图片url在解析正文时就拼接成绝对url

4 years ago · 6cd2ba86b0
parent 2332302e7e
commit 6cd2ba86b0
10 changed files with 54 additions and 45 deletions
--- a/app/src/main/java/io/legado/app/help/JsExtensions.kt
+++ b/app/src/main/java/io/legado/app/help/JsExtensions.kt
@ -227,7 +227,7 @@ interface JsExtensions {
    }

    fun htmlFormat(str: String): String {
-        return HtmlFormat.formatKeepImg(str)
+        return HtmlFormatter.formatKeepImg(str)
    }

    /**
--- a/app/src/main/java/io/legado/app/model/Debug.kt
+++ b/app/src/main/java/io/legado/app/model/Debug.kt
@ -8,7 +8,7 @@ import io.legado.app.data.entities.RssSource
 import io.legado.app.help.coroutine.CompositeCoroutine
 import io.legado.app.model.rss.Rss
 import io.legado.app.model.webBook.WebBook
-import io.legado.app.utils.HtmlFormat
+import io.legado.app.utils.HtmlFormatter
 import io.legado.app.utils.isAbsUrl
 import io.legado.app.utils.msg
 import kotlinx.coroutines.CoroutineScope
@ -36,7 +36,7 @@ object Debug {
        if (debugSource != sourceUrl || callback == null || !print) return
        var printMsg = msg ?: ""
        if (isHtml) {
-            printMsg = HtmlFormat.format(msg)
+            printMsg = HtmlFormatter.format(msg)
        }
        if (showTime) {
            printMsg =
--- a/app/src/main/java/io/legado/app/model/analyzeRule/AnalyzeRule.kt
+++ b/app/src/main/java/io/legado/app/model/analyzeRule/AnalyzeRule.kt
@ -67,11 +67,11 @@ class AnalyzeRule(val ruleData: RuleDataInterface) : JsExtensions {
        return this
    }

-    fun setRedirectUrl(url: String): AnalyzeRule {
+    fun setRedirectUrl(url: String): URL? {
        kotlin.runCatching {
            redirectUrl = URL(url.split(AnalyzeUrl.splitUrlRegex, 1)[0])
        }
-        return this
+        return redirectUrl
    }

    /**
--- a/app/src/main/java/io/legado/app/model/localBook/EPUBFile.kt
+++ b/app/src/main/java/io/legado/app/model/localBook/EPUBFile.kt
@ -181,7 +181,7 @@ class EPUBFile(var book: io.legado.app.data.entities.Book) {
            if (book.getDelTag(tag)) {
                html = html.replace("<ruby>\\s?([\\u4e00-\\u9fa5])\\s?.*?</ruby>".toRegex(), "$1")
            }
-            return HtmlFormat.formatKeepImg(html)
+            return HtmlFormatter.formatKeepImg(html)
        }
        return null
    }
--- a/app/src/main/java/io/legado/app/model/webBook/BookContent.kt
+++ b/app/src/main/java/io/legado/app/model/webBook/BookContent.kt
@ -10,7 +10,7 @@ import io.legado.app.help.BookHelp
 import io.legado.app.model.Debug
 import io.legado.app.model.analyzeRule.AnalyzeRule
 import io.legado.app.model.analyzeRule.AnalyzeUrl
-import io.legado.app.utils.HtmlFormat
+import io.legado.app.utils.HtmlFormatter
 import io.legado.app.utils.NetworkUtils
 import kotlinx.coroutines.CoroutineScope
 import kotlinx.coroutines.withContext
@ -26,6 +26,7 @@ object BookContent {
        bookChapter: BookChapter,
        bookSource: BookSource,
        baseUrl: String,
+        redirectUrl: String,
        nextChapterUrl: String? = null
    ): String {
        body ?: throw Exception(
@ -44,7 +45,7 @@ object BookContent {
        analyzeRule.setRedirectUrl(baseUrl)
        analyzeRule.nextChapterUrl = mNextChapterUrl
        var contentData = analyzeContent(
-            book, baseUrl, body, contentRule, bookChapter, bookSource, mNextChapterUrl
+            book, baseUrl, redirectUrl, body, contentRule, bookChapter, bookSource, mNextChapterUrl
        )
        content.append(contentData.content).append("\n")

