[GitHub] drill pull request #600: DRILL-4373: Drill and Hive have incompatible timest...

[parthchandra]

Github user parthchandra commented on a diff in the pull request:

https://github.com/apache/drill/pull/600#discussion_r85449218
  
--- Diff: 
exec/java-exec/src/test/java/org/apache/drill/exec/physical/impl/writer/TestParquetWriter.java
 ---
@@ -739,30 +741,76 @@ public void runTestAndValidate(String selection, 
String validationSelection, Str
   }
 
   /*
-  Test the reading of an int96 field. Impala encodes timestamps as int96 
fields
+Impala encodes timestamp values as int96 fields. Test the reading of 
an int96 field with two converters:
+the first one converts parquet INT96 into drill VARBINARY and the 
second one (works while
+store.parquet.reader.int96_as_timestamp option is enabled) converts 
parquet INT96 into drill TIMESTAMP.
*/
   @Test
   public void testImpalaParquetInt96() throws Exception {
 compareParquetReadersColumnar("field_impala_ts", 
"cp.`parquet/int96_impala_1.parquet`");
+try {
+  test("alter session set %s = true", 
ExecConstants.PARQUET_READER_INT96_AS_TIMESTAMP);
+  compareParquetReadersColumnar("field_impala_ts", 
"cp.`parquet/int96_impala_1.parquet`");
+} finally {
+  test("alter session reset %s", 
ExecConstants.PARQUET_READER_INT96_AS_TIMESTAMP);
+}
   }
 
   /*
-  Test the reading of a binary field where data is in dicationary _and_ 
non-dictionary encoded pages
+  Test the reading of a binary field as drill varbinary where data is in 
dicationary _and_ non-dictionary encoded pages
*/
   @Test
-  public void testImpalaParquetVarBinary_DictChange() throws Exception {
+  public void testImpalaParquetBinaryAsVarBinary_DictChange() throws 
Exception {
 compareParquetReadersColumnar("field_impala_ts", 
"cp.`parquet/int96_dict_change.parquet`");
   }
 
   /*
+  Test the reading of a binary field as drill timestamp where data is in 
dicationary _and_ non-dictionary encoded pages
+   */
+  @Test
+  public void testImpalaParquetBinaryAsTimeStamp_DictChange() throws 
Exception {
+final String WORKING_PATH = TestTools.getWorkingPath();
+final String TEST_RES_PATH = WORKING_PATH + "/src/test/resources";
+try {
+  testBuilder()
+  .sqlQuery("select int96_ts from 
dfs_test.`%s/parquet/int96_dict_change`", TEST_RES_PATH)
+  .optionSettingQueriesForTestQuery(
+  "alter session set `%s` = true", 
ExecConstants.PARQUET_READER_INT96_AS_TIMESTAMP)
+  .ordered()
+  
.csvBaselineFile("testframework/testParquetReader/testInt96DictChange/q1.tsv")
+  .baselineTypes(TypeProtos.MinorType.TIMESTAMP)
+  .baselineColumns("int96_ts")
+  .build().run();
+} finally {
+  test("alter system reset `%s`", 
ExecConstants.PARQUET_READER_INT96_AS_TIMESTAMP);
+}
+  }
+
+  /*
  Test the conversion from int96 to impala timestamp
*/
   @Test
-  public void testImpalaParquetTimestampAsInt96() throws Exception {
+  public void testTimestampImpalaConvertFrom() throws Exception {
 compareParquetReadersColumnar("convert_from(field_impala_ts, 
'TIMESTAMP_IMPALA')", "cp.`parquet/int96_impala_1.parquet`");
   }
 
   /*
+ Test reading parquet Int96 as TimeStamp and comparing obtained values 
with the
+ old results (reading the same values as VarBinary and 
convert_fromTIMESTAMP_IMPALA function using)
+   */
+  @Test
+  public void testImpalaParquetTimestampInt96AsTimeStamp() throws 
Exception {
--- End diff --

The test testImpalaParquetTimestampInt96AsTimeStamp fails when run in  a 
different timezone. Can you mark this as @Ignore unless you can fix the test to 
run across different timezones?


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

[GitHub] drill pull request #600: DRILL-4373: Drill and Hive have incompatible timest...

