..

2025-09-23 16:18:41 +08:00
parent 4013fd72e6
commit b3d641d1f2
5 changed files with 1120 additions and 0 deletions
@@ -0,0 +1,273 @@
+package com.budwk.app.base.utils;
+
+import cn.hutool.core.util.ArrayUtil;
+import cn.hutool.core.util.StrUtil;
+import org.nutz.lang.util.NutMap;
+
+import java.util.Arrays;
+import java.util.HashSet;
+import java.util.Set;
+
+/**
+ * 文本相似度计算工具类
+ */
+public class TextSimilarityUtil {
+
+    /**
+     * 提案与关键词的相似度
+     *
+     * @param proposal   提案数据
+     * @param keywords   关键词
+     * @param matchScope 匹配范围
+     * @return 相似度百分比 (0-100)
+     */
+    public static double calculateProposalSimilarity(NutMap proposal, String keywords, String[] matchScope) {
+        String title = proposal.getString("name", "");
+        String brief = proposal.getString("brief", "");
+        String measures = proposal.getString("measures", "");
+
+        StringBuilder fullText = new StringBuilder();
+        if (ArrayUtil.isEmpty(matchScope)) {
+            fullText.append(title);
+            fullText.append(" ");
+            fullText.append(brief);
+            fullText.append(" ");
+            fullText.append(measures);
+        } else {
+            for (String scope : matchScope) {
+                String value = removeHtmlTags(proposal.getString(scope, ""));
+                fullText.append(value);
+                fullText.append(" ");
+            }
+        }
+
+        // 计算关键词匹配度
+        return calculateTextSimilarity(fullText.toString(), keywords);
+    }
+
+    /**
+     * 计算两个文本的相似度
+     *
+     * @param text1 文本1
+     * @param text2 文本2（关键词）
+     * @return 相似度百分比 (0-100)
+     */
+    public static double calculateTextSimilarity(String text1, String text2) {
+        if (StrUtil.isBlank(text1) || StrUtil.isBlank(text2)) {
+            return 0.0;
+        }
+
+        text1 = text1.toLowerCase().trim();
+        text2 = text2.toLowerCase().trim();
+
+        // 使用多种算法综合计算相似度
+        double keywordSimilarity = calculateKeywordSimilarity(text1, text2);
+        double jaccardSimilarity = calculateJaccardSimilarity(text1, text2);
+
+        // 加权平均
+        return (keywordSimilarity * 0.7 + jaccardSimilarity * 0.3);
+    }
+
+    /**
+     * 基于关键词匹配的相似度计算
+     */
+    private static double calculateKeywordSimilarity(String text1, String text2) {
+        // 分词处理
+        String[] keywords = text2.split("[\\s\\p{Punct}]+");
+        int matchCount = 0;
+        int totalKeywords = 0;
+
+        for (String keyword : keywords) {
+            keyword = keyword.trim();
+            if (!keyword.isEmpty()) {
+                totalKeywords++;
+                if (text1.contains(keyword)) {
+                    matchCount++;
+                }
+            }
+        }
+
+        if (totalKeywords == 0) {
+            return 0.0;
+        }
+
+        // 计算匹配率
+        double matchRate = (double) matchCount / totalKeywords;
+
+        // 考虑文本长度因素，避免短文本获得过高分数
+        double lengthFactor = Math.min(text2.length() / (double) Math.max(text1.length(), 1), 1.0);
+        lengthFactor = Math.max(lengthFactor, 0.3); // 最小权重0.3
+
+        return matchRate * lengthFactor * 100;
+    }
+
+    /**
+     * Jaccard相似度计算
+     * 基于集合的交集与并集比值
+     */
+    private static double calculateJaccardSimilarity(String text1, String text2) {
+        Set<String> set1 = new HashSet<>(Arrays.asList(text1.split("[\\s\\p{Punct}]+")));
+        Set<String> set2 = new HashSet<>(Arrays.asList(text2.split("[\\s\\p{Punct}]+")));
+
+        // 移除空字符串
+        set1.removeIf(String::isEmpty);
+        set2.removeIf(String::isEmpty);
+
+        if (set1.isEmpty() && set2.isEmpty()) {
+            return 0.0;
+        }
+
+        // 计算交集
+        Set<String> intersection = new HashSet<>(set1);
+        intersection.retainAll(set2);
+
+        // 计算并集
+        Set<String> union = new HashSet<>(set1);
+        union.addAll(set2);
+
+        return union.isEmpty() ? 0.0 : (double) intersection.size() / union.size() * 100;
