HBase
  1. HBase
  2. HBASE-4410

FilterList.filterKeyValue can return suboptimal ReturnCodes

    Details

    • Type: Improvement Improvement
    • Status: Open
    • Priority: Minor Minor
    • Resolution: Unresolved
    • Affects Version/s: None
    • Fix Version/s: None
    • Component/s: Filters
    • Labels:
      None

      Description

      FilterList.filterKeyValue does not always return the most optimal ReturnCode in both the AND and OR conditions.

      For example, if you have F1 AND F2, F1 returns SKIP. It immediately returns the SKIP. However, if F2 would have returned NEXT_COL or NEXT_ROW or SEEK_NEXT_USING_HINT, we would actually be able to return the more optimal ReturnCode from F2.

      For AND conditions, we can always pick the most restrictive return code.

      For OR conditions, we must always pick the least restrictive return code.

      This JIRA is to review the FilterList.filterKeyValue() method to try and make it more optimal and to add a new unit test which verifies the correct behavior.

        Activity

        Hide
        stack added a comment -

        Moving an improvement out of 0.96; move it back if you disagree JG

        Show
        stack added a comment - Moving an improvement out of 0.96; move it back if you disagree JG
        Hide
        Jonathan Gray added a comment -

        Not working on this right now, punt it! Thanks Lars

        On Mar 21, 2012, at 3:31 PM, "Lars Hofhansl (Updated) (JIRA)"

        Show
        Jonathan Gray added a comment - Not working on this right now, punt it! Thanks Lars On Mar 21, 2012, at 3:31 PM, "Lars Hofhansl (Updated) (JIRA)"
        Hide
        Lars Hofhansl added a comment -

        This would have been nice for 0.94. I think it's too late, though.
        @Jon: Are you still working on this?

        Show
        Lars Hofhansl added a comment - This would have been nice for 0.94. I think it's too late, though. @Jon: Are you still working on this?
        Hide
        Ted Yu added a comment -

        Optimization for 0.94

        Show
        Ted Yu added a comment - Optimization for 0.94
        Hide
        Jonathan Gray added a comment -

        My comeback is that Lars is right and I f-ed it up. I was supposed to make a new patch but forgot about this. I was a bit angry I came up with such a nice elegant solution that was fundamentally broken.

        Will try to get to this next week.

        Show
        Jonathan Gray added a comment - My comeback is that Lars is right and I f-ed it up. I was supposed to make a new patch but forgot about this. I was a bit angry I came up with such a nice elegant solution that was fundamentally broken. Will try to get to this next week.
        Hide
        stack added a comment -

        Any comeback on the above feedback?

        Show
        stack added a comment - Any comeback on the above feedback?
        Hide
        jiraposter@reviews.apache.org added a comment -

        -----------------------------------------------------------
        This is an automatically generated e-mail. To reply, visit:
        https://reviews.apache.org/r/1908/#review1960
        -----------------------------------------------------------

        /src/main/java/org/apache/hadoop/hbase/filter/Filter.java
        <https://reviews.apache.org/r/1908/#comment4439>

        Would you ever want to insert a value in between any of these? Or do the values of these ordinals have no meaning?

        /src/main/java/org/apache/hadoop/hbase/filter/FilterList.java
        <https://reviews.apache.org/r/1908/#comment4441>

        Is this comment right? MUST_PASS_ALL seems 'most restrictive', not 'least restrictive'.

        • Michael

        On 2011-09-14 21:52:34, Jonathan Gray wrote:

        -----------------------------------------------------------

        This is an automatically generated e-mail. To reply, visit:

        https://reviews.apache.org/r/1908/

        -----------------------------------------------------------

        (Updated 2011-09-14 21:52:34)

        Review request for hbase, Dhruba Borthakur, Michael Stack, Prakash Khemani, and Kannan Muthukkaruppan.

