Fix：Added support for preview of txt, md, excel, csv, ppt, image, doc and other files (#8712)

### What problem does this PR solve?

Added support for preview of txt, md, excel, csv, ppt, image, doc and
other files [#3221](https://github.com/infiniflow/ragflow/issues/3221)

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

web/package.json

@@ -81,6 +81,7 @@
     "mammoth": "^1.7.2",
     "next-themes": "^0.4.6",
     "openai-speech-stream-player": "^1.0.8",
+    "pptx-preview": "^1.0.5",
     "rc-tween-one": "^3.0.6",
     "react-copy-to-clipboard": "^5.1.0",
     "react-dropzone": "^14.3.5",

web/src/pages/chunk/parsed-result/add-knowledge/components/knowledge-chunk/components/document-preview/csv-preview.tsx

@@ -0,0 +1,114 @@
+import message from '@/components/ui/message';
+import { Spin } from '@/components/ui/spin';
+import request from '@/utils/request';
+import classNames from 'classnames';
+import React, { useEffect, useRef, useState } from 'react';
+import { useGetDocumentUrl } from './hooks';
+
+interface CSVData {
+  rows: string[][];
+  headers: string[];
+}
+
+interface FileViewerProps {
+  className?: string;
+}
+
+const CSVFileViewer: React.FC<FileViewerProps> = () => {
+  const [data, setData] = useState<CSVData | null>(null);
+  const [isLoading, setIsLoading] = useState<boolean>(true);
+  const containerRef = useRef<HTMLDivElement>(null);
+  const url = useGetDocumentUrl();
+  const parseCSV = (csvText: string): CSVData => {
+    console.log('Parsing CSV data:', csvText);
+    const lines = csvText.split('\n');
+    const headers = lines[0].split(',').map((header) => header.trim());
+    const rows = lines
+      .slice(1)
+      .map((line) => line.split(',').map((cell) => cell.trim()));
+
+    return { headers, rows };
+  };
+
+  useEffect(() => {
+    const loadCSV = async () => {
+      try {
+        const res = await request(url, {
+          method: 'GET',
+          responseType: 'blob',
+          onError: (err) => {
+            message.error('file load failed');
+            setIsLoading(false);
+          },
+        });
+
+        // parse CSV file
+        const reader = new FileReader();
+        reader.readAsText(res.data);
+        reader.onload = () => {
+          const parsedData = parseCSV(reader.result as string);
+          console.log('file loaded successfully', reader.result);
+          setData(parsedData);
+        };
+      } catch (error) {
+        message.error('CSV file parse failed');
+        console.error('Error loading CSV file:', error);
+      } finally {
+        setIsLoading(false);
+      }
+    };
+
+    loadCSV();
+
+    return () => {
+      setData(null);
+    };
+  }, [url]);
+
+  return (
+    <div
+      ref={containerRef}
+      className={classNames(
+        'relative w-full h-full p-4 bg-background-paper border border-border-normal rounded-md',
+        'overflow-auto max-h-[80vh] p-2',
+      )}
+    >
+      {isLoading ? (
+        <div className="absolute inset-0 flex items-center justify-center">
+          <Spin />
+        </div>
+      ) : data ? (
+        <table className="min-w-full divide-y divide-border-normal">
+          <thead className="bg-background-header-bar">
+            <tr>
+              {data.headers.map((header, index) => (
+                <th
+                  key={`header-${index}`}
+                  className="px-6 py-3 text-left text-sm font-medium text-text-title"
+                >
+                  {header}
+                </th>
+              ))}
+            </tr>
+          </thead>
+          <tbody className="bg-background-paper divide-y divide-border-normal">
+            {data.rows.map((row, rowIndex) => (
+              <tr key={`row-${rowIndex}`}>
+                {row.map((cell, cellIndex) => (
+                  <td
+                    key={`cell-${rowIndex}-${cellIndex}`}
+                    className="px-6 py-4 whitespace-nowrap text-sm text-text-secondary"
+                  >
+                    {cell || '-'}
+                  </td>
+                ))}
+              </tr>
+            ))}
+          </tbody>
+        </table>
+      ) : null}
+    </div>
+  );
+};
+
+export default CSVFileViewer;

web/src/pages/chunk/parsed-result/add-knowledge/components/knowledge-chunk/components/document-preview/doc-preview.tsx