@ -56,20 +57,15 @@ object BookContent {
                    == NetworkUtils.getAbsoluteURL(baseUrl, mNextChapterUrl)
                ) break
                nextUrlList.add(nextUrl)
-                AnalyzeUrl(
+                val res = AnalyzeUrl(
                    ruleUrl = nextUrl,
                    book = book,
                    headerMapF = bookSource.getHeaderMap()
-                ).getStrResponse(bookSource.bookSourceUrl).body?.let { nextBody ->
+                ).getStrResponse(bookSource.bookSourceUrl)
+                res.body?.let { nextBody ->
                    contentData = analyzeContent(
-                        book,
-                        nextUrl,
-                        nextBody,
-                        contentRule,
-                        bookChapter,
-                        bookSource,
-                        mNextChapterUrl,
-                        false
+                        book, nextUrl, res.url, nextBody, contentRule,
+                        bookChapter, bookSource, mNextChapterUrl, false
                    )
                    nextUrl =
                        if (contentData.nextUrl.isNotEmpty()) contentData.nextUrl[0] else ""
@ -85,20 +81,15 @@ object BookContent {
            }
            for (item in contentDataList) {
                withContext(scope.coroutineContext) {
-                    AnalyzeUrl(
+                    val res = AnalyzeUrl(
                        ruleUrl = item.nextUrl,
                        book = book,
                        headerMapF = bookSource.getHeaderMap()
-                    ).getStrResponse(bookSource.bookSourceUrl).body?.let {
+                    ).getStrResponse(bookSource.bookSourceUrl)
+                    res.body?.let { nextBody ->
                        contentData = analyzeContent(
-                            book,
-                            item.nextUrl,
-                            it,
-                            contentRule,
-                            bookChapter,
-                            bookSource,
-                            mNextChapterUrl,
-                            false
+                            book, item.nextUrl, res.url, nextBody, contentRule,
+                            bookChapter, bookSource, mNextChapterUrl, false
                        )
                        item.content = contentData.content
                    }
@ -109,7 +100,7 @@ object BookContent {
            }
        }
        content.deleteCharAt(content.length - 1)
-        var contentStr = HtmlFormat.formatKeepImg(content.toString())
+        var contentStr = content.toString()
        val replaceRegex = contentRule.replaceRegex
        if (!replaceRegex.isNullOrEmpty()) {
            contentStr = analyzeRule.getString(replaceRegex, value = contentStr)
@ -128,6 +119,7 @@ object BookContent {
    private fun analyzeContent(
        book: Book,
        baseUrl: String,
+        redirectUrl: String,
        body: String,
        contentRule: ContentRule,
        chapter: BookChapter,
@ -136,13 +128,14 @@ object BookContent {
        printLog: Boolean = true
    ): ContentData<List<String>> {
        val analyzeRule = AnalyzeRule(book)
-        analyzeRule.setContent(body).setBaseUrl(baseUrl)
-        analyzeRule.setRedirectUrl(baseUrl)
+        analyzeRule.setContent(body, baseUrl)
+        val rUrl = analyzeRule.setRedirectUrl(redirectUrl)
        analyzeRule.nextChapterUrl = nextChapterUrl
        val nextUrlList = arrayListOf<String>()
        analyzeRule.chapter = chapter
        //获取正文
-        val content = analyzeRule.getString(contentRule.content)
+        var content = analyzeRule.getString(contentRule.content)
+        content = HtmlFormatter.formatKeepImg(content, rUrl)
        //获取下一页链接
        val nextUrlRule = contentRule.nextContentUrl
        if (!nextUrlRule.isNullOrEmpty()) {
--- a/app/src/main/java/io/legado/app/model/webBook/BookInfo.kt
+++ b/app/src/main/java/io/legado/app/model/webBook/BookInfo.kt
@ -6,7 +6,7 @@ import io.legado.app.data.entities.BookSource
 import io.legado.app.help.BookHelp
 import io.legado.app.model.Debug
 import io.legado.app.model.analyzeRule.AnalyzeRule
-import io.legado.app.utils.HtmlFormat
+import io.legado.app.utils.HtmlFormatter
 import io.legado.app.utils.NetworkUtils
 import io.legado.app.utils.StringUtils.wordCountFormat
 import kotlinx.coroutines.CoroutineScope
@ -93,7 +93,7 @@ object BookInfo {
        scope.ensureActive()
        Debug.log(bookSource.bookSourceUrl, "┌获取简介")
        analyzeRule.getString(infoRule.intro).let {
-            if (it.isNotEmpty()) book.intro = HtmlFormat.format(it)
+            if (it.isNotEmpty()) book.intro = HtmlFormatter.format(it)
        }
        Debug.log(bookSource.bookSourceUrl, "└${book.intro}")
        scope.ensureActive()
--- a/app/src/main/java/io/legado/app/model/webBook/BookList.kt
+++ b/app/src/main/java/io/legado/app/model/webBook/BookList.kt
@ -9,7 +9,7 @@ import io.legado.app.help.BookHelp
 import io.legado.app.model.Debug
 import io.legado.app.model.analyzeRule.AnalyzeRule
 import io.legado.app.model.analyzeRule.AnalyzeUrl
-import io.legado.app.utils.HtmlFormat
+import io.legado.app.utils.HtmlFormatter
 import io.legado.app.utils.NetworkUtils
 import io.legado.app.utils.StringUtils.wordCountFormat
 import kotlinx.coroutines.CoroutineScope
@ -197,7 +197,7 @@ object BookList {
            Debug.log(bookSource.bookSourceUrl, "└${searchBook.latestChapterTitle}", log)
            scope.ensureActive()
            Debug.log(bookSource.bookSourceUrl, "┌获取简介", log)
-            searchBook.intro = HtmlFormat.format(analyzeRule.getString(ruleIntro))
+            searchBook.intro = HtmlFormatter.format(analyzeRule.getString(ruleIntro))
            Debug.log(bookSource.bookSourceUrl, "└${searchBook.intro}", log)
            scope.ensureActive()
            Debug.log(bookSource.bookSourceUrl, "┌获取封面链接", log)
--- a/app/src/main/java/io/legado/app/model/webBook/WebBook.kt
+++ b/app/src/main/java/io/legado/app/model/webBook/WebBook.kt
@ -228,6 +228,7 @@ class WebBook(val bookSource: BookSource) {
                bookChapter,
                bookSource,
                bookChapter.getAbsoluteURL(),
+                bookChapter.getAbsoluteURL(),
                nextChapterUrl
            )
        } else {
@ -249,6 +250,7 @@ class WebBook(val bookSource: BookSource) {
                bookChapter,
                bookSource,
                bookChapter.getAbsoluteURL(),
+                res.url,
                nextChapterUrl
            )
        }
--- a/app/src/main/java/io/legado/app/ui/book/read/page/provider/ChapterProvider.kt
+++ b/app/src/main/java/io/legado/app/ui/book/read/page/provider/ChapterProvider.kt
@ -90,13 +90,12 @@ object ChapterProvider {
        textPages.add(TextPage())
        contents.forEachIndexed { index, content ->
            if (book.getImageStyle() == Book.imgStyleText) {
-                var text = content.replace(srcReplaceChar, "画")
+                var text = content.replace(srcReplaceChar, "▣")
                val srcList = LinkedList<String>()
                val sb = StringBuffer()
                val matcher = AppPattern.imgPattern.matcher(text)
                while (matcher.find()) {
-                    matcher.group(1)?.let { it ->
-                        val src = NetworkUtils.getAbsoluteURL(bookChapter.getAbsoluteURL(), it)
+                    matcher.group(1)?.let { src ->
                        srcList.add(src)
                        ImageProvider.getImage(book, bookChapter.index, src)
                        matcher.appendReplacement(sb, srcReplaceChar)
@ -118,12 +117,8 @@ object ChapterProvider {
                        if (text.isNotBlank()) {
                            val matcher = AppPattern.imgPattern.matcher(text)
                            if (matcher.find()) {
-                                matcher.group(1)?.let { it ->
+                                matcher.group(1)?.let { src ->
                                    if (!book.isEpub()) {
-                                        val src = NetworkUtils.getAbsoluteURL(
-                                            bookChapter.getAbsoluteURL(),
-                                            it
-                                        )
                                        durY = setTypeImage(
                                            book, bookChapter, src,
                                            durY, textPages, book.getImageStyle()
--- a/app/src/main/java/io/legado/app/utils/HtmlFormatter.kt
+++ b/app/src/main/java/io/legado/app/utils/HtmlFormatter.kt
@ -1,9 +1,13 @@
 package io.legado.app.utils

