perf: doc2x plugins (#3162)

2025-07-22 12:20:34 +00:00 · 2024-11-14 21:56:13 +08:00
parent be59c2f6a7
commit 3f72f88591
10 changed files with 158 additions and 285 deletions
--- a/packages/plugins/register.ts
+++ b/packages/plugins/register.ts
@@ -87,7 +87,3 @@ export const getCommunityCb = async () => {
    {}
  );
 };
-
-export const getSystemPluginCb = async () => {
-  return global.systemPluginCb;
-};
--- a/packages/plugins/src/Doc2X/PDF2text/index.ts
+++ b/packages/plugins/src/Doc2X/PDF2text/index.ts
@@ -1,108 +1,79 @@
 import { delay } from '@fastgpt/global/common/system/utils';
 import axios from 'axios';
-import { addLog } from '@fastgpt/service/common/system/log';
-import { result } from 'lodash';
+import { getErrText } from '@fastgpt/global/common/error/utils';

 type Props = {
  apikey: string;
-  files: any;
-  ocr: boolean;
+  files: string[];
 };

 // Response type same as HTTP outputs
 type Response = Promise<{
  result: string;
-  failreason: string;
  success: boolean;
+  error?: Record<string, any>;
 }>;

 const main = async ({ apikey, files }: Props): Response => {
  // Check the apikey
  if (!apikey) {
-    return {
-      result: '',
-      failreason: `API key is required`,
-      success: false
-    };
-  }
-  let final_result = '';
-  let fail_reason = '';
-  let flag = false;
-  //Convert the String to Array<String> or String
-  let All_URL: Array<string>;
-  try {
-    const parsed = JSON.parse(files);
-    if (Array.isArray(parsed)) {
-      All_URL = parsed;
-    } else {
-      All_URL = [String(parsed)];
-    }
-  } catch (e) {
-    // Set it as String
-    All_URL = [String(files)];
+    return Promise.reject(`API key is required`);
  }
+  const successResult = [];
+  const failedResult = [];
+
  const axiosInstance = axios.create({
    timeout: 30000 // 30 seconds timeout
  });

  //Process each file one by one
-  for await (const url of All_URL) {
-    //Fetch the pdf and check its content type
-    let PDFResponse;
+  for await (const url of files) {
    try {
-      PDFResponse = await axiosInstance.get(url, { responseType: 'arraybuffer' });
-    } catch (e) {
-      fail_reason += `\n---\nFile:${url} \n<Content>\nFailed to fetch image from URL: ${e}\n</Content>\n`;
-      flag = true;
-      continue;
-    }
-    if (PDFResponse.status !== 200) {
-      fail_reason += `\n---\nFile:${url} \n<Content>\nFailed to fetch PDF from URL: ${PDFResponse.statusText}\n</Content>\n`;
-      flag = true;
-      continue;
-    }
+      //Fetch the pdf and check its content type
+      const PDFResponse = await axiosInstance.get(url, { responseType: 'arraybuffer' });
+      if (PDFResponse.status !== 200) {
+        throw new Error(
+          `File:${url} \n<Content>\nFailed to fetch PDF from URL: ${PDFResponse.statusText}\n</Content>`
+        );
+      }

-    const contentType = PDFResponse.headers['content-type'];
-    const file_name = url.match(/read\/([^?]+)/)?.[1] || 'unknown.pdf';
-    if (!contentType || !contentType.startsWith('application/pdf')) {
-      fail_reason += `\n---\nFile:${file_name}\n<Content>\nThe provided file does not point to a PDF: ${contentType}\n</Content>\n`;
-      flag = true;
-      continue;
-    }
-    const blob = new Blob([PDFResponse.data], { type: 'application/pdf' });
+      const contentType = PDFResponse.headers['content-type'];
+      const file_name = url.match(/read\/([^?]+)/)?.[1] || 'unknown.pdf';
+      if (!contentType || !contentType.startsWith('application/pdf')) {
+        throw new Error(
+          `File:${file_name}\n<Content>\nThe provided file does not point to a PDF: ${contentType}\n</Content>`
+        );
+      }