@@ -0,0 +1,67 @@
+import message from '@/components/ui/message';
+import { Spin } from '@/components/ui/spin';
+import request from '@/utils/request';
+import classNames from 'classnames';
+import mammoth from 'mammoth';
+import { useEffect, useState } from 'react';
+import { useGetDocumentUrl } from './hooks';
+
+interface DocPreviewerProps {
+  className?: string;
+}
+
+export const DocPreviewer: React.FC<DocPreviewerProps> = ({ className }) => {
+  const url = useGetDocumentUrl();
+  const [htmlContent, setHtmlContent] = useState<string>('');
+  const [loading, setLoading] = useState(false);
+  const fetchDocument = async () => {
+    setLoading(true);
+    const res = await request(url, {
+      method: 'GET',
+      responseType: 'blob',
+      onError: () => {
+        message.error('Document parsing failed');
+        console.error('Error loading document:', url);
+      },
+    });
+    try {
+      const arrayBuffer = await res.data.arrayBuffer();
+      const result = await mammoth.convertToHtml(
+        { arrayBuffer },
+        { includeDefaultStyleMap: true },
+      );
+
+      const styledContent = result.value
+        .replace(/<p>/g, '<p class="mb-2">')
+        .replace(/<h(\d)>/g, '<h$1 class="font-semibold mt-4 mb-2">');
+
+      setHtmlContent(styledContent);
+    } catch (err) {
+      message.error('Document parsing failed');
+      console.error('Error parsing document:', err);
+    }
+    setLoading(false);
+  };
+
+  useEffect(() => {
+    if (url) {
+      fetchDocument();
+    }
+  }, [url]);
+  return (
+    <div
+      className={classNames(
+        'relative w-full h-full p-4 bg-background-paper border border-border-normal rounded-md',
+        className,
+      )}
+    >
+      {loading && (
+        <div className="absolute inset-0 flex items-center justify-center">
+          <Spin />
+        </div>
+      )}
+
+      {!loading && <div dangerouslySetInnerHTML={{ __html: htmlContent }} />}
+    </div>
+  );
+};

web/src/pages/chunk/parsed-result/add-knowledge/components/knowledge-chunk/components/document-preview/excel-preview.tsx

@@ -0,0 +1,24 @@
+import { useFetchExcel } from '@/pages/document-viewer/hooks';
+import classNames from 'classnames';
+import { useGetDocumentUrl } from './hooks';
+
+interface ExcelCsvPreviewerProps {
+  className?: string;
+}
+
+export const ExcelCsvPreviewer: React.FC<ExcelCsvPreviewerProps> = ({
+  className,
+}) => {
+  const url = useGetDocumentUrl();
+  const { containerRef } = useFetchExcel(url);
+
+  return (
+    <div
+      ref={containerRef}
+      className={classNames(
+        'relative w-full h-full p-4 bg-background-paper border border-border-normal rounded-md excel-csv-previewer',
+        className,
+      )}
+    ></div>
+  );
+};

web/src/pages/chunk/parsed-result/add-knowledge/components/knowledge-chunk/components/document-preview/image-preview.tsx

@@ -0,0 +1,72 @@
+import message from '@/components/ui/message';
+import { Spin } from '@/components/ui/spin';
+import request from '@/utils/request';
+import classNames from 'classnames';
+import { useEffect, useState } from 'react';
+import { useGetDocumentUrl } from './hooks';
+
+interface ImagePreviewerProps {
+  className?: string;
+}
+
+export const ImagePreviewer: React.FC<ImagePreviewerProps> = ({
+  className,
+}) => {
+  const url = useGetDocumentUrl();
+  const [imageSrc, setImageSrc] = useState<string | null>(null);
+  const [isLoading, setIsLoading] = useState<boolean>(true);
+
+  const fetchImage = async () => {
+    setIsLoading(true);
+    const res = await request(url, {
+      method: 'GET',
+      responseType: 'blob',
+      onError: () => {
+        message.error('Failed to load image');
+        setIsLoading(false);
+      },
+    });
+    const objectUrl = URL.createObjectURL(res.data);
+    setImageSrc(objectUrl);
+    setIsLoading(false);
+  };
+  useEffect(() => {
+    if (url) {
+      fetchImage();
+    }
+  }, [url]);
+
+  useEffect(() => {
+    return () => {
+      if (imageSrc) {
+        URL.revokeObjectURL(imageSrc);
+      }
+    };
+  }, [imageSrc]);
+
+  return (
+    <div
+      className={classNames(
+        'relative w-full h-full p-4 bg-background-paper border border-border-normal rounded-md image-previewer',
+        className,
+      )}
+    >
+      {isLoading && (
+        <div className="absolute inset-0 flex items-center justify-center">
+          <Spin />
+        </div>
+      )}
+
+      {!isLoading && imageSrc && (
+        <div className="max-h-[80vh] overflow-auto p-2">
+          <img
+            src={imageSrc}
+            alt={'image'}
+            className="w-full h-auto max-w-full object-contain"
+            onLoad={() => URL.revokeObjectURL(imageSrc!)}
+          />
+        </div>
+      )}
+    </div>
+  );
+};

web/src/pages/chunk/parsed-result/add-knowledge/components/knowledge-chunk/components/document-preview/index.tsx

