修改自动推广的相关任务

.env

@@ -33,6 +33,9 @@ UNSUBSCRIBE_BASE_URL=https://submission.tmrjournals.com/api/Unsubscribe/index
 [yboard]
 APPLY_URL="https://submission.tmrjournals.com/youthBoardRegister"
 
+[plagiarism]
+static_root="/home/wwwroot/api.tmrjournals.com/public"
+
 [journal]
 ;官网服务器地址
 base_url = http://journalapi.tmrjournals.com/public/index.php

application/api/controller/Plagiarism.php

@@ -51,7 +51,7 @@ class Plagiarism extends Base
             $localPath = $fileUrl !== ''
                 ? $svc->resolveFileUrlToLocal($fileUrl)
                 : $svc->locateArticleManuscript($articleId);
-
+            echo $localPath;
             $checkId = $svc->submit($articleId, $localPath, $editorId, 'manual');
             return jsonSuccess(['check_id' => $checkId]);
         } catch (\Throwable $e) {
@@ -59,6 +59,15 @@ class Plagiarism extends Base
         }
     }
 
+
+    public function testccone(){
+        $svc = new PlagiarismService();
+        $checkId = 9;
+        $filePath = "/home/wwwroot/api.tmrjournals.com/public/manuscirpt/20260509/6832a56e8ace38fe99df390ab5221deb.docx";
+        $svc->runUploadAndTrigger($checkId,$filePath);
+
+    }
+
     /**
      * 重试 = 提交一次新查重（保留历史）
      */

application/api/controller/Preaccept.php

@@ -6,6 +6,7 @@ use think\Db;
 use think\Env;
 use think\Queue;
 use think\Validate;
+use app\common\CrossrefService;
 
 class Preaccept extends Base
 {
@@ -708,36 +709,66 @@ class Preaccept extends Base
     }
 
 
+    /**
+     * 通过 DOI 获取文献元数据（Crossref REST API）。
+     *
+     * POST 参数:
+     *   doi  必填，可为纯 DOI（10.xxxx/...）或 https://doi.org/10.xxxx/...
+     *
+     * 返回 data.formate 与旧版字段兼容: author, title, joura, dateno, doilink
+     * 另附 data.crossref: 原始摘要字段（不含 raw message，避免体积过大）
+     */
     public function searchDoi()
     {
         $data = $this->request->post();
         $rule = new Validate([
-            "doi" => "require"
+            'doi' => 'require',
         ]);
         if (!$rule->check($data)) {
             return jsonError($rule->getError());
         }
-        $doi = str_replace('/', '%2F', $data['doi']);
+
-//        $url = "https://citation.crosscite.org/format?doi=$doi&style=cancer-translational-medicine&lang=en-US";
+        $doiInput = trim((string)$data['doi']);
-        $url = "https://citation.doi.org/format?doi=$doi&style=cancer-translational-medicine&lang=en-US";
+        if ($doiInput === '') {
-        $res = myGet($url);
+            return jsonError('doi empty');
-        $frag = trim(substr($res, strpos($res, '.') + 1));
-        if ($frag == "") {
-            return jsonError("not find");
         }
-        if (mb_substr_count($frag, '.') != 3) {
+        // 去掉 URL 前缀，得到裸 DOI
-            return jsonError("formate fail");
+        $doiNorm = preg_replace('#^https?://(dx\.)?doi\.org/#i', '', $doiInput);
+        $doiNorm = trim($doiNorm, " \t\n\r\0\x0B/");
+
+        $svc = new CrossrefService([
+            'mailto' => trim((string)Env::get('crossref_mailto', '')),
+        ]);
+        $summary = $svc->fetchWorkSummary($doiNorm);
+        if ($summary === null || empty($summary['doi'])) {
+            return jsonError('DOI not found or invalid (Crossref)');
         }
-        $res = explode('.', $frag);
+
-        $f['author'] = prgeAuthor($res[0]);
+        $title = trim((string)($summary['title'] ?? ''));
-        $f['title'] = trim($res[1]);
+        $jouraRaw = trim((string)($summary['joura'] ?? ''));
-        $bj = bekjournal($res[2]);
+        $authorStr = trim((string)($summary['author_str'] ?? ''));
-        $joura = formateJournal(trim($bj[0]));
+        $dateno = trim((string)($summary['dateno'] ?? ''));
-        $f['joura'] = $joura;
+        $doilink = trim((string)($summary['doilink'] ?? ''));
-        $f['dateno'] = str_replace(' ', '', str_replace('-', '–', trim($bj[1])));
+        if ($doilink === '') {
-        $f['doilink'] = strpos($data['doi'], "http") === false ? "http://doi.org/" . $data['doi'] : $data['doi'];
+            $doilink = 'https://doi.org/' . $summary['doi'];
-        $re['formate'] = $f;
+        }
-        return jsonSuccess($re);
+
+        $f = [
+            'author'  => $authorStr !== '' ? prgeAuthor($authorStr) : '',
+            'title'   => $title,
+            'joura'   => $jouraRaw !== '' ? formateJournal($jouraRaw) : '',
+            'dateno'  => str_replace(' ', '', str_replace('-', '–', $dateno)),
+            'doilink' => $doilink,
+        ];
+
+        $crossrefOut = $summary;
+        unset($crossrefOut['raw']);
+
+        return jsonSuccess([
+            'formate'   => $f,
+            'crossref'  => $crossrefOut,
+            'doi'       => $summary['doi'],
+        ]);
     }
 
