mattermost-community-enterp.../platform/services/docextractor/pdf.go
Claude ec1f89217a Merge: Complete Mattermost Server with Community Enterprise
Full Mattermost server source with integrated Community Enterprise features.
Includes vendor directory for offline/air-gapped builds.

Structure:
- enterprise-impl/: Enterprise feature implementations
- enterprise-community/: Init files that register implementations
- enterprise/: Bridge imports (community_imports.go)
- vendor/: All dependencies for offline builds

Build (online):
  go build ./cmd/mattermost

Build (offline/air-gapped):
  go build -mod=vendor ./cmd/mattermost

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2025-12-17 23:59:07 +09:00

63 lines
1.2 KiB
Go

// Copyright (c) 2015-present Mattermost, Inc. All Rights Reserved.
// See LICENSE.txt for license information.
package docextractor
import (
"bytes"
"errors"
"fmt"
"io"
"os"
"path"
"strings"
"github.com/ledongthuc/pdf"
)
type pdfExtractor struct{}
func (pe *pdfExtractor) Name() string {
return "pdfExtractor"
}
func (pe *pdfExtractor) Match(filename string) bool {
supportedExtensions := map[string]bool{
"pdf": true,
}
extension := strings.TrimPrefix(path.Ext(filename), ".")
return supportedExtensions[extension]
}
func (pe *pdfExtractor) Extract(filename string, r io.ReadSeeker) (out string, outErr error) {
defer func() {
if r := recover(); r != nil {
out = ""
outErr = errors.New("error extracting pdf text")
}
}()
f, err := os.CreateTemp(os.TempDir(), "pdflib")
if err != nil {
return "", fmt.Errorf("error creating temporary file: %v", err)
}
defer f.Close()
defer os.Remove(f.Name())
size, err := io.Copy(f, r)
if err != nil {
return "", fmt.Errorf("error copying data into temporary file: %v", err)
}
reader, err := pdf.NewReader(f, size)
if err != nil {
return "", err
}
var buf bytes.Buffer
b, err := reader.GetPlainText()
if err != nil {
return "", err
}
buf.ReadFrom(b)
return buf.String(), nil
}