前后端基本架构和完全excel表的解析及统计图表的生成以及excel表的到出

2026-03-19 01:51:34 +08:00
parent c23b93bb70
commit 2f630695ff
194 changed files with 23354 additions and 174 deletions
--- a/frontend/supabase/config.toml
+++ b/frontend/supabase/config.toml
@@ -0,0 +1,5 @@
+[auth.phone]
+enable_confirmations = false
+
+[auth.email]
+enable_confirmations = false
--- a/frontend/supabase/functions/_shared/cors.ts
+++ b/frontend/supabase/functions/_shared/cors.ts
@@ -0,0 +1,4 @@
+export const corsHeaders = {
+  'Access-Control-Allow-Origin': '*',
+  'Access-Control-Allow-Headers': 'authorization, x-client-info, apikey, content-type',
+};
--- a/frontend/supabase/functions/chat-assistant/index.ts
+++ b/frontend/supabase/functions/chat-assistant/index.ts
@@ -0,0 +1,61 @@
+import { serve } from 'https://deno.land/std@0.168.0/http/server.ts';
+import { createClient } from 'https://esm.sh/@supabase/supabase-js@2';
+import { corsHeaders } from '../_shared/cors.ts';
+
+serve(async (req) => {
+  if (req.method === 'OPTIONS') {
+    return new Response(null, { headers: corsHeaders });
+  }
+
+  try {
+    const { messages, userId } = await req.json();
+    if (!messages || !userId) throw new Error('Missing messages or userId');
+
+    const supabaseUrl = Deno.env.get('SUPABASE_URL')!;
+    const supabaseKey = Deno.env.get('SUPABASE_SERVICE_ROLE_KEY')!;
+    const supabase = createClient(supabaseUrl, supabaseKey);
+
+    const miniMaxApiKey = Deno.env.get('INTEGRATIONS_API_KEY');
+    const miniMaxResponse = await fetch(
+      'https://app-a6ww9j3ja3nl-api-Aa2PqMJnJGwL-gateway.appmiaoda.com/v1/text/chatcompletion_v2',
+      {
+        method: 'POST',
+        headers: {
+          'X-Gateway-Authorization': `Bearer ${miniMaxApiKey}`,
+          'Content-Type': 'application/json',
+        },
+        body: JSON.stringify({
+          model: 'MiniMax-M2.5',
+          messages: [
+            {
+              role: 'system',
+              content: `你是一个专业的文档智能交互助手。你的目标是协助用户完成文档管理、信息提取和表格填写。
+你所在的系统支持：
+1. 列出当前用户的文档列表。
+2. 查看某个文档提取到的关键实体。
+3. 协助用户通过自然语言发起文档编辑、排版等指令（当前模拟执行）。
+请以专业且友好的语气回复。`
+            },
+            ...messages
+          ]
+        })
+      }
+    );
+
+    const miniMaxData = await miniMaxResponse.json();
+    if (!miniMaxResponse.ok) throw new Error('MiniMax chat failed');
+
+    return new Response(JSON.stringify({ 
+      choices: miniMaxData.choices 
+    }), {
+      headers: { ...corsHeaders, 'Content-Type': 'application/json' }
+    });
+
+  } catch (error) {
+    console.error('Error in chat assistant:', error);
+    return new Response(JSON.stringify({ error: error.message }), {
+      status: 500,
+      headers: { ...corsHeaders, 'Content-Type': 'application/json' }
+    });
+  }
+});
--- a/frontend/supabase/functions/fill-template/index.ts
+++ b/frontend/supabase/functions/fill-template/index.ts
@@ -0,0 +1,94 @@
+import { serve } from 'https://deno.land/std@0.168.0/http/server.ts';
+import { createClient } from 'https://esm.sh/@supabase/supabase-js@2';
+import { corsHeaders } from '../_shared/cors.ts';
+
+serve(async (req) => {
+  if (req.method === 'OPTIONS') {
+    return new Response(null, { headers: corsHeaders });
+  }
+
+  try {
+    const { taskId } = await req.json();
+    if (!taskId) throw new Error('Missing taskId');
+
+    const supabaseUrl = Deno.env.get('SUPABASE_URL')!;
+    const supabaseKey = Deno.env.get('SUPABASE_SERVICE_ROLE_KEY')!;
