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

Add a new configuration option to control tabler header row inference (#161)

Browse Source 
Add option to infer first table row as table header (defaults to false)
This commit is contained in:
SomeBottle
2025-01-19 21:13:24 +08:00
committed by GitHub
parent 1783995cb2
commit 3bf0b527a4
4 changed files with 81 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

33
markdownify/__init__.py
View File

@@ -102,6 +102,7 @@ class MarkdownConverter(object):
strong_em_symbol = ASTERISK
sub_symbol = ''
sup_symbol = ''
table_infer_header = False
wrap = False
wrap_width = 80
@@ -518,13 +519,24 @@ class MarkdownConverter(object):
cells = el.find_all(['td', 'th'])
is_headrow = (
all([cell.name == 'th' for cell in cells])
or (not el.previous_sibling and not el.parent.name == 'tbody')
or (el.parent.name == 'thead'
# avoid multiple tr in thead
and len(el.parent.find_all('tr')) == 1)
)
is_head_row_missing = (
(not el.previous_sibling and not el.parent.name == 'tbody')
or (not el.previous_sibling and el.parent.name == 'tbody' and len(el.parent.parent.find_all(['thead'])) < 1)
)
overline = ''
underline = ''
if is_headrow and not el.previous_sibling:
# first row and is headline: print headline underline
if ((is_headrow
or (is_head_row_missing
and self.options['table_infer_header']))
and not el.previous_sibling):
# first row and:
# - is headline or
# - headline is missing and header inference is enabled
# print headline underline
full_colspan = 0
for cell in cells:
if 'colspan' in cell.attrs and cell['colspan'].isdigit():
@@ -532,13 +544,16 @@ class MarkdownConverter(object):
else:
full_colspan += 1
underline += '| ' + ' | '.join(['---'] * full_colspan) + ' |' + '\n'
elif (not el.previous_sibling
and (el.parent.name == 'table'
or (el.parent.name == 'tbody'
and not el.parent.previous_sibling))):
elif ((is_head_row_missing
and not self.options['table_infer_header'])
or (not el.previous_sibling
and (el.parent.name == 'table'
or (el.parent.name == 'tbody'
and not el.parent.previous_sibling)))):
# headline is missing and header inference is disabled or:
# first row, not headline, and:
# - the parent is table or
# - the parent is tbody at the beginning of a table.
# - the parent is table or
# - the parent is tbody at the beginning of a table.
# print empty headline above this row
overline += '| ' + ' | '.join([''] * len(cells)) + ' |' + '\n'
overline += '| ' + ' | '.join(['---'] * len(cells)) + ' |' + '\n'

4
markdownify/main.py
View File

@@ -61,6 +61,10 @@ def main(argv=sys.argv[1:]):
"should be converted to markdown images instead, this option can "
"be set to a list of parent tags that should be allowed to "
"contain inline images.")
parser.add_argument('--table-infer-header', dest='table_infer_header',
action='store_true',
help="When a table has no header row (as indicated by '<thead>' "
"or '<th>'), use the first body row as the header row.")
parser.add_argument('-w', '--wrap', action='store_true',
help="Wrap all text paragraphs at --wrap-width characters.")
parser.add_argument('--wrap-width', type=int, default=80)