@@ -0,0 +1,65 @@
+import { memo } from 'react';
+
+import CSVFileViewer from './csv-preview';
+import { DocPreviewer } from './doc-preview';
+import { ExcelCsvPreviewer } from './excel-preview';
+import { ImagePreviewer } from './image-preview';
+import styles from './index.less';
+import PdfPreviewer, { IProps } from './pdf-preview';
+import { PptPreviewer } from './ppt-preview';
+import { TxtPreviewer } from './txt-preview';
+
+type PreviewProps = {
+  fileType: string;
+  className?: string;
+};
+const Preview = ({
+  fileType,
+  className,
+  highlights,
+  setWidthAndHeight,
+}: PreviewProps & Partial<IProps>) => {
+  return (
+    <>
+      {fileType === 'pdf' && highlights && setWidthAndHeight && (
+        <section className={styles.documentPreview}>
+          <PdfPreviewer
+            highlights={highlights}
+            setWidthAndHeight={setWidthAndHeight}
+          ></PdfPreviewer>
+        </section>
+      )}
+      {['doc', 'docx'].indexOf(fileType) > -1 && (
+        <section>
+          <DocPreviewer className={className} />
+        </section>
+      )}
+      {['txt', 'md'].indexOf(fileType) > -1 && (
+        <section>
+          <TxtPreviewer className={className} />
+        </section>
+      )}
+      {['visual'].indexOf(fileType) > -1 && (
+        <section>
+          <ImagePreviewer className={className} />
+        </section>
+      )}
+      {['pptx'].indexOf(fileType) > -1 && (
+        <section>
+          <PptPreviewer className={className} />
+        </section>
+      )}
+      {['xlsx'].indexOf(fileType) > -1 && (
+        <section>
+          <ExcelCsvPreviewer className={className} />
+        </section>
+      )}
+      {['csv'].indexOf(fileType) > -1 && (
+        <section>
+          <CSVFileViewer className={className} />
+        </section>
+      )}
+    </>
+  );
+};
+export default memo(Preview);

web/src/pages/chunk/parsed-result/add-knowledge/components/knowledge-chunk/components/document-preview/pdf-preview.tsx

@@ -14,7 +14,7 @@ import { useCatchDocumentError } from '@/components/pdf-previewer/hooks';
 import FileError from '@/pages/document-viewer/file-error';
 import styles from './index.less';
 
-interface IProps {
+export interface IProps {
   highlights: IHighlight[];
   setWidthAndHeight: (width: number, height: number) => void;
 }
@@ -30,7 +30,7 @@ const HighlightPopup = ({
   ) : null;
 
 // TODO: merge with DocumentPreviewer
-const Preview = ({ highlights: state, setWidthAndHeight }: IProps) => {
+const PdfPreview = ({ highlights: state, setWidthAndHeight }: IProps) => {
   const url = useGetDocumentUrl();
 
   const ref = useRef<(highlight: IHighlight) => void>(() => {});
@@ -120,4 +120,4 @@ const Preview = ({ highlights: state, setWidthAndHeight }: IProps) => {
   );
 };
 
-export default memo(Preview);
+export default memo(PdfPreview);

web/src/pages/chunk/parsed-result/add-knowledge/components/knowledge-chunk/components/document-preview/ppt-preview.tsx

@@ -0,0 +1,67 @@
+import message from '@/components/ui/message';
+import request from '@/utils/request';
+import classNames from 'classnames';
+import { init } from 'pptx-preview';
+import { useEffect, useRef } from 'react';
+import { useGetDocumentUrl } from './hooks';
+interface PptPreviewerProps {
+  className?: string;
+}
+
+export const PptPreviewer: React.FC<PptPreviewerProps> = ({ className }) => {
+  const url = useGetDocumentUrl();
+  const wrapper = useRef<HTMLDivElement>(null);
+  const containerRef = useRef<HTMLDivElement>(null);
+  const fetchDocument = async () => {
+    const res = await request(url, {
+      method: 'GET',
+      responseType: 'blob',
+      onError: () => {
+        message.error('Document parsing failed');
+        console.error('Error loading document:', url);
+      },
+    });
+    console.log(res);
+    try {
+      const arrayBuffer = await res.data.arrayBuffer();
+
+      if (containerRef.current) {
+        let width = 500;
+        let height = 900;
+        if (containerRef.current) {
+          width = containerRef.current.clientWidth - 50;
+          height = containerRef.current.clientHeight - 50;
+        }
+        let pptxPrviewer = init(containerRef.current, {
+          width: width,
+          height: height,
+        });
+        pptxPrviewer.preview(arrayBuffer);
+      }
+    } catch (err) {
+      message.error('ppt parse failed');
+    }
+  };
+
+  useEffect(() => {
+    if (url) {
+      fetchDocument();
+    }
+  }, [url]);
+
+  return (
+    <div
+      ref={containerRef}
+      className={classNames(
+        'relative w-full h-full p-4 bg-background-paper border border-border-normal rounded-md ppt-previewer',
+        className,
+      )}
+    >
+      <div className="overflow-auto p-2">
+        <div className="flex flex-col gap-4">
+          <div ref={wrapper} />
+        </div>
+      </div>
+    </div>
+  );
+};

web/src/pages/chunk/parsed-result/add-knowledge/components/knowledge-chunk/components/document-preview/txt-preview.tsx

