fix #308: ignore parsing page

pdf2docx/table/Cell.py

@@ -76,7 +76,7 @@ def make_docx(self, table, indexes):
         n_row, n_col = self.merged_cells
         i, j = indexes
         docx_cell = table.cell(i, j)
-        if n_row*n_col!=1:
+        if n_row*n_col != 1 and ((i+n_row-1) * table._column_count + j+n_col-1) < len(table._cells): # check whether index is over length of cells
             _cell = table.cell(i+n_row-1, j+n_col-1)
             try:
                 docx_cell.merge(_cell)
@@ -133,7 +133,8 @@ def _set_style(self, table, indexes):
         # merged cells are assumed to have same borders with the main cell
         for m in range(i, i+n_row):
             for n in range(j, j+n_col):
-                docx.set_cell_border(table.cell(m, n), **kwargs)
+                if len(table._cells) > m * table._column_count + n: # check whether index is over length of cells
+                    docx.set_cell_border(table.cell(m, n), **kwargs)
 
         # ---------------------
         # cell bg-color

pdf2docx/table/Row.py

@@ -66,7 +66,10 @@ def make_docx(self, table, idx_row:int):
         # to control the layout precisely, set `exact` value, rather than `at least` value
         # the associated steps in MS word: Table Properties -> Row -> Row height -> exactly
         docx_row.height_rule = WD_ROW_HEIGHT.EXACTLY
-
+
+        if self.height < 0: # to prevent negative height validation
+            self.height = 0
+
         # NOTE: row height is counted from center-line of top border to center line of bottom border
         docx_row.height = Pt(self.height)