-    // Get pre-upload URL first
-    let preupload_url = 'https://v2.doc2x.noedgeai.com/api/v2/parse/preupload';
-    let preupload_response;
-    try {
-      preupload_response = await axiosInstance.post(preupload_url, null, {
-        headers: {
-          Authorization: `Bearer ${apikey}`
+      const blob = new Blob([PDFResponse.data], { type: 'application/pdf' });
+      // Get pre-upload URL first
+      const preupload_response = await axiosInstance.post(
+        'https://v2.doc2x.noedgeai.com/api/v2/parse/preupload',
+        null,
+        {
+          headers: {
+            Authorization: `Bearer ${apikey}`
+          }
        }
-      });
-    } catch (e) {
-      fail_reason += `\n---\nFile:${file_name}\n<Content>\nFailed to get pre-upload URL: ${e}\n</Content>\n`;
-      flag = true;
-      continue;
-    }
+      );

-    if (preupload_response.status !== 200) {
-      fail_reason += `\n---\nFile:${file_name}\n<Content>\nFailed to get pre-upload URL: ${preupload_response.statusText}\n</Content>\n`;
-      flag = true;
-      continue;
-    }
+      if (preupload_response.status !== 200) {
+        throw new Error(
+          `File:${file_name}\n<Content>\nFailed to get pre-upload URL: ${preupload_response.statusText}\n</Content>`
+        );
+      }

-    const preupload_data = preupload_response.data;
-    if (preupload_data.code !== 'success') {
-      fail_reason += `\n---\nFile:${file_name}\n<Content>\nFailed to get pre-upload URL: ${JSON.stringify(preupload_data)}\n</Content>\n`;
-      flag = true;
-      continue;
-    }
+      const preupload_data = preupload_response.data;
+      if (preupload_data.code !== 'success') {
+        throw new Error(
+          `File:${file_name}\n<Content>\nFailed to get pre-upload URL: ${JSON.stringify(preupload_data)}\n</Content>`
+        );
+      }
+
+      const upload_url = preupload_data.data.url;
+      const uid = preupload_data.data.uid;
+      // Upload file to pre-signed URL with binary stream

-    const upload_url = preupload_data.data.url;
-    const uid = preupload_data.data.uid;
-    // Upload file to pre-signed URL with binary stream
-    try {
      const response = await axiosInstance.put(upload_url, blob, {
        headers: {
          'Content-Type': 'application/pdf'
@@ -111,71 +82,75 @@ const main = async ({ apikey, files }: Props): Response => {
      if (response.status !== 200) {
        throw new Error(`Upload failed with status ${response.status}: ${response.statusText}`);
      }
-    } catch (e) {
-      fail_reason += `\n---\nFile:${file_name}\n<Content>\nFailed to upload file (uid: ${uid}): ${e}\n</Content>\n`;
-      flag = true;
-      continue;
-    }

-    // Get the result by uid
-    const result_url = `https://v2.doc2x.noedgeai.com/api/v2/parse/status?uid=${uid}`;
-    let required_flag = true;
-    let result = '';
+      // Get the result by uid

-    // Wait for the result, at most 90s
-    const maxAttempts = 30;
-    for await (const _ of Array(maxAttempts).keys()) {
-      let result_response;
-      try {
-        result_response = await axiosInstance.get(result_url, {
-          headers: {
-            Authorization: `Bearer ${apikey}`
+      // Wait for the result, at most 90s
+      const checkResult = async (retry = 30) => {
+        if (retry <= 0)
+          return Promise.reject(
+            `File:${file_name}\n<Content>\nFailed to get result (uid: ${uid}): Get result timeout\n</Content>`
+          );
+
+        try {
+          const result_response = await axiosInstance.get(
+            `https://v2.doc2x.noedgeai.com/api/v2/parse/status?uid=${uid}`,
+            {
+              headers: {
+                Authorization: `Bearer ${apikey}`
+              }
+            }
+          );
+
+          const result_data = result_response.data;
+          if (!['ok', 'success'].includes(result_data.code)) {
+            return Promise.reject(
+              `File:${file_name}\n<Content>\nFailed to get result (uid: ${uid}): ${JSON.stringify(result_data)}\n</Content>`
+            );
          }
-        });
-      } catch (e) {
-        fail_reason += `\n---\nFile:${file_name}\n<Content>\nFailed to get result (uid: ${uid}): ${e}\n</Content>\n`;
-        flag = true;
-        required_flag = false;
-        break;
-      }

