交大智能研究院

app/Services/Crawl/Adapters/AiSjtuResearchCenterAdapter.php

@@ -0,0 +1,219 @@
+<?php
+
+namespace App\Services\Crawl\Adapters;
+
+use App\Models\CrawlSource;
+use App\Services\Crawl\Contracts\CrawlerAdapterInterface;
+use App\Services\Crawl\CrawlAuthorParser;
+use App\Services\Crawl\CrawlItemDto;
+use App\Services\Crawl\CrawlKeywordParser;
+use Illuminate\Support\Facades\Http;
+
+/**
+ * 上海交大人工智能研究院研究中心页（Vue SPA）：
+ * GET /api/researchCenter 列表，各中心「研究团队」tab 对应 teams 字段。
+ */
+class AiSjtuResearchCenterAdapter implements CrawlerAdapterInterface
+{
+    protected const API_BASE = 'https://ai.sjtu.edu.cn/api';
+
+    protected const LIST_PATH = '/researchCenter';
+
+    protected const UNIVERSITY_NAME = '上海交通大学';
+
+    public function fetch(string $requestUrl, CrawlSource $source, array $params): array
+    {
+        $keywords = CrawlKeywordParser::parse((string) ($params['keyword'] ?? ''));
+        $maxResults = min(500, max(1, (int) ($params['max_results'] ?? 200)));
+
+        $response = Http::timeout(30)
+            ->connectTimeout(10)
+            ->retry(1, 300, fn ($exception) => $exception instanceof \Illuminate\Http\Client\ConnectionException, throw: false)
+            ->withHeaders([
+                'User-Agent' => 'SlakeSchool-Crawler/1.0',
+                'Accept' => 'application/json',
+            ])
+            ->get(self::API_BASE.self::LIST_PATH, [
+                'page' => 1,
+                'limit' => 99999,
+            ]);
+
+        if (! $response->successful()) {
+            throw new \RuntimeException('无法访问人工智能研究院研究中心 API：HTTP '.$response->status());
+        }
+
+        $json = $response->json();
+        if (! is_array($json)) {
+            throw new \RuntimeException('研究中心 API 返回格式异常');
+        }
+
+        $centers = $json['researchCenters'] ?? [];
+        if (! is_array($centers)) {
+            throw new \RuntimeException('研究中心 API 缺少 researchCenters 字段');
+        }
+
+        $items = [];
+        $seen = [];
+
+        foreach ($centers as $center) {
+            if (! is_array($center)) {
+                continue;
+            }
+
+            $centerId = (int) ($center['id'] ?? 0);
+            $centerName = trim((string) ($center['name'] ?? ''));
+            if ($centerId <= 0 || $centerName === '') {
+                continue;
+            }
+
+            $teams = $center['teams'] ?? [];
+            if (! is_array($teams)) {
+                continue;
+            }
+
+            foreach ($teams as $member) {
+                if (! is_array($member)) {
+                    continue;
+                }
+
+                $item = $this->memberToItem($member, $centerId, $centerName, $keywords, $requestUrl);
+                if ($item === null || isset($seen[$item->externalId])) {
+                    continue;
+                }
+
+                $seen[$item->externalId] = true;
+                $items[] = $item;
+
+                if (count($items) >= $maxResults) {
+                    return $items;
+                }
+            }
+        }
+
+        return $items;
+    }
+
+    /**
+     * @param  list<string>  $keywords
+     */
+    protected function memberToItem(
+        array $member,
+        int $centerId,
+        string $centerName,
+        array $keywords,
+        string $requestUrl,
+    ): ?CrawlItemDto {
+        $name = trim((string) ($member['name'] ?? ''));
+        if ($name === '' || ! $this->looksLikePersonName($name)) {
+            return null;
+        }
+
+        $email = CrawlAuthorParser::normalizeEmail(trim((string) ($member['email'] ?? '')));
+        $phone = $this->normalizePhone((string) ($member['phone'] ?? ''));
