diff --git a/app/build.gradle b/app/build.gradle index b76d28272..6d576b873 100644 --- a/app/build.gradle +++ b/app/build.gradle @@ -178,7 +178,8 @@ dependencies { implementation 'com.github.gedoor:rhino-android:1.5' //网络 - implementation 'com.squareup.okhttp3:okhttp:4.9.0' + //noinspection GradleDependency + implementation 'com.squareup.okhttp3:okhttp:4.9.1' implementation 'com.ljx.rxhttp:rxhttp:2.5.7' kapt 'com.ljx.rxhttp:rxhttp-compiler:2.5.7' diff --git a/app/src/main/java/io/legado/app/help/BookHelp.kt b/app/src/main/java/io/legado/app/help/BookHelp.kt index aa89a5505..dba76f97f 100644 --- a/app/src/main/java/io/legado/app/help/BookHelp.kt +++ b/app/src/main/java/io/legado/app/help/BookHelp.kt @@ -1,5 +1,6 @@ package io.legado.app.help +import android.net.Uri import io.legado.app.constant.AppPattern import io.legado.app.constant.EventBus import io.legado.app.data.appDb @@ -55,6 +56,28 @@ object BookHelp { } } + fun getEpubFile(book: Book,): File { + val file = FileUtils.getFile( + downloadDir, + cacheFolderName, + book.getFolderName(), + "index.epubx" + ) + if(!file.exists()){ + val input = if (book.bookUrl.isContentScheme()) { + val uri = Uri.parse(book.bookUrl) + appCtx.contentResolver.openInputStream(uri) + } else { + File(book.bookUrl).inputStream() + } + if (input != null) { + FileUtils.writeInputStream(file, input) + } + + } + return file + } + suspend fun saveContent(book: Book, bookChapter: BookChapter, content: String) { if (content.isEmpty()) return //保存文本 @@ -113,7 +136,7 @@ object BookHelp { ) } - private fun getImageSuffix(src: String): String { + fun getImageSuffix(src: String): String { var suffix = src.substringAfterLast(".").substringBefore(",") if (suffix.length > 5) { suffix = ".jpg" diff --git a/app/src/main/java/io/legado/app/model/localBook/EpubFile.kt b/app/src/main/java/io/legado/app/model/localBook/EpubFile.kt index d076187f3..b80540f74 100644 --- a/app/src/main/java/io/legado/app/model/localBook/EpubFile.kt +++ b/app/src/main/java/io/legado/app/model/localBook/EpubFile.kt @@ -2,16 +2,16 @@ package io.legado.app.model.localBook import android.graphics.Bitmap import android.graphics.BitmapFactory -import android.net.Uri import android.text.TextUtils import io.legado.app.data.entities.Book import io.legado.app.data.entities.BookChapter -import io.legado.app.utils.* +import io.legado.app.help.BookHelp +import io.legado.app.utils.FileUtils +import io.legado.app.utils.HtmlFormatter +import io.legado.app.utils.MD5Utils +import io.legado.app.utils.externalFilesDir import me.ag2s.epublib.domain.EpubBook -import me.ag2s.epublib.domain.MediaTypes -import me.ag2s.epublib.domain.Resources import me.ag2s.epublib.epub.EpubReader -import me.ag2s.epublib.util.ResourceUtil import org.jsoup.Jsoup import splitties.init.appCtx import java.io.File @@ -20,8 +20,7 @@ import java.io.IOException import java.io.InputStream import java.nio.charset.Charset import java.util.* -import java.util.zip.ZipEntry -import java.util.zip.ZipInputStream +import java.util.zip.ZipFile class EpubFile(var book: Book) { @@ -30,6 +29,7 @@ class EpubFile(var book: Book) { @Synchronized private fun getEFile(book: Book): EpubFile { + BookHelp.getEpubFile(book) if (eFile == null || eFile?.book?.bookUrl != book.bookUrl) { eFile = EpubFile(book) return eFile!! @@ -101,32 +101,12 @@ class EpubFile(var book: Book) { /*重写epub文件解析代码，直接读出压缩包文件生成Resources给epublib，这样的好处是可以逐一修改某些文件的格式错误*/ private fun readEpub(): EpubBook? { try { - val input = if (book.bookUrl.isContentScheme()) { - val uri = Uri.parse(book.bookUrl) - appCtx.contentResolver.openInputStream(uri) - } else { - File(book.bookUrl).inputStream() - } - input ?: return null - val inZip = ZipInputStream(input) - var zipEntry: ZipEntry? - val resources = Resources() - do { - zipEntry = inZip.nextEntry - if ((zipEntry == null) || zipEntry.isDirectory || zipEntry == ZipEntry("")) continue - val resource = ResourceUtil.createResource(zipEntry, inZip) - if (resource.mediaType == MediaTypes.XHTML) resource.inputEncoding = "UTF-8" - if (zipEntry.name.endsWith(".opf")) { - /*掌上书苑有很多自制书OPF的nameSpace格式不标准，强制修复成正确的格式*/ - val newS = String(resource.data).replace( - "\\smlns=\"http://www.idpf.org/2007/opf\"".toRegex(), - " xmlns=\"http://www.idpf.org/2007/opf\"" - ) - resource.data = newS.toByteArray() - } - resources.add(resource) - } while (zipEntry != null) - if (resources.size() > 0) return EpubReader().readEpub(resources) + + val file = BookHelp.getEpubFile(book) + //通过懒加载读取epub + return EpubReader().readEpubLazy(ZipFile(file), "utf-8") + + } catch (e: Exception) { e.printStackTrace() } diff --git a/app/src/main/java/io/legado/app/model/localBook/LocalBook.kt b/app/src/main/java/io/legado/app/model/localBook/LocalBook.kt index a01efaee0..001cc93b5 100644 --- a/app/src/main/java/io/legado/app/model/localBook/LocalBook.kt +++ b/app/src/main/java/io/legado/app/model/localBook/LocalBook.kt @@ -138,6 +138,13 @@ object LocalBook { val bookFile = FileUtils.getFile(cacheFolder, book.originName) bookFile.delete() } + if(book.isEpub()){ + val bookFile=BookHelp.getEpubFile(book).parentFile + if (bookFile!=null&&bookFile.exists()){ + FileUtils.delete(bookFile,true) + } + + } if (deleteOriginal) { if (book.bookUrl.isContentScheme()) { diff --git a/app/src/main/java/io/legado/app/ui/book/cache/CacheViewModel.kt b/app/src/main/java/io/legado/app/ui/book/cache/CacheViewModel.kt index 55ecb17eb..819686dcf 100644 --- a/app/src/main/java/io/legado/app/ui/book/cache/CacheViewModel.kt +++ b/app/src/main/java/io/legado/app/ui/book/cache/CacheViewModel.kt @@ -24,7 +24,6 @@ import me.ag2s.epublib.epub.EpubWriter import me.ag2s.epublib.util.ResourceUtil import java.io.ByteArrayOutputStream import java.io.File -import java.io.FileInputStream import java.io.FileOutputStream import java.nio.charset.Charset @@ -184,6 +183,7 @@ class CacheViewModel(application: Application) : BaseViewModel(application) { } } + private fun exportEpub(file: File, book: Book) { val filename = "${book.name} by ${book.author}.epub" val epubBook = EpubBook() @@ -254,9 +254,12 @@ class CacheViewModel(application: Application) : BaseViewModel(application) { } private fun setPic(src: String, book: Book, epubBook: EpubBook) { + + val href = "${MD5Utils.md5Encode16(src)}${BookHelp.getImageSuffix(src)}" val vFile = BookHelp.getImage(book, src) + val fp = FileResourceProvider(vFile.parent) if (vFile.exists()) { - val img = Resource(FileInputStream(vFile), MD5Utils.md5Encode16(src) + ".jpg") + val img = LazyResource(fp, href) epubBook.resources.add(img) } } @@ -275,7 +278,10 @@ class CacheViewModel(application: Application) : BaseViewModel(application) { matchResult.groupValues[1].let { val src = NetworkUtils.getAbsoluteURL(chapter.url, it) setPic(src, book, epubBook) - text1 = text1.replace(src, MD5Utils.md5Encode16(src) + ".jpg") + text1 = text1.replace( + src, + "${MD5Utils.md5Encode16(src)}${BookHelp.getImageSuffix(src)}" + ) } } diff --git a/app/src/main/java/io/legado/app/ui/book/read/page/provider/ImageProvider.kt b/app/src/main/java/io/legado/app/ui/book/read/page/provider/ImageProvider.kt index 88bea027c..0bed03ea1 100644 --- a/app/src/main/java/io/legado/app/ui/book/read/page/provider/ImageProvider.kt +++ b/app/src/main/java/io/legado/app/ui/book/read/page/provider/ImageProvider.kt @@ -54,7 +54,9 @@ object ImageProvider { ChapterProvider.visibleWidth, ChapterProvider.visibleHeight ) - setCache(chapterIndex, src, bitmap) + if (bitmap != null) { + setCache(chapterIndex, src, bitmap) + } bitmap } catch (e: Exception) { null diff --git a/app/src/main/java/io/legado/app/utils/BitmapUtils.kt b/app/src/main/java/io/legado/app/utils/BitmapUtils.kt index ddd29063e..78491d904 100644 --- a/app/src/main/java/io/legado/app/utils/BitmapUtils.kt +++ b/app/src/main/java/io/legado/app/utils/BitmapUtils.kt @@ -9,6 +9,7 @@ import android.renderscript.RenderScript import android.renderscript.ScriptIntrinsicBlur import android.view.View import splitties.init.appCtx +import java.io.FileInputStream import java.io.IOException import kotlin.math.