        Summary

        -------

        FilterList.filterKeyValue does not always return the most optimal ReturnCode in both the AND and OR conditions.

        For example, if you have F1 AND F2, F1 returns SKIP. It immediately returns the SKIP. However, if F2 would have returned NEXT_COL or NEXT_ROW or SEEK_NEXT_USING_HINT, we would actually be able to return the more optimal ReturnCode from F2.

        For AND conditions, we can always pick the most restrictive return code.

        For OR conditions, we must always pick the least restrictive return code.

        This JIRA is to review the FilterList.filterKeyValue() method to try and make it more optimal and to add a new unit test which verifies the correct behavior.

        This addresses bug HBASE-4410.

        https://issues.apache.org/jira/browse/HBASE-4410

        Diffs

        -----

        /src/main/java/org/apache/hadoop/hbase/filter/Filter.java 1170860

        /src/main/java/org/apache/hadoop/hbase/filter/FilterList.java 1170860

        /src/test/java/org/apache/hadoop/hbase/filter/TestFilterList.java 1170860

        Diff: https://reviews.apache.org/r/1908/diff

        Testing

        -------

        Adds new tests to TestFilterList.

        Thanks,

        Jonathan

        Show
        jiraposter@reviews.apache.org added a comment - ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/1908/#review1960 ----------------------------------------------------------- /src/main/java/org/apache/hadoop/hbase/filter/Filter.java < https://reviews.apache.org/r/1908/#comment4439 > Would you ever want to insert a value in between any of these? Or do the values of these ordinals have no meaning? /src/main/java/org/apache/hadoop/hbase/filter/FilterList.java < https://reviews.apache.org/r/1908/#comment4441 > Is this comment right? MUST_PASS_ALL seems 'most restrictive', not 'least restrictive'. Michael On 2011-09-14 21:52:34, Jonathan Gray wrote: ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/1908/ ----------------------------------------------------------- (Updated 2011-09-14 21:52:34) Review request for hbase, Dhruba Borthakur, Michael Stack, Prakash Khemani, and Kannan Muthukkaruppan. Summary ------- FilterList.filterKeyValue does not always return the most optimal ReturnCode in both the AND and OR conditions. For example, if you have F1 AND F2, F1 returns SKIP. It immediately returns the SKIP. However, if F2 would have returned NEXT_COL or NEXT_ROW or SEEK_NEXT_USING_HINT, we would actually be able to return the more optimal ReturnCode from F2. For AND conditions, we can always pick the most restrictive return code. For OR conditions, we must always pick the least restrictive return code. This JIRA is to review the FilterList.filterKeyValue() method to try and make it more optimal and to add a new unit test which verifies the correct behavior. This addresses bug HBASE-4410 . https://issues.apache.org/jira/browse/HBASE-4410 Diffs ----- /src/main/java/org/apache/hadoop/hbase/filter/Filter.java 1170860 /src/main/java/org/apache/hadoop/hbase/filter/FilterList.java 1170860 /src/test/java/org/apache/hadoop/hbase/filter/TestFilterList.java 1170860 Diff: https://reviews.apache.org/r/1908/diff Testing ------- Adds new tests to TestFilterList. Thanks, Jonathan
        Hide
        jiraposter@reviews.apache.org added a comment -

        -----------------------------------------------------------
        This is an automatically generated e-mail. To reply, visit:
        https://reviews.apache.org/r/1908/#review1961
        -----------------------------------------------------------

        Ship it!

        Looks good. Can address nits on commit.

        • Michael

        On 2011-09-14 21:52:34, Jonathan Gray wrote:

        -----------------------------------------------------------

        This is an automatically generated e-mail. To reply, visit:

        https://reviews.apache.org/r/1908/

        -----------------------------------------------------------

        (Updated 2011-09-14 21:52:34)

        Review request for hbase, Dhruba Borthakur, Michael Stack, Prakash Khemani, and Kannan Muthukkaruppan.

