Cyclic0007/python-markdownify
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

implemented table parsing correctly

Browse Source 
instead of manually walking down the dom tree
in a table, we now rely on the main descent loop
and just implement conversion for rows and cells
correctly. this enables the use of html inside a
table cell.
This commit is contained in:
AlexVonB
2021-05-17 14:00:00 +02:00
parent e6da15c173
commit ea81407b87
3 changed files with 64 additions and 35 deletions
Download Patch File Download Diff File Expand all files Collapse all files

31
tests/test_conversions.py
View File

@@ -1,5 +1,4 @@
from markdownify import markdownify as md, ATX, ATX_CLOSED, BACKSLASH, UNDERSCORE
import re
nested_uls = """
@@ -60,6 +59,25 @@ table = """<table>
</table>"""
table_with_html_content = """<table>
<tr>
<th>Firstname</th>
<th>Lastname</th>
<th>Age</th>
</tr>
<tr>
<td><b>Jill</b></td>
<td><i>Smith</i></td>
<td><a href="#">50</a></td>
</tr>
<tr>
<td>Eve</td>
<td>Jackson</td>
<td>94</td>
</tr>
</table>"""
table_with_header_column = """<table>
<tr>
<th>Firstname</th>
@@ -353,11 +371,12 @@ def test_div():
def test_table():
assert md(table) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |'
assert md(table_with_header_column) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |'
assert md(table_head_body) == '| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |'
assert md(table_missing_text) == '| | Lastname | Age |\n| --- | --- | --- |\n| Jill | | 50 |\n| Eve | Jackson | 94 |'
assert md(table_missing_head) == '| | | |\n| --- | --- | --- |\n| Firstname | Lastname | Age |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |'
assert md(table) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
assert md(table_with_html_content) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| **Jill** | *Smith* | [50](#) |\n| Eve | Jackson | 94 |\n\n'
assert md(table_with_header_column) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
assert md(table_head_body) == '\n\n| Firstname | Lastname | Age |\n| --- | --- | --- |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
assert md(table_missing_text) == '\n\n| | Lastname | Age |\n| --- | --- | --- |\n| Jill | | 50 |\n| Eve | Jackson | 94 |\n\n'
assert md(table_missing_head) == '\n\n| | | |\n| --- | --- | --- |\n| Firstname | Lastname | Age |\n| Jill | Smith | 50 |\n| Eve | Jackson | 94 |\n\n'
def test_strong_em_symbol():