-      const result_data = result_response.data;
-      if (!['ok', 'success'].includes(result_data.code)) {
-        fail_reason += `\n---\nFile:${file_name}\n<Content>\nFailed to get result (uid: ${uid}): ${JSON.stringify(result_data)}\n</Content>\n`;
-        flag = true;
-        required_flag = false;
-        break;
-      }
-      if (['ready', 'processing'].includes(result_data.data.status)) {
-        await delay(3000);
-      } else if (result_data.data.status === 'success') {
-        result = await Promise.all(
-          result_data.data.result.pages.map((page: { md: any }) => page.md)
-        ).then((pages) => pages.join('\n'));
-        // Do some post-processing
-        result = result.replace(/\\[\(\)]/g, '$').replace(/\\[\[\]]/g, '$$');
-        result = result.replace(/<img\s+src="([^"]+)"(?:\s*\?[^>]*)?(?:\s*\/>|>)/g, '![img]($1)');
-        final_result += `\n---\nFile:${file_name}\n<Content>\n${result}\n</Content>\n`;
-        required_flag = false;
-        break;
-      } else {
-        fail_reason += `\n---\nFile:${file_name}\n<Content>\nFailed to get result (uid: ${uid}): ${result_data.data.status}\n</Content>\n`;
-        flag = true;
-        required_flag = false;
-        break;
-      }
-    }
+          if (['ready', 'processing'].includes(result_data.data.status)) {
+            await delay(3000);
+            return checkResult(retry - 1);
+          }

-    if (required_flag) {
-      fail_reason += `\n---\nFile:${file_name}\n<Content>\nTimeout for uid ${uid}\n</Content>\n`;
-      flag = true;
+          if (result_data.data.status === 'success') {
+            const result = (
+              await Promise.all(
+                result_data.data.result.pages.map((page: { md: any }) => page.md)
+              ).then((pages) => pages.join('\n'))
+            )
+              // Do some post-processing
+              .replace(/\\[\(\)]/g, '$')
+              .replace(/\\[\[\]]/g, '$$')
+              .replace(/<img\s+src="([^"]+)"(?:\s*\?[^>]*)?(?:\s*\/>|>)/g, '![img]($1)');
+
+            return `File:${file_name}\n<Content>\n${result}\n</Content>`;
+          }
+
+          await delay(100);
+          return checkResult(retry - 1);
+        } catch (error) {
+          await delay(100);
+          return checkResult(retry - 1);
+        }
+      };
+
+      const result = await checkResult();
+      successResult.push(result);
+    } catch (error) {
+      failedResult.push(
+        `File:${url} \n<Content>\nFailed to fetch image from URL: ${getErrText(error)}\n</Content>`
+      );
    }
  }

  return {
-    result: final_result,
-    failreason: fail_reason,
-    success: !flag
+    result: successResult.join('\n******\n'),
+    error: {
+      message: failedResult.join('\n******\n')
+    },
+    success: failedResult.length === 0
  };
 };

--- a/packages/plugins/src/Doc2X/PDF2text/template.json
+++ b/packages/plugins/src/Doc2X/PDF2text/template.json
@@ -1,10 +1,10 @@
 {
  "author": "Menghuan1918",
  "version": "488",
-  "name": "Doc2X PDF识别",
+  "name": "PDF识别",
  "avatar": "plugins/doc2x",
  "intro": "将PDF文件发送至Doc2X进行解析，返回结构化的LaTeX公式的文本(markdown)，支持传入String类型的URL或者流程输出中的文件链接变量",
-  "inputExplanationUrl": "https://fael3z0zfze.feishu.cn/wiki/Rkc5witXWiJoi5kORd2cofh6nDg?fromScene=spaceOverview",
+  "courseUrl": "https://fael3z0zfze.feishu.cn/wiki/Rkc5witXWiJoi5kORd2cofh6nDg?fromScene=spaceOverview",
  "showStatus": true,
  "weight": 10,