+    const supabase = createClient(supabaseUrl, supabaseKey);
+
+    // Get task and documents
+    const { data: task, error: taskError } = await supabase
+      .from('fill_tasks')
+      .select('*, templates(*)')
+      .eq('id', taskId)
+      .single();
+
+    if (taskError || !task) throw new Error('Task not found');
+
+    const { data: entities, error: entitiesError } = await supabase
+      .from('extracted_entities')
+      .select('*')
+      .in('document_id', task.document_ids);
+
+    if (entitiesError) throw new Error('Failed to fetch entities');
+
+    // Aggregate entities for context
+    const context = entities.map(e => `${e.entity_type}: ${e.entity_value}`).join('\n');
+
+    // Get template content (assume docx/xlsx text extraction for template)
+    // Actually, MiniMax can help generate the final filled content if we provide the template structure
+    const miniMaxApiKey = Deno.env.get('INTEGRATIONS_API_KEY');
+    const miniMaxResponse = await fetch(
+      'https://app-a6ww9j3ja3nl-api-Aa2PqMJnJGwL-gateway.appmiaoda.com/v1/text/chatcompletion_v2',
+      {
+        method: 'POST',
+        headers: {
+          'X-Gateway-Authorization': `Bearer ${miniMaxApiKey}`,
+          'Content-Type': 'application/json',
+        },
+        body: JSON.stringify({
+          model: 'MiniMax-M2.5',
+          messages: [
+            {
+              role: 'system',
+              content: '你是一个专业的文档处理助手。请根据提供的源数据实体，自动填充到给定的表格模板中。生成的文档应当格式严谨，满足业务应用需求。'
+            },
+            {
+              role: 'user',
+              content: `源数据信息如下：\n\n${context}\n\n模板名称：${task.templates.name}\n模板类型：${task.templates.type}\n\n请生成填充后的文档内容预览（以 Markdown 格式呈现，确保包含所有源数据中的关键指标）。`
+            }
+          ]
+        })
+      }
+    );
+
+    const miniMaxData = await miniMaxResponse.json();
+    if (!miniMaxResponse.ok) throw new Error('MiniMax filling failed');
+
+    const resultText = miniMaxData.choices[0].message.content;
+
+    // Update task
+    const { error: updateError } = await supabase
+      .from('fill_tasks')
+      .update({
+        status: 'completed',
+        result_path: `results/${taskId}.md` // Save as md for simplicity in this demo
+      })
+      .eq('id', taskId);
+
+    // Store result in storage
+    await supabase.storage
+      .from('document_storage')
+      .upload(`results/${taskId}.md`, new Blob([resultText], { type: 'text/markdown' }));
+
+    return new Response(JSON.stringify({ success: true, result: resultText }), {
+      headers: { ...corsHeaders, 'Content-Type': 'application/json' }
+    });
+
+  } catch (error) {
+    console.error('Error filling template:', error);
+    return new Response(JSON.stringify({ error: error.message }), {
+      status: 500,
+      headers: { ...corsHeaders, 'Content-Type': 'application/json' }
+    });
+  }
+});
--- a/frontend/supabase/functions/process-document/index.ts
+++ b/frontend/supabase/functions/process-document/index.ts
@@ -0,0 +1,146 @@
+import { serve } from 'https://deno.land/std@0.168.0/http/server.ts';
+import { createClient } from 'https://esm.sh/@supabase/supabase-js@2';
+import { corsHeaders } from '../_shared/cors.ts';
+import * as XLSX from 'https://esm.sh/xlsx@0.18.5';
+import * as mammoth from 'https://esm.sh/mammoth@1.5.1';
+
+serve(async (req) => {
+  if (req.method === 'OPTIONS') {
+    return new Response(null, { headers: corsHeaders });
