diff --git a/parser/src/main/java/cn/qaiu/parser/impl/LzTool.java b/parser/src/main/java/cn/qaiu/parser/impl/LzTool.java index a677ada..1057c2b 100644 --- a/parser/src/main/java/cn/qaiu/parser/impl/LzTool.java +++ b/parser/src/main/java/cn/qaiu/parser/impl/LzTool.java @@ -82,31 +82,31 @@ public class LzTool extends PanBase { Pattern.compile("(?s)文件描述:
(.*?)|class=\"n_box_des\">(.*?)"); private static final Pattern P_FI_ID = Pattern.compile("\\?f=(.*?)&|fid = (.*?);"); private static final Pattern P_FI_TIME = Pattern.compile(">上传时间:(.*?)<"); - /** 分享页导航头,只读共享;需要改写时先 addAll 到新 MultiMap。 */ - private static final MultiMap PAGE_HEADERS = HeaderUtils.parseHeaders(""" + /** + * 蓝奏全部对外 HTTP 共用的移动端身份。 + * 桌面 Chrome UA 拉分享页常返回 off0 下线空壳(~665B,空 title,无 filemoreajax)。 + */ + private static final String MOBILE_UA = "Mozilla/5.0 (Linux; Android 6.0; Nexus 5 Build/MRA58N) AppleWebKit/537.36 " + + "(KHTML, like Gecko) Chrome/111.0.0.0 Mobile Safari/537.36"; + + /** 分享页 / iframe / 下载域导航头,只读共享;需要改写时先 addAll 到新 MultiMap。 */ + private static final MultiMap PAGE_HEADERS = applyMobileIdentity(HeaderUtils.parseHeaders(""" Accept: text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.7 Accept-Encoding: identity Accept-Language: zh-CN,zh;q=0.9,en;q=0.8,en-GB;q=0.7,en-US;q=0.6 Cache-Control: max-age=0 DNT: 1 Priority: u=0, i - Sec-CH-UA: "Chromium";v="134", "Not:A-Brand";v="24", "Google Chrome";v="134" - Sec-CH-UA-Mobile: ?0 - Sec-CH-UA-Platform: "Windows" + Sec-CH-UA: "Chromium";v="111", "Not:A-Brand";v="24", "Google Chrome";v="111" + Sec-CH-UA-Mobile: ?1 + Sec-CH-UA-Platform: "Android" Sec-Fetch-Dest: document Sec-Fetch-Mode: navigate Sec-Fetch-Site: cross-site Sec-Fetch-User: ?1 Upgrade-Insecure-Requests: 1 - User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/134.0.0.0 Safari/537.36 - """); - - /** - * 目录分享移动端身份。桌面 Chrome 拉目录页常返回 off0 空壳(~665B,空 title,无 filemoreajax); - * 仅 parseFileList 抽出列表参数失败时才用这个 UA 重试首屏,单文件 parse() 仍走 PAGE_HEADERS。 - */ - private static final String FOLDER_MOBILE_UA = "Mozilla/5.0 (Linux; Android 6.0; Nexus 5 Build/MRA58N) AppleWebKit/537.36 " - + "(KHTML, like Gecko) Chrome/111.0.0.0 Mobile Safari/537.36"; + User-Agent: Mozilla/5.0 (Linux; Android 6.0; Nexus 5 Build/MRA58N) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/111.0.0.0 Mobile Safari/537.36 + """)); private static final String DOWN_AJAX_HEADERS = """ Accept: application/json, text/javascript, */*; q=0.01 @@ -118,11 +118,7 @@ public class LzTool extends PanBase { Sec-Fetch-Dest: empty Sec-Fetch-Mode: cors Sec-Fetch-Site: same-origin - User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/134.0.0.0 Safari/537.36 X-Requested-With: XMLHttpRequest - sec-ch-ua: "Chromium";v="134", "Not:A-Brand";v="24", "Google Chrome";v="134" - sec-ch-ua-mobile: ?0 - sec-ch-ua-platform: "Windows" """; private static final String VERIFY_AJAX_HEADERS = """ @@ -582,7 +578,7 @@ public class LzTool extends PanBase { } private void getDownURL(String referer, AjaxCall call) { - MultiMap headers = HeaderUtils.parseHeaders(DOWN_AJAX_HEADERS); + MultiMap headers = applyMobileIdentity(HeaderUtils.parseHeaders(DOWN_AJAX_HEADERS)); // 个性域名 ajaxfile.php 会立刻 inf=已超时;POST 必须打 wwww。