SinMH: upload date logic, fixes, move YKMH source (#12189)

* SinMH: upload date logic and cleanup

* Fix chapter and page list parse

* SinMH: move, rename and fix Manhuadui -> YKMH

* Fetch categories on popular/latest manga parse

* Change SChapter List sorting

* [skip ci] refine filter prompt

multisrc/overrides/sinmh/gufengmh/src/Gufengmh.kt

@@ -0,0 +1,10 @@
+package eu.kanade.tachiyomi.extension.zh.gufengmh
+
+import eu.kanade.tachiyomi.multisrc.sinmh.SinMH
+
+class Gufengmh : SinMH("古风漫画网", "https://www.gufengmh9.com") {
+
+    override val dateSelector = ".pic_zi:nth-of-type(4) > dd"
+
+    override fun chapterListSelector() = ".list li > a"
+}

multisrc/overrides/sinmh/imitui/src/Imitui.kt

@@ -1,23 +1,15 @@
 package eu.kanade.tachiyomi.extension.zh.imitui
 
 import eu.kanade.tachiyomi.multisrc.sinmh.SinMH
-import eu.kanade.tachiyomi.network.GET
 import eu.kanade.tachiyomi.source.model.Page
-import eu.kanade.tachiyomi.source.model.SChapter
-import eu.kanade.tachiyomi.source.model.SManga
 import org.jsoup.nodes.Document
 
 class Imitui : SinMH("爱米推漫画", "https://www.imitui.com") {
-    private val mobileUrl = "https://m.imitui.com"
-
-    override fun chapterListRequest(manga: SManga) = GET(mobileUrl + manga.url, headers)
-
-    override fun pageListRequest(chapter: SChapter) = GET(mobileUrl + chapter.url, headers)
 
     override fun pageListParse(document: Document): List<Page> {
         val pageCount = document.select("div.image-content > p").text().removePrefix("1/").toInt()
         val prefix = document.location().removeSuffix(".html")
-        return (0 until pageCount).map { Page(it, "$prefix-${it + 1}.html") }
+        return (0 until pageCount).map { Page(it, url = "$prefix-${it + 1}.html") }
     }
 
     override fun imageUrlParse(document: Document): String =

multisrc/overrides/sinmh/manhuadui/src/YKMH.kt

@@ -0,0 +1,31 @@
+package eu.kanade.tachiyomi.extension.zh.manhuadui
+
+import eu.kanade.tachiyomi.multisrc.sinmh.SinMH
+import eu.kanade.tachiyomi.source.model.SChapter
+import eu.kanade.tachiyomi.source.model.SManga
+import org.jsoup.nodes.Document
+
+class YKMH : SinMH("优酷漫画", "http://www.ykmh.com") {
+    override val id = 1637952806167036168
+    override val mobileUrl = "http://wap.ykmh.com"
+
+    override val comicItemSelector = "li.list-comic"
+    override val comicItemTitleSelector = "h3 > a, p > a"
+
+    // DMZJ style
+    override fun mangaDetailsParse(document: Document) = SManga.create().apply {
+        title = document.selectFirst("h1").text()
+        val details = document.selectFirst("ul.comic_deCon_liO").children()
+        author = details[0].selectFirst("a").text()
+        status = when (details[1].selectFirst("a").text()) {
+            "连载中" -> SManga.ONGOING
+            "已完结" -> SManga.COMPLETED
+            else -> SManga.UNKNOWN
+        }
+        genre = (details[2].select("a") + details[3].select("a")).joinToString(", ") { it.text() }
+        description = document.selectFirst("p.comic_deCon_d").text()
+        thumbnail_url = document.selectFirst("div.comic_i_img > img").attr("src")
+    }
+
+    override fun List<SChapter>.sortedDescending() = this
+}

multisrc/src/main/java/eu/kanade/tachiyomi/multisrc/sinmh/SinMH.kt

@@ -1,9 +1,10 @@
 package eu.kanade.tachiyomi.multisrc.sinmh
 
-import android.util.Log
+import eu.kanade.tachiyomi.AppInfo
 import eu.kanade.tachiyomi.network.GET
 import eu.kanade.tachiyomi.source.model.Filter
 import eu.kanade.tachiyomi.source.model.FilterList
+import eu.kanade.tachiyomi.source.model.MangasPage
 import eu.kanade.tachiyomi.source.model.Page
 import eu.kanade.tachiyomi.source.model.SChapter
 import eu.kanade.tachiyomi.source.model.SManga
@@ -13,7 +14,8 @@ import okhttp3.Headers
 import okhttp3.Response
 import org.jsoup.nodes.Document
 import org.jsoup.nodes.Element
-import kotlin.concurrent.thread
+import java.text.SimpleDateFormat
+import java.util.Locale
 