* @@ -25,11 +26,13 @@ object BitmapUtils { * @param height 想要显示的图片的高度 * @return */ - fun decodeBitmap(path: String, width: Int, height: Int): Bitmap { + fun decodeBitmap(path: String, width: Int, height: Int): Bitmap? { val op = BitmapFactory.Options() + op.inPreferredConfig = Config.RGB_565 + var ips = FileInputStream(path) // inJustDecodeBounds如果设置为true,仅仅返回图片实际的宽和高,宽和高是赋值给opts.outWidth,opts.outHeight; op.inJustDecodeBounds = true - BitmapFactory.decodeFile(path, op) //获取尺寸信息 + BitmapFactory.decodeStream(ips, null, op) //获取比例大小 val wRatio = ceil((op.outWidth / width).toDouble()).toInt() val hRatio = ceil((op.outHeight / height).toDouble()).toInt() @@ -42,21 +45,26 @@ object BitmapUtils { } } op.inJustDecodeBounds = false - return BitmapFactory.decodeFile(path, op) + ips = FileInputStream(path) + return BitmapFactory.decodeStream(ips, null, op) } /** 从path中获取Bitmap图片 * @param path 图片路径 * @return */ - fun decodeBitmap(path: String): Bitmap { + fun decodeBitmap(path: String): Bitmap? { + val opts = BitmapFactory.Options() + opts.inPreferredConfig = Config.RGB_565 + var ips = FileInputStream(path) opts.inJustDecodeBounds = true - BitmapFactory.decodeFile(path, opts) + BitmapFactory.decodeStream(ips, null, opts) opts.inSampleSize = computeSampleSize(opts, -1, 128 * 128) opts.inJustDecodeBounds = false + ips = FileInputStream(path) - return BitmapFactory.decodeFile(path, opts) + return BitmapFactory.decodeStream(ips, null, opts) } /** diff --git a/app/src/main/java/io/legado/app/utils/FileUtils.kt b/app/src/main/java/io/legado/app/utils/FileUtils.kt index 40ef4e2b1..1e8c8b502 100644 --- a/app/src/main/java/io/legado/app/utils/FileUtils.kt +++ b/app/src/main/java/io/legado/app/utils/FileUtils.kt @@ -517,6 +517,44 @@ object FileUtils { closeSilently(fos) } } + /** + * 保存文件内容 + */ + fun writeInputStream(filepath: String, data: InputStream): Boolean { + val file = File(filepath) + return writeInputStream(file,data) + } + + /** + * 保存文件内容 + */ + fun writeInputStream(file: File, data: InputStream): Boolean { + var fos: FileOutputStream? = null + return try { + if (!file.exists()) { + file.parentFile?.mkdirs() + file.createNewFile() + } + val buffer=ByteArray(1024*4) + fos = FileOutputStream(file) + while (true) { + val len = data.read(buffer, 0, buffer.size) + if (len == -1) { + break + } else { + fos.write(buffer, 0, len) + } + } + data.close() + fos.flush() + + true + } catch (e: IOException) { + false + } finally { + closeSilently(fos) + } + } /** * 追加文本内容 diff --git a/epublib/src/main/java/me/ag2s/epublib/domain/FileResourceProvider.java b/epublib/src/main/java/me/ag2s/epublib/domain/FileResourceProvider.java new file mode 100644 index 000000000..a9db2b5d9 --- /dev/null +++ b/epublib/src/main/java/me/ag2s/epublib/domain/FileResourceProvider.java @@ -0,0 +1,38 @@ +package me.ag2s.epublib.domain; + +import java.io.File; +import java.io.FileInputStream; +import java.io.IOException; +import java.io.InputStream; + +/** + * 用于创建epub，添加大文件（如大量图片）时容易OOM，使用LazyResource，避免OOM. + * + */ + +public class FileResourceProvider implements LazyResourceProvider { + //需要导入资源的父目录 + String dir; + + /** + * 创建一个文件夹里面文件夹的LazyResourceProvider，用于LazyResource。 + * @param dir 文件的目录 + */ + public FileResourceProvider(String dir) { + this.dir = dir; + } + + /** + * 创建一个文件夹里面文件夹的LazyResourceProvider，用于LazyResource。 + * @param dirfile 文件夹 + */ + @SuppressWarnings("unused") + public FileResourceProvider(File dirfile) { + this.dir = dirfile.getPath(); + } + + @Override + public InputStream getResourceStream(String href) throws IOException { + return new FileInputStream(new File(dir, href)); + } +} diff --git a/epublib/src/main/java/me/ag2s/epublib/epub/DOMUtil.java b/epublib/src/main/java/me/ag2s/epublib/epub/DOMUtil.java index 34e826abc..e6ae05b3d 100644 --- a/epublib/src/main/java/me/ag2s/epublib/epub/DOMUtil.java +++ b/epublib/src/main/java/me/ag2s/epublib/epub/DOMUtil.java @@ -1,135 +1,177 @@ package me.ag2s.epublib.epub; -import me.ag2s.epublib.util.StringUtil; -import java.util.ArrayList; -import java.util.List; import org.w3c.dom.Document; import org.w3c.dom.Element; import org.w3c.dom.Node; import org.w3c.dom.NodeList; import org.w3c.dom.Text; +import java.util.ArrayList; +import java.util.List; + +import me.ag2s.epublib.util.StringUtil; + /** * Utility methods for working with the DOM. * * @author paul - * */ // package class DOMUtil { - /** - * First tries to get the attribute value by doing an getAttributeNS on the element, if that gets an empty element it does a getAttribute without namespace. - * - * @param element element - * @param namespace namespace - * @param attribute attribute - * @return String Attribute - */ - public static String getAttribute(Element element, String namespace, - String attribute) { - String result = element.getAttributeNS(namespace, attribute); - if (StringUtil.isEmpty(result)) { - result = element.getAttribute(attribute); + /** + * First tries to get the attribute value by doing an getAttributeNS on the element, if that gets an empty element it does a getAttribute without namespace. + * + * @param element element + * @param namespace namespace + * @param attribute attribute + * @return String Attribute + */ + public static String getAttribute(Element element, String namespace, + String attribute) { + String result = element.getAttributeNS(namespace, attribute); + if (StringUtil.isEmpty(result)) { + result = element.getAttribute(attribute); + } + return result; } - return result; - } - /** - * Gets all descendant elements of the given parentElement with the given namespace and tagname and returns their text child as a list of String. - * - * @param parentElement parentElement - * @param namespace namespace - * @param tagName tagName - * @return List - */ - public static List getElementsTextChild(Element parentElement, - String namespace, String tagName) { - NodeList elements = parentElement - .getElementsByTagNameNS(namespace, tagName); - //ArrayList 初始化时指定长度提高性能 - List result = new ArrayList<>(elements.getLength()); - for (int i = 0; i < elements.getLength(); i++) { - result.add(getTextChildrenContent((Element) elements.item(i))); + /** + * Gets all descendant elements of the given parentElement with the given namespace and tagname and returns their text child as a list of String. + * + * @param parentElement parentElement + * @param namespace namespace + * @param tagName tagName + * @return List + */ + public static List getElementsTextChild(Element parentElement, + String namespace, String tagName) { + NodeList elements = parentElement + .getElementsByTagNameNS(namespace, tagName); + //ArrayList 初始化时指定长度提高性能 + List result = new ArrayList<>(elements.getLength()); + for (int i = 0; i < elements.getLength(); i++) { + result.add(getTextChildrenContent((Element) elements.item(i))); + } + return result; } - return result; - } - /** - * Finds in the current document the first element with the given namespace and elementName and with the given findAttributeName and findAttributeValue. - * It then returns the value of the given resultAttributeName. - * - * @param document document - * @param namespace namespace - * @param elementName elementName - * @param findAttributeName findAttributeName - * @param findAttributeValue findAttributeValue - * @param resultAttributeName resultAttributeName - * @return String value - */ - public static String getFindAttributeValue(Document document, - String namespace, String elementName, String findAttributeName, - String findAttributeValue, String resultAttributeName) { - NodeList metaTags = document.getElementsByTagNameNS(namespace, elementName); - for (int i = 0; i < metaTags.getLength(); i++) { - Element metaElement = (Element) metaTags.item(i); - if (findAttributeValue - .equalsIgnoreCase(metaElement.getAttribute(findAttributeName)) - && StringUtil - .isNotBlank(metaElement.getAttribute(resultAttributeName))) { - return metaElement.getAttribute(resultAttributeName); - } + /** + * Finds in the current document the first element with the given namespace and elementName and with the given findAttributeName and findAttributeValue. + * It then returns the value of the given resultAttributeName. + * + * @param document document + * @param namespace namespace + * @param elementName elementName + * @param findAttributeName findAttributeName + * @param findAttributeValue