stevedlawrence commented on a change in pull request #210: Add midBitsToEnd layer transform URL: https://github.com/apache/incubator-daffodil/pull/210#discussion_r278114832
########## File path: daffodil-runtime1/src/main/scala/org/apache/daffodil/layers/midBitsToEndTransformer.scala ########## @@ -0,0 +1,345 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one or more + * contributor license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright ownership. + * The ASF licenses this file to You under the Apache License, Version 2.0 + * (the "License"); you may not use this file except in compliance with + * the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ +package org.apache.daffodil.layers + +import org.apache.daffodil.processors.TermRuntimeData +import org.apache.daffodil.util.Maybe +import org.apache.daffodil.processors.LayerCharsetEv +import org.apache.daffodil.schema.annotation.props.gen.LayerLengthKind +import org.apache.daffodil.processors.LayerLengthInBytesEv +import org.apache.daffodil.schema.annotation.props.gen.LayerLengthUnits +import org.apache.daffodil.processors.LayerBoundaryMarkEv +import org.apache.daffodil.processors.LayerTransformArgsEv +import java.io.InputStream +import java.io.OutputStream +import org.apache.daffodil.processors.unparsers.UState +import org.apache.daffodil.processors.parsers.PState +import org.apache.daffodil.exceptions.Assert +import org.apache.daffodil.processors.ParseOrUnparseState +import org.apache.daffodil.processors.parsers.Parser + +import org.apache.daffodil.processors.ExplicitLengthEv +import org.apache.daffodil.io.ExplicitLengthLimitingStream +import java.io.ByteArrayOutputStream +import java.io.ByteArrayInputStream +import org.apache.daffodil.util.MaybeULong +import org.apache.daffodil.processors.Failure + +class midBitsToEndTransformer( + layerLengthInBytesEv: LayerLengthInBytesEv, + layerTransformArgsEv: LayerTransformArgsEv) + extends LayerTransformer { + /* + * For unparsing, it is more or less unavoidable that we cannot stream, as we need to reach + * the end of the input before we can provide the mid-bits that were moved. + * In practice, it is not expected that this transform will be used on "large" blocks, + * so it is not worth the effort to figure out how to implement some form of streaming. + */ + + var length: Int = -1 + + var firstBitIndex0b: Int = -1 + var numBits: Int = -1 + + /* + * Note the extra padding bytes. One of these bytes is just to accomodate rounding from integer division. + * The other one is because, to simplify the code, a portion of the encoder may attempt to read up to 7 extra bits, + * then discard them. (This would happen when the last bit that gets moved is the first bit of a byte in the encoded form; + * the encoder will attempt to read enough bits to be able to fully replace the entire byte, then mask out the lower 7 bits) + */ + def tmpByteCount = numBits / 8 + 2 + def buffSize = length + tmpByteCount + + def byteShift = numBits / 8 + def bitShift = numBits % 8 + + def firstBitByteIndex0b = (firstBitIndex0b / 8) + def lastBitByteIndex0b = ((firstBitIndex0b + numBits) / 8) + def firstBitBitIndex0b = firstBitIndex0b % 8 + def lastBitBitIndex0b = (firstBitIndex0b + numBits) % 8 + + private def buffToHex(buff: Array[Byte]): String = { + buff.map(b => String.format("%02x", Byte.box(b))).mkString(" ") + } + + private def initialize(state: ParseOrUnparseState) { + length = layerLengthInBytesEv.evaluate(state).toInt + val argsString = layerTransformArgsEv.evaluate(state) + .trim() + .split("\\ +") + if (argsString.length != 2) { + state.SDE( + "midBitsToEnd layer transform requires exactly 2 arguments, recieved %s: %s}", + argsString.length, argsString.mkString(",")) + } + try { + //DFDL is 1-indexed, we are working 0-indexed + firstBitIndex0b = argsString(0).toInt - 1 + numBits = argsString(1).toInt + } catch { + case _: NumberFormatException => state.SDE( + "midBitsToEnd layer transform requires Integer arguements, recieved: %s", + argsString.mkString(",")) + } + + } Review comment: This function is essentially just parses and validates the arguments. Two thoughts: 1. We should probably rename this to make it more clear that the purpose here is to parse argments. 2. This will not get called until runtime so if someone uses invalid transformArgs, we aren't going to knowl until they try to parse/unparse something. Ideally that validation would be detected at schema compile time when the transform properties are constant. So maybe this needs to be split up into two functions? One that does just argument validation (probably a member of the factory object?) and one that actually extracts the values at runtime? ---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services