        Summary

        -------

        FilterList.filterKeyValue does not always return the most optimal ReturnCode in both the AND and OR conditions.

        For example, if you have F1 AND F2, F1 returns SKIP. It immediately returns the SKIP. However, if F2 would have returned NEXT_COL or NEXT_ROW or SEEK_NEXT_USING_HINT, we would actually be able to return the more optimal ReturnCode from F2.

        For AND conditions, we can always pick the most restrictive return code.

        For OR conditions, we must always pick the least restrictive return code.

        This JIRA is to review the FilterList.filterKeyValue() method to try and make it more optimal and to add a new unit test which verifies the correct behavior.

        This addresses bug HBASE-4410.

        https://issues.apache.org/jira/browse/HBASE-4410

        Diffs

        -----

        /src/main/java/org/apache/hadoop/hbase/filter/Filter.java 1170860

        /src/main/java/org/apache/hadoop/hbase/filter/FilterList.java 1170860

        /src/test/java/org/apache/hadoop/hbase/filter/TestFilterList.java 1170860

        Diff: https://reviews.apache.org/r/1908/diff

        Testing

        -------

        Adds new tests to TestFilterList.

        Thanks,

        Jonathan

        Show
        jiraposter@reviews.apache.org added a comment - ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/1908/#review1961 ----------------------------------------------------------- Ship it! Looks good. Can address nits on commit. Michael On 2011-09-14 21:52:34, Jonathan Gray wrote: ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/1908/ ----------------------------------------------------------- (Updated 2011-09-14 21:52:34) Review request for hbase, Dhruba Borthakur, Michael Stack, Prakash Khemani, and Kannan Muthukkaruppan. Summary ------- FilterList.filterKeyValue does not always return the most optimal ReturnCode in both the AND and OR conditions. For example, if you have F1 AND F2, F1 returns SKIP. It immediately returns the SKIP. However, if F2 would have returned NEXT_COL or NEXT_ROW or SEEK_NEXT_USING_HINT, we would actually be able to return the more optimal ReturnCode from F2. For AND conditions, we can always pick the most restrictive return code. For OR conditions, we must always pick the least restrictive return code. This JIRA is to review the FilterList.filterKeyValue() method to try and make it more optimal and to add a new unit test which verifies the correct behavior. This addresses bug HBASE-4410 . https://issues.apache.org/jira/browse/HBASE-4410 Diffs ----- /src/main/java/org/apache/hadoop/hbase/filter/Filter.java 1170860 /src/main/java/org/apache/hadoop/hbase/filter/FilterList.java 1170860 /src/test/java/org/apache/hadoop/hbase/filter/TestFilterList.java 1170860 Diff: https://reviews.apache.org/r/1908/diff Testing ------- Adds new tests to TestFilterList. Thanks, Jonathan
        Hide
        Jonathan Gray added a comment -

        Actually I think Lars is correct. It's a question of whether we should execute all filters in a list filterKeyValue() or not.

        I think the right behavior is actually just to make it execute how one would expect this type of conditional to execute:

        if (conditionA && conditionB)

        If conditionA fails, we don't expect conditionB to be executed.

        if (conditionA || conditionB)

        If conditionA passes, we don't expect conditionB to be executed.

        This was the previous behavior and my patch undoes it. I will work on a new patch.

        Show
        Jonathan Gray added a comment - Actually I think Lars is correct. It's a question of whether we should execute all filters in a list filterKeyValue() or not. I think the right behavior is actually just to make it execute how one would expect this type of conditional to execute: if (conditionA && conditionB) If conditionA fails, we don't expect conditionB to be executed. if (conditionA || conditionB) If conditionA passes, we don't expect conditionB to be executed. This was the previous behavior and my patch undoes it. I will work on a new patch.
        Hide
        stack added a comment -

        @Lars You OK with Jon's comeback?

