CPMT_EPUB_exporter.py

Go to the documentation of this file.

"""
SPDX-FileCopyrightText: 2018 Wolthera van Hövell tot Westerflier <griffinvalley@gmail.com>
 
This file is part of the Comics Project Management Tools(CPMT).
 
SPDX-License-Identifier: GPL-3.0-or-later
"""
 
"""
Create an epub folder, finally, package to a epubzip.
"""
 
import shutil
import os
from pathlib import Path
import zipfile
try:
    from PyQt6.QtXml import QDomDocument
    from PyQt6.QtCore import Qt, QDateTime, QPointF
    from PyQt6.QtGui import QImage, QPolygonF, QColor
except:
    from PyQt5.QtXml import QDomDocument
    from PyQt5.QtCore import Qt, QDateTime, QPointF
    from PyQt5.QtGui import QImage, QPolygonF, QColor
from builtins import i18n
 
def export(configDictionary = {}, projectURL = str(), pagesLocationList = [], pageData = []):
    path = Path(os.path.join(projectURL, configDictionary["exportLocation"]))
    exportPath = path / "EPUB-files"
    metaInf = exportPath / "META-INF"
    oebps = exportPath / "OEBPS"
    imagePath = oebps / "Images"
    # Don't write empty folders. Epubcheck doesn't like that.
    # stylesPath = oebps / "Styles"
    textPath = oebps / "Text"
 
    if exportPath.exists() is False:
        exportPath.mkdir()
        metaInf.mkdir()
        oebps.mkdir()
        imagePath.mkdir()
        # stylesPath.mkdir()
        textPath.mkdir()
 
    # Due the way EPUB verifies, the mimetype needs to be packaged in first.
    # Due the way zips are constructed, the only way to ensure that is to
    # Fill the zip as we go along...
    
    # Use the project name if there's no title to avoid sillyness with unnamed zipfiles.
    title = configDictionary["projectName"]
    if "title" in configDictionary.keys():
        title = str(configDictionary["title"]).replace(" ", "_")
 
    # Get the appropriate path.
    url = str(path / str(title + ".epub"))
 
    # Create a zip file.
    epubArchive = zipfile.ZipFile(url, mode="w", compression=zipfile.ZIP_STORED)
 
    mimetype = open(str(Path(exportPath / "mimetype")), mode="w")
    mimetype.write("application/epub+zip")
    mimetype.close()
    
    # Write to zip.
    epubArchive.write(Path(exportPath / "mimetype"), Path("mimetype"))
 
    container = QDomDocument()
    cRoot = container.createElement("container")
    cRoot.setAttribute("version", "1.0")
    cRoot.setAttribute("xmlns", "urn:oasis:names:tc:opendocument:xmlns:container")
    container.appendChild(cRoot)
    rootFiles = container.createElement("rootfiles")
    rootfile = container.createElement("rootfile")
    rootfile.setAttribute("full-path", "OEBPS/content.opf")
    rootfile.setAttribute("media-type", "application/oebps-package+xml")
    rootFiles.appendChild(rootfile)
    cRoot.appendChild(rootFiles)
    
    containerFileName = str(Path(metaInf / "container.xml"))
 
    containerFile = open(containerFileName, 'w', newline="", encoding="utf-8")
    containerFile.write(container.toString(indent=2))
    containerFile.close()
    
    # Write to zip.
    epubArchive.write(containerFileName, os.path.relpath(containerFileName, str(exportPath)))
 
    # copyimages to images
    pagesList = []
    if len(pagesLocationList)>0:
        if "cover" in configDictionary.keys():
            coverNumber = configDictionary["pages"].index(configDictionary["cover"])
        else:
            coverNumber = 0
        for p in pagesLocationList:
            if os.path.exists(p):
                shutil.copy2(p, str(imagePath))
                filename = str(Path(imagePath / os.path.basename(p)))
                pagesList.append(filename)
                epubArchive.write(filename, os.path.relpath(filename, str(exportPath)))
        if len(pagesLocationList) >= coverNumber:
            coverpageurl = pagesList[coverNumber]
    else:
        print("CPMT: Couldn't find the location for the epub images.")
        return False
 
    # for each image, make an xhtml file
    
    htmlFiles = []
    listOfNavItems = {}
    listofSpreads = []
    regions = []
    for i in range(len(pagesList)):
        pageName = "Page" + str(i) + ".xhtml"
        doc = QDomDocument()
        html = doc.createElement("html")
        doc.appendChild(html)
        html.setAttribute("xmlns", "http://www.w3.org/1999/xhtml")
        html.setAttribute("xmlns:epub", "http://www.idpf.org/2007/ops")
 
