lxbfYeaa/BookConverter
Archived
Template
1
0
Fork 0
forked from LiveCarta/BookConverter
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix heading in json 3.0

Browse Source
This commit is contained in:
shirshasa
2020-09-28 16:40:20 +03:00
parent e913a5f49a
commit 5917fe51e0
2 changed files with 9 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
src/html_preprocessor.py
View File

@@ -512,7 +512,7 @@ class HTMLPreprocessor:
for i in range(0, len(self.top_level_headers)): for i in range(0, len(self.top_level_headers)):
self.top_level_headers[i]['should_be_numbered'] = True self.top_level_headers[i]['should_be_numbered'] = True
def _dfs(self, tag, is_first_span=None): def _clean_header_by_children(self, tag, is_first_span=None):
children = tag.find_all(recursive=False) children = tag.find_all(recursive=False)
if not children: if not children:
text = tag.text text = tag.text
@@ -525,9 +525,9 @@ class HTMLPreprocessor:
for i, child in enumerate(tag.find_all(recursive=False)): for i, child in enumerate(tag.find_all(recursive=False)):
if is_first_span and i == 0: if is_first_span and i == 0:
self._dfs(child, True) self._clean_header_by_children(child, True)
else: else:
self._dfs(child) self._clean_header_by_children(child)
def _process_headings(self): def _process_headings(self):
""" """
@@ -548,7 +548,7 @@ class HTMLPreprocessor:
assert tag.name in LawCartaConfig.SUPPORTED_HEADERS, \ assert tag.name in LawCartaConfig.SUPPORTED_HEADERS, \
f'Preprocessing went wrong, there is still h{LawCartaConfig.SUPPORTED_LEVELS + 1}-h9 headings.' f'Preprocessing went wrong, there is still h{LawCartaConfig.SUPPORTED_LEVELS + 1}-h9 headings.'
self._dfs(tag, is_first_span=True) self._clean_header_by_children(tag, is_first_span=True)
span_with_style_font = tag.find_all("span", {'style': re.compile(r'^font.+')}) span_with_style_font = tag.find_all("span", {'style': re.compile(r'^font.+')})
if span_with_style_font: if span_with_style_font:

13
src/json_converter.py
View File

@@ -1,14 +1,10 @@
import logging import logging
import re import re
import codecs
import json
from copy import copy from copy import copy
from config import LawCartaConfig from config import LawCartaConfig
class JSONConverter: class JSONConverter:
def __init__(self, content, footnotes, top_level_headers, logger_object, book_api_status=None): def __init__(self, content, footnotes, top_level_headers, logger_object, book_api_status=None):
self.content_dict = None self.content_dict = None
@@ -37,9 +33,10 @@ class JSONConverter:
:param ind: Index of header in content list. :param ind: Index of header in content list.
""" """
if self.content[ind].name in LawCartaConfig.SUPPORTED_HEADERS: if self.content[ind].name in LawCartaConfig.SUPPORTED_HEADERS:
title = '' title = str(self.content[ind])
for child in self.content[ind].find_all(recursive=False): title = title.replace(f'<{self.content[ind].name}>', '')
title += str(child) title = title.replace(f'</{self.content[ind].name}>', '')
title = re.sub(r'^\n', '', title)
curr_outline = int(re.sub(r"^h", "", self.content[ind].name)) # extract outline from tag curr_outline = int(re.sub(r"^h", "", self.content[ind].name)) # extract outline from tag
result = { result = {
@@ -140,7 +137,7 @@ class JSONConverter:
same_first_titles = self.top_level_headers[0]['title'] == json_strc[0]['title'] same_first_titles = self.top_level_headers[0]['title'] == json_strc[0]['title']
is_first_header_introduction = not self.top_level_headers[0]['should_be_numbered'] is_first_header_introduction = not self.top_level_headers[0]['should_be_numbered']
json_strc[0]['is_introduction'] = is_first_header_introduction and same_first_titles json_strc[0]['is_introduction'] = is_first_header_introduction
self.content_dict = { self.content_dict = {
"content": json_strc, "content": json_strc,