        Show
        stack added a comment - @Lars You OK with Jon's comeback?
        Hide
        jiraposter@reviews.apache.org added a comment -

        On 2011-09-14 22:22:06, Lars Hofhansl wrote:

        > /src/main/java/org/apache/hadoop/hbase/filter/FilterList.java, line 199

        > <https://reviews.apache.org/r/1908/diff/1/?file=40598#file40598line199>

        >

        > You could break here, no?

        > There will never be more restrictive filter among the other filters.

        Jonathan Gray wrote:

        Another part of the changes here is ensuring that filterKeyValue() gets called on every filter, regardless of what the other filters are returning. Otherwise, filters down the list will have indeterminate behavior.

        (See my addition to TestFilterList.testHintPassThru() which would fail without this change)

        • Jonathan

        -----------------------------------------------------------
        This is an automatically generated e-mail. To reply, visit:
        https://reviews.apache.org/r/1908/#review1899
        -----------------------------------------------------------

        On 2011-09-14 21:52:34, Jonathan Gray wrote:

        -----------------------------------------------------------

        This is an automatically generated e-mail. To reply, visit:

        https://reviews.apache.org/r/1908/

        -----------------------------------------------------------

        (Updated 2011-09-14 21:52:34)

        Review request for hbase, Dhruba Borthakur, Michael Stack, Prakash Khemani, and Kannan Muthukkaruppan.

        Summary

        -------

        FilterList.filterKeyValue does not always return the most optimal ReturnCode in both the AND and OR conditions.

        For example, if you have F1 AND F2, F1 returns SKIP. It immediately returns the SKIP. However, if F2 would have returned NEXT_COL or NEXT_ROW or SEEK_NEXT_USING_HINT, we would actually be able to return the more optimal ReturnCode from F2.

        For AND conditions, we can always pick the most restrictive return code.

        For OR conditions, we must always pick the least restrictive return code.

        This JIRA is to review the FilterList.filterKeyValue() method to try and make it more optimal and to add a new unit test which verifies the correct behavior.

        This addresses bug HBASE-4410.

        https://issues.apache.org/jira/browse/HBASE-4410

        Diffs

        -----

        /src/main/java/org/apache/hadoop/hbase/filter/Filter.java 1170860

        /src/main/java/org/apache/hadoop/hbase/filter/FilterList.java 1170860

        /src/test/java/org/apache/hadoop/hbase/filter/TestFilterList.java 1170860

        Diff: https://reviews.apache.org/r/1908/diff

        Testing

        -------

        Adds new tests to TestFilterList.

        Thanks,

        Jonathan

        Show
        jiraposter@reviews.apache.org added a comment - On 2011-09-14 22:22:06, Lars Hofhansl wrote: > /src/main/java/org/apache/hadoop/hbase/filter/FilterList.java, line 199 > < https://reviews.apache.org/r/1908/diff/1/?file=40598#file40598line199 > > > You could break here, no? > There will never be more restrictive filter among the other filters. Jonathan Gray wrote: Another part of the changes here is ensuring that filterKeyValue() gets called on every filter, regardless of what the other filters are returning. Otherwise, filters down the list will have indeterminate behavior. (See my addition to TestFilterList.testHintPassThru() which would fail without this change) Jonathan ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/1908/#review1899 ----------------------------------------------------------- On 2011-09-14 21:52:34, Jonathan Gray wrote: ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/1908/ ----------------------------------------------------------- (Updated 2011-09-14 21:52:34) Review request for hbase, Dhruba Borthakur, Michael Stack, Prakash Khemani, and Kannan Muthukkaruppan. Summary ------- FilterList.filterKeyValue does not always return the most optimal ReturnCode in both the AND and OR conditions. For example, if you have F1 AND F2, F1 returns SKIP. It immediately returns the SKIP. However, if F2 would have returned NEXT_COL or NEXT_ROW or SEEK_NEXT_USING_HINT, we would actually be able to return the more optimal ReturnCode from F2. For AND conditions, we can always pick the most restrictive return code. For OR conditions, we must always pick the least restrictive return code. This JIRA is to review the FilterList.filterKeyValue() method to try and make it more optimal and to add a new unit test which verifies the correct behavior. This addresses bug HBASE-4410 . https://issues.apache.org/jira/browse/HBASE-4410 Diffs ----- /src/main/java/org/apache/hadoop/hbase/filter/Filter.java 1170860 /src/main/java/org/apache/hadoop/hbase/filter/FilterList.java 1170860 /src/test/java/org/apache/hadoop/hbase/filter/TestFilterList.java 1170860 Diff: https://reviews.apache.org/r/1908/diff Testing ------- Adds new tests to TestFilterList. Thanks, Jonathan
        Hide
        jiraposter@reviews.apache.org added a comment -