[vdiravka]

Github user vdiravka commented on a diff in the pull request:

https://github.com/apache/drill/pull/600#discussion_r85124582
  
--- Diff: 
exec/java-exec/src/test/java/org/apache/drill/exec/physical/impl/writer/TestParquetWriter.java
 ---
@@ -739,30 +739,54 @@ public void runTestAndValidate(String selection, 
String validationSelection, Str
   }
 
   /*
-  Test the reading of an int96 field. Impala encodes timestamps as int96 
fields
+Impala encodes timestamp values as int96 fields. Test the reading of 
an int96 field with two converters:
+the first one converts parquet INT96 into drill VARBINARY and the 
second one (works while
+store.parquet.reader.int96_as_timestamp option is enabled) converts 
parquet INT96 into drill TIMESTAMP.
*/
   @Test
   public void testImpalaParquetInt96() throws Exception {
 compareParquetReadersColumnar("field_impala_ts", 
"cp.`parquet/int96_impala_1.parquet`");
+try {
+  test("alter session set %s = true", 
ExecConstants.PARQUET_READER_INT96_AS_TIMESTAMP);
+  compareParquetReadersColumnar("field_impala_ts", 
"cp.`parquet/int96_impala_1.parquet`");
--- End diff --

The above comment was addressed to the 
[testImpalaParquetBinaryAsTimeStamp_DictChange](https://github.com/apache/drill/pull/600/commits/81c48c9cd5cdc3905ea78c6cad07a9d818d5026f#diff-aab74a5027942e775c846cebc06c32a4R771)
 method


Test was updated:
An old incorrect file int96_dict_change.parquet was replaced with the new 
two ones with int96 timestamp field and different encoded pages (dictionary and 
non-dictionary).
Csv baseline file also was added.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

[GitHub] drill pull request #600: DRILL-4373: Drill and Hive have incompatible timest...

[parthchandra]

Github user parthchandra commented on a diff in the pull request:

https://github.com/apache/drill/pull/600#discussion_r83908798
  
--- Diff: 
exec/java-exec/src/test/java/org/apache/drill/exec/physical/impl/writer/TestParquetWriter.java
 ---
@@ -739,30 +739,54 @@ public void runTestAndValidate(String selection, 
String validationSelection, Str
   }
 
   /*
-  Test the reading of an int96 field. Impala encodes timestamps as int96 
fields
+Impala encodes timestamp values as int96 fields. Test the reading of 
an int96 field with two converters:
+the first one converts parquet INT96 into drill VARBINARY and the 
second one (works while
+store.parquet.reader.int96_as_timestamp option is enabled) converts 
parquet INT96 into drill TIMESTAMP.
*/
   @Test
   public void testImpalaParquetInt96() throws Exception {
 compareParquetReadersColumnar("field_impala_ts", 
"cp.`parquet/int96_impala_1.parquet`");
+try {
+  test("alter session set %s = true", 
ExecConstants.PARQUET_READER_INT96_AS_TIMESTAMP);
+  compareParquetReadersColumnar("field_impala_ts", 
"cp.`parquet/int96_impala_1.parquet`");
--- End diff --

Github seems to have swallowed the previous comments so including 
@vdiravka's questions here:

>  1) Is it better to compare result with baseline columns and values from 
the file or it is ok to compare with sqlBaselineQuery and disabled new 
PARQUET_READER_INT96_AS_TIMESTAMP option?
> In the process of investigating this test I found that the primitive data 
type of the column in the file int96_dict_change.parquet is BINARY, not INT96.
> 2) I am a little bit confused with this. Do we need convert this BINARY 
to TIMESTAMP as well? CONVERT_FROM function with IMPALA_TIMESTAMP argument 
works properly for this field. I will investigate a little more about does 
impala and hive can store timestamps into parquet BINARY.

For 1) I think it is better to compare values from the file as opposed to 
running with the the PARQUET_READER_INT96_AS_TIMESTAMP disabled.
For 2) Can you correct the int96 data in the file? AFAIK, the data should 
be int96 for the test.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

[GitHub] drill pull request #600: DRILL-4373: Drill and Hive have incompatible timest...