 

application/api/controller/Production.php

@@ -10,6 +10,7 @@ use think\Db;
 use think\Queue;
 use think\Validate;
 use think\log;
+use app\common\ArticleSymbolNormalizer;
 
 /**
  * @title 公共管理相关
@@ -1380,6 +1381,10 @@ class Production extends Base
         return $html;
     }
 
+    public function testsym(){
+        ArticleSymbolNormalizer::normalize("");
+    }
+
 
     public function doTypeSettingNew()
     {
@@ -1399,7 +1404,7 @@ class Production extends Base
         $editor_info = $this->user_obj->where('user_id', $journal_info['editor_id'])->find();
 
         $typesetInfo = [];
-        $typesetInfo['info_title'] = $p_info['title'];
+        $typesetInfo['info_title'] = ArticleSymbolNormalizer::normalize($p_info['title']);
         $typesetInfo['info_type'] = $p_info['type'];
         $typesetInfo['doi'] = $p_info['doi'];
         $typesetInfo['topic'] = '';

application/api/job/PlagiarismRun.php

@@ -29,6 +29,7 @@ class PlagiarismRun
             return;
         }
         $svc = new PlagiarismService();
+        $svc->log("PlagiarismRun job act!!");
         $svc->runUploadAndTrigger($checkId, $filePath);
         $job->delete();
     }