+        $title = trim((string) ($member['title'] ?? ''));
+        $direction = trim((string) ($member['direction'] ?? ''));
+        $memberKey = (string) ($member['id'] ?? md5($name.$email));
+
+        $plain = implode(' ', array_filter([$name, $centerName, $title, $direction, $email, $phone]));
+        if (! $this->matchesKeywords($plain, $keywords)) {
+            return null;
+        }
+
+        $profileUrl = 'https://ai.sjtu.edu.cn/center?centerId='.$centerId;
+        $externalId = 'ai_sjtu_center_'.$centerId.'_team_'.$memberKey;
+        $researchDirectionNames = $this->parseResearchDirectionNames($direction);
+
+        $summaryParts = array_filter([
+            $title !== '' ? '职称：'.$title : null,
+            $phone !== '' ? '电话：'.$phone : null,
+            $direction !== '' ? '研究方向：'.$direction : null,
+            '所属中心：'.$centerName,
+        ]);
+
+        $lead = [
+            'name' => $name,
+            'email' => $email,
+            'phone' => $phone !== '' ? $phone : null,
+            'affiliation' => $centerName,
+            'college' => $centerName,
+            'university_name' => self::UNIVERSITY_NAME,
+            'academic_title' => $title !== '' ? $title : null,
+            'research_direction_names' => $researchDirectionNames,
+        ];
+
+        return new CrawlItemDto(
+            externalId: $externalId,
+            title: $name,
+            canonicalUrl: $profileUrl,
+            authors: $name,
+            summary: implode('；', $summaryParts),
+            schoolName: self::UNIVERSITY_NAME,
+            section: $centerName,
+            extra: [
+                'platform' => 'ai_sjtu_research_center',
+                'academic_title' => $title !== '' ? $title : null,
+                'college_name' => $centerName,
+                'profile_url' => $profileUrl,
+                'phone' => $phone !== '' ? $phone : null,
+                'research_direction_names' => $researchDirectionNames,
+                'lead_author' => $lead,
+            ],
+            authorsParsed: [[
+                'name' => $name,
+                'email' => $email,
+                'affiliation' => $centerName,
+                'university_name' => self::UNIVERSITY_NAME,
+                'academic_title' => $title !== '' ? $title : null,
+            ]],
+        );
+    }
+
+    /**
+     * @param  list<string>  $keywords
+     */
+    protected function matchesKeywords(string $plain, array $keywords): bool
+    {
+        if ($keywords === []) {
+            return true;
+        }
+
+        foreach ($keywords as $keyword) {
+            if ($keyword !== '' && mb_stripos($plain, $keyword) !== false) {
+                return true;
+            }
+        }
+
+        return false;
+    }
+
+    /**
+     * @return list<string>
+     */
+    protected function parseResearchDirectionNames(string $direction): array
+    {
+        $direction = trim($direction);
+        if ($direction === '') {
+            return [];
+        }
+
+        $parts = preg_split('/[、，,;；\/]+/u', $direction) ?: [];
+
+        return array_values(array_unique(array_filter(array_map(
+            fn (string $part) => trim($part),
+            $parts,
+        ))));
+    }
+
+    protected function normalizePhone(string $phone): string
+    {
+        $phone = trim(preg_replace('/\s+/u', ' ', $phone) ?? '');
+
+        return $phone;
+    }
+
+    protected function looksLikePersonName(string $name): bool
+    {
+        return (bool) preg_match('/^[\x{4e00}-\x{9fff}]{2,4}(?:·[\x{4e00}-\x{9fff}]{2,4})?$/u', $name)
+            || (bool) preg_match('/^[A-Za-z][A-Za-z\s\.\-]{1,40}$/', $name);
+    }
+}

