|
|
|
|
@ -13,32 +13,45 @@ import java.nio.charset.Charset |
|
|
|
|
import java.util.regex.Matcher |
|
|
|
|
import java.util.regex.Pattern |
|
|
|
|
|
|
|
|
|
object AnalyzeTxtFile { |
|
|
|
|
private const val folderName = "bookTxt" |
|
|
|
|
private const val BLANK: Byte = 0x0a |
|
|
|
|
//默认从文件中获取数据的长度 |
|
|
|
|
private const val BUFFER_SIZE = 512 * 1024 |
|
|
|
|
//没有标题的时候，每个章节的最大长度 |
|
|
|
|
private const val MAX_LENGTH_WITH_NO_CHAPTER = 10 * 1024 |
|
|
|
|
val cacheFolder: File by lazy { |
|
|
|
|
val rootFile = App.INSTANCE.getExternalFilesDir(null) |
|
|
|
|
?: App.INSTANCE.externalCacheDir |
|
|
|
|
?: App.INSTANCE.cacheDir |
|
|
|
|
FileUtils.createFolderIfNotExist(rootFile, subDirs = *arrayOf(folderName)) |
|
|
|
|
} |
|
|
|
|
class AnalyzeTxtFile { |
|
|
|
|
|
|
|
|
|
private val tocRules = arrayListOf<TxtTocRule>() |
|
|
|
|
private lateinit var charset: Charset |
|
|
|
|
|
|
|
|
|
@Throws(Exception::class) |
|
|
|
|
fun analyze(context: Context, book: Book): ArrayList<BookChapter> { |
|
|
|
|
val bookFile = getBookFile(context, book) |
|
|
|
|
book.charset = EncodingDetect.getEncode(bookFile) |
|
|
|
|
val charset = book.fileCharset() |
|
|
|
|
val toc = arrayListOf<BookChapter>() |
|
|
|
|
charset = book.fileCharset() |
|
|
|
|
val rulePattern = if (book.tocUrl.isNotEmpty()) { |
|
|
|
|
Pattern.compile(book.tocUrl, Pattern.MULTILINE) |
|
|
|
|
} else { |
|
|
|
|
tocRules.addAll(getTocRules()) |
|
|
|
|
null |
|
|
|
|
} |
|
|
|
|
//获取文件流 |
|
|
|
|
val bookStream = RandomAccessFile(bookFile, "r") |
|
|
|
|
val rulePattern = getTocRule(book, bookStream, charset) |
|
|
|
|
return analyze(bookStream, book, rulePattern) |
|
|
|
|
} |
|
|
|
|
|
|
|
|
|
@Throws(Exception::class) |
|
|
|
|
private fun analyze( |
|
|
|
|
bookStream: RandomAccessFile, |
|
|
|
|
book: Book, |
|
|
|
|
pattern: Pattern? |
|
|
|
|
): ArrayList<BookChapter> { |
|
|
|
|
val toc = arrayListOf<BookChapter>() |
|
|
|
|
var tocRule: TxtTocRule? = null |
|
|
|
|
val rulePattern = pattern ?: let { |
|
|
|
|
tocRule = getTocRule(bookStream) |
|
|
|
|
tocRule?.let { |
|
|
|
|
Pattern.compile(it.rule, Pattern.MULTILINE) |
|
|
|
|
} |
|
|
|
|
} |
|
|
|
|
//加载章节 |
|
|
|
|
val buffer = ByteArray(BUFFER_SIZE) |
|
|
|
|
//获取到的块起始点，在文件中的位置 |
|
|
|
|
bookStream.seek(0) |
|
|
|
|
var curOffset: Long = 0 |
|
|
|
|
//block的个数 |
|
|
|
|
var blockPos = 0 |
|
|
|
|
@ -47,13 +60,14 @@ object AnalyzeTxtFile { |
|
|
|
|
var allLength = 0 |
|
|
|
|
|
|
|
|
|
//获取文件中的数据到buffer，直到没有数据为止 |
|
|
|
|
while (bookStream.read(buffer, 0, buffer.size).also { length = it } > 0) { |