 /**
  * 圣樱漫画CMS https://gitee.com/shenl/SinMH-2.0-Guide
@@ -26,6 +28,7 @@ abstract class SinMH(
     override val lang: String = "zh",
 ) : ParsedHttpSource() {
     override val baseUrl = _baseUrl
+    protected open val mobileUrl = _baseUrl.replace("www", "m")
     override val supportsLatest = true
 
     override fun headersBuilder(): Headers.Builder = Headers.Builder()
@@ -34,13 +37,12 @@ abstract class SinMH(
     protected open val nextPageSelector = "ul.pagination > li.next:not(.disabled)"
     protected open val comicItemSelector = "#contList > li"
     protected open val comicItemTitleSelector = "p > a"
-    protected open fun mangaFromElement(element: Element) =
-        SManga.create().apply {
-            val titleElement = element.select(comicItemTitleSelector)
-            title = titleElement.text()
-            setUrlWithoutDomain(titleElement.attr("abs:href"))
-            thumbnail_url = element.select("img").attr("abs:src")
-        }
+    protected open fun mangaFromElement(element: Element) = SManga.create().apply {
+        val titleElement = element.selectFirst(comicItemTitleSelector)
+        title = titleElement.text()
+        setUrlWithoutDomain(titleElement.attr("abs:href"))
+        thumbnail_url = element.selectFirst("img").attr("abs:src")
+    }
 
     // Popular
 
@@ -49,6 +51,14 @@ abstract class SinMH(
     override fun popularMangaSelector() = comicItemSelector
     override fun popularMangaFromElement(element: Element) = mangaFromElement(element)
 
+    override fun popularMangaParse(response: Response): MangasPage {
+        val document = response.asJsoup()
+        parseCategories(document)
+        val mangas = document.select(popularMangaSelector()).map(::popularMangaFromElement)
+        val hasNextPage = popularMangaNextPageSelector()?.let { document.selectFirst(it) } != null
+        return MangasPage(mangas, hasNextPage)
+    }
+
     // Latest
 
     override fun latestUpdatesRequest(page: Int) = GET("$baseUrl/list/update/?page=$page", headers)
@@ -56,6 +66,14 @@ abstract class SinMH(
     override fun latestUpdatesSelector() = comicItemSelector
     override fun latestUpdatesFromElement(element: Element) = mangaFromElement(element)
 
+    override fun latestUpdatesParse(response: Response): MangasPage {
+        val document = response.asJsoup()
+        parseCategories(document)
+        val mangas = document.select(latestUpdatesSelector()).map(::latestUpdatesFromElement)
+        val hasNextPage = latestUpdatesNextPageSelector()?.let { document.selectFirst(it) } != null
+        return MangasPage(mangas, hasNextPage)
+    }
+
     // Search
 
     override fun searchMangaNextPageSelector(): String? = nextPageSelector
@@ -65,34 +83,48 @@ abstract class SinMH(
         if (query.isNotBlank()) {
             GET("$baseUrl/search/?keywords=$query&page=$page", headers)
         } else {
-            val categories = filters.filterIsInstance<UriPartFilter>()
-                .joinToString("-", transform = UriPartFilter::toUriPart) + "-"
+            val categories = filters.filterIsInstance<UriPartFilter>().map { it.toUriPart() }
+                .filter { it.isNotEmpty() }.joinToString("-") + "-"
             GET("$baseUrl/list/$categories/", headers)
         }
 
