Mostly working Reader

lib/books.py

@@ -1,15 +1,49 @@
 import json
 import os
 import xml.dom.minidom
+from typing import Dict, List, cast
+
+from PyQt6.QtSql import QSqlQuery
+
+from main import query_error
 
 
 class Book:
-    sections = []
-    metadata = {}
+    sections: List[str] = []
+    metadata: Dict[str, str] = {}
+    words = {}
 
     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:
+        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(f"Missing book? book_id={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:
@@ -53,15 +87,105 @@ class Book:
             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)
+            section_id = query.lastInsertId()
+        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.Node:
+            if elm.nodeType == xml.dom.Node.TEXT_NODE:
+                return cast(
+                    xml.dom.minidom.Node,
+                    newdom.createTextNode(cast(xml.dom.minidom.Text, elm).data),
+                )
+
+            newelm = newdom.createElement(elm.localName)
+            node = elm.firstChild
+            while node:
+                if node.nodeType == xml.dom.Node.TEXT_NODE:
+                    text = node.data
+                    if text:
+                        text = text.strip()
+                    if text and len(text) > 0:
+                        newelm.appendChild(newdom.createTextNode(text))
+                elif node.localName == "img":
+                    pass
+                elif node.localName == "a":
+                    a_node = node.firstChild
+                    while a_node:
+                        if a_node.nodeType == xml.dom.Node.TEXT_NODE:
+                            newelm.appendChild(newdom.createTextNode(a_node.data))
+                        else:
+                            newelm.appendChild(strip_node(a_node))
+                        a_node = a_node.nextSibling
+                else:
+                    newelm.appendChild(strip_node(node))
+                node = node.nextSibling
+            return newelm
+
+        def parse_node(parent: xml.dom.Node, elm: xml.dom.Node) -> None:
+            if elm.nodeType == xml.dom.Node.ELEMENT_NODE:
+                if elm.localName.startswith("h"):
+                    clone = strip_node(elm)
+                    parent.appendChild(clone)
+                elif elm.localName == "p":
+                    clone = strip_node(elm)
+                    clone.normalize()
+                    parent.appendChild(clone)
+                else:
+                    node = elm.firstChild
+                    while node:
+                        parse_node(parent, 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]
-        paragraphs = []
-        for p in body.getElementsByTagName("p"):
-            paragraphs.append(p.toxml())
-        self.sections.append({"title": title, "paragraphs": paragraphs})
+        section = newdom.createElement("body")
+        node = body.firstChild
+        while node:
+            parse_node(section, node)
+            node = node.nextSibling
+        self.sections.append(section.toxml())
         return