|
|
|
|
++blockPos |
|
|
|
|
while (bookStream.read(buffer).also { length = it } > 0) { |
|
|
|
|
blockPos++ |
|
|
|
|
//如果存在Chapter |
|
|
|
|
if (rulePattern != null) { //将数据转换成String |
|
|
|
|
if (rulePattern != null) { |
|
|
|
|
//将数据转换成String, 不能超过length |
|
|
|
|
var blockContent = String(buffer, 0, length, charset) |
|
|
|
|
val lastN = blockContent.lastIndexOf("\n") |
|
|
|
|
if (lastN != 0) { |
|
|
|
|
if (lastN > 0) { |
|
|
|
|
blockContent = blockContent.substring(0, lastN) |
|
|
|
|
length = blockContent.toByteArray(charset).size |
|
|
|
|
allLength += length |
|
|
|
|
@ -66,40 +80,49 @@ object AnalyzeTxtFile { |
|
|
|
|
//如果存在相应章节 |
|
|
|
|
while (matcher.find()) { //获取匹配到的字符在字符串中的起始位置 |
|
|
|
|
val chapterStart = matcher.start() |
|
|
|
|
//获取章节内容 |
|
|
|
|
val chapterContent = blockContent.substring(seekPos, chapterStart) |
|
|
|
|
val chapterLength = chapterContent.toByteArray(charset).size |
|
|
|
|
if (chapterLength > 30000 && pattern == null) { |
|
|
|
|
//移除不匹配的规则 |
|
|
|
|
tocRules.remove(tocRule) |
|
|
|
|
return analyze(bookStream, book, null) |
|
|
|
|
} |
|
|
|
|
//如果 seekPos == 0 && nextChapterPos != 0 表示当前block处前面有一段内容 |
|
|
|
|
//第一种情况一定是序章 第二种情况可能是上一个章节的内容 |
|
|
|
|
//第一种情况一定是序章 第二种情况是上一个章节的内容 |
|
|
|
|
if (seekPos == 0 && chapterStart != 0) { //获取当前章节的内容 |
|
|
|
|
val chapterContent = blockContent.substring(seekPos, chapterStart) |
|
|
|
|
//设置指针偏移 |
|
|
|
|
seekPos += chapterContent.length |
|
|
|
|
if (toc.size == 0) { //如果当前没有章节，那么就是序章 |
|
|
|
|
if (toc.isEmpty()) { //如果当前没有章节，那么就是序章 |
|
|
|
|
//加入简介 |
|
|
|
|
book.intro = chapterContent |
|
|
|
|
if (StringUtils.trim(chapterContent).isNotEmpty()) { |
|
|
|
|
val qyChapter = BookChapter() |
|
|
|
|
qyChapter.title = "前言" |
|
|
|
|
qyChapter.start = 0 |
|
|
|
|
qyChapter.end = chapterLength.toLong() |
|
|
|
|
toc.add(qyChapter) |
|
|
|
|
} |
|
|
|
|
//创建当前章节 |
|
|
|
|
val curChapter = BookChapter() |
|
|
|
|
curChapter.title = matcher.group() |
|
|
|
|
curChapter.start = chapterContent.toByteArray(charset).size.toLong() |
|
|
|
|
curChapter.start = chapterLength.toLong() |
|
|
|
|
toc.add(curChapter) |
|
|
|
|
} else { //否则就block分割之后，上一个章节的剩余内容 |
|
|
|
|
//获取上一章节 |
|
|
|
|
val lastChapter = toc.last() |
|
|
|
|
//将当前段落添加上一章去 |
|
|
|
|
lastChapter.end = |
|
|
|
|
lastChapter.end!! + chapterContent.toByteArray(charset).size |
|
|
|
|
lastChapter.end!! + chapterLength.toLong() |
|
|
|
|
//创建当前章节 |
|
|
|
|
val curChapter = BookChapter() |
|
|
|
|
curChapter.title = matcher.group() |
|
|
|
|
curChapter.start = lastChapter.end |
|
|
|
|