@@ -21,8 +21,8 @@
        "flowNodeType": "pluginInput",
        "showStatus": false,
        "position": {
-          "x": -139.66495007440972,
-          "y": -90.99689735553712
+          "x": -137.96875104510553,
+          "y": -90.9968973555371
        },
        "version": "481",
        "inputs": [
@@ -39,16 +39,19 @@
            "list": []
          },
          {
-            "renderTypeList": ["reference"],
+            "renderTypeList": ["fileSelect"],
            "selectedTypeIndex": 0,
-            "valueType": "any",
+            "valueType": "arrayString",
            "canEdit": true,
            "key": "files",
            "label": "files",
-            "description": "待处理的PDF文件变量或URL地址",
+            "description": "需要处理的PDF地址",
            "required": true,
-            "toolDescription": "待处理的PDF文件变量或URL地址",
-            "list": []
+            "list": [],
+            "canSelectFile": true,
+            "canSelectImg": false,
+            "maxFiles": 14,
+            "defaultValue": ""
          }
        ],
        "outputs": [
@@ -61,7 +64,7 @@
          },
          {
            "id": "url",
-            "valueType": "any",
+            "valueType": "arrayString",
            "key": "files",
            "label": "files",
            "type": "hidden"
@@ -76,8 +79,8 @@
        "flowNodeType": "pluginOutput",
        "showStatus": false,
        "position": {
-          "x": 1808.5347800638815,
-          "y": -105.67504356429907
+          "x": 1505.494975310334,
+          "y": -4.14668564643415
        },
        "version": "481",
        "inputs": [
@@ -92,12 +95,13 @@
          },
          {
            "renderTypeList": ["reference"],
-            "valueType": "string",
+            "valueType": "object",
            "canEdit": true,
-            "key": "failreason",
-            "label": "failreason",
-            "description": "文件处理失败原因，由文件名以及报错组成，多个文件之间由横线分隔开，如所有文件处理成功则为空",
-            "value": ["zHG5jJBkXmjB", "yDxzW5CFalGw"]
+            "key": "error",
+            "label": "error",
+            "description": "",
+            "value": ["zHG5jJBkXmjB", "httpRawResponse"],
+            "isToolOutput": true
          },
          {
            "renderTypeList": ["reference"],
@@ -106,7 +110,8 @@
            "key": "success",
            "label": "success",
            "description": "是否全部文件都处理成功，如有没有处理成功的文件，失败原因将会输出在failreason中",
-            "value": ["zHG5jJBkXmjB", "m6CJJj7GFud5"]
+            "value": ["zHG5jJBkXmjB", "m6CJJj7GFud5"],
+            "isToolOutput": false
          }
        ],
        "outputs": []
@@ -119,8 +124,8 @@
        "flowNodeType": "httpRequest468",
        "showStatus": true,
        "position": {
-          "x": 1077.7986740892777,
-          "y": -496.9521622173004
+          "x": 619.0661933308237,
+          "y": -472.91377894611503
        },
        "version": "481",
        "inputs": [
@@ -170,7 +175,7 @@
            "renderTypeList": ["custom"],
            "valueType": "number",
            "label": "",
-            "value": 30,
+            "value": 300,
            "min": 5,
            "max": 600,
            "required": true,
@@ -215,7 +220,7 @@
            "key": "system_httpJsonBody",
            "renderTypeList": ["hidden"],
            "valueType": "any",
-            "value": "{\n  \"apikey\": \"{{apikey}}\",\n  \"files\": \"{{files}}\"}",
+            "value": "{\n  \"apikey\": \"{{apikey}}\",\n  \"files\": {{files}}\n}",
            "label": "",
            "required": false,
            "debugLabel": "",
@@ -273,7 +278,7 @@
          },
          {
            "renderTypeList": ["reference"],
-            "valueType": "string",
+            "valueType": "arrayString",
            "canEdit": true,
            "key": "files",
            "label": "files",
@@ -299,7 +304,7 @@
              "showDefaultValue": true
            },
            "required": true,
-            "value": ["pMBi7J7vcsqB", "system_text"]
+            "value": [["pluginInput", "url"]]
          }
        ],
        "outputs": [
@@ -360,109 +365,6 @@
            "type": "dynamic",
            "key": "success",
            "label": "success"
-          },
-          {
-            "id": "yDxzW5CFalGw",
