fix(build): consolidate pdf parsing dependencies, remove extraneous html deps (#1212)

* fix(build): consolidate pdf parsing dependencies, remove extraneous html deps

* add types

apps/sim/app/api/files/parse/route.ts

@@ -553,22 +553,11 @@ function handleGenericBuffer(
  */
 async function parseBufferAsPdf(buffer: Buffer) {
   try {
-    try {
-      const { PdfParser } = await import('@/lib/file-parsers/pdf-parser')
-      const parser = new PdfParser()
-      logger.info('Using main PDF parser for buffer')
+    const { PdfParser } = await import('@/lib/file-parsers/pdf-parser')
+    const parser = new PdfParser()
+    logger.info('Using main PDF parser for buffer')
 
-      if (parser.parseBuffer) {
-        return await parser.parseBuffer(buffer)
-      }
-      throw new Error('PDF parser does not support buffer parsing')
-    } catch (error) {
-      logger.warn('Main PDF parser failed, using raw parser for buffer:', error)
-      const { RawPdfParser } = await import('@/lib/file-parsers/raw-pdf-parser')
-      const rawParser = new RawPdfParser()
-
-      return await rawParser.parseBuffer(buffer)
-    }
+    return await parser.parseBuffer(buffer)
   } catch (error) {
     throw new Error(`PDF parsing failed: ${(error as Error).message}`)
   }

apps/sim/lib/file-parsers/index.test.ts

@@ -141,17 +141,6 @@ describe('File Parsers', () => {
       })),
     }))
 
-    vi.doMock('@/lib/file-parsers/raw-pdf-parser', () => ({
-      RawPdfParser: vi.fn().mockImplementation(() => ({
-        parseFile: vi.fn().mockResolvedValue({
-          content: 'Raw parsed PDF content',
-          metadata: {
-            pageCount: 3,
-          },
-        }),
-      })),
-    }))
-
     vi.doMock('@/lib/file-parsers/txt-parser', () => ({
       TxtParser: vi.fn().mockImplementation(() => ({
         parseFile: mockTxtParseFile,

apps/sim/lib/file-parsers/index.ts

@@ -1,7 +1,5 @@
 import { existsSync } from 'fs'
-import { readFile } from 'fs/promises'
 import path from 'path'
-import { RawPdfParser } from '@/lib/file-parsers/raw-pdf-parser'
 import type { FileParseResult, FileParser, SupportedFileType } from '@/lib/file-parsers/types'
 import { createLogger } from '@/lib/logs/console/logger'
 
@@ -18,42 +16,12 @@ function getParserInstances(): Record<string, FileParser> {
 
     try {
       try {
-        logger.info('Attempting to load PDF parser...')
-        try {
-          const { PdfParser } = require('@/lib/file-parsers/pdf-parser')
-          parserInstances.pdf = new PdfParser()
-          logger.info('PDF parser loaded successfully')
-        } catch (pdfLibError) {
-          logger.error('Failed to load primary PDF parser:', pdfLibError)
-          logger.info('Falling back to raw PDF parser')
-          parserInstances.pdf = new RawPdfParser()
-          logger.info('Raw PDF parser loaded successfully')
-        }
+        logger.info('Loading PDF parser...')
+        const { PdfParser } = require('@/lib/file-parsers/pdf-parser')
+        parserInstances.pdf = new PdfParser()
+        logger.info('PDF parser loaded successfully')
       } catch (error) {
-        logger.error('Failed to load any PDF parser:', error)
-        parserInstances.pdf = {
-          async parseFile(filePath: string): Promise<FileParseResult> {
-            const buffer = await readFile(filePath)
-            return {
-              content: `PDF parsing is not available. File size: ${buffer.length} bytes`,
-              metadata: {
-                info: { Error: 'PDF parsing unavailable' },
-                pageCount: 0,
-                version: 'unknown',
-              },
-            }
-          },
-          async parseBuffer(buffer: Buffer): Promise<FileParseResult> {
-            return {
-              content: `PDF parsing is not available. File size: ${buffer.length} bytes`,
-              metadata: {
-                info: { Error: 'PDF parsing unavailable' },
-                pageCount: 0,
-                version: 'unknown',
-              },
-            }
-          },
-        }
+        logger.error('Failed to load PDF parser:', error)
       }
 
       try {

apps/sim/lib/file-parsers/pdf-parser.ts

@@ -1,11 +1,9 @@
 import { readFile } from 'fs/promises'
-import { PDFDocument } from 'pdf-lib'
+import pdfParse from 'pdf-parse'
 import type { FileParseResult, FileParser } from '@/lib/file-parsers/types'
 import { createLogger } from '@/lib/logs/console/logger'
-import { RawPdfParser } from './raw-pdf-parser'
 
 const logger = createLogger('PdfParser')
-const rawPdfParser = new RawPdfParser()
 
 export class PdfParser implements FileParser {
   async parseFile(filePath: string): Promise<FileParseResult> {
@@ -31,68 +29,23 @@ export class PdfParser implements FileParser {
     try {
       logger.info('Starting to parse buffer, size:', dataBuffer.length)
 