application/common/ArticleSymbolNormalizer.php

@@ -0,0 +1,194 @@
+<?php
+
+namespace app\common;
+
+/**
+ * 期刊文章内容「符号层」校对：只调整标点、空白、全角半角等，不增删语义文字。
+ *
+ * 设计原则：
+ * - 默认规则保守，可通过 $options 逐项关闭；
+ * - 纯文本用 normalize()；含 HTML 时用 normalizeHtml()（仅处理标签之间的文本段，避免破坏属性里的 URL）。
+ * - Abstract 常用：存储时被转义为 &gt; &lt; &amp; 等，可用 normalizeAbstract() 先解码再符号校对。
+ * - 英文期刊正文/摘要通常不含中文：设 english_journal=true（或 normalizeEnglishAbstract）可关闭仅针对汉字的规则。
+ */
+class ArticleSymbolNormalizer
+{
+    /** @var string 常用汉字 BMP 段（含扩展 A 前部，足够覆盖正文） */
+    private static $han = '\x{4E00}-\x{9FFF}\x{3400}-\x{4DBF}';
+
+    /**
+     * 纯文本符号校对。
+     *
+     * @param string $text
+     * @param array  $options 可选键（均为 bool，默认 true）：
+     *   - line_endings        CRLF / CR → LF
+     *   - fullwidth_space     U+3000 全角空格 → 普通空格
+     *   - collapse_spaces     连续半角空格（不含换行）压成单个空格
+     *   - remove_zwsp         删除零宽空格等不可见格式字符（不改变可见字）
+     *   - comma_cjk           两个汉字之间的英文逗号「,」→「，」
+     *   - comma_latin         两个 ASCII 字母/数字之间的全角逗号「，」→「,」
+     *   - period_cjk          汉字后的全角句点「．」(U+FF0E) →「。」
+     *   - bracket_latin       仅由 ASCII 标识包裹时「（）」→「()」（如 (a) 类简单情形，保守：仅当括号内全为 ASCII）
+     *   - decode_html_entities 将 &gt; &lt; &amp; &quot; &#39; 及数字实体等转为真实字符（默认 false；abstract 见 normalizeAbstract）
+     *   - english_journal      英文期刊：关闭「两汉字间英文逗号→，」「汉字后 FF0E→。」等中文专用规则（默认 false；见 normalizeEnglishAbstract）
+     *
+     * @return string
+     */
+    public static function normalize($text, array $options = [])
+    {
+        $text = (string)$text;
+        if ($text === '') {
+            return '';
+        }
+
+        $o = array_merge([
+            'line_endings'         => true,
+            'fullwidth_space'      => true,
+            'collapse_spaces'      => true,
+            'remove_zwsp'          => true,
+            'comma_cjk'            => true,
+            'comma_latin'          => true,
+            'period_cjk'           => true,
+            'bracket_latin'        => false,
+            'decode_html_entities' => false,
+            'english_journal'      => false,
+        ], $options);
+
+        if (!empty($o['english_journal'])) {
+            if (!array_key_exists('comma_cjk', $options)) {
+                $o['comma_cjk'] = false;
+            }
+            if (!array_key_exists('period_cjk', $options)) {
+                $o['period_cjk'] = false;
+            }
+        }
+
+        if (!empty($o['decode_html_entities'])) {
+            $text = html_entity_decode($text, ENT_QUOTES | ENT_HTML5, 'UTF-8');
+        }
+
+        if (!empty($o['line_endings'])) {
+            $text = str_replace(["\r\n", "\r"], "\n", $text);
+        }
+        if (!empty($o['fullwidth_space'])) {
+            $text = str_replace("\u{3000}", ' ', $text);
+        }
+        if (!empty($o['remove_zwsp'])) {
+            // 零宽空格、零宽非断空格、BOM、软连字符等（不改变可见字符）
+            $text = preg_replace('/[\x{200B}-\x{200D}\x{FEFF}\x{00AD}]/u', '', $text);
+        }
+        if (!empty($o['collapse_spaces'])) {
+            $text = preg_replace('/[ \t]{2,}/u', ' ', $text);
+        }
+
+        $han = self::$han;
+
+        if (!empty($o['comma_cjk'])) {
+            // 汉字 , 汉字 → 汉字 ， 汉字
+            $text = preg_replace('/(?<=[' . $han . ']),(?=[' . $han . '])/u', '，', $text);
+        }
+        if (!empty($o['comma_latin'])) {
+            // 字母/数字 ， 字母/数字 → ,
+            $text = preg_replace('/(?<=[0-9A-Za-z])，(?=[0-9A-Za-z])/u', ',', $text);
+        }
+        if (!empty($o['period_cjk'])) {
+            // 汉字后的全角英文句点 FF0E → 中文句号 。
+            $text = preg_replace('/(?<=[' . $han . '])．/u', '。', $text);
+        }
+        if (!empty($o['bracket_latin'])) {
+            // （ 仅 ASCII + 常见标点 + 空格 ）
+            $text = preg_replace_callback(
+                '/（([0-9A-Za-z\s\.,;:\-\+/=]+)）/u',
+                static function ($m) {
+                    return '(' . $m[1] . ')';
+                },
+                $text
+            );
+        }
+
+        return $text;
+    }
+
+    /**
+     * 对 HTML 片段做符号校对：只替换「标签外」的文本，不修改标签名与属性值。
+     *
+     * 实现：按 `<...>` 切分，对偶数段（文本）调用 normalize()，奇数段（标签）原样保留。
+     * 注意：畸形 HTML、属性值中含未转义 `<` 时可能误判，复杂场景请先抽纯文本再校对。
+     *
+     * @param string $html
+     * @param array  $options 同 normalize()
+     * @return string
+     */
+    public static function normalizeHtml($html, array $options = [])
+    {
+        $html = (string)$html;
+        if ($html === '') {
+            return '';
+        }
+
+        $parts = preg_split('/(<[^>]*>)/u', $html, -1, PREG_SPLIT_DELIM_CAPTURE);
+        if ($parts === false) {
+            return self::normalize($html, $options);
+        }
+
+        $out = '';
+        foreach ($parts as $i => $chunk) {
+            if ($chunk === '') {
+                continue;
+            }
+            // 偶数索引为文本，奇数索引且以 < 开头为标签
+            if ($i % 2 === 1 && isset($chunk[0]) && $chunk[0] === '<') {
+                $out .= $chunk;
+            } else {
+                $out .= self::normalize($chunk, $options);
+            }
+        }
+
+        return $out;
+    }
+
+    /**
+     * Abstract 专用：先 HTML 实体解码（&gt; → > 等），再执行与普通正文相同的符号校对。
+     *
+     * 适用于摘要字段在库中/接口中以 htmlspecialchars 形式存储的场景。
+     * 若摘要内本身含真实 HTML 标签且需保留标签结构，请改用 normalizeHtml() 并自行传入 decode_html_entities。
+     *
+     * @param string $abstract
+     * @param array  $options 同 normalize()，默认会合并 decode_html_entities=true（可被显式 false 覆盖）
+     * @return string
+     */
+    public static function normalizeAbstract($abstract, array $options = [])
+    {
+        $opts = array_merge(['decode_html_entities' => true], $options);
+        return self::normalize($abstract, $opts);
+    }
+
+    /**
+     * 带 HTML 标签的摘要：仅在「标签外文本」中做实体解码 + 符号校对，不改动标签与属性。
+     *
+     * @param string $html
+     * @param array  $options 同 normalize()，默认 decode_html_entities=true
+     * @return string
+     */
+    public static function normalizeAbstractHtml($html, array $options = [])
+    {
+        $opts = array_merge(['decode_html_entities' => true], $options);
+        return self::normalizeHtml($html, $opts);
+    }
+
+    /**
+     * 英文期刊 Abstract：实体解码 + 符号校对，且默认关闭中文专用标点规则。
+     */
+    public static function normalizeEnglishAbstract($abstract, array $options = [])
+    {
+        return self::normalizeAbstract($abstract, array_merge(['english_journal' => true], $options));
+    }
+
+    /**
+     * 英文期刊、带 HTML 的摘要（标签外文本）：实体解码 + 符号校对，且默认关闭中文专用规则。
+     */
+    public static function normalizeEnglishAbstractHtml($html, array $options = [])
+    {
+        return self::normalizeAbstractHtml($html, array_merge(['english_journal' => true], $options));
+    }
+}