     // Details
 
     override fun mangaDetailsParse(document: Document) = SManga.create().apply {
-        title = document.select(".book-title > h1 > span").text()
-        author = document.select(".detail-list strong:contains(作者) + a").text()
-        description = document.select("#intro-all").text().trim()
+        title = document.selectFirst(".book-title > h1 > span").text()
+        author = document.selectFirst(".detail-list strong:contains(作者) + a").text()
+        description = document.selectFirst("#intro-all").text().trim()
             .removePrefix("漫画简介：").trim()
             .removePrefix("漫画简介：").trim() // some sources have double prefix
-        genre = document.select(".detail-list strong:contains(类型) + a").text() + ", " +
+        genre = document.selectFirst(".detail-list strong:contains(类型) + a").text() + ", " +
             document.select(".breadcrumb-bar a[href*=/list/]").joinToString(", ") { it.text() }
-        status = when (document.select(".detail-list strong:contains(状态) + a").text()) {
+        status = when (document.selectFirst(".detail-list strong:contains(状态) + a").text()) {
             "连载中" -> SManga.ONGOING
             "已完结" -> SManga.COMPLETED
             else -> SManga.UNKNOWN
         }
-        thumbnail_url = document.select("p.cover > img").attr("abs:src")
-        // TODO: can use 更新时间：2022-05-23 to set default upload date
+        thumbnail_url = document.selectFirst("p.cover > img").attr("abs:src")
     }
 
     // Chapters
 
-    override fun chapterListSelector() = ".chapter-body li > a:not([href*=/comic/app/])"
-    override fun chapterListParse(response: Response) = super.chapterListParse(response).reversed()
+    override fun chapterListRequest(manga: SManga) = GET(mobileUrl + manga.url, headers)
+
+    protected open val dateSelector = ".date"
+
+    protected open fun List<SChapter>.sortedDescending() = this.asReversed()
+
+    override fun chapterListParse(response: Response): List<SChapter> {
+        val document = response.asJsoup()
+        return document.select(chapterListSelector()).map { chapterFromElement(it) }.sortedDescending().apply {
+            if (isNewDateLogic) {
+                val date = document.selectFirst(dateSelector).textNodes().last().text()
+                this[0].date_upload = DATE_FORMAT.parse(date)?.time ?: 0L
+            }
+        }
+    }
+
+    override fun chapterListSelector() = ".chapter-body li > a:not([href^=/comic/app/])"
     override fun chapterFromElement(element: Element) = SChapter.create().apply {
         setUrlWithoutDomain(element.attr("abs:href"))
         name = element.text()
@@ -100,13 +132,23 @@ abstract class SinMH(
 
     // Pages
 
+    override fun pageListRequest(chapter: SChapter) = GET(mobileUrl + chapter.url, headers)
+
     override fun pageListParse(document: Document): List<Page> {
-        val script = document.select("script:containsData(chapterImages)").html()
+        val script = document.selectFirst("body > script").html()
         val images = script.substringAfter("chapterImages = [\"").substringBefore("\"]").split("\",\"")
         val path = script.substringAfter("chapterPath = \"").substringBefore("\";")
-        // assume cover images are on the same server
+        // assume cover images are on the page image server
         val server = script.substringAfter("pageImage = \"").substringBefore("/images/cover")
-        return images.mapIndexed { i, image -> Page(i, "", "$server/$path/$image") }
+        return images.mapIndexed { i, image ->
+            val unescapedImage = image.replace("""\/""", "/")
+            val imageUrl = if (unescapedImage.startsWith("/")) {
+                "$server$unescapedImage"
+            } else {
+                "$server/$path$unescapedImage"
+            }
+            Page(i, imageUrl = imageUrl)
+        }
     }
 
     override fun imageUrlParse(document: Document): String = throw UnsupportedOperationException("Not Used.")
@@ -121,26 +163,11 @@ abstract class SinMH(
     }
 
     private lateinit var categories: List<Category>
-    private var isFetchingCategories = false
 