-      try {
-        logger.info('Attempting to parse with pdf-lib library...')
+      const pdfData = await pdfParse(dataBuffer)
 
-        logger.info('Starting PDF parsing...')
-        const pdfDoc = await PDFDocument.load(dataBuffer)
-        const pages = pdfDoc.getPages()
-        const pageCount = pages.length
+      logger.info(
+        'PDF parsed successfully, pages:',
+        pdfData.numpages,
+        'text length:',
+        pdfData.text.length
+      )
 
-        logger.info('PDF parsed successfully with pdf-lib, pages:', pageCount)
-
-        const metadata: Record<string, any> = {
-          pageCount,
-        }
-
-        try {
-          const title = pdfDoc.getTitle()
-          const author = pdfDoc.getAuthor()
-          const subject = pdfDoc.getSubject()
-          const creator = pdfDoc.getCreator()
-          const producer = pdfDoc.getProducer()
-          const creationDate = pdfDoc.getCreationDate()
-          const modificationDate = pdfDoc.getModificationDate()
-
-          if (title) metadata.title = title
-          if (author) metadata.author = author
-          if (subject) metadata.subject = subject
-          if (creator) metadata.creator = creator
-          if (producer) metadata.producer = producer
-          if (creationDate) metadata.creationDate = creationDate.toISOString()
-          if (modificationDate) metadata.modificationDate = modificationDate.toISOString()
-        } catch (metadataError) {
-          logger.warn('Could not extract PDF metadata:', metadataError)
-        }
-
-        logger.info(
-          'pdf-lib loaded successfully, but text extraction requires fallback to raw parser'
-        )
-        const rawResult = await rawPdfParser.parseBuffer(dataBuffer)
-
-        return {
-          content: rawResult.content,
-          metadata: {
-            ...rawResult.metadata,
-            ...metadata,
-            source: 'pdf-lib + raw-parser',
-          },
-        }
-      } catch (pdfLibError: unknown) {
-        logger.error('PDF-lib library failed:', pdfLibError)
-
-        logger.info('Falling back to raw PDF parser...')
-        const rawResult = await rawPdfParser.parseBuffer(dataBuffer)
-
-        return {
-          ...rawResult,
-          metadata: {
-            ...rawResult.metadata,
-            fallback: true,
-            source: 'raw-parser-only',
-            error: (pdfLibError as Error).message || 'Unknown error',
-          },
-        }
+      return {
+        content: pdfData.text,
+        metadata: {
+          pageCount: pdfData.numpages,
+          info: pdfData.info,
+          version: pdfData.version,
+          source: 'pdf-parse',
+        },
       }
     } catch (error) {
       logger.error('Error parsing buffer:', error)