+  }
+
+  try {
+    const { documentId } = await req.json();
+    if (!documentId) throw new Error('Missing documentId');
+
+    const supabaseUrl = Deno.env.get('SUPABASE_URL')!;
+    const supabaseKey = Deno.env.get('SUPABASE_SERVICE_ROLE_KEY')!;
+    const supabase = createClient(supabaseUrl, supabaseKey);
+
+    // Get document details
+    const { data: document, error: docError } = await supabase
+      .from('documents')
+      .select('*')
+      .eq('id', documentId)
+      .single();
+
+    if (docError || !document) throw new Error('Document not found');
+
+    // Download file
+    const { data: fileData, error: dlError } = await supabase.storage
+      .from('document_storage')
+      .download(document.storage_path);
+
+    if (dlError || !fileData) throw new Error('Failed to download file');
+
+    const arrayBuffer = await fileData.arrayBuffer();
+    const uint8Array = new Uint8Array(arrayBuffer);
+    let extractedText = '';
+
+    // Parse file based on type
+    const fileExt = document.type.toLowerCase();
+    if (fileExt === 'txt' || fileExt === 'md') {
+      extractedText = new TextDecoder().decode(uint8Array);
+    } else if (fileExt === 'docx') {
+      const result = await mammoth.extractRawText({ arrayBuffer });
+      extractedText = result.value;
+    } else if (fileExt === 'xlsx' || fileExt === 'xls') {
+      const workbook = XLSX.read(arrayBuffer, { type: 'array' });
+      extractedText = workbook.SheetNames.map(name => {
+        const sheet = workbook.Sheets[name];
+        return XLSX.utils.sheet_to_txt(sheet);
+      }).join('\n\n');
+    } else {
+      throw new Error(`Unsupported file type: ${fileExt}`);
+    }
+
+    if (!extractedText.trim()) throw new Error('Document is empty');
+
+    // Call MiniMax for entity extraction
+    const miniMaxApiKey = Deno.env.get('INTEGRATIONS_API_KEY');
+    const miniMaxResponse = await fetch(
+      'https://app-a6ww9j3ja3nl-api-Aa2PqMJnJGwL-gateway.appmiaoda.com/v1/text/chatcompletion_v2',
+      {
+        method: 'POST',
+        headers: {
+          'X-Gateway-Authorization': `Bearer ${miniMaxApiKey}`,
+          'Content-Type': 'application/json',
+        },
+        body: JSON.stringify({
+          model: 'MiniMax-M2.5',
+          messages: [
+            {
+              role: 'system',
+              content: '你是一个文档信息提取专家。请从提供的文档内容中提取关键实体信息（如姓名、日期、金额、项目名称、地址、关键指标等）。输出格式必须为 JSON 数组，包含 entity_type, entity_value, confidence 三个字段。'
+            },
+            {
+              role: 'user',
+              content: `文档内容如下：\n\n${extractedText.slice(0, 15000)}` // Limit text for token budget
+            }
+          ],
+          response_format: {
+            type: 'json_schema',
+            json_schema: {
+              name: 'extracted_entities',
+              schema: {
+                type: 'object',
+                properties: {
+                  entities: {
+                    type: 'array',
+                    items: {
+                      type: 'object',
+                      properties: {
+                        entity_type: { type: 'string' },
+                        entity_value: { type: 'string' },
+                        confidence: { type: 'number' }
+                      },
+                      required: ['entity_type', 'entity_value']
+                    }
+                  }
+                }
+              }