+    }
+
+    /**
+     * 提取相似内容片段
+     *
+     * @param proposal 提案数据
+     * @param keywords 关键词
+     * @return 相似内容片段
+     */
+    public static String extractSimilarContent(NutMap proposal, String keywords) {
+        String brief = removeHtmlTags(proposal.getString("brief", ""));
+        String measures = removeHtmlTags(proposal.getString("measures", ""));
+        String title = proposal.getString("name", "");
+
+        // 优先从标题中查找
+        if (containsKeywords(title, keywords)) {
+            return title;
+        }
+
+        // 查找包含关键词的句子
+        String[] sentences = (brief + " " + measures).split("[。！？\\.\\!\\?]");
+
+        for (String sentence : sentences) {
+            sentence = sentence.trim();
+            if (!sentence.isEmpty() && containsKeywords(sentence, keywords)) {
+                return truncateText(sentence, 150);
+            }
+        }
+
+        // 如果没找到，返回摘要的前150个字符
+        String fallback = !brief.isEmpty() ? brief : measures;
+        return truncateText(fallback, 150);
+    }
+
+    /**
+     * 检查文本是否包含关键词
+     */
+    public static boolean containsKeywords(String text, String keywords) {
+        if (StrUtil.isBlank(text) || StrUtil.isBlank(keywords)) {
+            return false;
+        }
+
+        String[] keywordArray = keywords.toLowerCase().split("[\\s\\p{Punct}]+");
+        String lowerText = text.toLowerCase();
+
+        for (String keyword : keywordArray) {
+            keyword = keyword.trim();
+            if (!keyword.isEmpty() && lowerText.contains(keyword)) {
+                return true;
+            }
+        }
+        return false;
+    }
+
+    /**
+     * 移除HTML标签
+     */
+    public static String removeHtmlTags(String html) {
+        if (StrUtil.isBlank(html)) {
+            return "";
+        }
+        return html.replaceAll("<[^>]+>", "")
+                .replaceAll("&nbsp;", " ")
+                .replaceAll("&lt;", "<")
+                .replaceAll("&gt;", ">")
+                .replaceAll("&amp;", "&")
+                .trim();
+    }
+
+    /**
+     * 截断文本到指定长度
+     */
+    public static String truncateText(String text, int maxLength) {
+        if (StrUtil.isBlank(text)) {
+            return "";
+        }
+
+        text = text.trim();
+        if (text.length() <= maxLength) {
+            return text;
+        }
+
+        return text.substring(0, maxLength) + "...";
+    }
+
+    /**
+     * 高亮关键词
+     *
+     * @param text     原文本
+     * @param keywords 关键词
+     * @return 高亮后的HTML文本
+     */
+    public static String highlightKeywords(String text, String keywords) {
+        if (StrUtil.isBlank(text) || StrUtil.isBlank(keywords)) {
+            return text;
+        }
+
+        String[] keywordArray = keywords.split("[\\s\\p{Punct}]+");
+        String result = text;
+
+        for (String keyword : keywordArray) {
+            keyword = keyword.trim();
+            if (!keyword.isEmpty()) {
+                // 使用正则表达式进行大小写不敏感的替换
+                result = result.replaceAll("(?i)(" + escapeRegex(keyword) + ")",
+                        "<span class=\"highlight\">$1</span>");
+            }
+        }
+
+        return result;
+    }
+
+    /**
+     * 转义正则表达式特殊字符
+     */
+    private static String escapeRegex(String text) {
+        return text.replaceAll("([\\\\\\[\\]{}()*+?.^$|])", "\\\\$1");
+    }
+
+    /**
+     * 计算文本的词频
+     *
+     * @param text 文本
+     * @return 词频统计Map
+     */
+    public static NutMap calculateWordFrequency(String text) {
+        NutMap frequency = new NutMap();
+
+        if (StrUtil.isBlank(text)) {
+            return frequency;
+        }
+
+        String[] words = text.toLowerCase().split("[\\s\\p{Punct}]+");
+
+        for (String word : words) {
+            word = word.trim();
+            if (word.length() > 1) { // 忽略单字符
+                frequency.put(word, frequency.getInt(word, 0) + 1);
+            }
+        }
+
+        return frequency;
