voussoir/epubfile
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

Fix toc from climbing too high when going from h2 to h4 to h2 etc.

Browse source 
Can you believe there was a bug amidst all these wonderful comments?
The code previously assumed that climbing .parent corresponded to
climbing from hX to hX-1, but this wasn't true when the headers
themselves skipped a level. After creating some toc for h2 then h4,
then encountering a new h2, it would try to climb back two levels to
go from 4 to 2, but actually there was only ever one level between them.
This commit is contained in:
Ethan Dalool 2020-02-04 16:40:14 -08:00
parent 92d595c00d
commit c56e0ec13e
1 changed files with 11 additions and 8 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

19
epubfile.py
View file

@ -1029,8 +1029,9 @@ class Epub:
# have different structural requirements. The attributes that I'm using # have different structural requirements. The attributes that I'm using
# in this initial toc object DO NOT represent any part of the epub format. # in this initial toc object DO NOT represent any part of the epub format.
toc = new_list(root=True) toc = new_list(root=True)
current_level = None
current_list = toc.ol current_list = toc.ol
current_list['level'] = None
spine = self.get_spine_order(linear_only=linear_only) spine = self.get_spine_order(linear_only=linear_only)
spine = [s for s in spine if s != nav_id] spine = [s for s in spine if s != nav_id]
@ -1055,11 +1056,10 @@ class Epub:
relative = file_path.relative_to(ncx_filepath.parent, simple=True) relative = file_path.relative_to(ncx_filepath.parent, simple=True)
toc_line['ncx_anchor'] = f'{relative}#{header["id"]}' toc_line['ncx_anchor'] = f'{relative}#{header["id"]}'
if current_level is None: if current_list['level'] is None:
current_level = level current_list['level'] = level
while level < current_level: while level < current_list['level']:
current_level -= 1
# Because the sub-<ol> are actually a child of the last # Because the sub-<ol> are actually a child of the last
# <li> of the previous <ol>, we must .parent twice. # <li> of the previous <ol>, we must .parent twice.
# The second .parent is conditional because if the current # The second .parent is conditional because if the current
@ -1079,11 +1079,10 @@ class Epub:
# In the resulting toc, that initial h4 would have the same # In the resulting toc, that initial h4 would have the same
# toc depth as the later h1 since it never had parents. # toc depth as the later h1 since it never had parents.
if current_list == toc: if current_list == toc:
current_level = level current_list['level'] = level
current_list = toc.ol current_list = toc.ol
if level > current_level: if level > current_list['level']:
current_level = level
# In order to properly render nested <ol>, you're supposed # In order to properly render nested <ol>, you're supposed
# to make the new <ol> a child of the last <li> of the # to make the new <ol> a child of the last <li> of the
# previous <ol>. NOT a child of the prev <ol> directly. # previous <ol>. NOT a child of the prev <ol> directly.
@ -1091,6 +1090,7 @@ class Epub:
# first <li> this condition can never occur, and new <ol>s # first <li> this condition can never occur, and new <ol>s
# always receive a child right after being created. # always receive a child right after being created.
_l = new_list() _l = new_list()
_l['level'] = level
final_li = list(current_list.children)[-1] final_li = list(current_list.children)[-1]
final_li.append(_l) final_li.append(_l)
current_list = _l current_list = _l
@ -1100,6 +1100,9 @@ class Epub:
# We have to save the id="toc_X" that we gave to all the headers. # We have to save the id="toc_X" that we gave to all the headers.
self.write_file(file_id, soup) self.write_file(file_id, soup)
for ol in toc.find_all('ol'):
del ol['level']
if nav_id: if nav_id:
self._set_nav_toc(nav_id, copy.copy(toc)) self._set_nav_toc(nav_id, copy.copy(toc))