findAttributeValue + * @param resultAttributeName resultAttributeName + * @return String value + */ + public static String getFindAttributeValue(Document document, + String namespace, String elementName, String findAttributeName, + String findAttributeValue, String resultAttributeName) { + NodeList metaTags = document.getElementsByTagNameNS(namespace, elementName); + for (int i = 0; i < metaTags.getLength(); i++) { + Element metaElement = (Element) metaTags.item(i); + if (findAttributeValue + .equalsIgnoreCase(metaElement.getAttribute(findAttributeName)) + && StringUtil + .isNotBlank(metaElement.getAttribute(resultAttributeName))) { + return metaElement.getAttribute(resultAttributeName); + } + } + return null; } - return null; - } - /** - * Gets the first element that is a child of the parentElement and has the given namespace and tagName - * - * @param parentElement parentElement - * @param namespace namespace - * @param tagName tagName - * @return Element - */ - public static Element getFirstElementByTagNameNS(Element parentElement, - String namespace, String tagName) { - NodeList nodes = parentElement.getElementsByTagNameNS(namespace, tagName); - if (nodes.getLength() != 0) { - return (Element) nodes.item(0); + /** + * Gets the first element that is a child of the parentElement and has the given namespace and tagName + * + * @param parentElement parentElement + * @param namespace namespace + * @param tagName tagName + * @return Element + */ + public static NodeList getElementsByTagNameNS(Element parentElement, + String namespace, String tagName) { + NodeList nodes = parentElement.getElementsByTagNameNS(namespace, tagName); + if (nodes.getLength() != 0) { + return nodes; + } + nodes = parentElement.getElementsByTagName(tagName); + if (nodes.getLength() == 0) { + return null; + } + return nodes; } - nodes= parentElement.getElementsByTagName(tagName); - if (nodes.getLength()==0){ - return null; + /** + * Gets the first element that is a child of the parentElement and has the given namespace and tagName + * + * @param parentElement parentElement + * @param namespace namespace + * @param tagName tagName + * @return Element + */ + public static NodeList getElementsByTagNameNS(Document parentElement, + String namespace, String tagName) { + NodeList nodes = parentElement.getElementsByTagNameNS(namespace, tagName); + if (nodes.getLength() != 0) { + return nodes; + } + nodes = parentElement.getElementsByTagName(tagName); + if (nodes.getLength() == 0) { + return null; + } + return nodes; } - return (Element) nodes.item(0); - } - /** - * The contents of all Text nodes that are children of the given parentElement. - * The result is trim()-ed. - * - * The reason for this more complicated procedure instead of just returning the data of the firstChild is that - * when the text is Chinese characters then on Android each Characater is represented in the DOM as - * an individual Text node. - * - * @param parentElement parentElement - * @return String value - */ - public static String getTextChildrenContent(Element parentElement) { - if (parentElement == null) { - return null; + /** + * Gets the first element that is a child of the parentElement and has the given namespace and tagName + * + * @param parentElement parentElement + * @param namespace namespace + * @param tagName tagName + * @return Element + */ + public static Element getFirstElementByTagNameNS(Element parentElement, + String namespace, String tagName) { + NodeList nodes = parentElement.getElementsByTagNameNS(namespace, tagName); + if (nodes.getLength() != 0) { + return (Element) nodes.item(0); + } + nodes = parentElement.getElementsByTagName(tagName); + if (nodes.getLength() == 0) { + return null; + } + return (Element) nodes.item(0); } - StringBuilder result = new StringBuilder(); - NodeList childNodes = parentElement.getChildNodes(); - for (int i = 0; i < childNodes.getLength(); i++) { - Node node = childNodes.item(i); - if ((node == null) || - (node.getNodeType() != Node.TEXT_NODE)) { - continue; - } - result.append(((Text) node).getData()); + + /** + * The contents of all Text nodes that are children of the given parentElement. + * The result is trim()-ed. + *

+ * The reason for this more complicated procedure instead of just returning the data of the firstChild is that + * when the text is Chinese characters then on Android each Characater is represented in the DOM as + * an individual Text node. + * + * @param parentElement parentElement + * @return String value + */ + public static String getTextChildrenContent(Element parentElement) { + if (parentElement == null) { + return null; + } + StringBuilder result = new StringBuilder(); + NodeList childNodes = parentElement.getChildNodes(); + for (int i = 0; i < childNodes.getLength(); i++) { + Node node = childNodes.item(i); + if ((node == null) || + (node.getNodeType() != Node.TEXT_NODE)) { + continue; + } + result.append(((Text) node).getData()); + } + return result.toString().trim(); } - return result.toString().trim(); - } } diff --git a/epublib/src/main/java/me/ag2s/epublib/epub/NCXDocumentV2.java b/epublib/src/main/java/me/ag2s/epublib/epub/NCXDocumentV2.java index 572f7f97b..2bf47de51 100644 --- a/epublib/src/main/java/me/ag2s/epublib/epub/NCXDocumentV2.java +++ b/epublib/src/main/java/me/ag2s/epublib/epub/NCXDocumentV2.java @@ -148,9 +148,9 @@ public class NCXDocumentV2 { if (resource == null) { Log.e(TAG, "Resource with href " + href + " in NCX document not found"); } - Log.d(TAG, "label:" + label); - Log.d(TAG, "href:" + href); - Log.d(TAG, "fragmentId:" + fragmentId); + Log.v(TAG, "label:" + label); + Log.v(TAG, "href:" + href); + Log.v(TAG, "fragmentId:" + fragmentId); TOCReference result = new TOCReference(label, resource, fragmentId); List childTOCReferences = readTOCReferences( navpointElement.getChildNodes(), book); diff --git a/epublib/src/main/java/me/ag2s/epublib/epub/NCXDocumentV3.java b/epublib/src/main/java/me/ag2s/epublib/epub/NCXDocumentV3.java index a538a43f2..845b999d1 100644 --- a/epublib/src/main/java/me/ag2s/epublib/epub/NCXDocumentV3.java +++ b/epublib/src/main/java/me/ag2s/epublib/epub/NCXDocumentV3.java @@ -107,12 +107,21 @@ public class NCXDocumentV3 { if (ncxResource == null) { return null; } - //Log.d(TAG, ncxResource.getHref()); + //一些epub 3 文件没有按照epub3的标准使用删除掉ncx目录文件 + if (ncxResource.getHref().endsWith(".ncx")){ + Log.v(TAG,"该epub文件不标准，使用了epub2的目录文件"); + return NCXDocumentV2.read(book, epubReader); + } + Log.d(TAG, ncxResource.getHref()); Document ncxDocument = ResourceUtil.getAsDocument(ncxResource); - //Log.d(TAG, ncxDocument.getNodeName()); + Log.d(TAG, ncxDocument.getNodeName()); Element navMapElement = (Element) ncxDocument.getElementsByTagName(XHTMLTgs.nav).item(0); + if(navMapElement==null){ + Log.d(TAG,"epub3目录文件未发现nav节点，尝试使用epub2的规则解析"); + return NCXDocumentV2.read(book, epubReader); + } navMapElement = (Element) navMapElement.getElementsByTagName(XHTMLTgs.ol).item(0); Log.d(TAG, navMapElement.getTagName()); diff --git a/epublib/src/main/java/me/ag2s/epublib/epub/PackageDocumentMetadataReader.java b/epublib/src/main/java/me/ag2s/epublib/epub/PackageDocumentMetadataReader.java index c9ce14429..86f3f95c0 100644 --- a/epublib/src/main/java/me/ag2s/epublib/epub/PackageDocumentMetadataReader.java +++ b/epublib/src/main/java/me/ag2s/epublib/epub/PackageDocumentMetadataReader.java @@ -22,7 +22,7 @@ import me.ag2s.epublib.util.StringUtil; /** * Reads the package document metadata. - * + *