        # The viewport is a prerequisite to get pre-paginated
        # layouts working. We'll make the layout the same size
        # as the image.
        
        head = doc.createElement("head")
        viewport = doc.createElement("meta")
        viewport.setAttribute("name", "viewport")
 
        img = QImage()
        img.load(pagesLocationList[i])
        w = img.width()
        h = img.height()
        
        widthHeight = "width="+str(w)+", height="+str(h)
        
        viewport.setAttribute("content", widthHeight)
        head.appendChild(viewport)
        html.appendChild(head)
        
        # Here, we process the region navigation data to percentages
        # because we have access here to the width and height of the viewport.
        
        data = pageData[i]
        transform = data["transform"]
        for v in data["vector"]:
            pointsList = []
            dominantColor = QColor(Qt.white)
            listOfColors = []
            for point in v["boundingBox"]:
                offset = QPointF(transform["offsetX"], transform["offsetY"])
                pixelPoint = QPointF(point.x() * transform["resDiff"], point.y() * transform["resDiff"])
                newPoint = pixelPoint - offset
                x = max(0, min(w, int(newPoint.x() * transform["scaleWidth"])))
                y = max(0, min(h, int(newPoint.y() * transform["scaleHeight"])))
                listOfColors.append(img.pixelColor(QPointF(x, y).toPoint()))
                pointsList.append(QPointF((x/w)*100, (y/h)*100))
            regionType = "panel"
            if "text" in v.keys():
                regionType = "text"
            if len(listOfColors)>0:
                dominantColor = listOfColors[-1]
                listOfColors = listOfColors[:-1]
                for color in listOfColors:
                    dominantColor.setRedF(0.5*(dominantColor.redF()+color.redF()))
                    dominantColor.setGreenF(0.5*(dominantColor.greenF()+color.greenF()))
                    dominantColor.setBlueF(0.5*(dominantColor.blueF()+color.blueF()))
            region = {}
            bounds = QPolygonF(pointsList).boundingRect()
            region["points"] = bounds
            region["type"] = regionType
            region["page"] = str(Path(textPath / pageName))
            region["primaryColor"] = dominantColor.name()
            regions.append(region)
 
        # We can also figureout here whether the page can be seen as a table of contents entry.
        
        if "acbf_title" in data["keys"]:
            listOfNavItems[str(Path(textPath / pageName))] = data["title"]
            
        # Or spreads...
        
        if "epub_spread" in data["keys"]:
            listofSpreads.append(str(Path(textPath / pageName)))
 
        body = doc.createElement("body")
 
        img = doc.createElement("img")
        img.setAttribute("src", os.path.relpath(pagesList[i], str(textPath)))
        body.appendChild(img)
 
        html.appendChild(body)
 
        filename = str(Path(textPath / pageName))
        docFile = open(filename, 'w', newline="", encoding="utf-8")
        docFile.write(doc.toString(indent=2))
        docFile.close()
 
        if pagesList[i] == coverpageurl:
            coverpagehtml = os.path.relpath(filename, str(oebps))
        htmlFiles.append(filename)
        
        # Write to zip.
        epubArchive.write(filename, os.path.relpath(filename, str(exportPath)))
 
    # metadata
    
    filename = write_opf_file(oebps, configDictionary, htmlFiles, pagesList, coverpageurl, coverpagehtml, listofSpreads)
    epubArchive.write(filename, os.path.relpath(filename, str(exportPath)))
    
    filename = write_region_nav_file(oebps, configDictionary, htmlFiles, regions)
    epubArchive.write(filename, os.path.relpath(filename, str(exportPath)))
    
    # toc
    filename = write_nav_file(oebps, configDictionary, htmlFiles, listOfNavItems)
    epubArchive.write(filename, os.path.relpath(filename, str(exportPath)))
    
    filename = write_ncx_file(oebps, configDictionary, htmlFiles, listOfNavItems)
    epubArchive.write(filename, os.path.relpath(filename, str(exportPath)))
    
    epubArchive.close()
 
    return True
 
"""
Write OPF metadata file
"""
 
 
def write_opf_file(path, configDictionary, htmlFiles, pagesList, coverpageurl, coverpagehtml, listofSpreads):
    