-    private fun tryFetchCategories() {
-        if (isFetchingCategories) return
-        isFetchingCategories = true
-        thread {
-            try {
-                fetchCategories()
-            } catch (e: Exception) {
-                Log.e("SinMH", "Failed to fetch categories ($e)")
-            } finally {
-                isFetchingCategories = false
-            }
-        }
-    }
-
-    protected open fun fetchCategories() {
-        val document = client.newCall(GET("$baseUrl/list/", headers)).execute().asJsoup()
-        categories = document.select(".page-main .filter-nav > .filter-item").map { element ->
-            val name = element.select("label").text()
+    protected open fun parseCategories(document: Document) {
+        if (::categories.isInitialized) return
+        categories = document.selectFirst(".filter-nav").children().map { element ->
+            val name = element.selectFirst("label").text()
             val tags = element.select("a")
             val values = tags.map { it.text() }.toTypedArray()
             val uriParts = tags.map { it.attr("href").removePrefix("/list/").removeSuffix("/") }.toTypedArray()
@@ -152,10 +179,17 @@ abstract class SinMH(
         if (::categories.isInitialized) FilterList(
             Filter.Header("如果使用文本搜索，将会忽略分类筛选"),
             *categories.map(Category::toUriPartFilter).toTypedArray()
-        ) else {
-            tryFetchCategories()
-            FilterList(
-                Filter.Header("分类尚未获取，请返回上一页后重试")
-            )
-        }
+        ) else FilterList(
+            Filter.Header("点击“重置”即可刷新分类，如果失败，"),
+            Filter.Header("请尝试重新从图源列表点击进入图源"),
+        )
+
+    private val DATE_FORMAT = SimpleDateFormat("yyyy-MM-dd", Locale.ENGLISH)
+    private val isNewDateLogic = run {
+        val commitCount = AppInfo.getVersionName().substringAfter('-', "")
+        if (commitCount.isNotEmpty()) // Preview
+            commitCount.toInt() >= 4442
+        else // Stable
+            AppInfo.getVersionCode() >= 81
+    }
 }

multisrc/src/main/java/eu/kanade/tachiyomi/multisrc/sinmh/SinMHGenerator.kt

@@ -6,7 +6,7 @@ import generator.ThemeSourceGenerator
 class SinMHGenerator : ThemeSourceGenerator {
     override val themeClass = "SinMH"
     override val themePkg = "sinmh"
-    override val baseVersionCode = 3
+    override val baseVersionCode = 4
     override val sources = listOf(
         SingleLang(
             name = "Gufeng Manhua", baseUrl = "https://www.gufengmh9.com", lang = "zh",
@@ -15,7 +15,11 @@ class SinMHGenerator : ThemeSourceGenerator {
         SingleLang(
             name = "Imitui Manhua", baseUrl = "https://www.imitui.com", lang = "zh",
             className = "Imitui", sourceName = "爱米推漫画", overrideVersionCode = 2
-        )
+        ),
+        SingleLang(
+            name = "YKMH", baseUrl = "http://www.ykmh.com", lang = "zh", className = "YKMH",
+            pkgName = "manhuadui", sourceName = "优酷漫画", overrideVersionCode = 17
+        ),
     )
 
     companion object {

src/zh/manhuadui/AndroidManifest.xml

@@ -1,2 +0,0 @@
-<?xml version="1.0" encoding="utf-8"?>
-<manifest package="eu.kanade.tachiyomi.extension" />

src/zh/manhuadui/build.gradle

@@ -1,11 +0,0 @@
-apply plugin: 'com.android.application'
-apply plugin: 'kotlin-android'
-
-ext {
-    extName = 'Manhuadui'
-    pkgNameSuffix = 'zh.manhuadui'
-    extClass = '.Manhuadui'
-    extVersionCode = 17
-}
-
-apply from: "$rootDir/common.gradle"

