def __init__()

in preprocess/htmltable.py [0:0]

• 25 lines of code
• 12 McCabe index (conditional complexity)

    def __init__(self, table, normalization=0, remove_hidden=True, first_row_as_caption=True):
        """Create table from a BeautifulSoup table Tag."""
        assert table.name == 'table'
        self.table = table
        self.caption = None if table.caption is None else table.caption.text
        if self.caption:
            self.caption = clean_cell_value(self.caption)

        if first_row_as_caption:
            rows = self.table.find_all('tr', recursive=False)
            col_num = max(len(row.find_all(['th', 'td'], recursive=False)) for row in rows)
            if len(rows) >= 2:
                row_1_cells = rows[0].find_all(['th', 'td'], recursive=False)
                if row_1_cells:
                    row_1_cell = row_1_cells[0]
                    if self.get_int(row_1_cell, 'colspan') == col_num:
                        self.caption = clean_cell_value(row_1_cell.text)
                        rows[0].decompose()

        if self.caption:
            self.caption = re.sub(r'(\[\d+\])+$', '', self.caption)

        if remove_hidden:
            self.remove_hidden()
        if normalization == HtmlTable.NORM_CORNER:
            self.normalize_table()
        elif normalization == HtmlTable.NORM_DUPLICATE:
            self.normalize_table(deep=True)

        self.get_cells()