Mailing-List: contact issues-help@drill.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@drill.apache.org
Date: Tue, 29 Aug 2017 18:23:00 +0000 (UTC)
From: "ASF GitHub Bot (JIRA)" <jira@apache.org>
To: issues@drill.apache.org
Message-ID: <JIRA.12929429.1452591635000.150299.1504030980549@Atlassian.JIRA>
In-Reply-To: <JIRA.12929429.1452591635000@Atlassian.JIRA>
References: <JIRA.12929429.1452591635000@Atlassian.JIRA> <JIRA.12929429.1452591635904@jira-lw-us.apache.org>
Subject: [jira] [Commented] (DRILL-4264) Allow field names to include dots
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
archived-at: Tue, 29 Aug 2017 18:23:07 -0000


    [ https://issues.apache.org/jira/browse/DRILL-4264?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16145834#comment-16145834 ] 

ASF GitHub Bot commented on DRILL-4264:
---------------------------------------

Github user paul-rogers commented on a diff in the pull request:

    https://github.com/apache/drill/pull/909#discussion_r135868684
  
    --- Diff: contrib/format-maprdb/src/main/java/org/apache/drill/exec/store/mapr/db/binary/CompareFunctionsProcessor.java ---
    @@ -82,466 +45,37 @@ public static CompareFunctionsProcessor process(FunctionCall call, boolean nullC
             LogicalExpression swapArg = valueArg;
             valueArg = nameArg;
             nameArg = swapArg;
    -        evaluator.functionName = COMPARE_FUNCTIONS_TRANSPOSE_MAP.get(functionName);
    +        evaluator.setFunctionName(COMPARE_FUNCTIONS_TRANSPOSE_MAP.get(functionName));
           }
    -      evaluator.success = nameArg.accept(evaluator, valueArg);
    +      evaluator.setSuccess(nameArg.accept(evaluator, valueArg));
         } else if (nullComparatorSupported && call.args.get(0) instanceof SchemaPath) {
    -      evaluator.success = true;
    -      evaluator.path = (SchemaPath) nameArg;
    +      evaluator.setSuccess(true);
    +      evaluator.setPath((SchemaPath) nameArg);
         }
     