+    }
+
+}
@@ -0,0 +1,124 @@
+package com.budwk.app.zhgh.democratic.proposal.controller.query;
+
+import cn.dev33.satoken.annotation.SaCheckPermission;
+import com.budwk.app.base.result.Result;
+import com.budwk.app.base.utils.TextSimilarityUtil;
+import com.budwk.app.zhgh.democratic.proposal.service.common.ProposalCommonService;
+import io.swagger.annotations.Api;
+import io.swagger.annotations.ApiOperation;
+import org.nutz.dao.Cnd;
+import org.nutz.dao.Sqls;
+import org.nutz.dao.sql.Sql;
+import org.nutz.ioc.loader.annotation.Inject;
+import org.nutz.ioc.loader.annotation.IocBean;
+import org.nutz.lang.util.NutMap;
+import org.nutz.mvc.annotation.At;
+import org.nutz.mvc.annotation.Ok;
+import org.nutz.mvc.annotation.Param;
+
+import java.util.ArrayList;
+import java.util.List;
+
+@IocBean
+@At("/platform/proposal/query/plagiarismCheck")
+@Api(tags = "提案查重")
+@Ok("json:full")
+public class ProposalPlagiarismCheckController {
+
+    @Inject
+    private ProposalCommonService proposalCommonService;
+
+    @At("")
+    @Ok("beetl:/platform/zhgh/democratic/proposal/query/plagiarismCheck/index.html")
+    @SaCheckPermission("proposal.query.plagiarismCheck")
+    public void index() {
+    }
+
+    /**
+     * 查询
+     *
+     * @param keywords            关键字
+     * @param sessionId           会话id
+     * @param similarityThreshold 相似度阈值
+     * @param typeId              类型id
+     * @return
+     */
+    @At
+    @SaCheckPermission("proposal.query.plagiarismCheck")
+    @ApiOperation(value = "查询")
+    public Result query(String keywords, String sessionId, Integer similarityThreshold, @Param("matchScope") String[] matchScope, String typeId) {
+        // 1. 获取所有提案数据
+        Sql sql = Sqls.create("""
+                SELECT
+                  info.*,
+                  COUNT(p.consolidationIds) > 0 AS isConsolidation,
+                  type.NAME AS typeName,
+                  tcs.fullName AS sessionName,
+                  tcd.`name` AS delegationName,
+                  ins.id AS instanceId,
+                  ins.businessNo,
+                  ins.state instanceState,
+                  ins.variable instanceVariable,
+                  ins.processDefineId instanceProcessDefineId,
+                  t.displayName curTaskName,
+                  JSON_UNQUOTE(JSON_EXTRACT(caseTasks.variable, '$.tf_hostUnitName')) AS masterUnitName,
+                  caseTasks.variable ->> '$.tf_helpUnitNameStr' AS slaveUnitNames
+                FROM
+                  proposal_info info
+                  LEFT JOIN proposal_consolidation p ON JSON_CONTAINS(p.consolidationIds, JSON_QUOTE(info.id))
+                  LEFT JOIN proposal_type type ON type.id = info.typeId
+                  LEFT JOIN teacher_congress_session tcs ON tcs.id = info.sessionId
+                  LEFT JOIN teacher_congress_delegation tcd ON tcd.id = info.delegationId
+                  LEFT JOIN teacher_congress_delegate tcde ON tcde.loginName = info.createUserLoginName
+                  LEFT JOIN wf_process_instance ins ON ins.businessNo = info.id
+                  LEFT JOIN (SELECT processInstanceId, MAX(finishTime) AS finishTime FROM wf_process_task WHERE taskName = '9846ab38-40c5-4093-bafc-a9b3b443338b' AND taskState = 20 GROUP BY processInstanceId) latestTasks ON latestTasks.processInstanceId = ins.id
+                  LEFT JOIN wf_process_task caseTasks ON caseTasks.processInstanceId = ins.id
+                  AND caseTasks.taskName = '9846ab38-40c5-4093-bafc-a9b3b443338b'
+                  AND caseTasks.taskState = 20
+                  AND caseTasks.finishTime = latestTasks.finishTime
+                  LEFT JOIN wf_process_task t ON t.processInstanceId = ins.id
+                  AND t.taskState = 10
+                $condition
+                """);
+        Cnd cnd = Cnd.NEW();
+        cnd.andEX("info.sessionId", "=", sessionId);
+        cnd.andEX("info.typeId", "=", typeId);
+        cnd.groupBy("info.id");