apps/sim/lib/file-parsers/raw-pdf-parser.ts

@@ -1,467 +0,0 @@
-import { readFile } from 'fs/promises'
-import { promisify } from 'util'
-import zlib from 'zlib'
-import type { FileParseResult, FileParser } from '@/lib/file-parsers/types'
-import { createLogger } from '@/lib/logs/console/logger'
-
-const logger = createLogger('RawPdfParser')
-
-const inflateAsync = promisify(zlib.inflate)
-const unzipAsync = promisify(zlib.unzip)
-
-export class RawPdfParser implements FileParser {
-  async parseFile(filePath: string): Promise<FileParseResult> {
-    try {
-      logger.info('Starting to parse file:', filePath)
-
-      if (!filePath) {
-        throw new Error('No file path provided')
-      }
-
-      logger.info('Reading file...')
-      const dataBuffer = await readFile(filePath)
-      logger.info('File read successfully, size:', dataBuffer.length)
-
-      return this.parseBuffer(dataBuffer)
-    } catch (error) {
-      logger.error('Error parsing PDF:', error)
-      return {
-        content: `Error parsing PDF: ${(error as Error).message}`,
-        metadata: {
-          error: (error as Error).message,
-          pageCount: 0,
-          version: 'unknown',
-        },
-      }
-    }
-  }
-
-  async parseBuffer(dataBuffer: Buffer): Promise<FileParseResult> {
-    try {
-      logger.info('Starting to parse buffer, size:', dataBuffer.length)
-
-      const rawContent = dataBuffer.toString('utf-8')
-
-      let version = 'Unknown'
-      let pageCount = 0
-
-      const versionMatch = rawContent.match(/%PDF-(\d+\.\d+)/)
-      if (versionMatch?.[1]) {
-        version = versionMatch[1]
-      }
-
-      const typePageMatches = rawContent.match(/\/Type\s*\/Page\b/gi)
-      if (typePageMatches) {
-        pageCount = typePageMatches.length
-        logger.info('Found page count using /Type /Page:', pageCount)
-      }
-
-      if (pageCount === 0) {
-        const pageMatches = rawContent.match(/\/Page\s*\//gi)
-        if (pageMatches) {
-          pageCount = pageMatches.length
-          logger.info('Found page count using /Page/ pattern:', pageCount)
-        }
-      }
-
-      if (pageCount === 0) {
-        const pagesObjMatches = rawContent.match(/\/Pages\s+\d+\s+\d+\s+R/gi)
-        if (pagesObjMatches && pagesObjMatches.length > 0) {
-          const pagesObjRef = pagesObjMatches[0].match(/\/Pages\s+(\d+)\s+\d+\s+R/i)
-          if (pagesObjRef?.[1]) {
-            const objNum = pagesObjRef[1]
-            const objRegex = new RegExp(`${objNum}\\s+0\\s+obj[\\s\\S]*?endobj`, 'i')
-            const objMatch = rawContent.match(objRegex)
-            if (objMatch) {
-              const countMatch = objMatch[0].match(/\/Count\s+(\d+)/i)
-              if (countMatch?.[1]) {
-                pageCount = Number.parseInt(countMatch[1], 10)
-                logger.info('Found page count using /Count in Pages object:', pageCount)
-              }
-            }
-          }
-        }
-      }
-
-      if (pageCount === 0) {
-        const trailerMatches = rawContent.match(/trailer/gi)
-        if (trailerMatches) {
-          pageCount = Math.max(1, Math.ceil(trailerMatches.length / 2))
-          logger.info('Estimated page count using trailer references:', pageCount)
-        }
-      }
-
-      if (pageCount === 0) {
-        pageCount = 1
-        logger.info('Defaulting to 1 page as no count was found')
-      }
-
-      let extractedText = ''
-
-      const textMatches = rawContent.match(/BT[\s\S]*?ET/g)
-      if (textMatches && textMatches.length > 0) {
-        logger.info('Found', textMatches.length, 'text blocks')
-
-        extractedText = textMatches
-          .map((textBlock) => {
-            const textObjects = textBlock.match(/(\([^)]*\)|\[[^\]]*\])\s*(Tj|TJ)/g)
-            if (textObjects && textObjects.length > 0) {
-              return textObjects
-                .map((obj) => {
-                  let text = ''
-                  if (obj.includes('Tj')) {
-                    const match = obj.match(/\(([^)]*)\)\s*Tj/)
-                    if (match?.[1]) {
-                      text = match[1]
-                    }
-                  } else if (obj.includes('TJ')) {
-                    const match = obj.match(/\[(.*)\]\s*TJ/)
-                    if (match?.[1]) {
-                      const parts = match[1].match(/\([^)]*\)/g)
-                      if (parts) {
-                        text = parts.map((p) => p.slice(1, -1)).join(' ')
-                      }
-                    }
-                  }
-
-                  return text
-                    .replace(/\\(\d{3})/g, (_, octal) =>
-                      String.fromCharCode(Number.parseInt(octal, 8))
-                    )
-                    .replace(/\\\\/g, '\\')
-                    .replace(/\\\(/g, '(')
-                    .replace(/\\\)/g, ')')
-                })
-                .join(' ')
-            }
-            return ''
-          })
-          .join('\n')
-          .trim()
-      }
-
-      let metadataText = ''
-      const xmlMatch = rawContent.match(/<x:xmpmeta[\s\S]*?<\/x:xmpmeta>/)
-      if (xmlMatch) {
-        const xmlContent = xmlMatch[0]
-        logger.info('Found XML metadata')
-
-        const titleMatch = xmlContent.match(/<dc:title>[\s\S]*?<rdf:li[^>]*>(.*?)<\/rdf:li>/i)
-        if (titleMatch?.[1]) {
-          const title = titleMatch[1].replace(/<[^>]+>/g, '').trim()
-          metadataText += `Document Title: ${title}\n\n`
-        }
-
-        const creatorMatch = xmlContent.match(/<dc:creator>[\s\S]*?<rdf:li[^>]*>(.*?)<\/rdf:li>/i)
-        if (creatorMatch?.[1]) {
-          const creator = creatorMatch[1].replace(/<[^>]+>/g, '').trim()
-          metadataText += `Author: ${creator}\n`
-        }
-
-        const dateMatch = xmlContent.match(/<xmp:CreateDate>(.*?)<\/xmp:CreateDate>/i)
-        if (dateMatch?.[1]) {
-          metadataText += `Created: ${dateMatch[1].trim()}\n`