    # marc relators
    # This has several entries removed to reduce it to the most relevant entries.
    marcRelators = {"abr":i18n("Abridger"), "acp":i18n("Art copyist"), "act":i18n("Actor"), "adi":i18n("Art director"), "adp":i18n("Adapter"), "ann":i18n("Annotator"), "ant":i18n("Bibliographic antecedent"), "arc":i18n("Architect"), "ard":i18n("Artistic director"), "art":i18n("Artist"), "asn":i18n("Associated name"), "ato":i18n("Autographer"), "att":i18n("Attributed name"), "aud":i18n("Author of dialog"), "aut":i18n("Author"), "bdd":i18n("Binding designer"), "bjd":i18n("Bookjacket designer"), "bkd":i18n("Book designer"), "bkp":i18n("Book producer"), "blw":i18n("Blurb writer"), "bnd":i18n("Binder"), "bpd":i18n("Bookplate designer"), "bsl":i18n("Bookseller"), "cll":i18n("Calligrapher"), "clr":i18n("Colorist"), "cns":i18n("Censor"), "cov":i18n("Cover designer"), "cph":i18n("Copyright holder"), "cre":i18n("Creator"), "ctb":i18n("Contributor"), "cur":i18n("Curator"), "cwt":i18n("Commentator for written text"), "drm":i18n("Draftsman"), "dsr":i18n("Designer"), "dub":i18n("Dubious author"), "edt":i18n("Editor"), "etr":i18n("Etcher"), "exp":i18n("Expert"), "fnd":i18n("Funder"), "ill":i18n("Illustrator"), "ilu":i18n("Illuminator"), "ins":i18n("Inscriber"), "lse":i18n("Licensee"), "lso":i18n("Licensor"), "ltg":i18n("Lithographer"), "mdc":i18n("Metadata contact"), "oth":i18n("Other"), "own":i18n("Owner"), "pat":i18n("Patron"), "pbd":i18n("Publishing director"), "pbl":i18n("Publisher"), "prt":i18n("Printer"), "sce":i18n("Scenarist"), "scr":i18n("Scribe"), "spn":i18n("Sponsor"), "stl":i18n("Storyteller"), "trc":i18n("Transcriber"), "trl":i18n("Translator"), "tyd":i18n("Type designer"), "tyg":i18n("Typographer"), "wac":i18n("Writer of added commentary"), "wal":i18n("Writer of added lyrics"), "wam":i18n("Writer of accompanying material"), "wat":i18n("Writer of added text"), "win":i18n("Writer of introduction"), "wpr":i18n("Writer of preface"), "wst":i18n("Writer of supplementary textual content")}
    
    # opf file
    opfFile = QDomDocument()
    opfRoot = opfFile.createElement("package")
    opfRoot.setAttribute("version", "3.0")
    opfRoot.setAttribute("unique-identifier", "BookId")
    opfRoot.setAttribute("xmlns", "http://www.idpf.org/2007/opf")
    opfRoot.setAttribute("prefix", "rendition: http://www.idpf.org/vocab/rendition/#")
    opfFile.appendChild(opfRoot)
    
    opfMeta = opfFile.createElement("metadata")
    opfMeta.setAttribute("xmlns:dc", "http://purl.org/dc/elements/1.1/")
    opfMeta.setAttribute("xmlns:dcterms", "http://purl.org/dc/terms/")
 
    # EPUB metadata requires a title, language and uuid
 
    langString = "en-US"
    if "language" in configDictionary.keys():
        langString = str(configDictionary["language"]).replace("_", "-")
    
    bookLang = opfFile.createElement("dc:language")
    bookLang.appendChild(opfFile.createTextNode(langString))
    opfMeta.appendChild(bookLang)
 
    bookTitle = opfFile.createElement("dc:title")
    if "title" in configDictionary.keys():
        bookTitle.appendChild(opfFile.createTextNode(str(configDictionary["title"])))
    else:
        bookTitle.appendChild(opfFile.createTextNode("Comic with no Name"))
    opfMeta.appendChild(bookTitle)
    
    # Generate series title and the like here too.
    if "seriesName" in configDictionary.keys():
        bookTitle.setAttribute("id", "main")
 
        refine = opfFile.createElement("meta")
        refine.setAttribute("refines", "#main")
        refine.setAttribute("property", "title-type")
        refine.appendChild(opfFile.createTextNode("main"))
        opfMeta.appendChild(refine)
 
        refine2 = opfFile.createElement("meta")
        refine2.setAttribute("refines", "#main")
        refine2.setAttribute("property", "display-seq")
        refine2.appendChild(opfFile.createTextNode("1"))
        opfMeta.appendChild(refine2)
 
        seriesTitle = opfFile.createElement("dc:title")
        seriesTitle.appendChild(opfFile.createTextNode(str(configDictionary["seriesName"])))
        seriesTitle.setAttribute("id", "series")
        opfMeta.appendChild(seriesTitle)
 
