hive-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Alexander Pivovarov" <apivova...@gmail.com>
Subject Re: Review Request 30335: Build UDF TRUNC to implement FIRST_DAY as compared with LAST_DAY
Date Sat, 21 Feb 2015 03:20:01 GMT

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/30335/#review73391
-----------------------------------------------------------



ql/src/java/org/apache/hadoop/hive/ql/udf/generic/GenericUDFTrunc.java
<https://reviews.apache.org/r/30335/#comment119699>

    according to java coding convention UPPER_CASE var names used only for constants (static
final)



ql/src/java/org/apache/hadoop/hive/ql/udf/generic/GenericUDFTrunc.java
<https://reviews.apache.org/r/30335/#comment119698>

    why you read FORMAT_INPUT only for the first row and then cache it?
    what if format is dynamic?
    
    As I mentioned before, use ConstantObjectInspector to get format in inintialize() method
    
    if it's not constant then you have to read format valus in evaluate() method for every
single row. don't cache format if it is not constant
    
    look at decode udf as an example



ql/src/java/org/apache/hadoop/hive/ql/udf/generic/GenericUDFTrunc.java
<https://reviews.apache.org/r/30335/#comment119700>

    I think ("MM".equals(formatInput) || "MON".equals(formatInput) || "MONTH".equals(formatInput))
is faster than calling contains() on ArrayList



ql/src/java/org/apache/hadoop/hive/ql/udf/generic/GenericUDFTrunc.java
<https://reviews.apache.org/r/30335/#comment119701>

    ("YEAR".equals(formatInput) || "YYYY".equals(formatInput) || "YY".equals(formatInput))


- Alexander Pivovarov


On Feb. 21, 2015, 2:54 a.m., XIAOBING ZHOU wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/30335/
> -----------------------------------------------------------
> 
> (Updated Feb. 21, 2015, 2:54 a.m.)
> 
> 
> Review request for hive.
> 
> 
> Repository: hive-git
> 
> 
> Description
> -------
> 
> Hive already supports LAST_DAY UDF, in some cases, FIRST_DAY is necessary to do date/timestamp
related computation. This JIRA is to track such an implementation.
> 
> https://issues.apache.org/jira/browse/HIVE-9480
> 
> 
> Diffs
> -----
> 
>   ql/src/java/org/apache/hadoop/hive/ql/exec/FunctionRegistry.java bfeb33c 
>   ql/src/java/org/apache/hadoop/hive/ql/udf/generic/GenericUDFTrunc.java PRE-CREATION

>   ql/src/test/org/apache/hadoop/hive/ql/udf/generic/TestGenericUDFTrunc.java PRE-CREATION

>   ql/src/test/queries/clientnegative/udf_trunc_error1.q PRE-CREATION 
>   ql/src/test/queries/clientnegative/udf_trunc_error2.q PRE-CREATION 
>   ql/src/test/queries/clientpositive/udf_trunc.q PRE-CREATION 
>   ql/src/test/results/clientnegative/udf_trunc_error1.q.out PRE-CREATION 
>   ql/src/test/results/clientnegative/udf_trunc_error2.q.out PRE-CREATION 
>   ql/src/test/results/clientpositive/show_functions.q.out d4b0650 
>   ql/src/test/results/clientpositive/udf_trunc.q.out PRE-CREATION 
> 
> Diff: https://reviews.apache.org/r/30335/diff/
> 
> 
> Testing
> -------
> 
> Unit tests done in ql/src/test/org/apache/hadoop/hive/ql/udf/generic/TestGenericUDFFirstDay.java
> 
> 
> Thanks,
> 
> XIAOBING ZHOU
> 
>


Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message