-        }
-
-        const producerMatch = xmlContent.match(/<pdf:Producer>(.*?)<\/pdf:Producer>/i)
-        if (producerMatch?.[1]) {
-          metadataText += `Producer: ${producerMatch[1].trim()}\n`
-        }
-      }
-
-      if (!extractedText || extractedText.length < 100 || extractedText.includes('/Type /Page')) {
-        logger.info('Trying advanced text extraction from content streams')
-
-        const contentRefs = rawContent.match(/\/Contents\s+\[?\s*(\d+)\s+\d+\s+R\s*\]?/g)
-        if (contentRefs && contentRefs.length > 0) {
-          logger.info('Found', contentRefs.length, 'content stream references')
-
-          const objNumbers = contentRefs
-            .map((ref) => {
-              const match = ref.match(/\/Contents\s+\[?\s*(\d+)\s+\d+\s+R\s*\]?/)
-              return match ? match[1] : null
-            })
-            .filter(Boolean)
-
-          logger.info('Content stream object numbers:', objNumbers)
-
-          if (objNumbers.length > 0) {
-            let textFromStreams = ''
-
-            for (const objNum of objNumbers) {
-              const objRegex = new RegExp(`${objNum}\\s+0\\s+obj[\\s\\S]*?endobj`, 'i')
-              const objMatch = rawContent.match(objRegex)
-
-              if (objMatch) {
-                const streamMatch = objMatch[0].match(/stream\r?\n([\s\S]*?)\r?\nendstream/)
-                if (streamMatch?.[1]) {
-                  const streamContent = streamMatch[1]
-
-                  const textFragments = streamContent.match(/\([^)]+\)\s*Tj|\[[^\]]*\]\s*TJ/g)
-                  if (textFragments && textFragments.length > 0) {
-                    const extractedFragments = textFragments
-                      .map((fragment) => {
-                        if (fragment.includes('Tj')) {
-                          return fragment
-                            .replace(/\(([^)]*)\)\s*Tj/, '$1')
-                            .replace(/\\(\d{3})/g, (_, octal) =>
-                              String.fromCharCode(Number.parseInt(octal, 8))
-                            )
-                            .replace(/\\\\/g, '\\')
-                            .replace(/\\\(/g, '(')
-                            .replace(/\\\)/g, ')')
-                        }
-                        if (fragment.includes('TJ')) {
-                          const parts = fragment.match(/\([^)]*\)/g)
-                          if (parts) {
-                            return parts
-                              .map((p) =>
-                                p
-                                  .slice(1, -1)
-                                  .replace(/\\(\d{3})/g, (_, octal) =>
-                                    String.fromCharCode(Number.parseInt(octal, 8))
-                                  )
-                                  .replace(/\\\\/g, '\\')
-                                  .replace(/\\\(/g, '(')
-                                  .replace(/\\\)/g, ')')
-                              )
-                              .join(' ')
-                          }
-                        }
-                        return ''
-                      })
-                      .filter(Boolean)
-                      .join(' ')
-
-                    if (extractedFragments.trim().length > 0) {
-                      textFromStreams += `${extractedFragments.trim()}\n`
-                    }
-                  }
-                }
-              }
-            }
-
-            if (textFromStreams.trim().length > 0) {
-              logger.info('Successfully extracted text from content streams')
-              extractedText = textFromStreams.trim()
-            }
-          }
-        }
-      }
-
-      if (!extractedText || extractedText.length < 100) {
-        logger.info('Trying to decompress PDF streams')
-
-        const compressedStreams = rawContent.match(
-          /\/Filter\s*\/FlateDecode[\s\S]*?stream[\s\S]*?endstream/g
-        )
-        if (compressedStreams && compressedStreams.length > 0) {
-          logger.info('Found', compressedStreams.length, 'compressed streams')
-
-          const decompressedContents = await Promise.all(
-            compressedStreams.map(async (stream) => {
-              try {
-                const streamMatch = stream.match(/stream\r?\n([\s\S]*?)\r?\nendstream/)
-                if (!streamMatch || !streamMatch[1]) return ''
-
-                const compressedData = Buffer.from(streamMatch[1], 'binary')
-
-                try {
-                  const decompressed = await inflateAsync(compressedData)
-                  const content = decompressed.toString('utf-8')
-
-                  const readable = content.replace(/[^\x20-\x7E\r\n]/g, ' ').trim()
-                  if (
-                    readable.length > 50 &&
-                    readable.includes(' ') &&
-                    (readable.includes('.') || readable.includes(',')) &&
-                    !/[\x00-\x1F\x7F]/.test(readable)
-                  ) {
-                    return readable
-                  }
-                } catch (_inflateErr) {
-                  try {
-                    const decompressed = await unzipAsync(compressedData)
-                    const content = decompressed.toString('utf-8')
-
-                    const readable = content.replace(/[^\x20-\x7E\r\n]/g, ' ').trim()
-                    if (
-                      readable.length > 50 &&
-                      readable.includes(' ') &&
-                      (readable.includes('.') || readable.includes(',')) &&
-                      !/[\x00-\x1F\x7F]/.test(readable)
-                    ) {
-                      return readable
-                    }
-                  } catch (_unzipErr) {
-                    return ''
-                  }
-                }
-              } catch (_error) {
-                return ''
-              }
-
-              return ''