        refineS = opfFile.createElement("meta")
        refineS.setAttribute("refines", "#series")
        refineS.setAttribute("property", "title-type")
        refineS.appendChild(opfFile.createTextNode("collection"))
        opfMeta.appendChild(refineS)
 
        refineS2 = opfFile.createElement("meta")
        refineS2.setAttribute("refines", "#series")
        refineS2.setAttribute("property", "display-seq")
        refineS2.appendChild(opfFile.createTextNode("2"))
        opfMeta.appendChild(refineS2)
 
        if "seriesNumber" in configDictionary.keys():
            refineS3 = opfFile.createElement("meta")
            refineS3.setAttribute("refines", "#series")
            refineS3.setAttribute("property", "group-position")
            refineS3.appendChild(opfFile.createTextNode(str(configDictionary["seriesNumber"])))
            opfMeta.appendChild(refineS3)
 
    uuid = str(configDictionary["uuid"])
    uuid = uuid.strip("{")
    uuid = uuid.strip("}")
 
    # Append the id, and assign it as the bookID.
    uniqueID = opfFile.createElement("dc:identifier")
    uniqueID.appendChild(opfFile.createTextNode("urn:uuid:"+uuid))
    uniqueID.setAttribute("id", "BookId")
    opfMeta.appendChild(uniqueID)
 
    if "authorList" in configDictionary.keys():
        authorEntry = 0
        for authorE in range(len(configDictionary["authorList"])):
            authorDict = configDictionary["authorList"][authorE]
            authorType = "dc:creator"
            if "role" in authorDict.keys():
                # This determines if someone was just a contributor, but might need a more thorough version.
                if str(authorDict["role"]).lower() in ["editor", "assistant editor", "proofreader", "beta", "patron", "funder"]:
                    authorType = "dc:contributor"
            author = opfFile.createElement(authorType)
            authorName = []
            if "last-name" in authorDict.keys():
                authorName.append(authorDict["last-name"])
            if "first-name" in authorDict.keys():
                authorName.append(authorDict["first-name"])
            if "initials" in authorDict.keys():
                authorName.append(authorDict["initials"])
            if "nickname" in authorDict.keys():
                authorName.append("(" + authorDict["nickname"] + ")")
            author.appendChild(opfFile.createTextNode(", ".join(authorName)))
            author.setAttribute("id", "cre" + str(authorE))
            opfMeta.appendChild(author)
            if "role" in authorDict.keys():
                role = opfFile.createElement("meta")
                role.setAttribute("refines", "#cre" + str(authorE))
                role.setAttribute("scheme", "marc:relators")
                role.setAttribute("property", "role")
                roleString = str(authorDict["role"])
                if roleString in marcRelators.values() or roleString in marcRelators.keys():
                    i = list(marcRelators.values()).index(roleString)
                    roleString = list(marcRelators.keys())[i]
                else:
                    roleString = "oth"
                role.appendChild(opfFile.createTextNode(roleString))
                opfMeta.appendChild(role)
            refine = opfFile.createElement("meta")
            refine.setAttribute("refines", "#cre"+str(authorE))
            refine.setAttribute("property", "display-seq")
            refine.appendChild(opfFile.createTextNode(str(authorE+1)))
            opfMeta.appendChild(refine)
 
    if "publishingDate" in configDictionary.keys():
        date = opfFile.createElement("dc:date")
        date.appendChild(opfFile.createTextNode(configDictionary["publishingDate"]))
        opfMeta.appendChild(date)
    
    #Creation date
    modified = opfFile.createElement("meta")
    modified.setAttribute("property", "dcterms:modified")
    modified.appendChild(opfFile.createTextNode(QDateTime.currentDateTimeUtc().toString(Qt.DateFormat.ISODate)))
    opfMeta.appendChild(modified)
    
    if "source" in configDictionary.keys():
        if len(configDictionary["source"])>0:
            source = opfFile.createElement("dc:source")
            source.appendChild(opfFile.createTextNode(configDictionary["source"]))
            opfMeta.appendChild(source)
    
    description = opfFile.createElement("dc:description")
    if "summary" in configDictionary.keys():
        description.appendChild(opfFile.createTextNode(configDictionary["summary"]))
    else:
        description.appendChild(opfFile.createTextNode("There was no summary upon generation of this file."))
    opfMeta.appendChild(description)
 