app/Services/Crawl/CrawlImportService.php

@@ -10,6 +10,7 @@ use App\Models\News;
 use App\Models\Paper;
 use App\Models\Teacher;
 use App\Models\University;
+use App\Services\ResearchDirectionResolver;
 use Carbon\Carbon;
 use Illuminate\Support\Facades\DB;
 use Illuminate\Support\Facades\Log;
@@ -192,6 +193,8 @@ class CrawlImportService
             return null;
         }
 
+        $phone = trim((string) ($lead['phone'] ?? $payload['phone'] ?? ''));
+
         $academicTitle = trim((string) ($lead['academic_title'] ?? $payload['academic_title'] ?? ''));
         $collegeName = trim((string) ($lead['college'] ?? $lead['affiliation'] ?? $payload['college_name'] ?? ''));
         $defaultDepartment = trim((string) ($defaults['department'] ?? ''));
@@ -257,13 +260,26 @@ class CrawlImportService
             'city' => $city ?: '待补充',
             'title' => $academicTitle !== '' ? $academicTitle : '待补充',
             'email' => $email,
+            'phone' => $phone !== '' ? $phone : null,
             'source_dict_item_id' => $sourceId,
             'status_dict_item_id' => $statusId,
             'remark' => implode('；', $remarkParts),
         ]);
 
+        $directionIds = [];
         if (! empty($defaults['research_direction_ids']) && is_array($defaults['research_direction_ids'])) {
-            $teacher->researchDirections()->sync($defaults['research_direction_ids']);
+            $directionIds = array_map('intval', $defaults['research_direction_ids']);
+        } else {
+            $directionNames = $payload['research_direction_names'] ?? $lead['research_direction_names'] ?? [];
+            if (is_string($directionNames)) {
+                $directionNames = preg_split('/[、，,;；\/]+/u', $directionNames) ?: [];
+            }
+            if (is_array($directionNames) && $directionNames !== []) {
+                $directionIds = app(ResearchDirectionResolver::class)->resolveIds([], $directionNames);
+            }
+        }
+        if ($directionIds !== []) {
+            $teacher->researchDirections()->sync($directionIds);
         }
 
         $paperExternalId = $payload['paper_external_id'] ?? null;

app/Services/Crawl/CrawlJobDispatcher.php

@@ -3,6 +3,7 @@
 namespace App\Services\Crawl;
 
 use App\Models\CrawlSource;
+use App\Services\Crawl\Adapters\AiSjtuResearchCenterAdapter;
 use App\Services\Crawl\Adapters\ArxivApiAdapter;
 use App\Services\Crawl\Adapters\FacultyListHtmlAdapter;
 use App\Services\Crawl\Adapters\GenericNewsHtmlAdapter;
@@ -20,6 +21,7 @@ class CrawlJobDispatcher
         protected HuxiuHtmlAdapter $huxiu,
         protected GenericPaperHtmlAdapter $genericPaper,
         protected FacultyListHtmlAdapter $facultyList,
+        protected AiSjtuResearchCenterAdapter $aiSjtuResearchCenter,
     ) {}
 
     /**
@@ -40,6 +42,7 @@ class CrawlJobDispatcher
             'generic_news_html' => $this->genericNews,
             'generic_paper_html' => $this->genericPaper,
             'faculty_list_html' => $this->facultyList,
+            'ai_sjtu_research_center_api' => $this->aiSjtuResearchCenter,
             default => throw new \InvalidArgumentException("未支持的适配器：{$code}"),
         };
     }

app/Services/Crawl/CrawlJobRunnerService.php

@@ -208,6 +208,9 @@ class CrawlJobRunnerService
                     'academic_title' => $dto->extra['academic_title'] ?? (is_array($lead) ? ($lead['academic_title'] ?? null) : null),
                     'college_name' => $dto->extra['college_name'] ?? (is_array($lead) ? ($lead['college'] ?? $lead['affiliation'] ?? null) : null),
                     'profile_url' => $dto->extra['profile_url'] ?? $dto->canonicalUrl,
+                    'phone' => $dto->extra['phone'] ?? (is_array($lead) ? ($lead['phone'] ?? null) : null),
+                    'research_direction_names' => $dto->extra['research_direction_names']
+                        ?? (is_array($lead) ? ($lead['research_direction_names'] ?? null) : null),
                 ],
                 'status' => $status,
                 'target_type' => 'teacher',

app/Services/Crawl/CrawlSourceResolver.php

@@ -97,6 +97,10 @@ class CrawlSourceResolver
             return $sources->firstWhere('adapter_code', 'arxiv_api');
         }
 