toc.add(curChapter) |
|
|
|
|
} |
|
|
|
|
} else { //是否存在章节 |
|
|
|
|
if (toc.size != 0) { //获取章节内容 |
|
|
|
|
val chapterContent = blockContent.substring(seekPos, matcher.start()) |
|
|
|
|
seekPos += chapterContent.length |
|
|
|
|
} else { |
|
|
|
|
if (toc.isNotEmpty()) { //获取章节内容 |
|
|
|
|
//获取上一章节 |
|
|
|
|
val lastChapter = toc.last() |
|
|
|
|
lastChapter.end = |
|
|
|
|
lastChapter.start!! + chapterContent.toByteArray(charset).size |
|
|
|
|
lastChapter.start!! + chapterContent.toByteArray(charset).size.toLong() |
|
|
|
|
//创建当前章节 |
|
|
|
|
val curChapter = BookChapter() |
|
|
|
|
curChapter.title = matcher.group() |
|
|
|
|
@ -108,11 +131,13 @@ object AnalyzeTxtFile { |
|
|
|
|
} else { //如果章节不存在则创建章节 |
|
|
|
|
val curChapter = BookChapter() |
|
|
|
|
curChapter.title = matcher.group() |
|
|
|
|
curChapter.start = 0L |
|
|
|
|
curChapter.end = 0L |
|
|
|
|
curChapter.start = 0 |
|
|
|
|
curChapter.end = 0 |
|
|
|
|
toc.add(curChapter) |
|
|
|
|
} |
|
|
|
|
} |
|
|
|
|
//设置指针偏移 |
|
|
|
|
seekPos += chapterContent.length |
|
|
|
|
} |
|
|
|
|
} else { //进行本地虚拟分章 |
|
|
|
|
//章节在buffer的偏移量 |
|
|
|
|
@ -156,7 +181,8 @@ object AnalyzeTxtFile { |
|
|
|
|
//block的偏移点 |
|
|
|
|
curOffset += length.toLong() |
|
|
|
|
|
|
|
|
|
if (rulePattern != null) { //设置上一章的结尾 |
|
|
|
|
if (rulePattern != null) { |
|
|
|
|
//设置上一章的结尾 |
|
|
|
|
val lastChapter = toc.last() |
|
|
|
|
lastChapter.end = curOffset |
|
|
|
|
} |
|
|
|
|
@ -178,21 +204,58 @@ object AnalyzeTxtFile { |
|
|
|
|
|
|
|
|
|
System.gc() |
|
|
|
|
System.runFinalization() |
|
|
|
|
tocRule?.let { |
|
|
|
|
book.tocUrl = it.rule |
|
|
|
|
} |
|
|
|
|
return toc |
|
|
|
|
} |
|
|
|
|
|
|
|
|
|
private fun getTocRule(bookStream: RandomAccessFile): TxtTocRule? { |
|
|
|
|
var txtTocRule: TxtTocRule? = null |
|
|
|
|
//首先获取128k的数据 |
|
|
|
|
val buffer = ByteArray(BUFFER_SIZE / 4) |
|
|
|
|
val length = bookStream.read(buffer, 0, buffer.size) |
|
|
|
|
val content = String(buffer, 0, length, charset) |
|
|
|
|
for (tocRule in tocRules) { |
|
|
|
|
val pattern = Pattern.compile(tocRule.rule, Pattern.MULTILINE) |
|
|
|
|
val matcher = pattern.matcher(content) |
|
|
|
|
if (matcher.find()) { |
|
|
|
|
txtTocRule = tocRule |
|
|
|
|
break |
|
|
|
|
} |
|
|
|
|
} |
|
|
|
|
|
|
|
|
|
return txtTocRule |
|
|
|
|
} |
|
|
|
|
|
|
|
|
|
companion object { |
|
|
|
|
private const val folderName = "bookTxt" |
|
|
|
|
private const val BLANK: Byte = 0x0a |
|
|
|
|
|
|
|
|
|
//默认从文件中获取数据的长度 |
|
|
|
|
private const val BUFFER_SIZE = 512 * 1024 |
|
|
|
|
|
|
|
|
|
//没有标题的时候，每个章节的最大长度 |
|