* In its own separate class because the PackageDocumentReader became a bit large and unwieldy. * * @author paul @@ -30,195 +30,196 @@ import me.ag2s.epublib.util.StringUtil; // package class PackageDocumentMetadataReader extends PackageDocumentBase { - private static final String TAG= PackageDocumentMetadataReader.class.getName(); - - public static Metadata readMetadata(Document packageDocument) { - Metadata result = new Metadata(); - Element metadataElement = DOMUtil - .getFirstElementByTagNameNS(packageDocument.getDocumentElement(), - NAMESPACE_OPF, OPFTags.metadata); - if (metadataElement == null) { - Log.e(TAG,"Package does not contain element " + OPFTags.metadata); - return result; - } - result.setTitles(DOMUtil - .getElementsTextChild(metadataElement, NAMESPACE_DUBLIN_CORE, - DCTags.title)); - result.setPublishers(DOMUtil - .getElementsTextChild(metadataElement, NAMESPACE_DUBLIN_CORE, - DCTags.publisher)); - result.setDescriptions(DOMUtil - .getElementsTextChild(metadataElement, NAMESPACE_DUBLIN_CORE, - DCTags.description)); - result.setRights(DOMUtil - .getElementsTextChild(metadataElement, NAMESPACE_DUBLIN_CORE, - DCTags.rights)); - result.setTypes(DOMUtil - .getElementsTextChild(metadataElement, NAMESPACE_DUBLIN_CORE, - DCTags.type)); - result.setSubjects(DOMUtil - .getElementsTextChild(metadataElement, NAMESPACE_DUBLIN_CORE, - DCTags.subject)); - result.setIdentifiers(readIdentifiers(metadataElement)); - result.setAuthors(readCreators(metadataElement)); - result.setContributors(readContributors(metadataElement)); - result.setDates(readDates(metadataElement)); - result.setOtherProperties(readOtherProperties(metadataElement)); - result.setMetaAttributes(readMetaProperties(metadataElement)); - Element languageTag = DOMUtil - .getFirstElementByTagNameNS(metadataElement, NAMESPACE_DUBLIN_CORE, - DCTags.language); - if (languageTag != null) { - result.setLanguage(DOMUtil.getTextChildrenContent(languageTag)); + private static final String TAG = PackageDocumentMetadataReader.class.getName(); + + public static Metadata readMetadata(Document packageDocument) { + Metadata result = new Metadata(); + Element metadataElement = DOMUtil + .getFirstElementByTagNameNS(packageDocument.getDocumentElement(), + NAMESPACE_OPF, OPFTags.metadata); + if (metadataElement == null) { + Log.e(TAG, "Package does not contain element " + OPFTags.metadata); + return result; + } + result.setTitles(DOMUtil + .getElementsTextChild(metadataElement, NAMESPACE_DUBLIN_CORE, + DCTags.title)); + result.setPublishers(DOMUtil + .getElementsTextChild(metadataElement, NAMESPACE_DUBLIN_CORE, + DCTags.publisher)); + result.setDescriptions(DOMUtil + .getElementsTextChild(metadataElement, NAMESPACE_DUBLIN_CORE, + DCTags.description)); + result.setRights(DOMUtil + .getElementsTextChild(metadataElement, NAMESPACE_DUBLIN_CORE, + DCTags.rights)); + result.setTypes(DOMUtil + .getElementsTextChild(metadataElement, NAMESPACE_DUBLIN_CORE, + DCTags.type)); + result.setSubjects(DOMUtil + .getElementsTextChild(metadataElement, NAMESPACE_DUBLIN_CORE, + DCTags.subject)); + result.setIdentifiers(readIdentifiers(metadataElement)); + result.setAuthors(readCreators(metadataElement)); + result.setContributors(readContributors(metadataElement)); + result.setDates(readDates(metadataElement)); + result.setOtherProperties(readOtherProperties(metadataElement)); + result.setMetaAttributes(readMetaProperties(metadataElement)); + Element languageTag = DOMUtil + .getFirstElementByTagNameNS(metadataElement, NAMESPACE_DUBLIN_CORE, + DCTags.language); + if (languageTag != null) { + result.setLanguage(DOMUtil.getTextChildrenContent(languageTag)); + } + + return result; } - return result; - } - - /** - * consumes meta tags that have a property attribute as defined in the standard. For example: - * <meta property="rendition:layout">pre-paginated</meta> - * @param metadataElement metadataElement - * @return Map - */ - private static Map readOtherProperties( - Element metadataElement) { - Map result = new HashMap<>(); - - NodeList metaTags = metadataElement.getElementsByTagName(OPFTags.meta); - for (int i = 0; i < metaTags.getLength(); i++) { - Node metaNode = metaTags.item(i); - Node property = metaNode.getAttributes() - .getNamedItem(OPFAttributes.property); - if (property != null) { - String name = property.getNodeValue(); - String value = metaNode.getTextContent(); - result.put(new QName(name), value); - } + /** + * consumes meta tags that have a property attribute as defined in the standard. For example: + * <meta property="rendition:layout">pre-paginated</meta> + * + * @param metadataElement metadataElement + * @return Map + */ + private static Map readOtherProperties( + Element metadataElement) { + Map result = new HashMap<>(); + + NodeList metaTags = metadataElement.getElementsByTagName(OPFTags.meta); + for (int i = 0; i < metaTags.getLength(); i++) { + Node metaNode = metaTags.item(i); + Node property = metaNode.getAttributes() + .getNamedItem(OPFAttributes.property); + if (property != null) { + String name = property.getNodeValue(); + String value = metaNode.getTextContent(); + result.put(new QName(name), value); + } + } + + return result; } - return result; - } - - /** - * consumes meta tags that have a property attribute as defined in the standard. For example: - * <meta property="rendition:layout">pre-paginated</meta> - * @param metadataElement metadataElement - * @return Map - */ - private static Map readMetaProperties( - Element metadataElement) { - Map result = new HashMap<>(); - - NodeList metaTags = metadataElement.getElementsByTagName(OPFTags.meta); - for (int i = 0; i < metaTags.getLength(); i++) { - Element metaElement = (Element) metaTags.item(i); - String name = metaElement.getAttribute(OPFAttributes.name); - String value = metaElement.getAttribute(OPFAttributes.content); - result.put(name, value); + /** + * consumes meta tags that have a property attribute as defined in the standard. For example: + * <meta property="rendition:layout">pre-paginated</meta> + * + * @param metadataElement metadataElement + * @return Map + */ + private static Map readMetaProperties( + Element metadataElement) { + Map result = new HashMap<>(); + + NodeList metaTags = metadataElement.getElementsByTagName(OPFTags.meta); + for (int i = 0; i < metaTags.getLength(); i++) { + Element metaElement = (Element) metaTags.item(i); + String name = metaElement.getAttribute(OPFAttributes.name); + String value = metaElement.getAttribute(OPFAttributes.content); + result.put(name, value); + } + + return result; } - return result; - } + private static String getBookIdId(Document document) { + Element packageElement = DOMUtil + .getFirstElementByTagNameNS(document.getDocumentElement(), + NAMESPACE_OPF, OPFTags.packageTag); + if (packageElement == null) { + return null; + } + return DOMUtil.getAttribute(packageElement, NAMESPACE_OPF, OPFAttributes.uniqueIdentifier); - private static String getBookIdId(Document document) { - Element packageElement = DOMUtil - .getFirstElementByTagNameNS(document.getDocumentElement(), - NAMESPACE_OPF, OPFTags.packageTag); - if (packageElement == null) { - return null; } - return packageElement - .getAttributeNS(NAMESPACE_OPF, OPFAttributes.uniqueIdentifier); - } - - private static List readCreators(Element metadataElement) { - return readAuthors(DCTags.creator, metadataElement); - } - - private static List readContributors(Element metadataElement) { - return readAuthors(DCTags.contributor, metadataElement); - } - - private static List readAuthors(String authorTag, - Element metadataElement) { - NodeList elements = metadataElement - .getElementsByTagNameNS(NAMESPACE_DUBLIN_CORE, authorTag); - List result = new ArrayList<>(elements.getLength()); - for (int i = 0; i < elements.getLength(); i++) { - Element authorElement = (Element) elements.item(i); - Author author = createAuthor(authorElement); - if (author != null) { - result.add(author); - } + + private static List readCreators(Element metadataElement) { + return readAuthors(DCTags.creator, metadataElement); } - return result; - - } - - private static List readDates(Element metadataElement) { - NodeList elements = metadataElement - .getElementsByTagNameNS(NAMESPACE_DUBLIN_CORE, DCTags.date); - List result = new ArrayList<>(elements.getLength()); - for (int i = 0; i < elements.getLength(); i++) { - Element dateElement = (Element) elements.item(i); - Date date; - try { - date = new Date(DOMUtil.getTextChildrenContent(dateElement), - dateElement.getAttributeNS(NAMESPACE_OPF, OPFAttributes.event)); - result.add(date); - } catch (IllegalArgumentException e) { - Log.e(TAG,e.getMessage()); - } + + private static List readContributors(Element metadataElement) { + return readAuthors(DCTags.contributor, metadataElement); } - return result; - } + private static List readAuthors(String authorTag, + Element metadataElement) { + NodeList elements = metadataElement + .getElementsByTagNameNS(NAMESPACE_DUBLIN_CORE, authorTag); + List result = new ArrayList<>(elements.getLength()); + for (int i = 0; i < elements.getLength(); i++) { + Element authorElement = (Element) elements.item(i); + Author author = createAuthor(authorElement); + if (author != null) { + result.add(author); + } + } + return result; - private static Author createAuthor(Element authorElement) { - String authorString = DOMUtil.getTextChildrenContent(authorElement); - if (StringUtil.isBlank(authorString)) { - return null; } - int spacePos = authorString.lastIndexOf(' '); - Author result; - if (spacePos < 0) { - result = new Author(authorString); - } else { - result = new Author(authorString.substring(0, spacePos), - authorString.substring(spacePos + 1)); + + private static List readDates(Element metadataElement) { + NodeList elements = metadataElement + .getElementsByTagNameNS(NAMESPACE_DUBLIN_CORE, DCTags.date); + List result = new ArrayList<>(elements.getLength()); + for (int i = 0; i < elements.getLength(); i++) { + Element dateElement = (Element) elements.item(i); + Date date; + try { + date = new Date(DOMUtil.getTextChildrenContent(dateElement), + DOMUtil.getAttribute(dateElement, NAMESPACE_OPF, OPFAttributes.event)); + result.add(date); + } catch (IllegalArgumentException e) { + Log.e(TAG, e.getMessage()); + } + } + return result; + } - result.setRole( - authorElement.getAttributeNS(NAMESPACE_OPF, OPFAttributes.role)); - return result; - } - - - private static List readIdentifiers(Element metadataElement) { - NodeList identifierElements = metadataElement - .getElementsByTagNameNS(NAMESPACE_DUBLIN_CORE, DCTags.identifier); - if (identifierElements.getLength() == 0) { - Log.e(TAG,"Package does not contain element " + DCTags.identifier); - return new ArrayList<>(); + + private static Author createAuthor(Element authorElement) { + String authorString = DOMUtil.getTextChildrenContent(authorElement); + if (StringUtil.isBlank(authorString)) { + return null; + } + int spacePos = authorString.lastIndexOf(' '); + Author result; + if (spacePos < 0) { + result = new Author(authorString); + } else { + result = new Author(authorString.substring(0, spacePos), + authorString.substring(spacePos + 1)); + } + result.setRole( + DOMUtil.getAttribute(authorElement, NAMESPACE_OPF, OPFAttributes.role)); + return result; } - String bookIdId = getBookIdId(metadataElement.getOwnerDocument()); - List result = new ArrayList<>( - identifierElements.getLength()); - for (int i = 0; i < identifierElements.getLength(); i++) { - Element identifierElement = (Element) identifierElements.item(i); - String schemeName = identifierElement - .getAttributeNS(NAMESPACE_OPF, DCAttributes.scheme); - String identifierValue = DOMUtil - .getTextChildrenContent(identifierElement); - if (StringUtil.isBlank(identifierValue)) { - continue; - } - Identifier identifier = new Identifier(schemeName, identifierValue); - if (identifierElement.getAttribute("id").equals(bookIdId)) { - identifier.setBookId(true); - } - result.add(identifier); + + + private static List readIdentifiers(Element metadataElement) { + NodeList identifierElements = metadataElement + .getElementsByTagNameNS(NAMESPACE_DUBLIN_CORE, DCTags.identifier); + if (identifierElements.getLength() == 0) { + Log.e(TAG, "Package does not contain element " + DCTags.identifier); + return new ArrayList<>(); + } + String bookIdId = getBookIdId(metadataElement.getOwnerDocument()); + List result = new ArrayList<>( + identifierElements.getLength()); + for (int i = 0; i < identifierElements.getLength(); i++) { + Element identifierElement = (Element) identifierElements.item(i); + String schemeName = DOMUtil.getAttribute(identifierElement, NAMESPACE_OPF, DCAttributes.scheme); + String identifierValue = DOMUtil + .getTextChildrenContent(identifierElement); + if (StringUtil.isBlank(identifierValue)) { + continue; + } + Identifier identifier = new Identifier(schemeName, identifierValue); + if (identifierElement.getAttribute("id").equals(bookIdId)) { + identifier.setBookId(true); + } + result.add(identifier); + } + return result; } - return result; - } } diff --git a/epublib/src/main/java/me/ag2s/epublib/epub/PackageDocumentReader.java b/epublib/src/main/java/me/ag2s/epublib/epub/PackageDocumentReader.java index 0b0c83848..eaf0c4687 100644 --- a/epublib/src/main/java/me/ag2s/epublib/epub/PackageDocumentReader.java +++ b/epublib/src/main/java/me/ag2s/epublib/epub/PackageDocumentReader.java @@ -36,42 +36,41 @@ import me.ag2s.epublib.util.StringUtil; * Reads the opf package document as defined by namespace http://www.idpf.org/2007/opf * * @author paul - * */ public class PackageDocumentReader extends PackageDocumentBase { - private static final String TAG= PackageDocumentReader.class.getName(); - private static final String[] POSSIBLE_NCX_ITEM_IDS = new String[]{"toc", - "ncx", "ncxtoc","htmltoc"}; - - - public static void read( - Resource packageResource, EpubReader epubReader, EpubBook book, - Resources resources) - throws SAXException, IOException { - Document packageDocument = ResourceUtil.getAsDocument(packageResource); - String packageHref = packageResource.getHref(); - resources = fixHrefs(packageHref, resources); - readGuide(packageDocument, epubReader, book, resources); - - // Books sometimes use non-identifier ids. We map these here to legal ones - Map idMapping = new HashMap<>(); - String version=DOMUtil.getAttribute(packageDocument.getDocumentElement(),PREFIX_OPF,PackageDocumentBase.version); - - resources = readManifest(packageDocument, packageHref, epubReader, - resources, idMapping); - book.setResources(resources); - book.setVersion(version); - readCover(packageDocument, book); - book.setMetadata( - PackageDocumentMetadataReader.readMetadata(packageDocument)); - book.setSpine(readSpine(packageDocument, book.getResources(), idMapping)); - - // if we did not find a cover page then we make the first page of the book the cover page - if (book.getCoverPage() == null && book.getSpine().size() > 0) { - book.setCoverPage(book.getSpine().getResource(0)); + private static final String TAG = PackageDocumentReader.class.getName(); + private static final String[] POSSIBLE_NCX_ITEM_IDS = new String[]{"toc", + "ncx", "ncxtoc", "htmltoc"}; + + + public static void read( + Resource packageResource, EpubReader epubReader, EpubBook book, + Resources resources) + throws SAXException, IOException { + Document packageDocument = ResourceUtil.getAsDocument(packageResource); + String packageHref = packageResource.getHref(); + resources = fixHrefs(packageHref, resources); + readGuide(packageDocument, epubReader, book, resources); + + // Books sometimes use non-identifier ids. We map these here to legal ones + Map idMapping = new HashMap<>(); + String version = DOMUtil.getAttribute(packageDocument.getDocumentElement(), PREFIX_OPF, PackageDocumentBase.version); + + resources = readManifest(packageDocument, packageHref, epubReader, + resources, idMapping); + book.setResources(resources); + book.setVersion(version); + readCover(packageDocument, book); + book.setMetadata( + PackageDocumentMetadataReader.readMetadata(packageDocument)); + book.setSpine(readSpine(packageDocument, book.getResources(), idMapping)); + + // if we did not find a cover page then we make the first page of the book the cover page + if (book.getCoverPage() == null && book.getSpine().size() > 0) { + book.setCoverPage(book.getSpine().getResource(0)); + } } - } // private static Resource readCoverImage(Element metadataElement, Resources resources) { // String coverResourceId = DOMUtil.getFindAttributeValue(metadataElement.getOwnerDocument(), NAMESPACE_OPF, OPFTags.meta, OPFAttributes.name, OPFValues.meta_cover, OPFAttributes.content); @@ -83,322 +82,326 @@ public class PackageDocumentReader extends PackageDocumentBase { // } - /** - * Reads the manifest containing the resource ids, hrefs and mediatypes. - * - * @param packageDocument e - * @param packageHref e - * @param epubReader e - * @param resources e - * @param idMapping e - * @return a Map with resources, with their id's as key. - */ - @SuppressWarnings("unused") - private static Resources readManifest(Document packageDocument, - String