+        if ($targetType === 'teacher' && str_contains($lower, 'ai.sjtu.edu.cn')) {
+            return $sources->firstWhere('adapter_code', 'ai_sjtu_research_center_api');
+        }
+
         return null;
     }
 }

database/seeders/CrawlSourcesSeeder.php

@@ -180,6 +180,19 @@ class CrawlSourcesSeeder extends Seeder
             ]
         );
 
+        CrawlSource::query()->updateOrCreate(
+            ['adapter_code' => 'ai_sjtu_research_center_api', 'target_type' => 'teacher'],
+            [
+                'name' => '交大人工智能研究院研究中心',
+                'entry_url' => 'https://ai.sjtu.edu.cn/center',
+                'match_domains' => ['ai.sjtu.edu.cn'],
+                'config' => ['api_base' => 'https://ai.sjtu.edu.cn/api'],
+                'param_schema' => $teacherSchema,
+                'status' => 1,
+                'sort' => 25,
+            ]
+        );
+
         $this->command?->info('采集源 arXiv / 虎嗅 / 投资界 / 通用 HTML / 师资列表 已写入。');
     }
 }

tests/Unit/AiSjtuResearchCenterAdapterTest.php

@@ -0,0 +1,64 @@
+<?php
+
+namespace Tests\Unit;
+
+use App\Models\CrawlSource;
+use App\Services\Crawl\Adapters\AiSjtuResearchCenterAdapter;
+use Illuminate\Support\Facades\Http;
+use Tests\TestCase;
+
+class AiSjtuResearchCenterAdapterTest extends TestCase
+{
+    public function test_fetches_research_team_members_from_api(): void
+    {
+        Http::fake([
+            'ai.sjtu.edu.cn/api/researchCenter*' => Http::response([
+                'researchCenters' => [
+                    [
+                        'id' => 3,
+                        'name' => '数学基础研究中心',
+                        'teams' => [
+                            [
+                                'id' => 0,
+                                'name' => '范金燕',
+                                'email' => 'jyfan@sjtu.edu.cn',
+                                'phone' => '54740206',
+                                'title' => '教授',
+                                'direction' => '最优化理论与方法、多项式优化',
+                            ],
+                        ],
+                    ],
+                ],
+            ], 200),
+        ]);
+
+        $adapter = new AiSjtuResearchCenterAdapter;
+        $source = new CrawlSource([
+            'adapter_code' => 'ai_sjtu_research_center_api',
+            'target_type' => 'teacher',
+        ]);
+
+        $items = $adapter->fetch('https://ai.sjtu.edu.cn/center', $source, [
+            'max_results' => 10,
+        ]);
+
+        $this->assertCount(1, $items);
+        $this->assertSame('范金燕', $items[0]->title);
+        $this->assertSame('jyfan@sjtu.edu.cn', $items[0]->extra['lead_author']['email']);
+        $this->assertSame('54740206', $items[0]->extra['phone']);
+        $this->assertSame(['最优化理论与方法', '多项式优化'], $items[0]->extra['research_direction_names']);
+        $this->assertSame('数学基础研究中心', $items[0]->extra['college_name']);
+    }
+
+    public function test_parse_research_direction_names(): void
+    {
+        $adapter = new AiSjtuResearchCenterAdapter;
+        $method = new \ReflectionMethod($adapter, 'parseResearchDirectionNames');
+        $method->setAccessible(true);
+
+        $this->assertSame(
+            ['图像与信号处理', '数据科学', '最优化方法'],
+            $method->invoke($adapter, '图像与信号处理，数据科学，最优化方法'),
+        );
+    }
+}