#include "transcript_cleaner.h"

#include <cassert>

int main() {
  assert(clean_transcript("language English<astext>What do you think?") ==
         "What do you think?");
  assert(clean_transcript("language Chinese<astext>\xe6\x9d\xa5\xe8\xae\xa9\xe4\xbd\xa0\xe5\x8f\x91\xe7\x8e\xb0\xe6\xa0\x87\xe7\xad\xbe\xe3\x80\x82") ==
         "\xe6\x9d\xa5\xe8\xae\xa9\xe4\xbd\xa0\xe5\x8f\x91\xe7\x8e\xb0\xe6\xa0\x87\xe7\xad\xbe\xe3\x80\x82");
  assert(clean_transcript("language German<asr_text>Ich habe keine Idee.") ==
         "Ich habe keine Idee.");
  assert(clean_transcript("language Japanese<asr_text>\xe3\x81\x82\xe3\x82\x8a\xe3\x81\x8c\xe3\x81\xa8\xe3\x81\x86\xe3\x81\x94\xe3\x81\x96\xe3\x81\x84\xe3\x81\xbe\xe3\x81\x99\xe3\x80\x82") ==
         "\xe3\x81\x82\xe3\x82\x8a\xe3\x81\x8c\xe3\x81\xa8\xe3\x81\x86\xe3\x81\x94\xe3\x81\x96\xe3\x81\x84\xe3\x81\xbe\xe3\x81\x99\xe3\x80\x82");
  assert(clean_transcript(" \n\tlanguage English<asr_text>Hello") == "Hello");
  assert(clean_transcript("Language matters.") == "Language matters.");
  assert(clean_transcript("language English") == "language English");
}