iframe 请求 Referer 用 iframe 地址。 headers.set("referer", referer != null && !referer.isBlank() ? referer : resolveShareUrl()); String url = joinUrl(SHARE_ORIGIN + "/", call.path()); @@ -623,7 +619,7 @@ public class LzTool extends PanBase { /** * 中间链:部分出口仍 302;多数(含机房 IP)固定出「验证并下载」页。 - * dmpdmp 等下载域仍会出 arg1 挑战(Windows UA 更常见),必须算 acw_sc__v2 再请求。 + * dmpdmp 等下载域仍会出 arg1 挑战,必须算 acw_sc__v2 再请求。 * 必须等约 2s 后 POST /file/ajax.php el=2 拿 CDN 直链;立刻 POST 会 ?SignError。 * 中间页不能当成功结果返回——用户侧裸打开永远是验证 HTML。 */ @@ -755,20 +751,17 @@ public class LzTool extends PanBase { } /** 最后一步 GET 用完整浏览器导航头,裸 curl / 无 UA 会固定落到验证页。 */ - private static MultiMap lanrarPageHeaders(String referer) { + static MultiMap lanrarPageHeaders(String referer) { MultiMap h = MultiMap.caseInsensitiveMultiMap(); h.addAll(PAGE_HEADERS); h.set("referer", referer); return h; } - private static MultiMap lanrarAjaxHeaders(String referer) { - MultiMap h = HeaderUtils.parseHeaders(VERIFY_AJAX_HEADERS); + static MultiMap lanrarAjaxHeaders(String referer) { + MultiMap h = applyMobileIdentity(HeaderUtils.parseHeaders(VERIFY_AJAX_HEADERS)); h.set("referer", referer); - copyHeader(h, "User-Agent", "User-Agent"); copyHeader(h, "Sec-CH-UA", "sec-ch-ua"); - copyHeader(h, "Sec-CH-UA-Mobile", "sec-ch-ua-mobile"); - copyHeader(h, "Sec-CH-UA-Platform", "sec-ch-ua-platform"); return h; } @@ -878,57 +871,34 @@ public class LzTool extends PanBase { complete(downloadUrl); } - /** 目录列表 filemoreajax.php 与失败重试首屏共用移动端身份。 */ - private static MultiMap folderMobileIdentity() { - MultiMap headers = MultiMap.caseInsensitiveMultiMap(); - headers.set("User-Agent", FOLDER_MOBILE_UA); - headers.set("sec-ch-ua-platform", "Android"); - headers.set("Accept-Language", "zh-CN,zh;q=0.8,zh-TW;q=0.7,zh-HK;q=0.5,en-US;q=0.3,en;q=0.2"); - headers.set("sec-ch-ua-mobile", "?1"); - return headers; - } - - /** 目录页失败重试用的移动端导航头(与 folderListHeaders 同一 UA)。 */ - static MultiMap folderPageHeaders(String referer) { - MultiMap headers = folderMobileIdentity(); - headers.set("Accept", "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8"); - headers.set("Accept-Encoding", "identity"); - headers.set("Upgrade-Insecure-Requests", "1"); - if (referer != null && !referer.isBlank()) { - headers.set("referer", referer); - } + /** 蓝奏对外请求统一写成 Android Mobile,避免桌面 UA 落到 off0 空壳。 */ + static MultiMap applyMobileIdentity(MultiMap headers) { + headers.set("User-Agent", MOBILE_UA); + headers.set("Sec-CH-UA-Mobile", "?1"); + headers.set("Sec-CH-UA-Platform", "\"Android\""); return headers; } static MultiMap folderListHeaders(String referer) { - MultiMap headers = folderMobileIdentity(); + MultiMap headers = applyMobileIdentity(MultiMap.caseInsensitiveMultiMap()); + headers.set("Accept-Language", "zh-CN,zh;q=0.8,zh-TW;q=0.7,zh-HK;q=0.5,en-US;q=0.3,en;q=0.2"); headers.set("referer", referer); return headers; } - static String desktopPageUserAgent() { - return PAGE_HEADERS.get("User-Agent"); + static String mobileUserAgent() { + return MOBILE_UA; } - static String folderMobileUserAgent() { - return FOLDER_MOBILE_UA; + static MultiMap sharePageHeaders() { + return PAGE_HEADERS; } - /** - * 桌面 UA 目录页抽不出 filemoreajax(常见是 off0 空壳)时,再用移动 UA 拉一次。 - * 已是真实单文件页则不重试,避免多余请求。 - */ - static boolean shouldRetryFolderPageWithMobileUa(String html, String shareUrl) { - if (isLzOfflineStub(html)) { - return true; - } - if (isLzFolderUrl(shareUrl)) { - return true; - } - return isLzFolderHtml(html); + static MultiMap downAjaxHeaders() { + return applyMobileIdentity(HeaderUtils.parseHeaders(DOWN_AJAX_HEADERS)); } - /** 蓝奏 off0 / off1 下线空壳:桌面 UA 拉目录页时常只有这个,没有 filemoreajax。 */ + /** 蓝奏 off0 / off1 下线空壳:桌面 UA 拉分享页时常只有这个,没有 filemoreajax。 */ static boolean isLzOfflineStub(String html) { if (html == null || html.isBlank()) { return true; @@ -948,64 +918,35 @@ public class LzTool extends PanBase { final String sUrl = resolveShareUrl(); final String pwd = shareLinkInfo.getSharePassword(); - // 成功路径仍用桌面 PAGE_HEADERS;抽不出目录参数再换移动 UA 重拉首屏。 getWithArg1Retry(sUrl, PAGE_HEADERS) - .compose(html -> { - AjaxCall call = tryExtractFolderAjaxCall(html, pwd, sUrl); - if (call != null) { - return postFolderList(sUrl, call); - } - if (!shouldRetryFolderPageWithMobileUa(html, sUrl) - && !isLzFolderShare(sUrl, html)) { - return Future.failedFuture(baseMsg() + "该链接为蓝奏云文件分享,请使用文件解析接口"); - } - log.warn("蓝奏目录页桌面 UA 未抽出列表参数,改用移动 UA 重试"); - return getWithArg1Retry(sUrl, folderPageHeaders(sUrl)) - .compose(mobileHtml -> { - AjaxCall mobileCall = tryExtractFolderAjaxCall(mobileHtml, pwd, sUrl); - if (mobileCall == null) { - if (!isLzFolderShare(sUrl, mobileHtml)) { - return Future.failedFuture( - baseMsg() + "该链接为蓝奏云文件分享,请使用文件解析接口"); - } - return Future.failedFuture(baseMsg() + "获取失败1, 可能分享已失效"); - } - return postFolderList(sUrl, mobileCall); - }); - }) - .onSuccess(listPromise::complete) + .onSuccess(html -> handleFileListParse(html, pwd, sUrl, listPromise)) .onFailure(listPromise::fail); return listPromise.future(); } - private AjaxCall tryExtractFolderAjaxCall(String html, String pwd, String sUrl) { - String webpage = extractWebpage(sUrl, html); - if (webpage != null) { - shareLinkInfo.getOtherParam().put("webpage", webpage); - } - AjaxCall call = extractFolderAjax(html, pwd); - if (call != null) { - return call; - } - if (!isLzFolderShare(sUrl, html) && !isLzFolderHtml(html)) { - return null; + private void handleFileListParse(String html, String pwd, String sUrl, Promise> listPromise) { + if (!isLzFolderShare(sUrl, html)) { + listPromise.fail(baseMsg() + "该链接为蓝奏云文件分享,请使用文件解析接口"); + return; } try { - return folderAjaxFromJs(html, pwd); - } catch (ScriptException | NoSuchMethodException | RuntimeException e) { - log.debug("目录页 JS 提取失败: {}", e.getMessage()); - return null; - } - } + String webpage = extractWebpage(sUrl, html); + if (webpage != null) { + shareLinkInfo.getOtherParam().put("webpage", webpage); + } + AjaxCall call = extractFolderAjax(html, pwd); + if (call == null) { + call = folderAjaxFromJs(html, pwd); + } + log.debug("解析参数: {}", call.form()); - private Future> postFolderList(String sUrl, AjaxCall call) { - log.debug("解析参数: {}", call.form()); - Promise> listPromise = Promise.promise(); - String url = joinUrl(originOf(sUrl, SHARE_ORIGIN) + "/", call.path()); - postFormWithArg1Retry(url, folderListHeaders(sUrl), call.toForm()) - .onSuccess(body -> handleFileListResponse(body, listPromise)) - .onFailure(listPromise::fail); - return listPromise.future(); + String url = joinUrl(originOf(sUrl, SHARE_ORIGIN) + "/", call.path()); + postFormWithArg1Retry(url, folderListHeaders(sUrl), call.toForm()) + .onSuccess(body -> handleFileListResponse(body, listPromise)) + .onFailure(listPromise::fail); + } catch (ScriptException | NoSuchMethodException | RuntimeException e) { + listPromise.fail(e); + } } /** 正则匹配不到 filemoreajax 块时,回退到执行页面 JS 取参数。 */ diff --git a/parser/src/test/java/cn/qaiu/parser/impl/LzToolFolderPageTest.java b/parser/src/test/java/cn/qaiu/parser/impl/LzToolFolderPageTest.java index 7c47a68..052838e 100644 --- a/parser/src/test/java/cn/qaiu/parser/impl/LzToolFolderPageTest.java +++ b/parser/src/test/java/cn/qaiu/parser/impl/LzToolFolderPageTest.java @@ -3,11 +3,13 @@ package cn.qaiu.parser.impl; import io.vertx.core.MultiMap; import org.junit.Test; +import java.util.List; + import static org.junit.Assert.*; /** - * 蓝奏目录首屏:桌面 UA 易返回 off0 空壳;抽出 filemoreajax 失败后再用移动 UA 重试。 - * 不访问真实蓝奏,只回归请求头选择与空壳/目录页识别。 + * 蓝奏 LzTool 对外请求一律移动 UA:桌面 Chrome 会触发 off0 下线空壳。 + * 不访问真实蓝奏,只回归请求头与空壳/目录页识别。 */ public class LzToolFolderPageTest { @@ -57,51 +59,45 @@ public class LzToolFolderPageTest { """; - private static final String FILE_HTML = """ - demo.apk - - - """; - @Test - public void testFolderRetryHeadersUseMobileUaNotDesktop() { - String desktopUa = LzTool.desktopPageUserAgent(); - String mobileUa = LzTool.folderMobileUserAgent(); - - assertTrue("单文件页仍是桌面 Chrome", desktopUa.contains("Windows NT 10.0")); - assertFalse("单文件页不能改成 Mobile", desktopUa.contains("Mobile")); - assertTrue("目录重试 UA 是 Android