-object HtmlFormat {
+import java.net.URL
+import java.util.regex.Pattern
+
+object HtmlFormatter {
    private val wrapHtmlRegex = "</?(?:div|p|br|hr|h\\d|article|dd|dl)[^>]*>".toRegex()
    private val notImgHtmlRegex = "</?(?!img)[a-zA-Z]+(?=[ >])[^<>]*>".toRegex()
    private val otherHtmlRegex = "</?[a-zA-Z]+(?=[ >])[^<>]*>".toRegex()
+    private val imgPattern = Pattern.compile("(<img )[^>]*src=\"([^\"]+)\"[^>]*(>)")

    fun format(html: String?): String {
        html ?: return ""
@ -23,5 +27,20 @@ object HtmlFormat {
            .replace("[\\n\\s]+$".toRegex(), "")
    }

+    fun formatKeepImg(html: String?, redirectUrl: URL?): String {
+        html ?: return ""
+        val sb = StringBuffer()
+        val matcher = imgPattern.matcher(html)
+        while (matcher.find()) {
+            val url = NetworkUtils.getAbsoluteURL(redirectUrl, matcher.group(2)!!)
+            matcher.appendReplacement(sb, "${matcher.group(1)} src=\"$url\"${matcher.group(3)}")
+        }
+        matcher.appendTail(sb)
+        return sb.replace(wrapHtmlRegex, "\n")
+            .replace(notImgHtmlRegex, "")
+            .replace("\\s*\\n+\\s*".toRegex(), "\n　　")
+            .replace("^[\\n\\s]+".toRegex(), "　　")
+            .replace("[\\n\\s]+$".toRegex(), "")
+    }

 }