Sha256: 0e356ca589e4016446f035656ae06486301d3c96050eb745e455947a348546eb
Contents?: true
Size: 672 Bytes
Versions: 8
Compression:
Stored size: 672 Bytes
Contents
module Embulk
class GuessCharset < GuessPlugin
Plugin.register_guess('charset', self)
def guess(config, sample_buffer)
# ICU4J
detector = com.ibm.icu.text.CharsetDetector.new
detector.setText(sample_buffer.to_java_bytes)
best_match = detector.detect
if best_match.getConfidence < 50
name = "UTF-8"
else
name = best_match.getName
if name == "ISO-8859-1"
# ISO-8859-1 means ASCII which is a subset
# of UTF-8 in most of cases due to lack of
# sample data set
name = "UTF-8"
end
end
return {"parser" => {"charset" => name}}
end
end
end
Version data entries
8 entries across 8 versions & 1 rubygems