-            "valueType": "string",
-            "type": "dynamic",
-            "key": "failreason",
-            "label": "failreason"
-          }
-        ]
-      },
-      {
-        "nodeId": "pMBi7J7vcsqB",
-        "name": "文本拼接",
-        "intro": "可对固定或传入的文本进行加工后输出，非字符串类型数据最终会转成字符串类型。",
-        "avatar": "core/workflow/template/textConcat",
-        "flowNodeType": "textEditor",
-        "position": {
-          "x": 469.8489508985863,
-          "y": -177.67504356429907
-        },
-        "version": "486",
-        "inputs": [
-          {
-            "key": "system_addInputParam",
-            "renderTypeList": ["addInputParam"],
-            "valueType": "dynamic",
-            "label": "",
-            "required": false,
-            "description": "workflow:dynamic_input_description_concat",
-            "customInputConfig": {
-              "selectValueTypeList": [
-                "string",
-                "number",
-                "boolean",
-                "object",
-                "arrayString",
-                "arrayNumber",
-                "arrayBoolean",
-                "arrayObject",
-                "arrayAny",
-                "any",
-                "chatHistory",
-                "datasetQuote",
-                "dynamic",
-                "selectApp",
-                "selectDataset"
-              ],
-              "showDescription": false,
-              "showDefaultValue": false
-            },
-            "debugLabel": "",
-            "toolDescription": ""
-          },
-          {
-            "key": "system_textareaInput",
-            "renderTypeList": ["textarea"],
-            "valueType": "string",
-            "required": true,
-            "label": "拼接文本",
-            "placeholder": "workflow:input_variable_list",
-            "value": "{{files}}",
-            "debugLabel": "",
-            "toolDescription": ""
-          },
-          {
-            "renderTypeList": ["reference"],
-            "valueType": "any",
-            "canEdit": true,
-            "key": "files",
-            "label": "files",
-            "customInputConfig": {
-              "selectValueTypeList": [
-                "string",
-                "number",
-                "boolean",
-                "object",
-                "arrayString",
-                "arrayNumber",
-                "arrayBoolean",
-                "arrayObject",
-                "arrayAny",
-                "any",
-                "chatHistory",
-                "datasetQuote",
-                "dynamic",
-                "selectApp",
-                "selectDataset"
-              ],
-              "showDescription": false,
-              "showDefaultValue": false
-            },
-            "required": true,
-            "value": ["pluginInput", "url"]
-          }
-        ],
-        "outputs": [
-          {
-            "id": "system_text",
-            "key": "system_text",
-            "label": "workflow:concatenation_result",
-            "type": "static",
-            "valueType": "string",
-            "description": ""
          }
        ]
      }
@@ -476,14 +378,8 @@
      },
      {
        "source": "pluginInput",
-        "target": "pMBi7J7vcsqB",
-        "sourceHandle": "pluginInput-source-right",
-        "targetHandle": "pMBi7J7vcsqB-target-left"
-      },
-      {
-        "source": "pMBi7J7vcsqB",
        "target": "zHG5jJBkXmjB",
-        "sourceHandle": "pMBi7J7vcsqB-source-right",
+        "sourceHandle": "pluginInput-source-right",
        "targetHandle": "zHG5jJBkXmjB-target-left"
      }
    ],
--- a/packages/service/core/workflow/dispatch/tools/http468.ts
+++ b/packages/service/core/workflow/dispatch/tools/http468.ts
@@ -18,7 +18,6 @@ import {
  textAdaptGptResponse,
  replaceEditorVariable
 } from '@fastgpt/global/core/workflow/runtime/utils';
-import { getSystemPluginCb } from '../../../../../plugins/register';
 import { ContentTypes } from '@fastgpt/global/core/workflow/constants';
 import { uploadFileFromBase64Img } from '../../../../common/file/gridfs/controller';
 import { ReadFileBaseUrl } from '@fastgpt/global/common/file/constants';
@@ -209,7 +208,8 @@ export const dispatchHttp468Request = async (props: HttpRequestProps): Promise<H

