| 1 | Genshi XPath is very limited and doesn't allow to do such things as selecting an empty row in a table. This recipe shows how to select and remove HTML elements using regular expressions in Transformers. |
| 2 | |
| 3 | {{{ |
| 4 | #!python |
| 5 | from genshi.input import HTML |
| 6 | from genshi.filters.transform import Transformer, StreamBuffer |
| 7 | import re |
| 8 | |
| 9 | html2 = HTML(''' |
| 10 | <table> |
| 11 | <tr><th></th><td></td><td></td></tr> |
| 12 | <tr><th>not empty</th><td></td><td></td></tr> |
| 13 | </table> |
| 14 | ''') |
| 15 | |
| 16 | buffer = StreamBuffer() |
| 17 | |
| 18 | def rowfilter(): # attention, closure |
| 19 | text = buffer.render() |
| 20 | text = re.sub(r'(?s)<tr>(\s*<t[hd](/>|>\s*</t[hd]>))+\s*</tr>', '', text) |
| 21 | #print(text) |
| 22 | return HTML(text) |
| 23 | |
| 24 | transtream = html2 | Transformer().select('.')\ |
| 25 | .copy(buffer).replace(rowfilter) |
| 26 | print transtream |
| 27 | }}} |