florin-akermann commented on a change in pull request #11456: URL: https://github.com/apache/kafka/pull/11456#discussion_r785328925
########## File path: core/src/main/scala/kafka/tools/ConsoleProducer.scala ########## @@ -264,39 +278,92 @@ object ConsoleProducer { var reader: BufferedReader = null var parseKey = false var keySeparator = "\t" + var parseHeader = false + var headersDelimiter = "\t" + var headersSeparator = "," + var headerKeySeparator = ":" var ignoreError = false var lineNumber = 0 var printPrompt = System.console != null + var headerSeparatorPattern: Pattern = _ override def init(inputStream: InputStream, props: Properties): Unit = { topic = props.getProperty("topic") if (props.containsKey("parse.key")) parseKey = props.getProperty("parse.key").trim.equalsIgnoreCase("true") if (props.containsKey("key.separator")) keySeparator = props.getProperty("key.separator") + if (props.containsKey("parse.headers")) + parseHeader = props.getProperty("parse.headers").trim.equalsIgnoreCase("true") + if (props.containsKey("headers.delimiter")) + headersDelimiter = props.getProperty("headers.delimiter") + if (props.containsKey("headers.separator")) + headersSeparator = props.getProperty("headers.separator") + headerSeparatorPattern = Pattern.compile(headersSeparator) + if (props.containsKey("headers.key.separator")) + headerKeySeparator = props.getProperty("headers.key.separator") if (props.containsKey("ignore.error")) ignoreError = props.getProperty("ignore.error").trim.equalsIgnoreCase("true") + if (headersDelimiter.equals(headersSeparator)) + throw new KafkaException("headers.delimiter and headers.separator may not be equal") + if (headersDelimiter.equals(headerKeySeparator)) + throw new KafkaException("headers.delimiter and headers.key.separator may not be equal") + if (headersSeparator.equals(headerKeySeparator)) + throw new KafkaException("headers.separator and headers.key.separator may not be equal") reader = new BufferedReader(new InputStreamReader(inputStream, StandardCharsets.UTF_8)) } - override def readMessage() = { + override def readMessage(): ProducerRecord[Array[Byte], Array[Byte]] = { lineNumber += 1 - if (printPrompt) - print(">") - (reader.readLine(), parseKey) match { - case (null, _) => null - case (line, true) => - line.indexOf(keySeparator) match { - case -1 => - if (ignoreError) new ProducerRecord(topic, line.getBytes(StandardCharsets.UTF_8)) - else throw new KafkaException(s"No key found on line $lineNumber: $line") - case n => - val value = (if (n + keySeparator.size > line.size) "" else line.substring(n + keySeparator.size)).getBytes(StandardCharsets.UTF_8) - new ProducerRecord(topic, line.substring(0, n).getBytes(StandardCharsets.UTF_8), value) + if (printPrompt) print(">") + val line = reader.readLine() + line match { + case null => null + case line => + + val headers = parse(parseHeader, line, headersDelimiter, "headers delimiter") + val key = parse(parseKey, line.substring(offset(headers)), keySeparator, "key separator") Review comment: Thanks, I do not follow. Would you prefer to pass the index of the starting char as the second parameter for parse(..., index, ... )? In this case parse() would need to have acces to 'line'? Which in turn would mean an additional parameter to parse()? Moreover, wouldn't this then mean the start index needs to passed as well. An adavantage of this might be a more detailed error messages as the full malformed line can be added to the exception instead of just the segment of the line and line number. I guess an other option could be to completley inline parse() at the cost of a 'somewhat' redundant match expression. Anyhow, before we go any further on this: As it is 'just' a 'nit' I dare asking whether it is ok to leave as is? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: jira-unsubscr...@kafka.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org