application/common/PlagiarismService.php

@@ -34,6 +34,13 @@ class PlagiarismService
      */
     const MAX_POLL_ATTEMPTS = 60;
 
+    private $logFile;
+
+    public function __construct()
+    {
+        $this->logFile = ROOT_PATH . 'runtime' . DS . 'plagiarism_task.log';
+    }
+
     // ---------- 顶层入口 ----------
 
     /**
@@ -55,6 +62,7 @@ class PlagiarismService
             ->where('article_id', $articleId)
             ->value('journal_id');
 
+
         $now = time();
         $checkId = Db::name('plagiarism_check')->insertGetId([
             'article_id'       => $articleId,
@@ -67,12 +75,12 @@ class PlagiarismService
             'ctime'            => $now,
             'utime'            => $now,
         ]);
-
+        $this->log("submit service act");
         // 入队执行：上传 + 触发 similarity
         Queue::push(
             'app\\api\\job\\PlagiarismRun',
             ['check_id' => $checkId, 'file_path' => $filePath],
-            'plagiarism'
+            'PlagiarismRun'
         );
 
         return (int)$checkId;
@@ -84,7 +92,7 @@ class PlagiarismService
     public function runUploadAndTrigger($checkId, $filePath)
     {
         $check = $this->mustGetCheck($checkId);
-
+        $this->log("runUploadAndTrigger is act0");
         try {
             $tii = new TurnitinService();
 
@@ -95,7 +103,7 @@ class PlagiarismService
             if ($articleTitle === '') {
                 $articleTitle = 'Article #' . $check['article_id'];
             }
-
+            $this->log("runUploadAndTrigger is act1");
             $createResp = $tii->createSubmission([
                 'title'     => mb_substr($articleTitle, 0, 250),
                 'owner'     => 'editor_' . $check['triggered_by'],
@@ -114,7 +122,7 @@ class PlagiarismService
                 'tii_submission_id' => $submissionId,
                 'raw_response'      => json_encode($createResp, JSON_UNESCAPED_UNICODE),
             ]);
-
+            $this->log("runUploadAndTrigger is act2");
             // 2. 上传文件
             $tii->uploadFile($submissionId, $filePath, basename($filePath));
 
@@ -127,12 +135,14 @@ class PlagiarismService
                 'raw_response'      => json_encode($simResp, JSON_UNESCAPED_UNICODE),
             ]);
 
+            $this->log("runUploadAndTrigger is act3");
+
             // 4. 排队首次轮询（晚一点开始，让 Turnitin 先处理）
             Queue::later(
                 self::POLL_INTERVAL,
                 'app\\api\\job\\PlagiarismPoll',
                 ['check_id' => $checkId, 'attempt' => 1],
-                'plagiarism'
+                'PlagiarismPoll'
             );
         } catch (\Throwable $e) {
             $this->markFailed($checkId, '[upload] ' . $e->getMessage());
@@ -320,6 +330,7 @@ class PlagiarismService
 
     private function markFailed($checkId, $errMsg)
     {
+        $this->log("markFailed act");
         $this->updateCheck($checkId, [
             'state'     => 4,
             'error_msg' => mb_substr($errMsg, 0, 1000),
@@ -337,7 +348,7 @@ class PlagiarismService
         $row = Db::name('article_file')
             ->where('article_id', $articleId)
             ->where('type_name', 'manuscirpt')   // 历史拼写
-            ->order('article_file_id desc')
+            ->order('file_id desc')
             ->find();
         if (!$row || empty($row['file_url'])) {
             throw new Exception("article #{$articleId} has no manuscirpt file");
@@ -420,4 +431,10 @@ class PlagiarismService
     {
         return Db::name('plagiarism_check')->where('check_id', $checkId)->find();
     }
+
+    public function log($msg)
+    {
+        $line = date('Y-m-d H:i:s') . ' ' . $msg . PHP_EOL;
+        @file_put_contents($this->logFile, $line, FILE_APPEND);
+    }
 }