From 744594ef1a2f530fc3e86ea9dc48b62247f79420 Mon Sep 17 00:00:00 2001 From: admin <weikou2014> Date: 星期二, 19 五月 2020 17:13:23 +0800 Subject: [PATCH] 饿了么绘图,添加口碑 --- fanli/src/main/java/com/yeshi/fanli/util/pinduoduo/PinDuoDuoUtil.java | 111 ++++++++++++++++++++++++++++++++++++++++++++++++++++--- 1 files changed, 104 insertions(+), 7 deletions(-) diff --git a/fanli/src/main/java/com/yeshi/fanli/util/pinduoduo/PinDuoDuoUtil.java b/fanli/src/main/java/com/yeshi/fanli/util/pinduoduo/PinDuoDuoUtil.java index e4fc9c2..3979375 100644 --- a/fanli/src/main/java/com/yeshi/fanli/util/pinduoduo/PinDuoDuoUtil.java +++ b/fanli/src/main/java/com/yeshi/fanli/util/pinduoduo/PinDuoDuoUtil.java @@ -9,12 +9,17 @@ import java.net.URLEncoder; import java.util.ArrayList; import java.util.List; +import java.util.regex.Matcher; +import java.util.regex.Pattern; import javax.script.Invocable; import javax.script.ScriptEngine; import javax.script.ScriptEngineManager; +import javax.script.ScriptException; +import org.apache.commons.httpclient.Header; import org.apache.commons.httpclient.HttpClient; +import org.apache.commons.httpclient.HttpException; import org.apache.commons.httpclient.methods.PostMethod; import org.jsoup.Jsoup; import org.jsoup.nodes.Document; @@ -367,7 +372,8 @@ HttpClient client = new HttpClient(); PostMethod pm = new PostMethod("https://jinbao.pinduoduo.com/network/api/common/queryTopGoodsList"); - pm.setRequestHeader("User-Agent", "Mozilla/5.0 (Windows NT 6.3; WOW64; Trident/7.0; rv:11.0) like Gecko"); + pm.setRequestHeader("User-Agent", + "Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/70.0.3538.25 Safari/537.36 Core/1.70.3730.400 QQBrowser/10.5.3805.400"); pm.setRequestHeader("Content-Type", "application/json;charset=UTF-8"); pm.setRequestHeader("Referer", "https://jinbao.pinduoduo.com/promotion/hot-promotion"); pm.setRequestBody(params.toString()); @@ -496,19 +502,46 @@ } static String jdGoodsJs = null; - - public static String getPDDGoodsId(String url) { + static ScriptEngine engine = null; + static { if (jdGoodsJs == null) jdGoodsJs = getJDGoodsJS(); ScriptEngineManager manager = new ScriptEngineManager(); - ScriptEngine engine = manager.getEngineByName("javascript"); + engine = manager.getEngineByName("javascript"); try { engine.eval(jdGoodsJs); + } catch (ScriptException e) { + e.printStackTrace(); + } + + } + + public static String getPDDGoodsId(String url) { + if (url == null) + return null; + String link = url; + if (link.contains("://p.pinduoduo.com/")) {// 鎷煎澶氱殑鐭摼 + HttpClient client = new HttpClient(); + client.getHttpConnectionManager().getParams().setConnectionTimeout(5000); + PostMethod pm = new PostMethod(link); + try { + client.executeMethod(pm); + Header location = pm.getResponseHeader("Location"); + if (location != null) + link = location.getValue(); + } catch (HttpException e) { + e.printStackTrace(); + } catch (IOException e) { + e.printStackTrace(); + } + } + + try { if (engine instanceof Invocable) { Invocable in = (Invocable) engine; - String goodsId = in.invokeFunction("getGoodsId", url).toString(); - if (!StringUtil.isNullOrEmpty(goodsId)) - return goodsId.trim(); + Object goodsId = in.invokeFunction("getGoodsId", link); + if (goodsId != null) + return goodsId.toString().trim(); } } catch (Exception e) { e.printStackTrace(); @@ -516,4 +549,68 @@ return null; } + public static PDDGoodsDetail getPDDGoodsInfo(String id) { + try { + Document doc = Jsoup.connect("http://yangkeduo.com/goods.html?goods_id=" + id) + .userAgent( + "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/75.0.3770.100 Safari/537.36") + .get(); + Elements els = doc.getElementsByTag("script"); + for (int i = 0; i < els.size(); i++) { + if (els.get(i).html().contains("window.rawData")) { + String dataJS = els.get(i).html().replace("window.", "var "); + dataJS += "function getData(){return JSON.stringify(rawData);}"; + + ScriptEngineManager manager = new ScriptEngineManager(); + ScriptEngine engine = manager.getEngineByName("javascript"); + try { + engine.eval(dataJS); + if (engine instanceof Invocable) { + Invocable in = (Invocable) engine; + String jsonStr = in.invokeFunction("getData").toString(); + JSONObject json = JSONObject.fromObject(jsonStr); + + JSONObject goods = json.optJSONObject("store").optJSONObject("initDataObj") + .optJSONObject("goods"); + + String imageData = goods.optString("viewImageData"); + imageData = imageData.replace("[", "").replace("]", ""); + String[] images = imageData.split(","); + + String image = images[0].trim(); + if (image.startsWith("\"")) + image = image.substring(1, image.length() - 1); + if (image.endsWith("\"")) + image = image.substring(0, image.length() - 2); + + PDDGoodsDetail goodsDetail = new PDDGoodsDetail(); + goodsDetail.setGoodsImageUrl("http:" + image); + goodsDetail.setGoodsThumbnailUrl("http:" + image); + goodsDetail.setGoodsName(goods.optString("goodsName")); + goodsDetail.setGoodsId(goods.optLong("goodsID")); + return goodsDetail; + } + } catch (Exception e) { + e.printStackTrace(); + } + break; + } + } + } catch (Exception e1) { + e1.printStackTrace(); + } + return null; + } + + public static List<String> getPDDShortLinksFromText(String text) { + String regex = "(https://p\\.pinduoduo\\.com/)[0-9A-Za-z]{1,20}"; + Pattern pattern = Pattern.compile(regex); + Matcher m = pattern.matcher(text); + List<String> urlList = new ArrayList<>(); + while (m.find()) { + urlList.add(m.group()); + } + return urlList; + } + } -- Gitblit v1.8.0