Files
esl-reader/lib/books.py
Christopher T. Johnson 7f01c8d040 Setup LSP and clear all lint
2024-03-11 17:06:25 -04:00

208 lines
7.4 KiB
Python

import xml.dom.minidom
from typing import Dict, List, cast
from PyQt6.QtCore import QCoreApplication
from PyQt6.QtSql import QSqlQuery
from lib import query_error
class Book:
sections: List[str] = []
metadata: Dict[str, str] = {}
def __init__(self, src: str) -> None:
super(Book, self).__init__()
self.parse_book(src)
book_id = self.store() # Does nothing if already in database
self.load(book_id)
return
def load(self, book_id: int) -> None:
translate = QCoreApplication.translate
query = QSqlQuery()
query.prepare("SELECT * FROM books where book_id = :book_id")
query.bindValue(":book_id", book_id)
if not query.exec():
query_error(query)
if not query.next():
raise Exception(
translate("Book", "Missing book? book_id=") + f"{book_id}"
)
self.metadata = {
"title": query.value("title"),
"creator": query.value("author"),
"identifier": query.value("uuid"),
"level": query.value("level"),
}
self.sections = []
query.prepare(
"SELECT * FORM sections WHERE book_id = :book_id "
"ORDER BY sequence"
)
while query.next():
self.sections.append(query.value("contents"))
#
# Load words!
#
return
def parse_book(self, src: str) -> None:
with open(f"{src}/content.opf", "r") as f:
dom = xml.dom.minidom.parse(f)
metadata = dom.getElementsByTagName("metadata")[0]
for meta in metadata.childNodes:
if meta.nodeType != xml.dom.Node.ELEMENT_NODE:
continue
if meta.prefix == "dc":
self.metadata[meta.localName] = meta.firstChild.data
#
# The manifest contains a list of all the files contained in this
# EPUB
#
manifest = dom.getElementsByTagName("manifest")[0]
#
# The spine contains the documents in order they are read
#
spine = dom.getElementsByTagName("spine")[0]
for itemref in spine.childNodes:
if itemref.nodeType != xml.dom.Node.ELEMENT_NODE:
continue
# If linear == "no" skip
if (
itemref.hasAttribute("linear")
and itemref.getAttribute("linear") == "no"
):
continue
idref = itemref.getAttribute("idref")
for item in manifest.childNodes:
if item.nodeType != xml.dom.Node.ELEMENT_NODE:
continue
id = item.getAttribute("id")
if id == idref:
break
#
# A properly created *.opf will always have a correct
# spin/manifest.
#
href = item.getAttribute("href")
print(f"{idref}: {href}")
self.parse_section(src, href)
#
# "sections" is now loaded
#
return
def store(self) -> int:
uuid = self.metadata["identifier"]
query = QSqlQuery()
query.prepare(
"SELECT COUNT(*) AS number, book_id FROM books b "
"WHERE b.uuid = :uuid"
)
query.bindValue(":uuid", uuid)
if not query.exec():
query_error(query)
query.next()
if query.value("number") > 0:
book_id: int = query.value("book_id")
return book_id
query.prepare(
"INSERT INTO books (title, author, uuid, level) VALUES ("
":title, :author, :uuid, 0)"
)
query.bindValue(":title", self.metadata["title"])
query.bindValue(":author", self.metadata["creator"])
query.bindValue(":uuid", uuid)
if not query.exec():
query_error(query)
book_id = query.lastInsertId()
query.prepare(
"INSERT INTO sections (sequence, book_id, content) "
"VALUES (:sequence, :book_id, :content)"
)
query.bindValue(":book_id", book_id)
for seq, section in enumerate(self.sections):
query.bindValue(":sequence", seq)
query.bindValue(":content", section)
if not query.exec():
query_error(query)
return book_id
def parse_section(self, src: str, href: str) -> None:
newdom = xml.dom.getDOMImplementation().createDocument("", "html", None)
def strip_node(elm: xml.dom.minidom.Element) -> xml.dom.minidom.Element:
if elm.nodeType == xml.dom.Node.TEXT_NODE:
return cast(
xml.dom.minidom.Element,
newdom.createTextNode(cast(xml.dom.minidom.Text, elm).data),
)
newelm: xml.dom.minidom.Element = newdom.createElement(
elm.localName
)
node = elm.firstChild
while node:
elm = cast(xml.dom.minidom.Element, node)
if elm.nodeType == xml.dom.Node.TEXT_NODE:
text = cast(xml.dom.minidom.Text, elm).data
if text:
text = text.strip()
if text and len(text) > 0:
newelm.appendChild(newdom.createTextNode(text))
elif elm.localName == "img":
pass
elif elm.nodeType == xml.dom.Node.COMMENT_NODE:
pass
elif elm.localName == "a":
a_node = cast(xml.dom.minidom.Element, elm.firstChild)
while a_node:
if a_node.nodeType == xml.dom.Node.TEXT_NODE:
text = cast(xml.dom.minidom.Text, a_node)
newelm.appendChild(newdom.createTextNode(text.data))
else:
newelm.appendChild(strip_node(a_node))
a_node = a_node.nextSibling
else:
newelm.appendChild(strip_node(elm))
node = node.nextSibling
return newelm
def parse_node(
parent: xml.dom.minidom.Element, elm: xml.dom.minidom.Element
) -> None:
if elm.nodeType == xml.dom.Node.ELEMENT_NODE:
tag: str = cast(str, elm.localName)
if tag.startswith("h"):
clone = strip_node(elm)
parent.appendChild(clone)
elif tag == "p":
clone = strip_node(elm)
clone.normalize()
parent.appendChild(clone)
else:
node = elm.firstChild
while node:
parse_node(parent, cast(xml.dom.minidom.Element, node))
node = node.nextSibling
return
with open(f"{src}/{href}") as f:
dom = xml.dom.minidom.parse(f)
# title = dom.getElementsByTagName("title")[0].firstChild.data
body = dom.getElementsByTagName("body")[0]
section = newdom.createElement("body")
node = body.firstChild
while node:
parse_node(section, cast(xml.dom.minidom.Element, node))
node = node.nextSibling
try:
self.sections.append(section.toxml())
except Exception as e:
print(src, href)
raise e
return