|
|
|
private const val MAX_LENGTH_WITH_NO_CHAPTER = 10 * 1024 |
|
|
|
|
val cacheFolder: File by lazy { |
|
|
|
|
val rootFile = App.INSTANCE.getExternalFilesDir(null) |
|
|
|
|
?: App.INSTANCE.externalCacheDir |
|
|
|
|
?: App.INSTANCE.cacheDir |
|
|
|
|
FileUtils.createFolderIfNotExist(rootFile, subDirs = *arrayOf(folderName)) |
|
|
|
|
} |
|
|
|
|
|
|
|
|
|
fun getContent(book: Book, bookChapter: BookChapter): String { |
|
|
|
|
val bookFile = getBookFile(App.INSTANCE, book) |
|
|
|
|
//获取文件流 |
|
|
|
|
val bookStream = RandomAccessFile(bookFile, "r") |
|
|
|
|
bookStream.seek(bookChapter.start ?: 0) |
|
|
|
|
val extent = (bookChapter.end!! - bookChapter.start!!).toInt() |
|
|
|
|
val content = ByteArray(extent) |
|
|
|
|
bookStream.read(content, 0, extent) |
|
|
|
|
val content = ByteArray((bookChapter.end!! - bookChapter.start!!).toInt()) |
|
|
|
|
bookStream.seek(bookChapter.start!!) |
|
|
|
|
bookStream.read(content) |
|
|
|
|
return String(content, book.fileCharset()) |
|
|
|
|
} |
|
|
|
|
|
|
|
|
|
private fun getBookFile(context: Context, book: Book): File { |
|
|
|
|
if (book.bookUrl.isContentPath()) { |
|
|
|
|
val uri = Uri.parse(book.bookUrl) |
|
|
|
|
val bookFile = FileUtils.getFile(cacheFolder, book.originName, subDirs = *arrayOf()) |
|
|
|
|
if (!bookFile.exists()) { |
|
|
|
|
@ -203,28 +266,7 @@ object AnalyzeTxtFile { |
|
|
|
|
} |
|
|
|
|
return bookFile |
|
|
|
|
} |
|
|
|
|
|
|
|
|
|
private fun getTocRule(book: Book, bookStream: RandomAccessFile, charset: Charset): Pattern? { |
|
|
|
|
if (book.tocUrl.isNotEmpty()) { |
|
|
|
|
return Pattern.compile(book.tocUrl, Pattern.MULTILINE) |
|
|
|
|
} |
|
|
|
|
val tocRules = getTocRules() |
|
|
|
|
var rulePattern: Pattern? = null |
|
|
|
|
//首先获取128k的数据 |
|
|
|
|
val buffer = ByteArray(BUFFER_SIZE / 4) |
|
|
|
|
val length = bookStream.read(buffer, 0, buffer.size) |
|
|
|
|
val content = String(buffer, 0, length, charset) |
|
|
|
|
for (tocRule in tocRules) { |
|
|
|
|
val pattern = Pattern.compile(tocRule.rule, Pattern.MULTILINE) |
|
|
|
|
val matcher = pattern.matcher(content) |
|
|
|
|
if (matcher.find()) { |
|
|
|
|
book.tocUrl = tocRule.rule |
|
|
|
|
rulePattern = pattern |
|
|
|
|
break |
|
|
|
|
} |
|
|
|
|
} |
|
|
|
|
bookStream.seek(0) |
|
|
|
|
return rulePattern |
|
|
|
|
return File(book.bookUrl) |
|
|
|
|
} |
|
|
|
|
|
|
|
|
|
private fun getTocRules(): List<TxtTocRule> { |
|
|
|
|
@ -245,3 +287,5 @@ object AnalyzeTxtFile { |
|
|
|
|
return emptyList() |
|
|
|
|
} |
|
|
|
|
} |
|
|
|
|
|
|
|
|
|
} |
Celeter
5 years ago
committed by
GitHub