vangarstadler
commented
3 months ago Hi @mstevens72 I've added the following functionality to my branch of this project. Replacing the "handle_li" function with the following should yield the desired behavior. Be aware that this relies on every "li" element possessing an attribute "id" unique within the html you're parsing.
Good luck!
def handle_li(self):
def __is_first_ol_element(HTMLParser__startag_text: str) -> bool:
"""determines if ol is first of ol list using starttag with id.
Args:
HTMLParser__startag_text (str): starttag as obtained by parser.
Returns:
bool: True if the case, False otherwise
"""
all_ordered_lists_in_html_snippet = self.soup.find_all("ol")
for ol in all_ordered_lists_in_html_snippet:
try:
if ol.contents[0].attrs["id"] in HTMLParser__startag_text:
return True
except Exception:
return False
return False
def __restart_numbering(paragraph: Paragraph) -> Paragraph:
"""Private method to reset list numbering of a given paragraph.
Implementation from: https://github.com/python-openxml/python-docx/pull/582#issuecomment-1717139576
Args:
paragraph (Paragraph): paragraph object
Returns:
Paragraph: paragraph object with reset list numbering
"""
# Getting the abstract number of paragraph
abstract_num_id = (
paragraph.part.document.part.numbering_part.element.num_having_numId(
paragraph.style.element.get_or_add_pPr()
.get_or_add_numPr()
.numId.val
).abstractNumId.val
)
# Add abstract number to numbering part and reset
num = paragraph.part.numbering_part.element.add_num(abstract_num_id)
num.add_lvlOverride(ilvl=0).add_startOverride(1)
# Get or add elements to paragraph
p_pr = paragraph._p.get_or_add_pPr()
num_pr = p_pr.get_or_add_numPr()
ilvl = num_pr.get_or_add_ilvl()
ilvl.val = int("0")
num_id = num_pr.get_or_add_numId()
num_id.val = int(num.numId)
return paragraph
# check list stack to determine style and depth
list_depth = len(self.tags["list"])
# list type assignment
if list_depth:
list_type = self.tags["list"][-1]
else:
list_type = "ul" # assign unordered if no tag
if list_type == "ol":
list_style = styles[f"LIST_N{list_depth}"]
else:
list_style = styles[f"LIST_B{list_depth}"]
self.paragraph = self.doc.add_paragraph(style=list_style)
if list_type == "ol" and __is_first_ol_element(
HTMLParser__startag_text=self._HTMLParser__starttag_text, soup=self.soup
):
self.paragraph = __restart_numbering(paragraph=self.paragraph)
# Indentation: default = no indent
if list_depth > 1:
self.paragraph.paragraph_format.left_indent = Inches(
min(list_depth * LIST_INDENT, MAX_INDENT)
)
self.paragraph.paragraph_format.line_spacing = 1Relies on imports:
from docx.table import Table, _Cell, _Row
from docx.document import Document as DocumentClass
from docx.text.paragraph import Paragraph
I have a question for you regarding the ordered list function for your class htmldocx. I can not seem to get this to work correctly in word, I have attached two files, one is a screenshot of the continued numbering in word of an ordered list. The second file is a test html file that I am using that has a nested ordered list in it and would like to convert the information from into a word document with the numbering correct. The numbering in the word document should be the same as in test.html.
(the html file name was appended with .txt to allow for upload)
Please let me know if you have any questions or need any additional information.
test.html.txt