+        sql.setCondition(cnd);
+        List<NutMap> list = proposalCommonService.listMap(sql);
+
+        // 2. 进行查重分析
+        List<NutMap> plagiarismResults = new ArrayList<>();
+
+        for (NutMap proposal : list) {
+            // 计算相似度
+            double similarity = TextSimilarityUtil.calculateProposalSimilarity(proposal, keywords, matchScope);
+            System.out.println(similarity);
+
+            // 只返回相似度大于阈值的结果
+            if (similarity >= similarityThreshold) {
+                // 构建查重结果
+                NutMap result = new NutMap();
+                result.put("id", proposal.getString("id"));
+                result.put("title", proposal.getString("name")); // 提案标题
+                result.put("sessionName", proposal.getString("sessionName"));
+                result.put("typeName", proposal.getString("typeName"));
+                result.put("createUserName", proposal.getString("createUserName"));
+                result.put("createTime", proposal.getString("createTime"));
+                result.put("similarity", Math.round(similarity)); // 相似度百分比
+                result.put("similarContent", TextSimilarityUtil.extractSimilarContent(proposal, keywords));
+                result.put("delegationName", proposal.getString("delegationName"));
+                result.put("unitName", proposal.getString("unitName"));
+
+                plagiarismResults.add(result);
+            }
+        }
+
+        // 3. 按相似度降序排序
+        plagiarismResults.sort((a, b) -> Integer.compare(b.getInt("similarity"), a.getInt("similarity")));
+
+        return Result.success(plagiarismResults);
+    }
+
+
+}
@@ -0,0 +1,142 @@
+package com.budwk.app.zhgh.democratic.proposal.controller.query;
+
+import cn.dev33.satoken.annotation.SaCheckPermission;
+import com.budwk.app.base.result.Result;
+import com.budwk.app.flow.entity.ProcessInstance;
+import com.budwk.app.flow.entity.ProcessTask;
+import com.budwk.app.flow.enums.ProcessTaskStateEnum;
+import com.budwk.app.sys.models.Sys_dict;
+import com.budwk.app.sys.services.SysDictService;
+import com.budwk.app.zhgh.democratic.proposal.models.ProposalInfo;
+import com.budwk.app.zhgh.democratic.proposal.models.ProposalUndertake;
+import io.swagger.annotations.Api;
+import org.nutz.dao.Cnd;
+import org.nutz.dao.Dao;
+import org.nutz.ioc.loader.annotation.Inject;
+import org.nutz.ioc.loader.annotation.IocBean;
+import org.nutz.json.Json;
+import org.nutz.lang.util.NutMap;
+import org.nutz.mvc.annotation.At;
+import org.nutz.mvc.annotation.Ok;
+
+import java.util.ArrayList;
+import java.util.List;
+import java.util.Map;
+import java.util.stream.Collectors;
+
+@IocBean
+@At("/platform/proposal/query/underTake/satisfaction")
+@Ok("json:full")
+@Api(tags = "提案承办单位满意度")
+public class ProposalUnderTakeSatisfactionController {
+
+    @Inject
+    private Dao dao;
+    @Inject
+    private SysDictService sysDictService;
+
+    @At("")
+    @Ok("beetl:/platform/zhgh/democratic/proposal/query/underTakeSatisfaction/index.html")
+    @SaCheckPermission("proposal.query.underTake.satisfaction")
+    public void index() {
+    }
+
+
+    @At
+    @SaCheckPermission("proposal.query.underTake.satisfaction")
+    public Result data(String sessionId){
+        List<ProposalInfo> infos = dao.query(ProposalInfo.class, Cnd.where(ProposalInfo::getSessionId, "=", sessionId));
+        List<String> infoIds = infos.stream().map(v -> v.getId()).toList();
+        List<ProcessInstance> instances = dao.query(ProcessInstance.class, Cnd.where(ProcessInstance::getBusinessNo, "in", infoIds));
+        List<Long> instIds = instances.stream().map(v -> v.getId()).toList();
+
+        // 立案任务记录
+        List<ProcessTask> caseTasks = dao.query(ProcessTask.class, Cnd.where(ProcessTask::getProcessInstanceId, "in", instIds)
+                .and(ProcessTask::getTaskState, "=", ProcessTaskStateEnum.FINISHED.getCode())
+                .and(ProcessTask::getTaskName, "=", "9846ab38-40c5-4093-bafc-a9b3b443338b"));
