
+
+ if colspan == 1:
+ # Initially, when iterating for the first time through the loop, the width of all the cells is None.
+ # As we start filling in contents, the initial width of the cell (which can be retrieved by:
+ # worksheet.column_dimensions[get_column_letter(column)].width) is equal to the width of the previous
+ # cell in the same column (i.e. width of A2 = width of A1)
+ width = max(worksheet.column_dimensions[get_column_letter(column)].width or 0, len(table_cell.value) + 2)
+ if max_width and width > max_width:
+ width = max_width
+ elif min_width and width < min_width:
+ width = min_width
+ worksheet.column_dimensions[get_column_letter(column)].width = width
+ column += colspan
+ row += 1
+ column = initial_column
+ return row
+
+
+def table_to_sheet(table, wb):
+ """
+ Takes a table and workbook and writes the table to a new sheet.
+ The sheet title will be the same as the table attribute name.
+ """
+ ws = wb.create_sheet(title=table.element.get('name'))
+ insert_table(table, ws, 1, 1)
+
+
+def document_to_workbook(doc, wb=None, base_url=None):
+ """
+ Takes a string representation of an html document and writes one sheet for
+ every table in the document.
+ The workbook is returned
+ """
+ if not wb:
+ wb = Workbook()
+ wb.remove(wb.active)
+
+ inline_styles_doc = Premailer(doc, base_url=base_url, remove_classes=False).transform()
+ tables = get_Tables(inline_styles_doc)
+
+ for table in tables:
+ table_to_sheet(table, wb)
+
+ return wb
+
+
+def document_to_xl(doc, filename, base_url=None):
+ """
+ Takes a string representation of an html document and writes one sheet for
+ every table in the document. The workbook is written out to a file called filename
+ """
+ wb = document_to_workbook(doc, base_url=base_url)
+ wb.save(filename)
+
+
+def insert_table(table, worksheet, column, row):
+ if table.head:
+ row = write_rows(worksheet, table.head, row, column)
+ if table.body:
+ row = write_rows(worksheet, table.body, row, column)
+
+
+def insert_table_at_cell(table, cell):
+ """
+ Inserts a table at the location of an openpyxl Cell object.
+ """
+ ws = cell.parent
+ column, row = cell.column, cell.row
+ insert_table(table, ws, column, row)
\ No newline at end of file
|