-            })
-          )
-
-          const decompressedText = decompressedContents
-            .filter((text) => text && text.length > 0)
-            .join('\n\n')
-
-          if (decompressedText && decompressedText.length > 0) {
-            logger.info('Successfully decompressed text content, length:', decompressedText.length)
-            extractedText = decompressedText
-          }
-        }
-      }
-
-      if (!extractedText || extractedText.length < 50) {
-        logger.info('Trying alternative text extraction method with streams')
-
-        const streamMatches = rawContent.match(/stream[\s\S]*?endstream/g)
-        if (streamMatches && streamMatches.length > 0) {
-          logger.info('Found', streamMatches.length, 'streams')
-
-          const textContent = streamMatches
-            .map((stream) => {
-              const content = stream.replace(/^stream\r?\n|\r?\nendstream$/g, '')
-
-              const readable = content.replace(/[^\x20-\x7E\r\n]/g, ' ').trim()
-
-              if (
-                readable.length > 20 &&
-                readable.includes(' ') &&
-                (readable.includes('.') || readable.includes(',')) &&
-                !/[\x00-\x1F\x7F]/.test(readable)
-              ) {
-                return readable
-              }
-              return ''
-            })
-            .filter((text) => text.length > 0 && text.split(' ').length > 5)
-            .join('\n\n')
-
-          if (textContent.length > 0) {
-            extractedText = textContent
-          }
-        }
-      }
-
-      if (!extractedText || extractedText.length < 50) {
-        logger.info('Trying object streams for text')
-
-        const objMatches = rawContent.match(/\d+\s+\d+\s+obj[\s\S]*?endobj/g)
-        if (objMatches && objMatches.length > 0) {
-          logger.info('Found', objMatches.length, 'objects')
-
-          const textContent = objMatches
-            .map((obj) => {
-              const readable = obj.replace(/[^\x20-\x7E\r\n]/g, ' ').trim()
-
-              if (
-                readable.length > 50 &&
-                readable.includes(' ') &&
-                !readable.includes('/Filter') &&
-                readable.split(' ').length > 10 &&
-                (readable.includes('.') || readable.includes(','))
-              ) {
-                return readable
-              }
-              return ''
-            })
-            .filter((text) => text.length > 0)
-            .join('\n\n')
-
-          if (textContent.length > 0) {
-            extractedText += (extractedText ? '\n\n' : '') + textContent
-          }
-        }
-      }
-
-      if (
-        extractedText &&
-        (extractedText.includes('endobj') ||
-          extractedText.includes('/Type /Page') ||
-          extractedText.match(/\d+\s+\d+\s+obj/g)) &&
-        metadataText
-      ) {
-        logger.info(
-          'Extracted content appears to be PDF structure information, using metadata instead'
-        )
-        extractedText = metadataText
-      } else if (metadataText && !extractedText.includes('Document Title:')) {
-        extractedText = metadataText + (extractedText ? `\n\n${extractedText}` : '')
-      }
-
-      const validCharCount = (extractedText || '').replace(/[^\x20-\x7E\r\n]/g, '').length
-      const totalCharCount = (extractedText || '').length
-      const validRatio = validCharCount / (totalCharCount || 1)
-
-      const hasBinaryArtifacts =
-        extractedText &&
-        (extractedText.includes('\\u') ||
-          extractedText.includes('\\x') ||
-          extractedText.includes('\0') ||
-          /[\x00-\x08\x0B\x0C\x0E-\x1F\x7F-\xFF]{10,}/g.test(extractedText) ||
-          validRatio < 0.7)
-
-      const looksLikeGibberish =
-        extractedText &&
-        (extractedText.replace(/[a-zA-Z0-9\s.,:'"()[\]{}]/g, '').length / extractedText.length >
-          0.3 ||
-          extractedText.split(' ').length < extractedText.length / 20)
-
-      if (!extractedText || extractedText.length < 50 || hasBinaryArtifacts || looksLikeGibberish) {
-        logger.info('Could not extract meaningful text, providing fallback message')
-        logger.info('Valid character ratio:', validRatio)
-        logger.info('Has binary artifacts:', hasBinaryArtifacts)
-        logger.info('Looks like gibberish:', looksLikeGibberish)
-
-        if (metadataText) {
-          extractedText = `${metadataText}\n`
-        } else {
-          extractedText = ''
-        }
-
-        extractedText += `This is a PDF document with ${pageCount} page(s) and version ${version}.\n\n`
-
-        const titleInStructure =
-          rawContent.match(/title\s*:\s*([^\n]+)/i) ||
-          rawContent.match(/Microsoft Word -\s*([^\n]+)/i)
-
-        if (titleInStructure?.[1] && !extractedText.includes('Document Title:')) {
-          const title = titleInStructure[1].trim()
-          extractedText = `Document Title: ${title}\n\n${extractedText}`
-        }
-
-        extractedText += `The text content could not be properly extracted due to encoding or compression issues.\nFile size: ${dataBuffer.length} bytes.\n\nTo view this PDF properly, please download the file and open it with a PDF reader.`
-      }
-
-      logger.info('PDF parsed with basic extraction, found text length:', extractedText.length)
-
-      return {
-        content: extractedText,
-        metadata: {
-          pageCount,
-          info: {
-            RawExtraction: true,
-            Version: version,
-            Size: dataBuffer.length,
-          },
-          version,
-        },
-      }
-    } catch (error) {
-      logger.error('Error parsing buffer:', error)
-      return {
-        content: `Error parsing PDF buffer: ${(error as Error).message}`,
-        metadata: {
-          error: (error as Error).message,
-          pageCount: 0,
-          version: 'unknown',
-        },
-      }
-    }
-  }
-}

