From 842e52bd6fca4c7f7e50f31fe1a9c279cd775bd2 Mon Sep 17 00:00:00 2001 From: Horis <8674809+821938089@users.noreply.github.com> Date: Wed, 3 Dec 2025 21:27:57 +0800 Subject: [PATCH] =?UTF-8?q?=E4=BC=98=E5=8C=96=20#5589?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .../io/legado/app/help/book/ContentHelp.kt | 50 +++++++++++-------- 1 file changed, 30 insertions(+), 20 deletions(-) diff --git a/app/src/main/java/io/legado/app/help/book/ContentHelp.kt b/app/src/main/java/io/legado/app/help/book/ContentHelp.kt index ed49cb9a2..1f2cd8fd4 100644 --- a/app/src/main/java/io/legado/app/help/book/ContentHelp.kt +++ b/app/src/main/java/io/legado/app/help/book/ContentHelp.kt @@ -20,13 +20,13 @@ object ContentHelp { var p = content1 .replace(""".toRegex(), "“") .replace("[:：]['\"‘”“]+".toRegex(), "：“") - .replace("[\"”“]+[\\s]*[\"”“][\\s\"”“]*".toRegex(), "”\n“") + .replace("[\"”“]+\\s*[\"”“][\\s\"”“]*".toRegex(), "”\n“") .split("\n(\\s*)".toRegex()).toTypedArray() //初始化StringBuilder的长度,在原content的长度基础上做冗余 var buffer = StringBuilder((content1.length * 1.15).toInt()) // 章节的文本格式为章节标题-空行-首段，所以处理段落时需要略过第一行文本。 - buffer.append(" ") + buffer.append(" ") if (chapterName.trim { it <= ' ' } != p[0].trim { it <= ' ' }) { // 去除段落内空格。unicode 3000 象形字间隔（中日韩符号和标点），不包含在\s内 buffer.append(p[0].replace("[\u3000\\s]+".toRegex(), "")) @@ -34,7 +34,12 @@ object ContentHelp { //如果原文存在分段错误，需要把段落重新黏合 for (i in 1 until p.size) { - if (match(MARK_SENTENCES_END, buffer[buffer.length - 1])) buffer.append("\n") + if (match(MARK_SENTENCES_END, buffer.last()) + || (match(MARK_QUOTATION_RIGHT, buffer.last()) + && match(MARK_SENTENCES_END, buffer[buffer.lastIndex - 1])) + ) { + buffer.append("\n") + } // 段落开头以外的地方不应该有空格 // 去除段落内空格。unicode 3000 象形字间隔（中日韩符号和标点），不包含在\s内 buffer.append(p[i].replace("[\u3000\\s]".toRegex(), "")) @@ -44,7 +49,7 @@ object ContentHelp { // ”。“处理为”。\n“。不考虑“？” “！”的情况。 // ”。xxx处理为 ”。\n xxx p = buffer.toString() - .replace("[\"”“]+[\\s]*[\"”“]+".toRegex(), "”\n“") + .replace("[\"”“]+\\s*[\"”“]+".toRegex(), "”\n“") .replace("[\"”“]+(？。！?!~)[\"”“]+".toRegex(), "”$1\n“") .replace("[\"”“]+(？。！?!~)([^\"”“])".toRegex(), "”$1\n$2") .replace( @@ -60,7 +65,7 @@ object ContentHelp { buffer = reduceLength(buffer) content1 = (buffer.toString() // 处理章节头部空格和换行 .replaceFirst("^\\s+".toRegex(), "") - .replace("\\s*[\"”“]+[\\s]*[\"”“][\\s\"”“]*".toRegex(), "”\n“") + .replace("\\s*[\"”“]+\\s*[\"”“][\\s\"”“]*".toRegex(), "”\n“") .replace("[:：][”“\"\\s]+".toRegex(), "：“") .replace("\n[\"“”]([^\n\"“”]+)([,:，：][\"”“])([^\n\"“”]+)".toRegex(), "\n$1：“$3") .replace("\n(\\s*)".toRegex(), "\n")) @@ -113,13 +118,13 @@ object ContentHelp { if (match(MARK_QUOTATION, str[0])) { val i = seekIndex(str, MARK_QUOTATION, 1, length - 2, true) + 1 if (i > 1) if (!match(MARK_QUOTATION_BEFORE, str[i - 1])) { - return "${str.substring(0, i)}\n${str.substring(i)}" + return "${str.take(i)}\n${str.substring(i)}" } } else if (match(MARK_QUOTATION, str[length - 1])) { val i = length - 1 - seekIndex(str, MARK_QUOTATION, 1, length - 2, false) if (i > 1) { if (!match(MARK_QUOTATION_BEFORE, str[i - 1])) { - return "${str.substring(0, i)}\n${str.substring(i)}" + return "${str.take(i)}\n${str.substring(i)}" } } } @@ -142,8 +147,8 @@ object ContentHelp { tigger: Int ): ArrayList { val result = ArrayList() - val arrayEnd = seekIndexs(str, MARK_SENTENCES_END_P, 0, str.length - 2, true) - val arrayMid = seekIndexs(str, MARK_SENTENCES_MID, 0, str.length - 2, true) + val arrayEnd = seekIndexes(str, MARK_SENTENCES_END_P, 0, str.length - 2, true) + val arrayMid = seekIndexes(str, MARK_SENTENCES_MID, 0, str.length - 2, true) if (arrayEnd.size < tigger && arrayMid.size < tigger * 3) return result var j = 0 var i = min @@ -347,7 +352,7 @@ object ContentHelp { var j = 0 var progress = 0 var nextLine = -1 - if (insN.size > 0) nextLine = insN[j] + if (insN.isNotEmpty()) nextLine = insN[j] var gain = 3 var min = 0 var trigger = 2 @@ -425,24 +430,24 @@ object ContentHelp { var j = 0 var progress = 0 var nextLine = -1 - if (insN.size > 0) nextLine = insN[j] + if (insN.isNotEmpty()) nextLine = insN[j] for (i in arrayQuote.indices) { - val qutoe = arrayQuote[i] + val quote = arrayQuote[i] // 把引号前的换行符与内容相间插入 while (j < insN.size) { // 如果下一个换行符在当前引号前，那么需要此次处理.如果紧挨当前引号，需要考虑插入引号的情况 - if (nextLine >= qutoe) break + if (nextLine >= quote) break nextLine = insN[j] buffer.append(string, progress, nextLine + 1) buffer.append('\n') progress = nextLine + 1 j++ } - if (progress < qutoe) { - buffer.append(string, progress, qutoe + 1) - progress = qutoe + 1 + if (progress < quote) { + buffer.append(string, progress, quote + 1) + progress = quote + 1 } if (insQuote[i] && buffer.length > 2) { if (buffer[buffer.length - 1] == '\n') buffer.append('“') else buffer.insert( @@ -504,7 +509,7 @@ object ContentHelp { * @param inOrder 是否按照从前向后的顺序匹配 * @return 返回距离构成的ArrayList */ - private fun seekIndexs( + private fun seekIndexes( str: String, key: String, from: Int, @@ -514,14 +519,18 @@ object ContentHelp { val list = ArrayList() if (str.length - from < 1) return list var i = 0 - if (from > i) i = from + if (from > 0) i = from var t = str.length if (to > 0) t = min(t, to) var c: Char while (i < t) { c = if (inOrder) str[i] else str[str.length - i - 1] if (key.indexOf(c) != -1) { - list.add(i) + if (list.isNotEmpty() && i - list.last() == 1) { + list[list.lastIndex] = i + } else { + list.add(i) + } } i++ } @@ -567,7 +576,7 @@ object ContentHelp { private fun seekIndex(str: String, key: String, from: Int, to: Int, inOrder: Boolean): Int { if (str.length - from < 1) return -1 var i = 0 - if (from > i) i = from + if (from > 0) i = from var t = str.length if (to > 0) t = min(t, to) var c: Char @@ -609,6 +618,7 @@ object ContentHelp { // 引号 private const val MARK_QUOTATION = "\"“”" + private const val MARK_QUOTATION_RIGHT = "\"”" private val PARAGRAPH_DIAGLOG = "^[\"”“][^\"”“]+[\"”“]$".toRegex() // 限制字典的长度