  try {
    const { formatResponse, rawResponse } = await (async () => {
-      const systemPluginCb = await getSystemPluginCb();
+      const systemPluginCb = global.systemPluginCb;
+      console.log(systemPluginCb, '-=', httpReqUrl);
      if (systemPluginCb[httpReqUrl]) {
        const pluginResult = await replaceSystemPluginResponse({
          response: await systemPluginCb[httpReqUrl](requestBody),
--- a/projects/app/src/pages/app/detail/components/WorkflowComponents/Flow/nodes/NodePluginIO/InputEditModal.tsx
+++ b/projects/app/src/pages/app/detail/components/WorkflowComponents/Flow/nodes/NodePluginIO/InputEditModal.tsx
@@ -215,6 +215,7 @@ const FieldEditModal = ({
  );
  const onSubmitError = useCallback(
    (e: Object) => {
+      console.log('e', e);
      for (const item of Object.values(e)) {
        if (item.message) {
          toast({
--- a/projects/app/src/pages/app/detail/components/WorkflowComponents/Flow/nodes/NodePluginIO/InputTypeConfig.tsx
+++ b/projects/app/src/pages/app/detail/components/WorkflowComponents/Flow/nodes/NodePluginIO/InputTypeConfig.tsx
@@ -511,22 +511,14 @@ const InputTypeConfig = ({
              <FormLabel flex={'0 0 132px'} fontWeight={'medium'}>
                {t('app:document_upload')}
              </FormLabel>
-              <Switch
-                {...register('canSelectFile', {
-                  required: true
-                })}
-              />
+              <Switch {...register('canSelectFile')} />
            </Flex>
            <Box w={'full'} minH={'40px'}>
              <Flex alignItems={'center'}>
                <FormLabel flex={'0 0 132px'} fontWeight={'medium'}>
                  {t('app:image_upload')}
                </FormLabel>
-                <Switch
-                  {...register('canSelectImg', {
-                    required: true
-                  })}
-                />
+                <Switch {...register('canSelectImg')} />
              </Flex>
              <Flex color={'myGray.500'}>
                <Box fontSize={'xs'}>{t('app:image_upload_tip')}</Box>
--- a/projects/app/src/pages/app/detail/components/WorkflowComponents/Flow/nodes/render/NodeCard.tsx
+++ b/projects/app/src/pages/app/detail/components/WorkflowComponents/Flow/nodes/render/NodeCard.tsx
@@ -115,6 +115,16 @@ const NodeCard = (props: Props) => {
      }
    },
    {
+      onSuccess(res) {
+        if (!res) return;
+        // Execute forcibly updates the courseUrl field
+        onChangeNode({
+          nodeId,
+          type: 'attr',
+          key: 'courseUrl',
+          value: res?.courseUrl
+        });
+      },
      manual: false
    }
  );
--- a/projects/app/src/service/common/system/volumnMongoWatch.ts
+++ b/projects/app/src/service/common/system/volumnMongoWatch.ts
@@ -1,4 +1,4 @@
-import { getSystemPlugins } from '@/service/core/app/plugin';
+import { getSystemPluginCb, getSystemPlugins } from '@/service/core/app/plugin';
 import { initSystemConfig } from '.';
 import { createDatasetTrainingMongoWatch } from '@/service/core/dataset/training/utils';
 import { MongoSystemConfigs } from '@fastgpt/service/common/system/config/schema';
--- a/projects/app/src/service/core/app/plugin.ts
+++ b/projects/app/src/service/core/app/plugin.ts
@@ -22,6 +22,8 @@ export const getSystemPlugins = async (refresh = false) => {

    addLog.info(`Load system plugin successfully: ${global.systemPlugins.length}`);

+    getSystemPluginCb();
+
    return cloneDeep(global.systemPlugins);
  } catch (error) {
    //@ts-ignore
--- a/projects/app/src/service/mongo.ts
+++ b/projects/app/src/service/mongo.ts
@@ -12,7 +12,7 @@ import { startMongoWatch } from './common/system/volumnMongoWatch';
 import { startTrainingQueue } from './core/dataset/training/utils';
 import { systemStartCb } from '@fastgpt/service/common/system/tools';
 import { addLog } from '@fastgpt/service/common/system/log';
-import { getSystemPluginCb } from './core/app/plugin';
+import { getSystemPlugins } from './core/app/plugin';

 /**
 * This function is equivalent to the entry to the service
@@ -32,8 +32,9 @@ export function connectToDatabase() {
      systemStartCb();

      //init system config；init vector database；init root user
-      await Promise.all([getInitConfig(), getSystemPluginCb(), initVectorStore(), initRootUser()]);
+      await Promise.all([getInitConfig(), initVectorStore(), initRootUser()]);

+      getSystemPlugins();
      startMongoWatch();
      // cron
      startCron();