apps/sim/lib/webhooks/outlook-polling-service.ts

@@ -385,7 +385,6 @@ async function processOutlookEmails(
           if (type === 'text' || type === 'text/plain') {
             return content
           }
-          // Default to converting HTML or unknown types
           return convertHtmlToPlainText(content)
         })(),
         bodyHtml: email.body?.content || '',

apps/sim/package.json

@@ -69,6 +69,7 @@
     "@react-email/components": "^0.0.34",
     "@sentry/nextjs": "^9.15.0",
     "@trigger.dev/sdk": "4.0.1",
+    "@types/pdf-parse": "1.1.5",
     "@types/three": "0.177.0",
     "@vercel/og": "^0.6.5",
     "@vercel/speed-insights": "^1.2.0",
@@ -84,6 +85,7 @@
     "dat.gui": "0.7.9",
     "date-fns": "4.1.0",
     "drizzle-orm": "^0.41.0",
+    "entities": "6.0.1",
     "framer-motion": "^12.5.0",
     "fuse.js": "7.1.0",
     "geist": "1.4.2",
@@ -105,7 +107,6 @@
     "officeparser": "^5.2.0",
     "openai": "^4.91.1",
     "papaparse": "5.5.3",
-    "pdf-lib": "^1.17.1",
     "postgres": "^3.4.5",
     "prismjs": "^1.30.0",
     "react": "19.1.0",
@@ -122,6 +123,7 @@
     "resend": "^4.1.2",
     "rtf-parser": "1.3.3",
     "rtf-stream-parser": "3.8.0",
+    "sharp": "0.34.3",
     "socket.io": "^4.8.1",
     "stripe": "^17.7.0",
     "tailwind-merge": "^2.6.0",
@@ -137,7 +139,7 @@
     "@testing-library/react": "^16.3.0",
     "@testing-library/user-event": "^14.6.1",
     "@trigger.dev/build": "4.0.1",
-    "@types/html-to-text": "^9.0.4",
+    "@types/html-to-text": "9.0.4",
     "@types/iconv-lite": "0.0.1",
     "@types/js-yaml": "4.0.9",
     "@types/jsdom": "21.1.7",

bun.lock

@@ -97,6 +97,7 @@
         "@react-email/components": "^0.0.34",
         "@sentry/nextjs": "^9.15.0",
         "@trigger.dev/sdk": "4.0.1",
+        "@types/pdf-parse": "1.1.5",
         "@types/three": "0.177.0",
         "@vercel/og": "^0.6.5",
         "@vercel/speed-insights": "^1.2.0",
@@ -112,6 +113,7 @@
         "dat.gui": "0.7.9",
         "date-fns": "4.1.0",
         "drizzle-orm": "^0.41.0",
+        "entities": "6.0.1",
         "framer-motion": "^12.5.0",
         "fuse.js": "7.1.0",
         "geist": "1.4.2",
@@ -133,7 +135,6 @@
         "officeparser": "^5.2.0",
         "openai": "^4.91.1",
         "papaparse": "5.5.3",
-        "pdf-lib": "^1.17.1",
         "postgres": "^3.4.5",
         "prismjs": "^1.30.0",
         "react": "19.1.0",
@@ -150,6 +151,7 @@
         "resend": "^4.1.2",
         "rtf-parser": "1.3.3",
         "rtf-stream-parser": "3.8.0",
+        "sharp": "0.34.3",
         "socket.io": "^4.8.1",
         "stripe": "^17.7.0",
         "tailwind-merge": "^2.6.0",
@@ -165,7 +167,7 @@
         "@testing-library/react": "^16.3.0",
         "@testing-library/user-event": "^14.6.1",
         "@trigger.dev/build": "4.0.1",
-        "@types/html-to-text": "^9.0.4",
+        "@types/html-to-text": "9.0.4",
         "@types/iconv-lite": "0.0.1",
         "@types/js-yaml": "4.0.9",
         "@types/jsdom": "21.1.7",
@@ -830,10 +832,6 @@
 