[vdiravka]

Github user vdiravka commented on a diff in the pull request:

https://github.com/apache/drill/pull/600#discussion_r83852721
  
--- Diff: 
exec/java-exec/src/test/java/org/apache/drill/exec/physical/impl/writer/TestParquetWriter.java
 ---
@@ -899,18 +883,21 @@ public void testLastPageOneNull() throws Exception {
 "cp.`parquet/last_page_one_null.parquet`");
   }
 
-  private void compareParquetInt96Converters(String newInt96ConverterQuery,
-  String oldInt96ConverterAndConvertFromFunctionQuery) throws 
Exception {
-testBuilder()
-.ordered()
-.sqlQuery(newInt96ConverterQuery)
-.optionSettingQueriesForTestQuery(
-"alter session set `%s` = true", 
ExecConstants.PARQUET_READER_INT96_AS_TIMESTAMP)
-.sqlBaselineQuery(oldInt96ConverterAndConvertFromFunctionQuery)
-.optionSettingQueriesForBaseline(
-"alter session set `%s` = false", 
ExecConstants.PARQUET_READER_INT96_AS_TIMESTAMP)
-.build()
-.run();
+  private void compareParquetInt96Converters(String selection, String 
table) throws Exception {
+try {
--- End diff --

I refactored my helped method with more clear code.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

[GitHub] drill pull request #600: DRILL-4373: Drill and Hive have incompatible timest...

[vdiravka]

Github user vdiravka commented on a diff in the pull request:

https://github.com/apache/drill/pull/600#discussion_r83853471
  
--- Diff: 
exec/java-exec/src/main/java/org/apache/drill/exec/store/parquet/ParquetReaderUtility.java
 ---
@@ -45,4 +53,34 @@ public static int getIntFromLEBytes(byte[] input, int 
start) {
 }
 return out;
   }
+
+  /**
+   * Utilities for converting from parquet INT96 binary (impala, hive 
timestamp)
+   * to date time value. This utilizes the Joda library.
+   */
+  public static class NanoTimeUtils {
+
+public static final long NANOS_PER_DAY = TimeUnit.DAYS.toNanos(1);
+public static final long NANOS_PER_HOUR = TimeUnit.HOURS.toNanos(1);
+public static final long NANOS_PER_MINUTE = 
TimeUnit.MINUTES.toNanos(1);
+public static final long NANOS_PER_SECOND = 
TimeUnit.SECONDS.toNanos(1);
+public static final long NANOS_PER_MILLISECOND =  
TimeUnit.MILLISECONDS.toNanos(1);
+
+  /**
+   * @param binaryTimeStampValue
+   *  hive, impala timestamp values with nanoseconds precision
+   *  are stored in parquet Binary as INT96
+   *
+   * @return  the number of milliseconds since January 1, 1970, 00:00:00 
GMT
+   *  represented by @param binaryTimeStampValue .
+   */
+public static long getDateTimeValueFromBinary(Binary 
binaryTimeStampValue) {
+  NanoTime nt = NanoTime.fromBinary(binaryTimeStampValue);
+  int julianDay = nt.getJulianDay();
+  long nanosOfDay = nt.getTimeOfDayNanos();
+  return DateTimeUtils.fromJulianDay(julianDay-0.5d) + 
nanosOfDay/NANOS_PER_MILLISECOND;
--- End diff --

The comment is removed. And numbers are replaced with constants from 
ParquetReaderUtility and DateTimeConstants.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

[GitHub] drill pull request #600: DRILL-4373: Drill and Hive have incompatible timest...

[bitblender]

Github user bitblender commented on a diff in the pull request:

https://github.com/apache/drill/pull/600#discussion_r83761501
  
--- Diff: 
exec/java-exec/src/main/java/org/apache/drill/exec/store/parquet/ParquetReaderUtility.java
 ---
@@ -45,4 +53,34 @@ public static int getIntFromLEBytes(byte[] input, int 
start) {
 }
 return out;
   }
+
+  /**
+   * Utilities for converting from parquet INT96 binary (impala, hive 
timestamp)
+   * to date time value. This utilizes the Joda library.
+   */
+  public static class NanoTimeUtils {
+
+public static final long NANOS_PER_DAY = TimeUnit.DAYS.toNanos(1);
+public static final long NANOS_PER_HOUR = TimeUnit.HOURS.toNanos(1);
+public static final long NANOS_PER_MINUTE = 
TimeUnit.MINUTES.toNanos(1);
+public static final long NANOS_PER_SECOND = 
TimeUnit.SECONDS.toNanos(1);
+public static final long NANOS_PER_MILLISECOND =  
TimeUnit.MILLISECONDS.toNanos(1);
+
+  /**
+   * @param binaryTimeStampValue
+   *  hive, impala timestamp values with nanoseconds precision
+   *  are stored in parquet Binary as INT96
+   *
+   * @return  the number of milliseconds since January 1, 1970, 00:00:00 
GMT
+   *  represented by @param binaryTimeStampValue .
+   */
+public static long getDateTimeValueFromBinary(Binary 
binaryTimeStampValue) {
+  NanoTime nt = NanoTime.fromBinary(binaryTimeStampValue);
+  int julianDay = nt.getJulianDay();
+  long nanosOfDay = nt.getTimeOfDayNanos();
+  return DateTimeUtils.fromJulianDay(julianDay-0.5d) + 
nanosOfDay/NANOS_PER_MILLISECOND;
--- End diff --

