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

Add replacer(-two_columns,-font_to_span, preproc_headings)

Browse Source
This commit is contained in:
Kiryl
2022-09-05 17:23:03 +03:00
parent 3fcff462d3
commit 9e31d3152c
2 changed files with 106 additions and 121 deletions
Download Patch File Download Diff File Expand all files Collapse all files

68
presets/docx_presets.json
View File

@@ -11,12 +11,63 @@
{
"name": "title",
"value": "footer"
},
{
"name": "id",
"value": "^Table of Contents\\d+"
}
]
}
}
]
},
{
"preset_name": "replacer",
"rules": [
{
"tags": ["^font$"],
"condition": null,
"tag_to_replace": "span"
},
{
"tags": ["^h[6-9]$"],
"condition": null,
"tag_to_replace": "p"
},
{
"tags": ["^div$"],
"condition": {
"parent_tags": null,
"child_tags": null,
"attrs": [
{
"name": "style",
"value": "column-count: 2"
}
]
},
"tag_to_replace": "p"
}
]
},
{
"preset_name": "attr_replacer",
"rules": [
{
"attr": {
"name": "style",
"value": "column-count: 2"
},
"condition": {
"tags": ["^p$"]
},
"attr_to_replace": {
"name": null,
"value": "columns2"
}
}
]
},
{
"preset_name": "unwrapper",
"rules": [
@@ -33,16 +84,7 @@
{
"name": "lang",
"value": "^ru-RU$"
}
]
}
},
{
"tags": ["^font$"],
"condition": {
"parent_tags": null,
"child_tags": null,
"attrs": [
},
{
"name": "face",
"value": "^Times New Roman[\\w, ]+$"
@@ -53,7 +95,7 @@
{
"tags": ["^p$"],
"condition": {
"parent_tags": "li",
"parent_tags": ":is(li)",
"child_tags": null,
"attrs": null
}
@@ -74,8 +116,8 @@
{
"tags": ["^u$"],
"condition": {
"parent_tags": "a",
"child_tags": "a",
"parent_tags": ":is(a)",
"child_tags": ":is(a)",
"attrs": null
}
},