@@ -0,0 +1,57 @@
+import message from '@/components/ui/message';
+import request from '@/utils/request';
+import { Spin } from 'antd';
+import classNames from 'classnames';
+import { useEffect, useState } from 'react';
+import { useGetDocumentUrl } from './hooks';
+
+type TxtPreviewerProps = { className?: string };
+export const TxtPreviewer = ({ className }: TxtPreviewerProps) => {
+  const url = useGetDocumentUrl();
+  const [loading, setLoading] = useState(false);
+  const [data, setData] = useState<string>('');
+  const fetchTxt = async () => {
+    setLoading(true);
+    const res = await request(url, {
+      method: 'GET',
+      responseType: 'blob',
+      onError: (err: any) => {
+        message.error('Failed to load file');
+        console.error('Error loading file:', err);
+      },
+    });
+    // blob to string
+    const reader = new FileReader();
+    reader.readAsText(res.data);
+    reader.onload = () => {
+      setData(reader.result as string);
+      setLoading(false);
+      console.log('file loaded successfully', reader.result);
+    };
+    console.log('file data:', res);
+  };
+  useEffect(() => {
+    if (url) {
+      fetchTxt();
+    } else {
+      setLoading(false);
+      setData('');
+    }
+  }, [url]);
+  return (
+    <div
+      className={classNames(
+        'relative w-full h-full p-4 bg-background-paper border border-border-normal rounded-md',
+        className,
+      )}
+    >
+      {loading && (
+        <div className="absolute inset-0 flex items-center justify-center">
+          <Spin />
+        </div>
+      )}
+
+      {!loading && <pre className="whitespace-pre-wrap p-2 ">{data}</pre>}
+    </div>
+  );
+};

web/src/pages/chunk/parsed-result/add-knowledge/components/knowledge-chunk/index.less

@@ -35,7 +35,8 @@
 
   .documentPreview {
     // width: 40%;
-    height: 100%;
+    height: calc(100vh - 130px);
+    overflow: auto;
   }
 
   .chunkContainer {

web/src/pages/chunk/parsed-result/add-knowledge/components/knowledge-chunk/index.tsx

@@ -3,11 +3,11 @@ import {
   useSwitchChunk,
 } from '@/hooks/use-chunk-request';
 import classNames from 'classnames';
-import { useCallback, useEffect, useState } from 'react';
+import { useCallback, useEffect, useMemo, useState } from 'react';
 import { useTranslation } from 'react-i18next';
 import ChunkCard from './components/chunk-card';
 import CreatingModal from './components/chunk-creating-modal';
-import DocumentPreview from './components/document-preview/preview';
+import DocumentPreview from './components/document-preview';
 import {
   useChangeChunkTextMode,
   useDeleteChunkByIds,
@@ -143,6 +143,20 @@ const Chunk = () => {
   const { highlights, setWidthAndHeight } =
     useGetChunkHighlights(selectedChunkId);
 
+  const fileType = useMemo(() => {
+    switch (documentInfo?.type) {
+      case 'doc':
+        return documentInfo?.name.split('.').pop() || 'doc';
+      case 'visual':
+      case 'docx':
+      case 'txt':
+      case 'md':
+      case 'pdf':
+        return documentInfo?.type;
+    }
+    return 'unknown';
+  }, [documentInfo]);
+
   return (
     <>
       <div className={styles.chunkPage}>
@@ -151,14 +165,14 @@ const Chunk = () => {
             <div className="h-[100px] flex flex-col justify-end pb-[5px]">
               <DocumentHeader {...documentInfo} />
             </div>
-            {isPdf && (
+            <section className={styles.documentPreview}>
-              <section className={styles.documentPreview}>
+              <DocumentPreview
-                <DocumentPreview
+                className={styles.documentPreview}
-                  highlights={highlights}
+                fileType={fileType}
-                  setWidthAndHeight={setWidthAndHeight}
+                highlights={highlights}
-                ></DocumentPreview>
+                setWidthAndHeight={setWidthAndHeight}
-              </section>
+              ></DocumentPreview>
-            )}
+            </section>
           </div>
           <div
             className={classNames(