    # Type can be dictionary or index, or one of those edupub thingies. Not necessary for comics.
    # typeE = opfFile.createElement("dc:type")
    # opfMeta.appendChild(typeE)
    
    if "publisherName" in configDictionary.keys():
        publisher = opfFile.createElement("dc:publisher")
        publisher.appendChild(opfFile.createTextNode(configDictionary["publisherName"]))
        opfMeta.appendChild(publisher)
    
    
    if "isbn-number" in configDictionary.keys():
        isbnnumber = configDictionary["isbn-number"]
 
        if len(isbnnumber)>0:
            publishISBN = opfFile.createElement("dc:identifier")
            publishISBN.appendChild(opfFile.createTextNode(str("urn:isbn:") + isbnnumber))
            opfMeta.appendChild(publishISBN)
 
    if "license" in configDictionary.keys():
 
        if len(configDictionary["license"])>0:
            rights = opfFile.createElement("dc:rights")
            rights.appendChild(opfFile.createTextNode(configDictionary["license"]))
            opfMeta.appendChild(rights)
    
    """
    Not handled
    Relation -  This is for whether the work has a relationship with another work.
                It could be fanart, but also adaptation, an academic work, etc.
    Coverage -  This is for the time/place that the work covers. Typically to determine
                whether an academic work deals with a certain time period or place.
                For comics you could use this to mark historical comics, but other than
                that we'd need a much better ui to define this.
    """
 
    # These are all dublin core subjects.
    # 3.1 defines the ability to use an authority, but that
    # might be a bit too complicated right now.
 
    if "genre" in configDictionary.keys():
        genreListConf = configDictionary["genre"]
        if isinstance(configDictionary["genre"], dict):
            genreListConf = configDictionary["genre"].keys()
        for g in genreListConf:
            subject = opfFile.createElement("dc:subject")
            subject.appendChild(opfFile.createTextNode(g))
            opfMeta.appendChild(subject)
    if "characters" in configDictionary.keys():
        for name in configDictionary["characters"]:
            char = opfFile.createElement("dc:subject")
            char.appendChild(opfFile.createTextNode(name))
            opfMeta.appendChild(char)
    if "format" in configDictionary.keys():
        for formatF in configDictionary["format"]:
            f = opfFile.createElement("dc:subject")
            f.appendChild(opfFile.createTextNode(formatF))
            opfMeta.appendChild(f)
    if "otherKeywords" in configDictionary.keys():
        for key in configDictionary["otherKeywords"]:
            word = opfFile.createElement("dc:subject")
            word.appendChild(opfFile.createTextNode(key))
            opfMeta.appendChild(word)
 
    # Pre-pagination and layout
    # Comic are always prepaginated.
    
    elLayout = opfFile.createElement("meta")
    elLayout.setAttribute("property", "rendition:layout")
    elLayout.appendChild(opfFile.createTextNode("pre-paginated"))
    opfMeta.appendChild(elLayout)
    
    # We should figure out if the pages are portrait or not...
    elOrientation = opfFile.createElement("meta")
    elOrientation.setAttribute("property", "rendition:orientation")
    elOrientation.appendChild(opfFile.createTextNode("portrait"))
    opfMeta.appendChild(elOrientation)
    
    elSpread = opfFile.createElement("meta")
    elSpread.setAttribute("property", "rendition:spread")
    elSpread.appendChild(opfFile.createTextNode("landscape"))
    opfMeta.appendChild(elSpread)
    
    opfRoot.appendChild(opfMeta)
    
    # Manifest
 
    opfManifest = opfFile.createElement("manifest")
    toc = opfFile.createElement("item")
    toc.setAttribute("id", "ncx")
    toc.setAttribute("href", "toc.ncx")
    toc.setAttribute("media-type", "application/x-dtbncx+xml")
    opfManifest.appendChild(toc)
    
    region = opfFile.createElement("item")
    region.setAttribute("id", "regions")
    region.setAttribute("href", "region-nav.xhtml")
    region.setAttribute("media-type", "application/xhtml+xml")
    region.setAttribute("properties", "data-nav") # Set the propernavmap to use this later)
    opfManifest.appendChild(region)
    
    nav = opfFile.createElement("item")
    nav.setAttribute("id", "nav")
    nav.setAttribute("href", "nav.xhtml")
    nav.setAttribute("media-type", "application/xhtml+xml")
    nav.setAttribute("properties", "nav") # Set the propernavmap to use this later)
    opfManifest.appendChild(nav)
    
    ids = 0
    for p in pagesList:
        item = opfFile.createElement("item")
        item.setAttribute("id", "img"+str(ids))
        ids +=1
        item.setAttribute("href", os.path.relpath(p, str(path)))
        item.setAttribute("media-type", "image/png")
        if os.path.basename(p) == os.path.basename(coverpageurl):
            item.setAttribute("properties", "cover-image")
        opfManifest.appendChild(item)
 
 
    ids = 0
    for p in htmlFiles:
        item = opfFile.createElement("item")
        item.setAttribute("id", "p"+str(ids))
        ids +=1
        item.setAttribute("href", os.path.relpath(p, str(path)))
        item.setAttribute("media-type", "application/xhtml+xml")
        opfManifest.appendChild(item)
    
 
    opfRoot.appendChild(opfManifest)
    