     "@orama/orama": ["@orama/orama@3.1.12", "", {}, "sha512-U7PY8FwXHuJ6bNBpbsqe0KLzb91IcJuORDggqHHkFy1waokY5SpWLN9tzB3AOW776awp6s1bjwts9I9Davy3lw=="],
 
-    "@pdf-lib/standard-fonts": ["@pdf-lib/standard-fonts@1.0.0", "", { "dependencies": { "pako": "^1.0.6" } }, "sha512-hU30BK9IUN/su0Mn9VdlVKsWBS6GyhVfqjwl1FjZN4TxP6cCw0jP2w7V3Hf5uX7M0AZJ16vey9yE0ny7Sa59ZA=="],
-
-    "@pdf-lib/upng": ["@pdf-lib/upng@1.0.1", "", { "dependencies": { "pako": "^1.0.10" } }, "sha512-dQK2FUMQtowVP00mtIksrlZhdFXQZPC+taih1q4CvPZ5vqdxR/LKBaFg0oAfzd1GlHZXXSPdQfzQnt+ViGvEIQ=="],
-
     "@peculiar/asn1-android": ["@peculiar/asn1-android@2.4.0", "", { "dependencies": { "@peculiar/asn1-schema": "^2.4.0", "asn1js": "^3.0.6", "tslib": "^2.8.1" } }, "sha512-YFueREq97CLslZZBI8dKzis7jMfEHSLxM+nr0Zdx1POiXFLjqqwoY5s0F1UimdBiEw/iKlHey2m56MRDv7Jtyg=="],
 
     "@peculiar/asn1-ecc": ["@peculiar/asn1-ecc@2.4.0", "", { "dependencies": { "@peculiar/asn1-schema": "^2.4.0", "@peculiar/asn1-x509": "^2.4.0", "asn1js": "^3.0.6", "tslib": "^2.8.1" } }, "sha512-fJiYUBCJBDkjh347zZe5H81BdJ0+OGIg0X9z06v8xXUoql3MFeENUX0JsjCaVaU9A0L85PefLPGYkIoGpTnXLQ=="],
@@ -1440,6 +1438,8 @@
 
     "@types/papaparse": ["@types/papaparse@5.3.16", "", { "dependencies": { "@types/node": "*" } }, "sha512-T3VuKMC2H0lgsjI9buTB3uuKj3EMD2eap1MOuEQuBQ44EnDx/IkGhU6EwiTf9zG3za4SKlmwKAImdDKdNnCsXg=="],
 
+    "@types/pdf-parse": ["@types/pdf-parse@1.1.5", "", { "dependencies": { "@types/node": "*" } }, "sha512-kBfrSXsloMnUJOKi25s3+hRmkycHfLK6A09eRGqF/N8BkQoPUmaCr+q8Cli5FnfohEz/rsv82zAiPz/LXtOGhA=="],
+
     "@types/pg": ["@types/pg@8.6.1", "", { "dependencies": { "@types/node": "*", "pg-protocol": "*", "pg-types": "^2.2.0" } }, "sha512-1Kc4oAGzAl7uqUStZCDvaLFqZrW9qWSjXOmBfdgyBP5La7Us6Mg4GBvRlSoaZMhQF/zSj1C8CtKMBkoiT8eL8w=="],
 
     "@types/pg-pool": ["@types/pg-pool@2.0.6", "", { "dependencies": { "@types/pg": "*" } }, "sha512-TaAUE5rq2VQYxab5Ts7WZhKNmuN78Q6PiFonTDdpbx8a1H0M1vhy3rhiMjl+e2iHmogyMw7jZF4FrE6eJUy5HQ=="],
@@ -2644,8 +2644,6 @@
 
     "pathval": ["pathval@2.0.1", "", {}, "sha512-//nshmD55c46FuFw26xV/xFAaB5HF9Xdap7HJBBnrKdAd6/GxDBaNA1870O79+9ueg61cZLSVc+OaFlfmObYVQ=="],
 
-    "pdf-lib": ["pdf-lib@1.17.1", "", { "dependencies": { "@pdf-lib/standard-fonts": "^1.0.0", "@pdf-lib/upng": "^1.0.1", "pako": "^1.0.11", "tslib": "^1.11.1" } }, "sha512-V/mpyJAoTsN4cnP31vc0wfNA1+p20evqqnap0KLoRUN0Yk/p3wN52DOEsL4oBFcLdb76hlpKPtzJIgo67j/XLw=="],
-
     "pdfjs-dist": ["pdfjs-dist@5.4.54", "", { "optionalDependencies": { "@napi-rs/canvas": "^0.1.74" } }, "sha512-TBAiTfQw89gU/Z4LW98Vahzd2/LoCFprVGvGbTgFt+QCB1F+woyOPmNNVgLa6djX9Z9GGTnj7qE1UzpOVJiINw=="],
 