Sorry for the late reply. For some reason, I did not see these comments 
till now. 
About 1) Yes, you are correct. I just want the comments in 
ConvertFromImpalaTimestamp to be removed.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

[GitHub] drill pull request #600: DRILL-4373: Drill and Hive have incompatible timest...

[vdiravka]

Github user vdiravka commented on a diff in the pull request:

https://github.com/apache/drill/pull/600#discussion_r83710133
  
--- Diff: 
exec/java-exec/src/test/java/org/apache/drill/exec/physical/impl/writer/TestParquetWriter.java
 ---
@@ -754,15 +764,45 @@ public void testImpalaParquetVarBinary_DictChange() 
throws Exception {
 compareParquetReadersColumnar("field_impala_ts", 
"cp.`parquet/int96_dict_change.parquet`");
   }
 
+  @Test
+  public void testImpalaParquetBinaryTimeStamp_DictChange() throws 
Exception {
+try {
+  test("alter session set %s = true", 
ExecConstants.PARQUET_READER_INT96_AS_TIMESTAMP);
+  compareParquetReadersColumnar("field_impala_ts", 
"cp.`parquet/int96_dict_change.parquet`");
--- End diff --

1. Is it better to compare result with baseline columns and values from the 
file or it is ok to compare with `sqlBaselineQuery` and disabled new 
`PARQUET_READER_INT96_AS_TIMESTAMP` option?
2. In the process of investigating this test I found that the primitive 
data type of the column in the file `int96_dict_change.parquet`  is BINARY, not 
INT96.  
I am a little bit confused with this. Do we need convert this BINARY to 
TIMESTAMP as well?
CONVERT_FROM function with IMPALA_TIMESTAMP argument works properly for 
this field.
I will investigate a little more about does impala and hive can store 
timestamps into parquet BINARY. 


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

[GitHub] drill pull request #600: DRILL-4373: Drill and Hive have incompatible timest...

[parthchandra]

Github user parthchandra commented on a diff in the pull request:

https://github.com/apache/drill/pull/600#discussion_r83284350
  
--- Diff: 
exec/java-exec/src/test/java/org/apache/drill/exec/physical/impl/writer/TestParquetWriter.java
 ---
@@ -754,15 +764,45 @@ public void testImpalaParquetVarBinary_DictChange() 
throws Exception {
 compareParquetReadersColumnar("field_impala_ts", 
"cp.`parquet/int96_dict_change.parquet`");
   }
 
+  @Test
+  public void testImpalaParquetBinaryTimeStamp_DictChange() throws 
Exception {
+try {
+  test("alter session set %s = true", 
ExecConstants.PARQUET_READER_INT96_AS_TIMESTAMP);
+  compareParquetReadersColumnar("field_impala_ts", 
"cp.`parquet/int96_dict_change.parquet`");
--- End diff --

This is not a good enough test. Both the baseline and test case queries 
will use the getDateTimeValueFromBinary method and if there is a bug in that 
method, the test will still pass as both will produce the same incorrect value. 
Better to compare with the actual baseline value in the file.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

[GitHub] drill pull request #600: DRILL-4373: Drill and Hive have incompatible timest...

[parthchandra]

Github user parthchandra commented on a diff in the pull request:

https://github.com/apache/drill/pull/600#discussion_r83281827
  
--- Diff: 
exec/java-exec/src/main/java/org/apache/drill/exec/ExecConstants.java ---
@@ -132,6 +132,8 @@
   OptionValidator PARQUET_VECTOR_FILL_CHECK_THRESHOLD_VALIDATOR = new 
PositiveLongValidator(PARQUET_VECTOR_FILL_CHECK_THRESHOLD, 100l, 10l);
   String PARQUET_NEW_RECORD_READER = "store.parquet.use_new_reader";
   OptionValidator PARQUET_RECORD_READER_IMPLEMENTATION_VALIDATOR = new 
BooleanValidator(PARQUET_NEW_RECORD_READER, false);
+  String PARQUET_READER_INT96_AS_TIMESTAMP = 
"store.parquet.int96_as_timestamp";
--- End diff --

Should rename this to store.parquet.reader.abc to make it clear this is a 
reader only property


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

[GitHub] drill pull request #600: DRILL-4373: Drill and Hive have incompatible timest...