src/zh/manhuadui/src/eu/kanade/tachiyomi/extension/zh/manhuadui/Manhuadui.kt

@@ -1,205 +0,0 @@
-package eu.kanade.tachiyomi.extension.zh.manhuadui
-
-import eu.kanade.tachiyomi.network.GET
-import eu.kanade.tachiyomi.source.model.Filter
-import eu.kanade.tachiyomi.source.model.FilterList
-import eu.kanade.tachiyomi.source.model.Page
-import eu.kanade.tachiyomi.source.model.SChapter
-import eu.kanade.tachiyomi.source.model.SManga
-import eu.kanade.tachiyomi.source.online.ParsedHttpSource
-import okhttp3.HttpUrl.Companion.toHttpUrlOrNull
-import okhttp3.OkHttpClient
-import okhttp3.Request
-import okhttp3.Response
-import org.jsoup.nodes.Document
-import org.jsoup.nodes.Element
-
-class Manhuadui : ParsedHttpSource() {
-
-    override val name = "漫画堆"
-    override val baseUrl = "https://ykmh.com"
-    override val lang = "zh"
-    override val supportsLatest = true
-    // Servers can be found from resHost in baseUrl/js/config.js
-    private val imageServer = arrayOf("https://js.tingliu.cc")
-
-    override val client: OkHttpClient = super.client.newBuilder()
-        .followRedirects(true)
-        .build()
-
-    override fun popularMangaSelector() = "li.list-comic"
-    override fun searchMangaSelector() = popularMangaSelector()
-    override fun latestUpdatesSelector() = popularMangaSelector()
-    override fun chapterListSelector() = "ul[id^=chapter-list] > li a"
-
-    override fun searchMangaNextPageSelector() = "li.next a"
-    override fun popularMangaNextPageSelector() = searchMangaNextPageSelector()
-    override fun latestUpdatesNextPageSelector() = searchMangaNextPageSelector()
-
-    override fun popularMangaRequest(page: Int) = GET("$baseUrl/list_$page/", headers)
-    override fun latestUpdatesRequest(page: Int) = GET("$baseUrl/update/$page/", headers)
-    override fun searchMangaRequest(page: Int, query: String, filters: FilterList): Request {
-        return if (query != "") {
-            val url = "$baseUrl/search/?keywords=$query&page=$page".toHttpUrlOrNull()?.newBuilder()
-            GET(url.toString(), headers)
-        } else {
-            val params = filters.map {
-                if (it is UriPartFilter) {
-                    it.toUriPart()
-                } else ""
-            }.filter { it != "" }.joinToString("-")
-            val url = when {
-                params.isEmpty() -> "$baseUrl/list_$page/".toHttpUrlOrNull()?.newBuilder()
-                else -> "$baseUrl/list/$params/$page/".toHttpUrlOrNull()?.newBuilder()
-            }
-            GET(url.toString(), headers)
-        }
-    }
-
-    override fun popularMangaFromElement(element: Element) = mangaFromElement(element)
-    override fun latestUpdatesFromElement(element: Element) = mangaFromElement(element)
-    private fun mangaFromElement(element: Element): SManga {
-        val manga = SManga.create()
-        element.select("a.comic_img").first().let {
-            manga.setUrlWithoutDomain(it.attr("href"))
-            manga.title = it.select("img").attr("alt").trim()
-            manga.thumbnail_url = if (it.select("img").attr("src").trim().indexOf("http") == -1)
-                "https:${it.select("img").attr("src").trim()}"
-            else it.select("img").attr("src").trim()
-        }
-        manga.author = element.select("span.comic_list_det > p").first()?.text()?.substring(3)
-        return manga
-    }
-
-    override fun searchMangaFromElement(element: Element): SManga {
-        val manga = SManga.create()
-        val els = element.select("a.image-link")
-        if (els.size == 0) {
-            element.select("li.list-comic").first().let {
-                manga.setUrlWithoutDomain(it.select("a").attr("href"))
-                manga.title = it.select("span").attr("title").trim()
-                manga.thumbnail_url = it.select("a > img").attr("src").trim()
-                manga.author = it.select("span > p").first().text().split("：")[1].trim()
-            }
-        } else {
-            element.select("a.image-link").first().let {
-                manga.setUrlWithoutDomain(it.attr("href"))
-                manga.title = it.attr("title").trim()