Mobile", mobileUa.contains("Android")); + public void testAllOutboundHeadersUseAndroidMobileUa() { + String mobileUa = LzTool.mobileUserAgent(); + assertTrue(mobileUa.contains("Android")); assertTrue(mobileUa.contains("Mobile Safari")); - assertNotEquals(desktopUa, mobileUa); + assertFalse("蓝奏路径不应再带桌面 Chrome UA", mobileUa.contains("Windows NT")); - MultiMap page = LzTool.folderPageHeaders(FOLDER_URL); - MultiMap ajax = LzTool.folderListHeaders(FOLDER_URL); - assertEquals(mobileUa, page.get("User-Agent")); - assertEquals(mobileUa, ajax.get("User-Agent")); - assertEquals("?1", page.get("sec-ch-ua-mobile")); - assertEquals("Android", page.get("sec-ch-ua-platform")); - assertEquals(FOLDER_URL, page.get("referer")); + List outbound = List.of( + LzTool.sharePageHeaders(), + LzTool.folderListHeaders(FOLDER_URL), + LzTool.downAjaxHeaders(), + LzTool.lanrarPageHeaders("https://developer2.lanrar.com"), + LzTool.lanrarAjaxHeaders(FILE_URL) + ); + for (MultiMap headers : outbound) { + assertEquals(mobileUa, headers.get("User-Agent")); + assertEquals("?1", headers.get("Sec-CH-UA-Mobile")); + assertFalse(String.valueOf(headers.get("User-Agent")).contains("Windows NT")); + assertFalse("?0".equals(headers.get("Sec-CH-UA-Mobile"))); + } + assertEquals("\"Android\"", LzTool.sharePageHeaders().get("Sec-CH-UA-Platform")); } @Test - public void testOff0StubDetectedAndTriggersMobileRetry() { + public void testOff0StubDetectedWithoutFilemoreajax() { assertTrue(LzTool.isLzOfflineStub(OFF0_STUB)); assertTrue(LzTool.isLzOfflineStub("")); assertTrue(LzTool.isLzOfflineStub(null)); assertFalse(LzTool.isLzOfflineStub(FOLDER_HTML)); assertFalse(LzTool.isLzFolderHtml(OFF0_STUB)); assertNull(LzTool.extractFolderAjax(OFF0_STUB, null)); - - assertTrue("off0 空壳应触发移动 UA 重试", - LzTool.shouldRetryFolderPageWithMobileUa(OFF0_STUB, FOLDER_URL)); - assertTrue("文件链接到空壳也值得重试一次", - LzTool.shouldRetryFolderPageWithMobileUa(OFF0_STUB, FILE_URL)); } @Test public void testRealFolderHtmlExtractsFilemoreajax() { assertTrue(LzTool.isLzFolderHtml(FOLDER_HTML)); assertFalse(LzTool.isLzOfflineStub(FOLDER_HTML)); + assertTrue(LzTool.isLzFolderUrl(FOLDER_URL)); + assertFalse(LzTool.isLzFolderUrl(FILE_URL)); LzTool.AjaxCall call = LzTool.extractFolderAjax(FOLDER_HTML, "pwd1"); assertNotNull(call); @@ -111,14 +107,4 @@ public class LzToolFolderPageTest { assertEquals("KEY_TOKEN", call.form().get("k")); assertEquals("pwd1", call.form().get("pwd")); } - - @Test - public void testRealFilePageDoesNotRetryMobileUa() { - assertFalse(LzTool.isLzFolderUrl(FILE_URL)); - assertFalse(LzTool.isLzFolderHtml(FILE_HTML)); - assertFalse(LzTool.isLzOfflineStub(FILE_HTML)); - assertFalse("真实单文件页不要为了目录去换移动 UA", - LzTool.shouldRetryFolderPageWithMobileUa(FILE_HTML, FILE_URL)); - assertTrue(LzTool.isLzFolderUrl(FOLDER_URL)); - } }