     "peberminta": ["peberminta@0.9.0", "", {}, "sha512-XIxfHpEuSJbITd1H3EeQwpcZbTLHc+VVr8ANI9t5sit565tsI4/xK3KWTUFE2e6QiangUkh3B0jihzmGnNrRsQ=="],
@@ -3680,6 +3678,8 @@
 
     "@types/papaparse/@types/node": ["@types/node@24.2.1", "", { "dependencies": { "undici-types": "~7.10.0" } }, "sha512-DRh5K+ka5eJic8CjH7td8QpYEV6Zo10gfRkjHCO3weqZHWDtAaSTFtl4+VMqOJ4N5jcuhZ9/l+yy8rVgw7BQeQ=="],
 
+    "@types/pdf-parse/@types/node": ["@types/node@24.2.1", "", { "dependencies": { "undici-types": "~7.10.0" } }, "sha512-DRh5K+ka5eJic8CjH7td8QpYEV6Zo10gfRkjHCO3weqZHWDtAaSTFtl4+VMqOJ4N5jcuhZ9/l+yy8rVgw7BQeQ=="],
+
     "@types/pg/@types/node": ["@types/node@24.2.1", "", { "dependencies": { "undici-types": "~7.10.0" } }, "sha512-DRh5K+ka5eJic8CjH7td8QpYEV6Zo10gfRkjHCO3weqZHWDtAaSTFtl4+VMqOJ4N5jcuhZ9/l+yy8rVgw7BQeQ=="],
 
     "@types/tedious/@types/node": ["@types/node@24.2.1", "", { "dependencies": { "undici-types": "~7.10.0" } }, "sha512-DRh5K+ka5eJic8CjH7td8QpYEV6Zo10gfRkjHCO3weqZHWDtAaSTFtl4+VMqOJ4N5jcuhZ9/l+yy8rVgw7BQeQ=="],
@@ -3816,8 +3816,6 @@
 
     "parse-entities/@types/unist": ["@types/unist@2.0.11", "", {}, "sha512-CmBKiL6NNo/OqgmMn95Fk9Whlp2mtvIv+KNpQKN2F4SjvrEesubTRWGYSg+BnWZOnlCaSTU1sMpsBOzgbYhnsA=="],
 
-    "pdf-lib/tslib": ["tslib@1.14.1", "", {}, "sha512-Xni35NKzjgMrwevysHTCArtLDpPvye8zV/0E4EyYn43P7/7qvQwPh9BGkHewbMulVntbigmcT7rdX3BNo9wRJg=="],
-
     "playwright/fsevents": ["fsevents@2.3.2", "", { "os": "darwin" }, "sha512-xiqMQR4xAeHTuB9uWm+fFRcIOgKBMiOBP+eXiyT7jsgVCq1bkVygt00oASowB7EdtpOHaaPgKt812P9ab+DDKA=="],
 
     "postcss-nested/postcss-selector-parser": ["postcss-selector-parser@6.1.2", "", { "dependencies": { "cssesc": "^3.0.0", "util-deprecate": "^1.0.2" } }, "sha512-Q8qQfPiZ+THO/3ZrOrO0cJJKfpYCagtMUkXbnEfmgUjwXg6z/WBeOyS9APBBPCTSiDV+s4SwQGu8yFsiMRIudg=="],
@@ -4262,6 +4260,8 @@
 
     "@types/papaparse/@types/node/undici-types": ["undici-types@7.10.0", "", {}, "sha512-t5Fy/nfn+14LuOc2KNYg75vZqClpAiqscVvMygNnlsHBFpSXdJaYtXMcdNLpl/Qvc3P2cB3s6lOV51nqsFq4ag=="],
 
+    "@types/pdf-parse/@types/node/undici-types": ["undici-types@7.10.0", "", {}, "sha512-t5Fy/nfn+14LuOc2KNYg75vZqClpAiqscVvMygNnlsHBFpSXdJaYtXMcdNLpl/Qvc3P2cB3s6lOV51nqsFq4ag=="],
+
     "@types/pg/@types/node/undici-types": ["undici-types@7.10.0", "", {}, "sha512-t5Fy/nfn+14LuOc2KNYg75vZqClpAiqscVvMygNnlsHBFpSXdJaYtXMcdNLpl/Qvc3P2cB3s6lOV51nqsFq4ag=="],
 
     "@types/tedious/@types/node/undici-types": ["undici-types@7.10.0", "", {}, "sha512-t5Fy/nfn+14LuOc2KNYg75vZqClpAiqscVvMygNnlsHBFpSXdJaYtXMcdNLpl/Qvc3P2cB3s6lOV51nqsFq4ag=="],