+            }
+          }
+        })
+      }
+    );
+
+    const miniMaxData = await miniMaxResponse.json();
+    if (!miniMaxResponse.ok) {
+      console.error('MiniMax Error:', miniMaxData);
+      throw new Error('MiniMax extraction failed');
+    }
+
+    const extractionResult = JSON.parse(miniMaxData.choices[0].message.content);
+    const entities = extractionResult.entities || [];
+
+    // Save entities
+    if (entities.length > 0) {
+      const entitiesToInsert = entities.map((e: any) => ({
+        document_id: documentId,
+        entity_type: e.entity_type,
+        entity_value: e.entity_value,
+        confidence: e.confidence || 1.0
+      }));
+
+      await supabase.from('extracted_entities').insert(entitiesToInsert);
+    }
+
+    // Update document
+    await supabase.from('documents').update({
+      content_text: extractedText,
+      status: 'completed'
+    }).eq('id', documentId);
+
+    return new Response(JSON.stringify({ success: true, entities }), {
+      headers: { ...corsHeaders, 'Content-Type': 'application/json' }
+    });
+
+  } catch (error) {
+    console.error('Error processing document:', error);
+    return new Response(JSON.stringify({ error: error.message }), {
+      status: 500,
+      headers: { ...corsHeaders, 'Content-Type': 'application/json' }
+    });
+  }
+});
--- a/frontend/supabase/migrations/00001_init_database_v1.sql
+++ b/frontend/supabase/migrations/00001_init_database_v1.sql
@@ -0,0 +1,141 @@
+-- Create user_role enum
+CREATE TYPE public.user_role AS ENUM ('user', 'admin');
+
+-- Create profiles table
+CREATE TABLE IF NOT EXISTS public.profiles (
+  id uuid PRIMARY KEY REFERENCES auth.users(id) ON DELETE CASCADE,
+  email text,
+  phone text,
+  role public.user_role DEFAULT 'user'::public.user_role,
+  created_at timestamp with time zone DEFAULT now()
+);
+
+-- Sync auth.users to profiles trigger
+CREATE OR REPLACE FUNCTION public.handle_new_user()
+RETURNS trigger
+LANGUAGE plpgsql
+SECURITY DEFINER SET search_path = public
+AS $$
+DECLARE
+  user_count int;
+BEGIN
+  SELECT COUNT(*) INTO user_count FROM profiles;
+  INSERT INTO public.profiles (id, email, phone, role)
+  VALUES (
+    NEW.id,
+    NEW.email,
+    NEW.phone,
+    CASE WHEN user_count = 0 THEN 'admin'::public.user_role ELSE 'user'::public.user_role END
+  );
+  RETURN NEW;
+END;
+$$;
+
+DROP TRIGGER IF EXISTS on_auth_user_confirmed ON auth.users;
+CREATE TRIGGER on_auth_user_confirmed
+  AFTER UPDATE ON auth.users
+  FOR EACH ROW
+  WHEN (OLD.confirmed_at IS NULL AND NEW.confirmed_at IS NOT NULL)
+  EXECUTE FUNCTION handle_new_user();
+
+-- Create documents table
+CREATE TABLE IF NOT EXISTS public.documents (
+  id uuid PRIMARY KEY DEFAULT gen_random_uuid(),
+  owner_id uuid REFERENCES public.profiles(id) ON DELETE CASCADE,
+  name text NOT NULL,
+  type text NOT NULL,
+  storage_path text NOT NULL,
+  content_text text,
+  metadata jsonb DEFAULT '{}'::jsonb,
+  status text DEFAULT 'pending',
+  created_at timestamp with time zone DEFAULT now()
+);
+
+-- Create extracted_entities table
+CREATE TABLE IF NOT EXISTS public.extracted_entities (
+  id uuid PRIMARY KEY DEFAULT gen_random_uuid(),
+  document_id uuid REFERENCES public.documents(id) ON DELETE CASCADE,
+  entity_type text NOT NULL,
+  entity_value text NOT NULL,
+  confidence float DEFAULT 1.0,
+  created_at timestamp with time zone DEFAULT now()
+);
+