+
+        // 主办任务记录
+        List<ProcessTask> masterUnitTasks = dao.query(ProcessTask.class, Cnd.where(ProcessTask::getProcessInstanceId, "in", instIds)
+                .and(ProcessTask::getTaskState, "in", List.of(ProcessTaskStateEnum.FINISHED.getCode(), ProcessTaskStateEnum.DOING.getCode()))
+                .and(ProcessTask::getTaskName, "in", List.of("85cf23da-2dd5-4007-a1e9-bd9bfddc68f0", "5ed5a1be-0cd3-4b47-a0d5-238d53b61f34")));
+
+        // 协办任务记录
+        List<ProcessTask> slaveUnitTasks = dao.query(ProcessTask.class, Cnd.where(ProcessTask::getProcessInstanceId, "in", instIds)
+                .and(ProcessTask::getTaskState, "in", List.of(ProcessTaskStateEnum.FINISHED.getCode(), ProcessTaskStateEnum.DOING.getCode()))
+                .and(ProcessTask::getTaskName, "=", "c9a1586e-272b-41b8-b438-0be9f9e3781d"));
+
+
+        // 所有承办单位任务记录
+        List<ProcessTask> allUnitTasks = new ArrayList<>();
+        allUnitTasks.addAll(masterUnitTasks);
+        allUnitTasks.addAll(slaveUnitTasks);
+
+
+        //该届次所有涉及到的承办单位
+        List<String> underTakeIds = caseTasks.stream().map(v -> {
+                    NutMap variable = Json.fromJson(NutMap.class, v.getVariable());
+                    String hostUnitId = variable.getString("tf_hostUnitId");
+                    List<String> helpUnitIds = variable.getAsList("tf_helpUnitIds", String.class);
+                    List<String> list = new ArrayList<>();
+                    list.add(hostUnitId);
+                    list.addAll(helpUnitIds);
+                    return list;
+                })
+                .flatMap(List::stream)
+                .distinct()
+                .toList();
+
+
+        List<ProposalUndertake> undertakes = dao.query(ProposalUndertake.class, Cnd.where(ProposalUndertake::getId, "in", underTakeIds));
+        Map<String, ProposalUndertake> undertakeMap = undertakes.stream().collect(Collectors.toMap(ProposalUndertake::getId, v -> v));
+
+        List<NutMap> tableData = new ArrayList<>();
+
+        // 反馈评分
+        List<Sys_dict> proposalFeedback = sysDictService.getSubListByCode("PROPOSAL_FEEDBACK");
+
+        for (String underTakeId : underTakeIds) {
+            NutMap tableRow = NutMap.NEW();
+            tableRow.put("id", underTakeId);
+            if (undertakeMap.get(underTakeId) == null) {
+                continue;
+            }
+            tableRow.put("name", undertakeMap.get(underTakeId).getName());
+
+            //找出承办提案总数
+            long sum = caseTasks.stream()
+                    .filter(v -> {
+                        NutMap variable = Json.fromJson(NutMap.class, v.getVariable());
+                        if (variable == null) {
+                            return false;
+                        }
+
+                        String hostUnitId = variable.getString("tf_hostUnitId");
+                        List<String> helpUnitIds = variable.getAsList("tf_helpUnitIds", String.class);
+
+                        boolean isHost = underTakeId != null && underTakeId.equals(hostUnitId);
+                        boolean isHelper = helpUnitIds != null && helpUnitIds.contains(underTakeId);
+
+                        return isHost || isHelper;
+                    })
+                    .map(ProcessTask::getProcessInstanceId)
+                    .distinct()
+                    .count();
+
+            // 主办提案数量
+            long hostSum = masterUnitTasks.stream().map(v -> Json.fromJson(NutMap.class, v.getVariable())).filter(variable -> underTakeId.equals(variable.getString("unitId")) && variable.getBoolean("isMaster")).count();
+
+            // 协办提案数量
+            long slaveSum = slaveUnitTasks.stream().map(v -> Json.fromJson(NutMap.class, v.getVariable())).filter(variable -> underTakeId.equals(variable.getString("unitId")) && !variable.getBoolean("isMaster")).count();
+
+            // 满意
+//            allUnitTasks
+
+        }
+
+
+        return Result.success();
+    }
+
+
+}