packageHref, - EpubReader epubReader, Resources resources, - Map idMapping) { - Element manifestElement = DOMUtil - .getFirstElementByTagNameNS(packageDocument.getDocumentElement(), - NAMESPACE_OPF, OPFTags.manifest); - Resources result = new Resources(); - if (manifestElement == null) { - Log.e(TAG, - "Package document does not contain element " + OPFTags.manifest); - return result; - } - NodeList itemElements = manifestElement - .getElementsByTagNameNS(NAMESPACE_OPF, OPFTags.item); - for (int i = 0; i < itemElements.getLength(); i++) { - Element itemElement = (Element) itemElements.item(i); - String id = DOMUtil - .getAttribute(itemElement, NAMESPACE_OPF, OPFAttributes.id); - String href = DOMUtil - .getAttribute(itemElement, NAMESPACE_OPF, OPFAttributes.href); - - try { - href = URLDecoder.decode(href, Constants.CHARACTER_ENCODING); - } catch (UnsupportedEncodingException e) { - Log.e(TAG,e.getMessage()); - } - String mediaTypeName = DOMUtil - .getAttribute(itemElement, NAMESPACE_OPF, OPFAttributes.media_type); - Resource resource = resources.remove(href); - if (resource == null) { - Log.e(TAG,"resource with href '" + href + "' not found"); - continue; - } - resource.setId(id); - //for epub3 - String properties=DOMUtil.getAttribute(itemElement,NAMESPACE_OPF,OPFAttributes.properties); - resource.setProperties(properties); - - MediaType mediaType = MediaTypes.getMediaTypeByName(mediaTypeName); - if (mediaType != null) { - resource.setMediaType(mediaType); - } - result.add(resource); - idMapping.put(id, resource.getId()); - } - return result; - } - - - /** - * Reads the book's guide. - * Here some more attempts are made at finding the cover page. - * - * @param packageDocument r - * @param epubReader r - * @param book r - * @param resources g - */ - @SuppressWarnings("unused") - private static void readGuide(Document packageDocument, - EpubReader epubReader, EpubBook book, Resources resources) { - Element guideElement = DOMUtil - .getFirstElementByTagNameNS(packageDocument.getDocumentElement(), - NAMESPACE_OPF, OPFTags.guide); - if (guideElement == null) { - return; - } - Guide guide = book.getGuide(); - NodeList guideReferences = guideElement - .getElementsByTagNameNS(NAMESPACE_OPF, OPFTags.reference); - for (int i = 0; i < guideReferences.getLength(); i++) { - Element referenceElement = (Element) guideReferences.item(i); - String resourceHref = DOMUtil - .getAttribute(referenceElement, NAMESPACE_OPF, OPFAttributes.href); - if (StringUtil.isBlank(resourceHref)) { - continue; - } - Resource resource = resources.getByHref(StringUtil - .substringBefore(resourceHref, Constants.FRAGMENT_SEPARATOR_CHAR)); - if (resource == null) { - Log.e(TAG,"Guide is referencing resource with href " + resourceHref - + " which could not be found"); - continue; - } - String type = DOMUtil - .getAttribute(referenceElement, NAMESPACE_OPF, OPFAttributes.type); - if (StringUtil.isBlank(type)) { - Log.e(TAG,"Guide is referencing resource with href " + resourceHref - + " which is missing the 'type' attribute"); - continue; - } - String title = DOMUtil - .getAttribute(referenceElement, NAMESPACE_OPF, OPFAttributes.title); - if (GuideReference.COVER.equalsIgnoreCase(type)) { - continue; // cover is handled elsewhere - } - GuideReference reference = new GuideReference(resource, type, title, - StringUtil - .substringAfter(resourceHref, Constants.FRAGMENT_SEPARATOR_CHAR)); - guide.addReference(reference); - } - } - - - /** - * Strips off the package prefixes up to the href of the packageHref. - * - * Example: - * If the packageHref is "OEBPS/content.opf" then a resource href like "OEBPS/foo/bar.html" will be turned into "foo/bar.html" - * - * @param packageHref f - * @param resourcesByHref g - * @return The stripped package href - */ - static Resources fixHrefs(String packageHref, - Resources resourcesByHref) { - int lastSlashPos = packageHref.lastIndexOf('/'); - if (lastSlashPos < 0) { - return resourcesByHref; - } - Resources result = new Resources(); - for (Resource resource : resourcesByHref.getAll()) { - if (StringUtil.isNotBlank(resource.getHref()) - && resource.getHref().length() > lastSlashPos) { - resource.setHref(resource.getHref().substring(lastSlashPos + 1)); - } - result.add(resource); - } - return result; - } - - /** - * Reads the document's spine, containing all sections in reading order. - * - * @param packageDocument b - * @param resources b - * @param idMapping b - * @return the document's spine, containing all sections in reading order. - */ - private static Spine readSpine(Document packageDocument, Resources resources, - Map idMapping) { - - Element spineElement = DOMUtil - .getFirstElementByTagNameNS(packageDocument.getDocumentElement(), - NAMESPACE_OPF, OPFTags.spine); - if (spineElement == null) { - Log.e(TAG,"Element " + OPFTags.spine - + " not found in package document, generating one automatically"); - return generateSpineFromResources(resources); + /** + * Reads the manifest containing the resource ids, hrefs and mediatypes. + * + * @param packageDocument e + * @param packageHref e + * @param epubReader e + * @param resources e + * @param idMapping e + * @return a Map with resources, with their id's as key. + */ + @SuppressWarnings("unused") + private static Resources readManifest(Document packageDocument, + String packageHref, + EpubReader epubReader, Resources resources, + Map idMapping) { + Element manifestElement = DOMUtil + .getFirstElementByTagNameNS(packageDocument.getDocumentElement(), + NAMESPACE_OPF, OPFTags.manifest); + Resources result = new Resources(); + if (manifestElement == null) { + Log.e(TAG, + "Package document does not contain element " + OPFTags.manifest); + return result; + } + NodeList itemElements = manifestElement + .getElementsByTagNameNS(NAMESPACE_OPF, OPFTags.item); + for (int i = 0; i < itemElements.getLength(); i++) { + Element itemElement = (Element) itemElements.item(i); + String id = DOMUtil + .getAttribute(itemElement, NAMESPACE_OPF, OPFAttributes.id); + String href = DOMUtil + .getAttribute(itemElement, NAMESPACE_OPF, OPFAttributes.href); + + try { + href = URLDecoder.decode(href, Constants.CHARACTER_ENCODING); + } catch (UnsupportedEncodingException e) { + Log.e(TAG, e.getMessage()); + } + String mediaTypeName = DOMUtil + .getAttribute(itemElement, NAMESPACE_OPF, OPFAttributes.media_type); + Resource resource = resources.remove(href); + if (resource == null) { + Log.e(TAG, "resource with href '" + href + "' not found"); + continue; + } + resource.setId(id); + //for epub3 + String properties = DOMUtil.getAttribute(itemElement, NAMESPACE_OPF, OPFAttributes.properties); + resource.setProperties(properties); + + MediaType mediaType = MediaTypes.getMediaTypeByName(mediaTypeName); + if (mediaType != null) { + resource.setMediaType(mediaType); + } + result.add(resource); + idMapping.put(id, resource.getId()); + } + return result; } - Spine result = new Spine(); - String tocResourceId = DOMUtil - .getAttribute(spineElement, NAMESPACE_OPF, OPFAttributes.toc); - result - .setTocResource(findTableOfContentsResource(tocResourceId, resources)); - NodeList spineNodes = packageDocument - .getElementsByTagNameNS(NAMESPACE_OPF, OPFTags.itemref); - List spineReferences = new ArrayList<>( - spineNodes.getLength()); - for (int i = 0; i < spineNodes.getLength(); i++) { - Element spineItem = (Element) spineNodes.item(i); - String itemref = DOMUtil - .getAttribute(spineItem, NAMESPACE_OPF, OPFAttributes.idref); - if (StringUtil.isBlank(itemref)) { - Log.e(TAG,"itemref with missing or empty idref"); // XXX - continue; - } - String id = idMapping.get(itemref); - if (id == null) { - id = itemref; - } - Resource resource = resources.getByIdOrHref(id); - if (resource == null) { - Log.e(TAG, "resource with id '" + id + "' not found"); - continue; - } - - SpineReference spineReference = new SpineReference(resource); - if (OPFValues.no.equalsIgnoreCase(DOMUtil - .getAttribute(spineItem, NAMESPACE_OPF, OPFAttributes.linear))) { - spineReference.setLinear(false); - } - spineReferences.add(spineReference); + + + /** + * Reads the book's guide. + * Here some more attempts are made at finding the cover page. + * + * @param packageDocument r + * @param