    # Spine
 
    opfSpine = opfFile.createElement("spine")
    # this sets the table of contents to use the ncx file
    opfSpine.setAttribute("toc", "ncx")
    # Reading Direction:
 
    spreadRight = True
    direction = 0
    if "readingDirection" in configDictionary.keys():
        if configDictionary["readingDirection"] == "rightToLeft":
            opfSpine.setAttribute("page-progression-direction", "rtl")
            spreadRight = False
            direction = 1
        else:
            opfSpine.setAttribute("page-progression-direction", "ltr")
 
    # Here we'd need to switch between the two and if spread keywrod use neither but combine with spread-none
    
    ids = 0
    for p in htmlFiles:
        item = opfFile.createElement("itemref")
        item.setAttribute("idref", "p"+str(ids))
        ids +=1
        props = []
        if p in listofSpreads:
            # Put this one in the center.
            props.append("rendition:page-spread-center")
            
            # Reset the spread boolean.
            # It needs to point at the first side after the spread.
            # So ltr -> spread-left, rtl->spread-right
            if direction == 0:
                spreadRight = False
            else:
                spreadRight = True
        else:
            if spreadRight:
                props.append("page-spread-right")
                spreadRight = False
            else:
                props.append("page-spread-left")
                spreadRight = True
        item.setAttribute("properties", " ".join(props))
        opfSpine.appendChild(item)
    opfRoot.appendChild(opfSpine)
 
    # Guide
    
    opfGuide = opfFile.createElement("guide")
    if coverpagehtml is not None and coverpagehtml.isspace() is False and len(coverpagehtml) > 0:
        item = opfFile.createElement("reference")
        item.setAttribute("type", "cover")
        item.setAttribute("title", "Cover")
        item.setAttribute("href", coverpagehtml)
        opfGuide.appendChild(item)
    opfRoot.appendChild(opfGuide)
 
    docFile = open(str(Path(path / "content.opf")), 'w', newline="", encoding="utf-8")
    docFile.write(opfFile.toString(indent=2))
    docFile.close()
    return str(Path(path / "content.opf"))
 
"""
Write a region navmap file.
"""
 
def write_region_nav_file(path, configDictionary, htmlFiles, regions = []):
    navDoc = QDomDocument()
    navRoot = navDoc.createElement("html")
    navRoot.setAttribute("xmlns", "http://www.w3.org/1999/xhtml")
    navRoot.setAttribute("xmlns:epub", "http://www.idpf.org/2007/ops")
    navDoc.appendChild(navRoot)
    
    head = navDoc.createElement("head")
    title = navDoc.createElement("title")
    title.appendChild(navDoc.createTextNode("Region Navigation"))
    head.appendChild(title)
    navRoot.appendChild(head)
    
    body = navDoc.createElement("body")
    navRoot.appendChild(body)
    
    nav = navDoc.createElement("nav")
    nav.setAttribute("epub:type", "region-based")
    nav.setAttribute("prefix", "ahl: http://idpf.org/epub/vocab/ahl")
    body.appendChild(nav)
    
    # Let's write the panels and balloons down now.
    
    olPanels = navDoc.createElement("ol")
    for region in regions:
        if region["type"] == "panel":
            pageName = os.path.relpath(region["page"], str(path))
            print("accessing panel")
            li = navDoc.createElement("li")
            li.setAttribute("epub:type", "panel")
            
            anchor = navDoc.createElement("a")
            bounds = region["points"]
            anchor.setAttribute("href", pageName+"#xywh=percent:"+str(bounds.x())+","+str(bounds.y())+","+str(bounds.width())+","+str(bounds.height()))
            
            if len(region["primaryColor"])>0:
                primaryC = navDoc.createElement("meta")
                primaryC.setAttribute("property","ahl:primary-color")
                primaryC.setAttribute("content", region["primaryColor"])
                anchor.appendChild(primaryC)
            
            li.appendChild(anchor)
            olBalloons = navDoc.createElement("ol")
            