+-- Create templates table
+CREATE TABLE IF NOT EXISTS public.templates (
+  id uuid PRIMARY KEY DEFAULT gen_random_uuid(),
+  owner_id uuid REFERENCES public.profiles(id) ON DELETE CASCADE,
+  name text NOT NULL,
+  type text NOT NULL,
+  storage_path text NOT NULL,
+  created_at timestamp with time zone DEFAULT now()
+);
+
+-- Create fill_tasks table
+CREATE TABLE IF NOT EXISTS public.fill_tasks (
+  id uuid PRIMARY KEY DEFAULT gen_random_uuid(),
+  owner_id uuid REFERENCES public.profiles(id) ON DELETE CASCADE,
+  template_id uuid REFERENCES public.templates(id) ON DELETE CASCADE,
+  document_ids uuid[] NOT NULL,
+  status text DEFAULT 'pending',
+  result_path text,
+  error_message text,
+  created_at timestamp with time zone DEFAULT now()
+);
+
+-- Enable RLS
+ALTER TABLE public.profiles ENABLE ROW LEVEL SECURITY;
+ALTER TABLE public.documents ENABLE ROW LEVEL SECURITY;
+ALTER TABLE public.extracted_entities ENABLE ROW LEVEL SECURITY;
+ALTER TABLE public.templates ENABLE ROW LEVEL SECURITY;
+ALTER TABLE public.fill_tasks ENABLE ROW LEVEL SECURITY;
+
+-- Helper function is_admin
+CREATE OR REPLACE FUNCTION is_admin(uid uuid)
+RETURNS boolean LANGUAGE sql SECURITY DEFINER AS $$
+  SELECT EXISTS (
+    SELECT 1 FROM profiles p
+    WHERE p.id = uid AND p.role = 'admin'::user_role
+  );
+$$;
+
+-- Policies
+-- Profiles: Admin full access, users view/update own (except role)
+CREATE POLICY "Admins have full access to profiles" ON profiles
+  FOR ALL TO authenticated USING (is_admin(auth.uid()));
+CREATE POLICY "Users can view their own profile" ON profiles
+  FOR SELECT TO authenticated USING (auth.uid() = id);
+CREATE POLICY "Users can update their own profile" ON profiles
+  FOR UPDATE TO authenticated USING (auth.uid() = id)
+  WITH CHECK (role IS NOT DISTINCT FROM (SELECT role FROM profiles WHERE id = auth.uid()));
+
+-- Documents: owner full access, admin view
+CREATE POLICY "Users can manage own documents" ON documents
+  FOR ALL TO authenticated USING (auth.uid() = owner_id);
+CREATE POLICY "Admins can view all documents" ON documents
+  FOR SELECT TO authenticated USING (is_admin(auth.uid()));
+
+-- Extracted Entities: owner (via document) full access
+CREATE POLICY "Users can manage entities of own docs" ON extracted_entities
+  FOR ALL TO authenticated USING (
+    EXISTS (SELECT 1 FROM documents d WHERE d.id = document_id AND d.owner_id = auth.uid())
+  );
+
+-- Templates: owner full access
+CREATE POLICY "Users can manage own templates" ON templates
+  FOR ALL TO authenticated USING (auth.uid() = owner_id);
+
+-- Fill Tasks: owner full access
+CREATE POLICY "Users can manage own tasks" ON fill_tasks
+  FOR ALL TO authenticated USING (auth.uid() = owner_id);
+
+-- Create storage buckets
+INSERT INTO storage.buckets (id, name, public) VALUES ('document_storage', 'document_storage', false) ON CONFLICT (id) DO NOTHING;
+
+-- Storage policies
+CREATE POLICY "Users can upload docs" ON storage.objects
+  FOR INSERT TO authenticated WITH CHECK (bucket_id = 'document_storage');
+CREATE POLICY "Users can view own docs" ON storage.objects
+  FOR SELECT TO authenticated USING (bucket_id = 'document_storage' AND (auth.uid()::text = (storage.foldername(name))[1]));
+CREATE POLICY "Users can delete own docs" ON storage.objects
+  FOR DELETE TO authenticated USING (bucket_id = 'document_storage' AND (auth.uid()::text = (storage.foldername(name))[1]));