[bitblender]

Github user bitblender commented on a diff in the pull request:

https://github.com/apache/drill/pull/600#discussion_r82314071
  
--- Diff: 
exec/java-exec/src/main/java/org/apache/drill/exec/store/parquet/ParquetReaderUtility.java
 ---
@@ -45,4 +53,34 @@ public static int getIntFromLEBytes(byte[] input, int 
start) {
 }
 return out;
   }
+
+  /**
+   * Utilities for converting from parquet INT96 binary (impala, hive 
timestamp)
+   * to date time value. This utilizes the Joda library.
+   */
+  public static class NanoTimeUtils {
+
+public static final long NANOS_PER_DAY = TimeUnit.DAYS.toNanos(1);
+public static final long NANOS_PER_HOUR = TimeUnit.HOURS.toNanos(1);
+public static final long NANOS_PER_MINUTE = 
TimeUnit.MINUTES.toNanos(1);
+public static final long NANOS_PER_SECOND = 
TimeUnit.SECONDS.toNanos(1);
+public static final long NANOS_PER_MILLISECOND =  
TimeUnit.MILLISECONDS.toNanos(1);
+
+  /**
+   * @param binaryTimeStampValue
+   *  hive, impala timestamp values with nanoseconds precision
+   *  are stored in parquet Binary as INT96
+   *
+   * @return  the number of milliseconds since January 1, 1970, 00:00:00 
GMT
+   *  represented by @param binaryTimeStampValue .
+   */
+public static long getDateTimeValueFromBinary(Binary 
binaryTimeStampValue) {
+  NanoTime nt = NanoTime.fromBinary(binaryTimeStampValue);
+  int julianDay = nt.getJulianDay();
+  long nanosOfDay = nt.getTimeOfDayNanos();
+  return DateTimeUtils.fromJulianDay(julianDay-0.5d) + 
nanosOfDay/NANOS_PER_MILLISECOND;
--- End diff --

1.  I would recommend not using Joda. Do the calculations directly, like in 
ConvertFromImpalaTimestamp. Joda uses non-standard, hence  confusing, 
terminology. What Joda calls and uses as JulianDay, is actually Julian Date. 
Seems like you have identified this discrepancy and adjusted for it by 
subtracting 0.5 from _julianDay_. 

Note: (I guess you have already figured this out) : The actual code and 
the Joda code in the comment, in ConvertFromImpalaTimestamp, are inconsistent. 
Took me a day to figure out the reason behind this ! A bug should be opened to 
delete the comment. 

2. Can you please also leave a comment stating that 2440588 is the JDN for 
the Unix Epoch.

3. Please leave a comment stating that the order of the calls to get 
_julianDay_ and _nanosOfDay_ matters. You can do this by just stating how 
timestamps are stored in INT96 i.e 32-bit JDN followed by 64-bit nanosOfDay.

4. Consistent(single or none) spacing for binary operators (+-/) used here 
would be nice. Single spacing would be preferable.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

[GitHub] drill pull request #600: DRILL-4373: Drill and Hive have incompatible timest...

[vdiravka]

GitHub user vdiravka opened a pull request:

https://github.com/apache/drill/pull/600

DRILL-4373: Drill and Hive have incompatible timestamp representations in 
parquet

- added sys/sess option "store.parquet.int96_as_timestamp";
- added int96 to timestamp converter for both readers;
- added unit tests;

You can merge this pull request into a Git repository by running:

$ git pull https://github.com/vdiravka/drill DRILL-4373

Alternatively you can review and apply these changes as the patch at:

https://github.com/apache/drill/pull/600.patch

To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:

This closes #600


commit 0d768c42f7c732360cafcacc91e29b67ae44fca4
Author: Vitalii Diravka 
Date:   2016-09-02T21:43:50Z

DRILL-4373: Drill and Hive have incompatible timestamp representations in 
parquet
- added sys/sess option "store.parquet.int96_as_timestamp";
- added int96 to timestamp converter for both readers;
- added unit tests;




---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---