28 lines
600 B
Markdown
28 lines
600 B
Markdown
|
```ruby title="Ruby"
|
||
|
|
require 'kreuzberg'
|
||
|
|
|
||
|
|
class PdfMetadataExtractor
|
||
|
|
def initialize
|
||
|
|
@count = 0
|
||
|
|
end
|
||
|
|
|
||
|
|
def call(result)
|
||
|
|
return result unless result['mime_type'] == 'application/pdf'
|
||
|
|
@count += 1
|
||
|
|
result['metadata'] ||= {}
|
||
|
|
result['metadata']['pdf_order'] = @count
|
||
|
|
result
|
||
|
|
end
|
||
|
|
end
|
||
|
|
|
||
|
|
extractor = PdfMetadataExtractor.new
|
||
|
|
Kreuzberg.register_post_processor('pdf_metadata', extractor)
|
||
|
|
|
||
|
|
config = Kreuzberg::ExtractionConfig.new(
|
||
|
|
postprocessor: { enabled: true }
|
||
|
|
)
|
||
|
|
|
||
|
|
result = Kreuzberg.extract_file_sync('report.pdf', config: config)
|
||
|
|
puts "Metadata: #{result.metadata.inspect}"
|
||
|
|
```
|