            """
            The region nav spec specifies that we should have text-areas/balloons as a refinement on
            the panel.
            For each panel, we'll check if there's balloons/text-areas inside, and we'll do that by
            checking whether the center point is inside the panel because some comics have balloons
            that overlap the gutters.
            """
            for balloon in regions:
                if balloon["type"] == "text" and balloon["page"] == region["page"] and bounds.contains(balloon["points"].center()):
                    liBalloon = navDoc.createElement("li")
                    liBalloon.setAttribute("epub:type", "text-area")
 
                    anchorBalloon = navDoc.createElement("a")
                    BBounds = balloon["points"]
                    anchorBalloon.setAttribute("href", pageName+"#xywh=percent:"+str(BBounds.x())+","+str(BBounds.y())+","+str(BBounds.width())+","+str(BBounds.height()))
 
                    liBalloon.appendChild(anchorBalloon)
                    olBalloons.appendChild(liBalloon)
 
            if olBalloons.hasChildNodes():
                li.appendChild(olBalloons)
            olPanels.appendChild(li)
    nav.appendChild(olPanels)
 
    navFile = open(str(Path(path / "region-nav.xhtml")), 'w', newline="", encoding="utf-8")
    navFile.write(navDoc.toString(indent=2))
    navFile.close()
    return str(Path(path / "region-nav.xhtml"))
 
"""
Write XHTML nav file.
 
This is virtually the same as the NCX file, except that
the navigation document can be styled, and is what 3.1 and
3.2 expect as a primary navigation document.
 
This function will both create a table of contents, using the
"acbf_title" feature, as well as a regular pageslist.
"""
 
def write_nav_file(path, configDictionary, htmlFiles, listOfNavItems):
    navDoc = QDomDocument()
    navRoot = navDoc.createElement("html")
    navRoot.setAttribute("xmlns", "http://www.w3.org/1999/xhtml")
    navRoot.setAttribute("xmlns:epub", "http://www.idpf.org/2007/ops")
    navDoc.appendChild(navRoot)
    
    head = navDoc.createElement("head")
    title = navDoc.createElement("title")
    title.appendChild(navDoc.createTextNode("Table of Contents"))
    head.appendChild(title)
    navRoot.appendChild(head)
    
    body = navDoc.createElement("body")
    navRoot.appendChild(body)
    
    # The Table of Contents
    
    toc = navDoc.createElement("nav")
    toc.setAttribute("epub:type", "toc")
    oltoc = navDoc.createElement("ol")
    li = navDoc.createElement("li")
    anchor = navDoc.createElement("a")
    anchor.setAttribute("href", os.path.relpath(htmlFiles[0], str(path)))
    anchor.appendChild(navDoc.createTextNode("Start"))
    li.appendChild(anchor)
    oltoc.appendChild(li)
    for fileName in listOfNavItems.keys():
        li = navDoc.createElement("li")
        anchor = navDoc.createElement("a")
        anchor.setAttribute("href", os.path.relpath(fileName, str(path)))
        anchor.appendChild(navDoc.createTextNode(listOfNavItems[fileName]))
        li.appendChild(anchor)
        oltoc.appendChild(li)
    
    toc.appendChild(oltoc)
    body.appendChild(toc)
    
    # The Pages List.
    
    pageslist = navDoc.createElement("nav")
    pageslist.setAttribute("epub:type", "page-list")
    olpages = navDoc.createElement("ol")
    
    entry = 1
    for i in range(len(htmlFiles)):
        li = navDoc.createElement("li")
        anchor = navDoc.createElement("a")
        anchor.setAttribute("href", os.path.relpath(htmlFiles[1], str(path)))
        anchor.appendChild(navDoc.createTextNode(str(i)))
        li.appendChild(anchor)
        olpages.appendChild(li)
    pageslist.appendChild(olpages)
    
    body.appendChild(pageslist)
    
    
 
    navFile = open(str(Path(path / "nav.xhtml")), 'w', newline="", encoding="utf-8")
    navFile.write(navDoc.toString(indent=2))
    navFile.close()
    return str(Path(path / "nav.xhtml"))
 
"""
Write a NCX file.
 