epubReader r + * @param book r + * @param resources g + */ + @SuppressWarnings("unused") + private static void readGuide(Document packageDocument, + EpubReader epubReader, EpubBook book, Resources resources) { + Element guideElement = DOMUtil + .getFirstElementByTagNameNS(packageDocument.getDocumentElement(), + NAMESPACE_OPF, OPFTags.guide); + if (guideElement == null) { + return; + } + Guide guide = book.getGuide(); + NodeList guideReferences = guideElement + .getElementsByTagNameNS(NAMESPACE_OPF, OPFTags.reference); + for (int i = 0; i < guideReferences.getLength(); i++) { + Element referenceElement = (Element) guideReferences.item(i); + String resourceHref = DOMUtil + .getAttribute(referenceElement, NAMESPACE_OPF, OPFAttributes.href); + if (StringUtil.isBlank(resourceHref)) { + continue; + } + Resource resource = resources.getByHref(StringUtil + .substringBefore(resourceHref, Constants.FRAGMENT_SEPARATOR_CHAR)); + if (resource == null) { + Log.e(TAG, "Guide is referencing resource with href " + resourceHref + + " which could not be found"); + continue; + } + String type = DOMUtil + .getAttribute(referenceElement, NAMESPACE_OPF, OPFAttributes.type); + if (StringUtil.isBlank(type)) { + Log.e(TAG, "Guide is referencing resource with href " + resourceHref + + " which is missing the 'type' attribute"); + continue; + } + String title = DOMUtil + .getAttribute(referenceElement, NAMESPACE_OPF, OPFAttributes.title); + if (GuideReference.COVER.equalsIgnoreCase(type)) { + continue; // cover is handled elsewhere + } + GuideReference reference = new GuideReference(resource, type, title, + StringUtil + .substringAfter(resourceHref, Constants.FRAGMENT_SEPARATOR_CHAR)); + guide.addReference(reference); + } } - result.setSpineReferences(spineReferences); - return result; - } - - /** - * Creates a spine out of all resources in the resources. - * The generated spine consists of all XHTML pages in order of their href. - * - * @param resources f - * @return a spine created out of all resources in the resources. - */ - private static Spine generateSpineFromResources(Resources resources) { - Spine result = new Spine(); - List resourceHrefs = new ArrayList<>(resources.getAllHrefs()); - Collections.sort(resourceHrefs, String.CASE_INSENSITIVE_ORDER); - for (String resourceHref : resourceHrefs) { - Resource resource = resources.getByHref(resourceHref); - if (resource.getMediaType() == MediaTypes.NCX) { - result.setTocResource(resource); - } else if (resource.getMediaType() == MediaTypes.XHTML) { - result.addSpineReference(new SpineReference(resource)); - } + + + /** + * Strips off the package prefixes up to the href of the packageHref. + *

+ * Example: + * If the packageHref is "OEBPS/content.opf" then a resource href like "OEBPS/foo/bar.html" will be turned into "foo/bar.html" + * + * @param packageHref f + * @param resourcesByHref g + * @return The stripped package href + */ + static Resources fixHrefs(String packageHref, + Resources resourcesByHref) { + int lastSlashPos = packageHref.lastIndexOf('/'); + if (lastSlashPos < 0) { + return resourcesByHref; + } + Resources result = new Resources(); + for (Resource resource : resourcesByHref.getAll()) { + if (StringUtil.isNotBlank(resource.getHref()) + && resource.getHref().length() > lastSlashPos) { + resource.setHref(resource.getHref().substring(lastSlashPos + 1)); + } + result.add(resource); + } + return result; } - return result; - } - - - /** - * The spine tag should contain a 'toc' attribute with as value the resource id of the table of contents resource. - * - * Here we try several ways of finding this table of contents resource. - * We try the given attribute value, some often-used ones and finally look through all resources for the first resource with the table of contents mimetype. - * - * @param tocResourceId g - * @param resources g - * @return the Resource containing the table of contents - */ - static Resource findTableOfContentsResource(String tocResourceId, - Resources resources) { - Resource tocResource = null; - if (StringUtil.isNotBlank(tocResourceId)) { - tocResource = resources.getByIdOrHref(tocResourceId); + + /** + * Reads the document's spine, containing all sections in reading order. + * + * @param packageDocument b + * @param resources b + * @param idMapping b + * @return the document's spine, containing all sections in reading order. + */ + private static Spine readSpine(Document packageDocument, Resources resources, + Map idMapping) { + + Element spineElement = DOMUtil + .getFirstElementByTagNameNS(packageDocument.getDocumentElement(), + NAMESPACE_OPF, OPFTags.spine); + if (spineElement == null) { + Log.e(TAG, "Element " + OPFTags.spine + + " not found in package document, generating one automatically"); + return generateSpineFromResources(resources); + } + Spine result = new Spine(); + String tocResourceId = DOMUtil.getAttribute(spineElement, NAMESPACE_OPF, OPFAttributes.toc); + Log.v(TAG,tocResourceId); + result.setTocResource(findTableOfContentsResource(tocResourceId, resources)); + NodeList spineNodes = DOMUtil.getElementsByTagNameNS(packageDocument, NAMESPACE_OPF, OPFTags.itemref); + if(spineNodes==null){ + Log.e(TAG,"spineNodes is null"); + return result; + } + List spineReferences = new ArrayList<>(spineNodes.getLength()); + for (int i = 0; i < spineNodes.getLength(); i++) { + Element spineItem = (Element) spineNodes.item(i); + String itemref = DOMUtil.getAttribute(spineItem, NAMESPACE_OPF, OPFAttributes.idref); + if (StringUtil.isBlank(itemref)) { + Log.e(TAG, "itemref with missing or empty idref"); // XXX + continue; + } + String id = idMapping.get(itemref); + if (id == null) { + id = itemref; + } + + Resource resource = resources.getByIdOrHref(id); + if (resource == null) { + Log.e(TAG, "resource with id '" + id + "' not found"); + continue; + } + + SpineReference spineReference = new SpineReference(resource); + if (OPFValues.no.equalsIgnoreCase(DOMUtil + .getAttribute(spineItem, NAMESPACE_OPF, OPFAttributes.linear))) { + spineReference.setLinear(false); + } + spineReferences.add(spineReference); + } + result.setSpineReferences(spineReferences); + return result; } - if (tocResource != null) { - return tocResource; + /** + * Creates a spine out of all resources in the resources. + * The generated spine consists of all XHTML pages in order of their href. + * + * @param resources f + * @return a spine created out of all resources in the resources. + */ + private static Spine generateSpineFromResources(Resources resources) { + Spine result = new Spine(); + List resourceHrefs = new ArrayList<>(resources.getAllHrefs()); + Collections.sort(resourceHrefs, String.CASE_INSENSITIVE_ORDER); + for (String resourceHref : resourceHrefs) { + Resource resource = resources.getByHref(resourceHref); + if (resource.getMediaType() == MediaTypes.NCX) { + result.setTocResource(resource); + } else if (resource.getMediaType() == MediaTypes.XHTML) { + result.addSpineReference(new SpineReference(resource)); + } + } + return result; } - // get the first resource with the NCX mediatype - tocResource = resources.findFirstResourceByMediaType(MediaTypes.NCX); - if (tocResource == null) { - for (String possibleNcxItemId : POSSIBLE_NCX_ITEM_IDS) { - tocResource = resources.getByIdOrHref(possibleNcxItemId); + /** + * The spine tag should contain a 'toc' attribute with as value the resource id of the table of contents resource. + *

