added wrap option

closes #66
This commit is contained in:
AlexVonB
2022-04-24 11:00:04 +02:00
parent 16acd2b763
commit 5f1b98e25d
3 changed files with 18 additions and 0 deletions

View File

@@ -130,6 +130,11 @@ keep_inline_images_in
that should be allowed to contain inline images, for example ``['td']``.
Defaults to an empty list.
wrap, wrap_width
If ``wrap`` is set to ``True``, all text paragraphs are wrapped at
``wrap_width`` characters. Defaults to ``False`` and ``80``.
Use with ``newline_style=BACKSLASH`` to keep line breaks in paragraphs.
Options may be specified as kwargs to the ``markdownify`` function, or as a
nested ``Options`` class in ``MarkdownConverter`` subclasses.

View File

@@ -1,4 +1,5 @@
from bs4 import BeautifulSoup, NavigableString, Comment, Doctype
from textwrap import fill
import re
import six
@@ -75,6 +76,8 @@ class MarkdownConverter(object):
strong_em_symbol = ASTERISK
sub_symbol = ''
sup_symbol = ''
wrap = False
wrap_width = 80
class Options(DefaultOptions):
pass
@@ -331,6 +334,11 @@ class MarkdownConverter(object):
def convert_p(self, el, text, convert_as_inline):
if convert_as_inline:
return text
if self.options['wrap']:
text = fill(text,
width=self.options['wrap_width'],
break_long_words=False,
break_on_hyphens=False)
return '%s\n\n' % text if text else ''
def convert_pre(self, el, text, convert_as_inline):

View File

@@ -177,6 +177,11 @@ def test_kbd():
def test_p():
assert md('<p>hello</p>') == 'hello\n\n'
assert md('<p>123456789 123456789</p>') == '123456789 123456789\n\n'
assert md('<p>123456789 123456789</p>', wrap=True, wrap_width=10) == '123456789\n123456789\n\n'
assert md('<p><a href="https://example.com">Some long link</a></p>', wrap=True, wrap_width=10) == '[Some long\nlink](https://example.com)\n\n'
assert md('<p>12345<br />67890</p>', wrap=True, wrap_width=10, newline_style=BACKSLASH) == '12345\\\n67890\n\n'
assert md('<p>12345678901<br />12345</p>', wrap=True, wrap_width=10, newline_style=BACKSLASH) == '12345678901\\\n12345\n\n'
def test_pre():