This is the same as the navigation document above, but then
for 2.0 backward compatibility.
"""
 
def write_ncx_file(path, configDictionary, htmlFiles, listOfNavItems):
    tocDoc = QDomDocument()
    ncx = tocDoc.createElement("ncx")
    ncx.setAttribute("version", "2005-1")
    ncx.setAttribute("xmlns", "http://www.daisy.org/z3986/2005/ncx/")
    tocDoc.appendChild(ncx)
 
    tocHead = tocDoc.createElement("head")
    
    # NCX also has some meta values that are in the head.
    # They are shared with the opf metadata document.
    
    uuid = str(configDictionary["uuid"])
    uuid = uuid.strip("{")
    uuid = uuid.strip("}")
    metaID = tocDoc.createElement("meta")
    metaID.setAttribute("content", uuid)
    metaID.setAttribute("name", "dtb:uid")
    tocHead.appendChild(metaID)
    metaDepth = tocDoc.createElement("meta")
    metaDepth.setAttribute("content", str(1))
    metaDepth.setAttribute("name", "dtb:depth")
    tocHead.appendChild(metaDepth)
    metaTotal = tocDoc.createElement("meta")
    metaTotal.setAttribute("content", str(len(htmlFiles)))
    metaTotal.setAttribute("name", "dtb:totalPageCount")
    tocHead.appendChild(metaTotal)
    metaMax = tocDoc.createElement("meta")
    metaMax.setAttribute("content", str(len(htmlFiles)))
    metaMax.setAttribute("name", "dtb:maxPageNumber")
    tocHead.appendChild(metaDepth)
    ncx.appendChild(tocHead)
 
    docTitle = tocDoc.createElement("docTitle")
    text = tocDoc.createElement("text")
    if "title" in configDictionary.keys():
        text.appendChild(tocDoc.createTextNode(str(configDictionary["title"])))
    else:
        text.appendChild(tocDoc.createTextNode("Comic with no Name"))
    docTitle.appendChild(text)
    ncx.appendChild(docTitle)
    
    # The navmap is a table of contents.
 
    navmap = tocDoc.createElement("navMap")
    navPoint = tocDoc.createElement("navPoint")
    navPoint.setAttribute("id", "navPoint-1")
    navPoint.setAttribute("playOrder", "1")
    navLabel = tocDoc.createElement("navLabel")
    navLabelText = tocDoc.createElement("text")
    navLabelText.appendChild(tocDoc.createTextNode("Start"))
    navLabel.appendChild(navLabelText)
    navContent = tocDoc.createElement("content")
    navContent.setAttribute("src", os.path.relpath(htmlFiles[0], str(path)))
    navPoint.appendChild(navLabel)
    navPoint.appendChild(navContent)
    navmap.appendChild(navPoint)
    entry = 1
    for fileName in listOfNavItems.keys():
        entry +=1
        navPointT = tocDoc.createElement("navPoint")
        navPointT.setAttribute("id", "navPoint-"+str(entry))
        navPointT.setAttribute("playOrder", str(entry))
        navLabelT = tocDoc.createElement("navLabel")
        navLabelTText = tocDoc.createElement("text")
        navLabelTText.appendChild(tocDoc.createTextNode(listOfNavItems[fileName]))
        navLabelT.appendChild(navLabelTText)
        navContentT = tocDoc.createElement("content")
        navContentT.setAttribute("src", os.path.relpath(fileName, str(path)))
        navPointT.appendChild(navLabelT)
        navPointT.appendChild(navContentT)
        navmap.appendChild(navPointT)
    ncx.appendChild(navmap)
    
    # The pages list on the other hand just lists all pages.
    
    pagesList = tocDoc.createElement("pageList")
    navLabelPages = tocDoc.createElement("navLabel")
    navLabelPagesText = tocDoc.createElement("text")
    navLabelPagesText.appendChild(tocDoc.createTextNode("Pages"))
    navLabelPages.appendChild(navLabelPagesText)
    pagesList.appendChild(navLabelPages)
    for i in range(len(htmlFiles)):
        pageTarget = tocDoc.createElement("pageTarget")
        pageTarget.setAttribute("type", "normal")
        pageTarget.setAttribute("id", "page-"+str(i))
        pageTarget.setAttribute("value", str(i))
        navLabelPagesTarget = tocDoc.createElement("navLabel")
        navLabelPagesTargetText = tocDoc.createElement("text")
        navLabelPagesTargetText.appendChild(tocDoc.createTextNode(str(i+1)))
        navLabelPagesTarget.appendChild(navLabelPagesTargetText)
        pageTarget.appendChild(navLabelPagesTarget)
        pageTargetContent = tocDoc.createElement("content")
        pageTargetContent.setAttribute("src", os.path.relpath(htmlFiles[i], str(path)))
        pageTarget.appendChild(pageTargetContent)
        pagesList.appendChild(pageTarget)
    ncx.appendChild(pagesList)
 
    # Save the document.
 
    docFile = open(str(Path(path / "toc.ncx")), 'w', newline="", encoding="utf-8")
    docFile.write(tocDoc.toString(indent=2))
    docFile.close()
    return str(Path(path / "toc.ncx"))