-                manga.thumbnail_url = it.select("img").attr("src").trim()
-            }
-            manga.author = element.select("p.auth").text().trim()
-        }
-        return manga
-    }
-
-    override fun chapterFromElement(element: Element): SChapter {
-        val chapter = SChapter.create()
-        chapter.setUrlWithoutDomain(element.attr("href"))
-        chapter.name = element.select("span:nth-child(2)").text().trim()
-        return chapter
-    }
-
-    override fun mangaDetailsParse(document: Document): SManga {
-        val manga = SManga.create()
-        manga.description = document.select("p.comic_deCon_d").text().trim()
-        manga.thumbnail_url = document.select("div.comic_i_img > img").attr("src")
-        manga.status = when (document.select("ul.comic_deCon_liO a")[1].attr("href")) {
-            "/list/lianzai/" -> SManga.ONGOING
-            "/list/wanjie/" -> SManga.COMPLETED
-            else -> SManga.UNKNOWN
-        }
-        manga.genre = document.select("ul.comic_deCon_liO li:nth-child(4) a").eachText().joinToString(", ")
-        return manga
-    }
-
-    override fun chapterListRequest(manga: SManga) = GET(baseUrl.replace("www", "m") + manga.url)
-    override fun chapterListParse(response: Response): List<SChapter> {
-        return super.chapterListParse(response).asReversed()
-    }
-
-    private val chapterImagesRegex = Regex("""var chapterImages =\s*\["(.*?)"\];""")
-    private val imgCodeCleanupRegex = Regex("""[\[\]"\\]""")
-
-    override fun pageListParse(document: Document): List<Page> {
-        val html = document.html()
-        val imgCodeStr = chapterImagesRegex.find(html)?.groups?.get(1)?.value ?: throw Exception("imgCodeStr not found")
-        val imgCode = imgCodeStr
-            .replace(imgCodeCleanupRegex, "")
-            .replace("%", "%25")
-        return imgCode.split(",").mapIndexed { i, imgStr ->
-            // The way image urls are processed by the website can be found in
-            // getChapterImage() in baseUrl/js/common.js
-            if (imgStr.startsWith("http://images.dmzj.com")) {
-                Page(i, "", "${imageServer[0]}/showImage.php?url=$imgStr")
-            } else {
-                Page(i, "", if (imgStr.indexOf("http") == -1) "${imageServer[0]}$imgStr" else imgStr)
-            }
-        }
-    }
-
-    override fun imageUrlParse(document: Document) = throw UnsupportedOperationException("Not used")
-
-    override fun getFilterList() = FilterList(
-        CategoryGroup(),
-        RegionGroup(),
-        GenreGroup(),
-        ProgressGroup()
-    )
-
-    private class CategoryGroup : UriPartFilter(
-        "按类型",
-        arrayOf(
-            Pair("全部", ""),
-            Pair("儿童漫画", "ertong"),
-            Pair("少年漫画", "shaonian"),
-            Pair("少女漫画", "shaonv"),
-            Pair("青年漫画", "qingnian")
-        )
-    )
-
-    private class ProgressGroup : UriPartFilter(
-        "按进度",
-        arrayOf(
-            Pair("全部", ""),
-            Pair("已完结", "wanjie"),
-            Pair("连载中", "lianzai")
-        )
-    )
-
-    private class RegionGroup : UriPartFilter(
-        "按地区",
-        arrayOf(
-            Pair("全部", ""),
-            Pair("日本", "riben"),
-            Pair("大陆", "dalu"),
-            Pair("香港", "hongkong"),
-            Pair("台湾", "taiwan"),
-            Pair("欧美", "oumei"),
-            Pair("韩国", "hanguo"),
-            Pair("其他", "qita")
-        )
-    )
-
-    private class GenreGroup : UriPartFilter(
-        "按剧情",
-        arrayOf(
-            Pair("全部", ""),
-            Pair("热血", "rexue"),
-            Pair("冒险", "maoxian"),
-            Pair("玄幻", "xuanhuan"),
-            Pair("搞笑", "gaoxiao"),
-            Pair("恋爱", "lianai"),
-            Pair("宠物", "chongwu"),
-            Pair("新作", "xinzuo")
-        )
-    )
-
-    private open class UriPartFilter(
-        displayName: String,
-        val vals: Array<Pair<String, String>>,
-        defaultValue: Int = 0
-    ) :
-        Filter.Select<String>(displayName, vals.map { it.first }.toTypedArray(), defaultValue) {
-        open fun toUriPart() = vals[state].second
-    }
-}