         return evaluator;
       }
     
    -  public CompareFunctionsProcessor(String functionName) {
    -    this.success = false;
    -    this.functionName = functionName;
    -    this.isEqualityFn = COMPARE_FUNCTIONS_TRANSPOSE_MAP.containsKey(functionName)
    -        && COMPARE_FUNCTIONS_TRANSPOSE_MAP.get(functionName).equals(functionName);
    -    this.isRowKeyPrefixComparison = false;
    -    this.sortOrderAscending = true;
    -  }
    -
    -  public byte[] getValue() {
    -    return value;
    -  }
    -
    -  public boolean isSuccess() {
    -    return success;
    -  }
    -
    -  public SchemaPath getPath() {
    -    return path;
    -  }
    -
    -  public String getFunctionName() {
    -    return functionName;
    -  }
    -
    -  public boolean isRowKeyPrefixComparison() {
    -	return isRowKeyPrefixComparison;
    -  }
    -
    -  public byte[] getRowKeyPrefixStartRow() {
    -    return rowKeyPrefixStartRow;
    -  }
    -
    -  public byte[] getRowKeyPrefixStopRow() {
    -  return rowKeyPrefixStopRow;
    -  }
    -
    -  public Filter getRowKeyPrefixFilter() {
    -  return rowKeyPrefixFilter;
    -  }
    -
    -  public boolean isSortOrderAscending() {
    -    return sortOrderAscending;
    -  }
    -
       @Override
    -  public Boolean visitCastExpression(CastExpression e, LogicalExpression valueArg) throws RuntimeException {
    -    if (e.getInput() instanceof CastExpression || e.getInput() instanceof SchemaPath) {
    -      return e.getInput().accept(this, valueArg);
    -    }
    -    return false;
    -  }
    -
    -  @Override
    -  public Boolean visitConvertExpression(ConvertExpression e, LogicalExpression valueArg) throws RuntimeException {
    -    if (e.getConvertFunction() == ConvertExpression.CONVERT_FROM) {
    -
    -      String encodingType = e.getEncodingType();
    -      int prefixLength    = 0;
    -
    -      // Handle scan pruning in the following scenario:
    -      // The row-key is a composite key and the CONVERT_FROM() function has byte_substr() as input function which is
    -      // querying for the first few bytes of the row-key(start-offset 1)
    -      // Example WHERE clause:
    -      // CONVERT_FROM(BYTE_SUBSTR(row_key, 1, 8), 'DATE_EPOCH_BE') < DATE '2015-06-17'
    -      if (e.getInput() instanceof FunctionCall) {
    -
    -        // We can prune scan range only for big-endian encoded data
    -        if (encodingType.endsWith("_BE") == false) {
    -          return false;
    -        }
    -
    -        FunctionCall call = (FunctionCall)e.getInput();
    -        String functionName = call.getName();
    -        if (!functionName.equalsIgnoreCase("byte_substr")) {
    -          return false;
    -        }
    -
    -        LogicalExpression nameArg = call.args.get(0);
    -        LogicalExpression valueArg1 = call.args.size() >= 2 ? call.args.get(1) : null;
    -        LogicalExpression valueArg2 = call.args.size() >= 3 ? call.args.get(2) : null;
    -
    -        if (((nameArg instanceof SchemaPath) == false) ||
    -             (valueArg1 == null) || ((valueArg1 instanceof IntExpression) == false) ||
    -             (valueArg2 == null) || ((valueArg2 instanceof IntExpression) == false)) {
    -          return false;
    -        }
    -
    -        boolean isRowKey = ((SchemaPath)nameArg).getAsUnescapedPath().equals(DrillHBaseConstants.ROW_KEY);
    -        int offset = ((IntExpression)valueArg1).getInt();
    -
    -        if (!isRowKey || (offset != 1)) {
    -          return false;
    -        }
    -
    -        this.path    = (SchemaPath)nameArg;
    -        prefixLength = ((IntExpression)valueArg2).getInt();
    -        this.isRowKeyPrefixComparison = true;
    -        return visitRowKeyPrefixConvertExpression(e, prefixLength, valueArg);
    -      }
    -
    -      if (e.getInput() instanceof SchemaPath) {
    -        ByteBuf bb = null;
    -
    -        switch (encodingType) {
    -        case "INT_BE":
    -        case "INT":
    -        case "UINT_BE":
    -        case "UINT":
    -        case "UINT4_BE":
    -        case "UINT4":
    -          if (valueArg instanceof IntExpression
    -              && (isEqualityFn || encodingType.startsWith("U"))) {
    -            bb = newByteBuf(4, encodingType.endsWith("_BE"));
    -            bb.writeInt(((IntExpression)valueArg).getInt());
    -          }
    -          break;
    -        case "BIGINT_BE":
    -        case "BIGINT":
    -        case "UINT8_BE":
    -        case "UINT8":
    -          if (valueArg instanceof LongExpression
    -              && (isEqualityFn || encodingType.startsWith("U"))) {
    -            bb = newByteBuf(8, encodingType.endsWith("_BE"));
    -            bb.writeLong(((LongExpression)valueArg).getLong());
    -          }
    -          break;
    -        case "FLOAT":
    -          if (valueArg instanceof FloatExpression && isEqualityFn) {
    -            bb = newByteBuf(4, true);
    -            bb.writeFloat(((FloatExpression)valueArg).getFloat());
    -          }
    -          break;
    -        case "DOUBLE":
    -          if (valueArg instanceof DoubleExpression && isEqualityFn) {
    -            bb = newByteBuf(8, true);
    -            bb.writeDouble(((DoubleExpression)valueArg).getDouble());
    -          }
    -          break;
    -        case "TIME_EPOCH":
    -        case "TIME_EPOCH_BE":
    -          if (valueArg instanceof TimeExpression) {
    -            bb = newByteBuf(8, encodingType.endsWith("_BE"));
    -            bb.writeLong(((TimeExpression)valueArg).getTime());
    -          }
    -          break;
    -        case "DATE_EPOCH":
    -        case "DATE_EPOCH_BE":
    -          if (valueArg instanceof DateExpression) {
    -            bb = newByteBuf(8, encodingType.endsWith("_BE"));
    -            bb.writeLong(((DateExpression)valueArg).getDate());
    -          }
    -          break;
    -        case "BOOLEAN_BYTE":
    -          if (valueArg instanceof BooleanExpression) {
    -            bb = newByteBuf(1, false /* does not matter */);
    -            bb.writeByte(((BooleanExpression)valueArg).getBoolean() ? 1 : 0);
    -          }
    -          break;
    -        case "DOUBLE_OB":
    -        case "DOUBLE_OBD":
    -          if (valueArg instanceof DoubleExpression) {
    -            bb = newByteBuf(9, true);
    -            PositionedByteRange br = new SimplePositionedMutableByteRange(bb.array(), 0, 9);
    -            if (encodingType.endsWith("_OBD")) {
    -              org.apache.hadoop.hbase.util.OrderedBytes.encodeFloat64(br,
    -                  ((DoubleExpression)valueArg).getDouble(), Order.DESCENDING);
    -              this.sortOrderAscending = false;
    -            } else {
    -              org.apache.hadoop.hbase.util.OrderedBytes.encodeFloat64(br,
    -                  ((DoubleExpression)valueArg).getDouble(), Order.ASCENDING);
    -            }
    +  protected ByteBuf getByteBuf(LogicalExpression valueArg, String encodingType) {
    --- End diff --
    
    Thanks!


> Allow field names to include dots
> ---------------------------------
>
>                 Key: DRILL-4264
>                 URL: https://issues.apache.org/jira/browse/DRILL-4264
>             Project: Apache Drill
>          Issue Type: Improvement
>          Components: Execution - Codegen
>            Reporter: Alex
>            Assignee: Volodymyr Vysotskyi
>              Labels: doc-impacting
>             Fix For: 1.12.0
>
>
> If you have some json data like this...
> {code:javascript}
>     {
>       "0.0.1":{
>         "version":"0.0.1",
>         "date_created":"2014-03-15"
>       },
>       "0.1.2":{
>         "version":"0.1.2",
>         "date_created":"2014-05-21"
>       }
>     }
> {code}
> ... there is no way to select any of the rows since their identifiers contain dots and when trying to select them, Drill throws the following error:
> Error: SYSTEM ERROR: UnsupportedOperationException: Unhandled field reference "0.0.1"; a field reference identifier must not have the form of a qualified name
> This must be fixed since there are many json data files containing dots in some of the keys (e.g. when specifying version numbers etc)


--
This message was sent by Atlassian JIRA
(v6.4.14#64029)