157
src/docx_converter/html_docx_processor.py
View File

@@ -22,6 +22,8 @@ class HTMLDocxProcessor:
self.style_processor = style_processor
self.name2action = {
"decomposer": self._decompose_tag,
"replacer": self._replace_tag,
"attr_replacer": self._replace_attr,
"unwrapper": self._unwrap_tag
}
self.top_level_headers = None
@@ -59,15 +61,34 @@ class HTMLDocxProcessor:
f"Tag name: {tag.name}")
@staticmethod
def _decompose_tag(tag):
tag.decompose()
def _decompose_tag(**kwargs):
kwargs["tag"].decompose()
@staticmethod
def _unwrap_tag(tag):
tag.unwrap()
def _replace_tag(**kwargs):
tag_to_replace: str = kwargs["rule"]["tag_to_replace"]
kwargs["tag"].name = tag_to_replace
@staticmethod
def _process_tags(body_tag: BeautifulSoup,
def _replace_attr(**kwargs):
attr, attr_value =\
kwargs["rule"]["attr"]["name"], kwargs["rule"]["attr"]["value"]
attr_to_replace, attr_value_to_replace =\
kwargs["rule"]["attr_to_replace"]["name"], kwargs["rule"]["attr_to_replace"]["value"]
if attr_to_replace:
kwargs["tag"][attr_to_replace] = kwargs["tag"][attr]
if attr_value_to_replace:
kwargs["tag"].attrs[attr_to_replace] = attr_value_to_replace
del kwargs["tag"][attr]
elif attr_value_to_replace:
kwargs["tag"].attrs[attr] = attr_value_to_replace
@staticmethod
def _unwrap_tag(**kwargs):
kwargs["tag"].unwrap()
@staticmethod
def _process_tags(body_tag: Tag,
rules: List[Dict[str, Union[List[str], str, Dict[str, Union[List[Dict[str, str]], int, str]]]]],
action):
"""
@@ -87,104 +108,32 @@ class HTMLDocxProcessor:
"""
for rule in rules:
tags: List[str] = rule["tags"]
tags: List[str] = rule["tags"] if rule.get("tags") else rule["condition"]["tags"]
if rule["condition"]:
for condition_on_tag in ((k, v) for k, v in rule["condition"].items() if v):
if condition_on_tag[0] == "parent_tags":
for tag in body_tag.select(', '.join([condition_on_tag[1] + " > " + re.sub(r"[\^$]", "", tag)
for tag in tags])):
tag.parent.attrs.update(tag.attrs)
action(tag)
action(body_tag=body_tag, tag=tag, rule=rule)
elif condition_on_tag[0] == "child_tags":
for tag in body_tag.select(', '.join([re.sub(r"[\^$]", "", tag) + condition_on_tag[1]
for tag in tags])):
action(tag)
action(body_tag=body_tag, tag=tag, rule=rule)
elif condition_on_tag[0] == "attrs":
for attr in rule["condition"]["attrs"]:
for tag in body_tag.find_all([re.compile(tag) for tag in tags],
{attr["name"]: re.compile(fr"{attr['value']}")}):
action(tag)
action(body_tag=body_tag, tag=tag, rule=rule)
# attr replacer
elif condition_on_tag[0] == "tags":
attr = rule["attr"]
for tag in body_tag.find_all([re.compile(tag) for tag in tags],
{attr['name']: re.compile(fr"{attr['value']}")}):
action(body_tag=body_tag, tag=tag, rule=rule)
else:
for tag in body_tag.find_all([re.compile(tag) for tag in tags]):
action(tag)
@classmethod
def convert_pt_to_px(cls, value: float) -> float:
value = float(value)
if value == LiveCartaConfig.WORD_DEFAULT_FONT_SIZE:
return LiveCartaConfig.LIVECARTA_DEFAULT_FONT_SIZE
else:
return value
@classmethod
def convert_font_pt_to_px(cls, style: str) -> str:
"""
Function converts point in the font-size to pixels.
Parameters
----------
style: str
str with style to proces
Returns
-------
: str
str with converted style
"""
size = re.search(r"font-size: (\d{1,3})pt", style)
if size is None:
return style
size = size.group(1)
new_size = cls.convert_pt_to_px(size)
if new_size == LiveCartaConfig.LIVECARTA_DEFAULT_FONT_SIZE:
return ""
return re.sub(size + "pt", str(new_size) + "px", style)
def _font_to_span(self):
"""
Function to convert <font> tag to <span>.
If font style is default, then remove this tag.
"""
fonts = self.body_tag.find_all("font")
for font in fonts:
face, style, color =\
font.get("face"), font.get("style"), font.get("color")
font.attrs, font.name = {}, "span"
if style:
style = self.convert_font_pt_to_px(style)
if style != "":
if color and color in LiveCartaConfig.COLORS_MAP:
style += f"; color: {color};"
font.attrs["style"] = style
elif color and color in LiveCartaConfig.COLORS_MAP:
font.attrs["style"] = f"color: {color};"
if len(font.attrs) == 0:
font.unwrap()
# on this step there should be no more <font> tags
assert len(self.body_tag.find_all("font")) == 0
def clean_trash(self):
"""Function to remove all styles and tags we don"t need."""
# todo replacer
self._font_to_span()
# replace toc with empty <TOC> tag
tables = self.body_tag.find_all(
"div", id=re.compile(r"^Table of Contents\d+"))
for table in tables:
table.wrap(self.html_soup.new_tag("TOC"))
table.decompose()
def _preprocessing_headings(self):
# todo replacer
"""Function to convert all lower level headings to p tags"""
pattern = f"^h[{LiveCartaConfig.SUPPORTED_LEVELS + 1}-9]$"
header_tags = self.body_tag.find_all(re.compile(pattern))
for tag in header_tags:
tag.name = "p"
action(body_tag=body_tag, tag=tag, rule=rule)
def _process_paragraph(self):
"""Function to process <p> tags (text-align and text-indent value)."""
@@ -237,16 +186,6 @@ class HTMLDocxProcessor:
if style:
p.attrs["style"] = style
def _process_two_columns(self):
"""Function to process paragraphs which has two columns layout."""
# todo replacer
two_columns = self.body_tag.find_all("div", style="column-count: 2")
for div in two_columns:
for child in div.children:
if child.name == "p":
child["class"] = "columns2"
div.unwrap()
def _process_quotes(self):
"""
Function to process block quotes.
@@ -285,6 +224,14 @@ class HTMLDocxProcessor:
table.replaceWith(new_div)
@staticmethod
def convert_pt_to_px(value: float) -> float:
value = float(value)
if value == LiveCartaConfig.WORD_DEFAULT_FONT_SIZE:
return LiveCartaConfig.LIVECARTA_DEFAULT_FONT_SIZE
else:
return value
def _process_tables(self):
"""Function to process tables. Set "border" attribute."""
tables = self.body_tag.find_all("table")
@@ -490,24 +437,20 @@ class HTMLDocxProcessor:
self.logger.log(f"Processing TOC and headers.")
self._process_toc_links()
for rule in self.preset:
self.logger.log(rule["preset_name"] + " process.")
action = self.name2action[rule["preset_name"]]
self._process_tags(self.body_tag, rule["rules"], action)
self.logger.log("CSS inline style preprocessing.")
self.style_processor.process_inline_styles_in_html_soup(self.html_soup)
self.logger.log("CSS inline style processing.")
modify_html_soup_with_css_styles(self.html_soup)
for rule in self.preset:
self.logger.log(rule["preset_name"] + " process.")
action = self.name2action[rule["preset_name"]]
self._process_tags(self.body_tag, rule["rules"], action)
self.clean_trash()
# process main elements of the .html doc
self.logger.log(f"Processing main elements of html.")
self._preprocessing_headings()
self._process_paragraph()
self._process_two_columns()
self.logger.log("Block quotes processing.")
self._process_quotes()