+ * Here we try several ways of finding this table of contents resource. + * We try the given attribute value, some often-used ones and finally look through all resources for the first resource with the table of contents mimetype. + * + * @param tocResourceId g + * @param resources g + * @return the Resource containing the table of contents + */ + static Resource findTableOfContentsResource(String tocResourceId, + Resources resources) { + Resource tocResource; + //一些epub3的文件为了兼容epub2,保留的epub2的目录文件，这里优先选择epub3的xml目录 + tocResource = resources.getByProperties("nav"); if (tocResource != null) { - break; + return tocResource; } - tocResource = resources - .getByIdOrHref(possibleNcxItemId.toUpperCase()); + + if (StringUtil.isNotBlank(tocResourceId)) { + tocResource = resources.getByIdOrHref(tocResourceId); + } + if (tocResource != null) { - break; + return tocResource; } - } - } - //For EPUB3 - if (tocResource==null){ - tocResource=resources.getByProperties("nav"); - } - if (tocResource == null) { - Log.e(TAG, - "Could not find table of contents resource. Tried resource with id '" - + tocResourceId + "', " + Constants.DEFAULT_TOC_ID + ", " - + Constants.DEFAULT_TOC_ID.toUpperCase() - + " and any NCX resource."); - } - return tocResource; - } - - - /** - * Find all resources that have something to do with the coverpage and the cover image. - * Search the meta tags and the guide references - * - * @param packageDocument s - * @return all resources that have something to do with the coverpage and the cover image. - */ - // package - static Set findCoverHrefs(Document packageDocument) { - - Set result = new HashSet<>(); - - // try and find a meta tag with name = 'cover' and a non-blank id - String coverResourceId = DOMUtil - .getFindAttributeValue(packageDocument, NAMESPACE_OPF, - OPFTags.meta, OPFAttributes.name, OPFValues.meta_cover, - OPFAttributes.content); - - if (StringUtil.isNotBlank(coverResourceId)) { - String coverHref = DOMUtil - .getFindAttributeValue(packageDocument, NAMESPACE_OPF, - OPFTags.item, OPFAttributes.id, coverResourceId, - OPFAttributes.href); - if (StringUtil.isNotBlank(coverHref)) { - result.add(coverHref); - } else { - result.add( - coverResourceId); // maybe there was a cover href put in the cover id attribute - } + // get the first resource with the NCX mediatype + tocResource = resources.findFirstResourceByMediaType(MediaTypes.NCX); + + if (tocResource == null) { + for (String possibleNcxItemId : POSSIBLE_NCX_ITEM_IDS) { + tocResource = resources.getByIdOrHref(possibleNcxItemId); + if (tocResource != null) { + break; + } + tocResource = resources + .getByIdOrHref(possibleNcxItemId.toUpperCase()); + if (tocResource != null) { + break; + } + } + } + + + if (tocResource == null) { + Log.e(TAG, + "Could not find table of contents resource. Tried resource with id '" + + tocResourceId + "', " + Constants.DEFAULT_TOC_ID + ", " + + Constants.DEFAULT_TOC_ID.toUpperCase() + + " and any NCX resource."); + } + return tocResource; } - // try and find a reference tag with type is 'cover' and reference is not blank - String coverHref = DOMUtil - .getFindAttributeValue(packageDocument, NAMESPACE_OPF, - OPFTags.reference, OPFAttributes.type, OPFValues.reference_cover, - OPFAttributes.href); - if (StringUtil.isNotBlank(coverHref)) { - result.add(coverHref); + + + /** + * Find all resources that have something to do with the coverpage and the cover image. + * Search the meta tags and the guide references + * + * @param packageDocument s + * @return all resources that have something to do with the coverpage and the cover image. + */ + // package + static Set findCoverHrefs(Document packageDocument) { + + Set result = new HashSet<>(); + + // try and find a meta tag with name = 'cover' and a non-blank id + String coverResourceId = DOMUtil + .getFindAttributeValue(packageDocument, NAMESPACE_OPF, + OPFTags.meta, OPFAttributes.name, OPFValues.meta_cover, + OPFAttributes.content); + + if (StringUtil.isNotBlank(coverResourceId)) { + String coverHref = DOMUtil + .getFindAttributeValue(packageDocument, NAMESPACE_OPF, + OPFTags.item, OPFAttributes.id, coverResourceId, + OPFAttributes.href); + if (StringUtil.isNotBlank(coverHref)) { + result.add(coverHref); + } else { + result.add( + coverResourceId); // maybe there was a cover href put in the cover id attribute + } + } + // try and find a reference tag with type is 'cover' and reference is not blank + String coverHref = DOMUtil + .getFindAttributeValue(packageDocument, NAMESPACE_OPF, + OPFTags.reference, OPFAttributes.type, OPFValues.reference_cover, + OPFAttributes.href); + if (StringUtil.isNotBlank(coverHref)) { + result.add(coverHref); + } + return result; } - return result; - } /** * Finds the cover resource in the packageDocument and adds it to the book if found. @@ -418,11 +421,11 @@ public class PackageDocumentReader extends PackageDocumentBase { } if (resource.getMediaType() == MediaTypes.XHTML) { book.setCoverPage(resource); - } else if (MediaTypes.isBitmapImage(resource.getMediaType())) { - book.setCoverImage(resource); - } + } else if (MediaTypes.isBitmapImage(resource.getMediaType())) { + book.setCoverImage(resource); + } + } } - } } diff --git a/epublib/src/main/java/me/ag2s/epublib/epub/ResourcesLoader.java b/epublib/src/main/java/me/ag2s/epublib/epub/ResourcesLoader.java index df53b0162..5da77b8d2 100644 --- a/epublib/src/main/java/me/ag2s/epublib/epub/ResourcesLoader.java +++ b/epublib/src/main/java/me/ag2s/epublib/epub/ResourcesLoader.java @@ -2,16 +2,6 @@ package me.ag2s.epublib.epub; import android.util.Log; -import me.ag2s.epublib.domain.EpubResourceProvider; -import me.ag2s.epublib.domain.LazyResource; -import me.ag2s.epublib.domain.LazyResourceProvider; -import me.ag2s.epublib.domain.MediaType; -import me.ag2s.epublib.domain.MediaTypes; -import me.ag2s.epublib.domain.Resource; -import me.ag2s.epublib.domain.Resources; -import me.ag2s.epublib.util.CollectionUtil; -import me.ag2s.epublib.util.ResourceUtil; - import java.io.IOException; import java.util.ArrayList; import java.util.Collection; @@ -22,6 +12,16 @@ import java.util.zip.ZipException; import java.util.zip.ZipFile; import java.util.zip.ZipInputStream; +import me.ag2s.epublib.domain.EpubResourceProvider; +import me.ag2s.epublib.domain.LazyResource; +import me.ag2s.epublib.domain.LazyResourceProvider; +import me.ag2s.epublib.domain.MediaType; +import me.ag2s.epublib.domain.MediaTypes; +import me.ag2s.epublib.domain.Resource; +import me.ag2s.epublib.domain.Resources; +import me.ag2s.epublib.util.CollectionUtil; +import me.ag2s.epublib.util.ResourceUtil; + /** * Loads Resources from inputStreams, ZipFiles, etc @@ -72,6 +72,12 @@ public class ResourcesLoader { } else { resource = ResourceUtil .createResource(zipEntry, zipFile.getInputStream(zipEntry)); + /*掌上书苑有很多自制书OPF的nameSpace格式不标准，强制修复成正确的格式*/ + if (href.endsWith("opf")) { + String string = new String(resource.getData()).replace("smlns=\"", "xmlns=\""); + resource.setData(string.getBytes()); + } + } if (resource.getMediaType() == MediaTypes.XHTML) { @@ -123,9 +129,15 @@ public class ResourcesLoader { if ((zipEntry == null) || zipEntry.isDirectory()) { continue; } + String href = zipEntry.getName(); // store resource Resource resource = ResourceUtil.createResource(zipEntry, zipInputStream); + ///*掌上书苑有很多自制书OPF的nameSpace格式不标准，强制修复成正确的格式*/ + if (href.endsWith("opf")) { + String string = new String(resource.getData()).replace("smlns=\"", "xmlns=\""); + resource.setData(string.getBytes()); + } if (resource.getMediaType() == MediaTypes.XHTML) { resource.setInputEncoding(defaultHtmlEncoding); } diff --git a/epublib/src/main/java/me/ag2s/epublib/util/IOUtil.java b/epublib/src/main/java/me/ag2s/epublib/util/IOUtil.java index c5d57d64f..dcbda6830 100644 --- a/epublib/src/main/java/me/ag2s/epublib/util/IOUtil.java +++ b/epublib/src/main/java/me/ag2s/epublib/util/IOUtil.java @@ -1,5 +1,7 @@ package me.ag2s.epublib.util; +import android.util.Log; + import java.io.ByteArrayOutputStream; import java.io.Closeable; import java.io.EOFException; @@ -18,6 +20,7 @@ import java.nio.CharBuffer; import java.nio.channels.ReadableByteChannel; import java.nio.charset.Charset; +import me.ag2s.epublib.epub.PackageDocumentReader; import me.ag2s.epublib.util.commons.io.IOConsumer; /** @@ -28,6 +31,7 @@ import me.ag2s.epublib.util.commons.io.IOConsumer; * and using my own implementation saves the inclusion of a 200Kb jar file. */ public class IOUtil { + private static final String TAG = IOUtil.class.getName(); /** * Represents the end-of-file (or stream). @@ -142,11 +146,7 @@ public class IOUtil { // public static void copy(InputStream in, OutputStream result) throws IOException { - int buffer=in.available(); - if(buffer>IOUtil.DEFAULT_BUFFER_SIZE||buffer==0){ - buffer=IOUtil.DEFAULT_BUFFER_SIZE; - } - copy(in, result,buffer); + copy(in, result,DEFAULT_BUFFER_SIZE); } /** @@ -450,6 +450,7 @@ public class IOUtil { output.write(buffer, 0, n); count += n; } + //input.close(); } return count; }