        On 2011-09-14 22:22:06, Lars Hofhansl wrote:

        > /src/main/java/org/apache/hadoop/hbase/filter/FilterList.java, line 199

        > <https://reviews.apache.org/r/1908/diff/1/?file=40598#file40598line199>

        >

        > You could break here, no?

        > There will never be more restrictive filter among the other filters.

        Another part of the changes here is ensuring that filterKeyValue() gets called on every filter, regardless of what the other filters are returning. Otherwise, filters down the list will have indeterminate behavior.

        • Jonathan

        -----------------------------------------------------------
        This is an automatically generated e-mail. To reply, visit:
        https://reviews.apache.org/r/1908/#review1899
        -----------------------------------------------------------

        On 2011-09-14 21:52:34, Jonathan Gray wrote:

        -----------------------------------------------------------

        This is an automatically generated e-mail. To reply, visit:

        https://reviews.apache.org/r/1908/

        -----------------------------------------------------------

        (Updated 2011-09-14 21:52:34)

        Review request for hbase, Dhruba Borthakur, Michael Stack, Prakash Khemani, and Kannan Muthukkaruppan.

        Summary

        -------

        FilterList.filterKeyValue does not always return the most optimal ReturnCode in both the AND and OR conditions.

        For example, if you have F1 AND F2, F1 returns SKIP. It immediately returns the SKIP. However, if F2 would have returned NEXT_COL or NEXT_ROW or SEEK_NEXT_USING_HINT, we would actually be able to return the more optimal ReturnCode from F2.

        For AND conditions, we can always pick the most restrictive return code.

        For OR conditions, we must always pick the least restrictive return code.

        This JIRA is to review the FilterList.filterKeyValue() method to try and make it more optimal and to add a new unit test which verifies the correct behavior.

        This addresses bug HBASE-4410.

        https://issues.apache.org/jira/browse/HBASE-4410

        Diffs

        -----

        /src/main/java/org/apache/hadoop/hbase/filter/Filter.java 1170860

        /src/main/java/org/apache/hadoop/hbase/filter/FilterList.java 1170860

        /src/test/java/org/apache/hadoop/hbase/filter/TestFilterList.java 1170860

        Diff: https://reviews.apache.org/r/1908/diff

        Testing

        -------

        Adds new tests to TestFilterList.

        Thanks,

        Jonathan

        Show
        jiraposter@reviews.apache.org added a comment - On 2011-09-14 22:22:06, Lars Hofhansl wrote: > /src/main/java/org/apache/hadoop/hbase/filter/FilterList.java, line 199 > < https://reviews.apache.org/r/1908/diff/1/?file=40598#file40598line199 > > > You could break here, no? > There will never be more restrictive filter among the other filters. Another part of the changes here is ensuring that filterKeyValue() gets called on every filter, regardless of what the other filters are returning. Otherwise, filters down the list will have indeterminate behavior. Jonathan ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/1908/#review1899 ----------------------------------------------------------- On 2011-09-14 21:52:34, Jonathan Gray wrote: ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/1908/ ----------------------------------------------------------- (Updated 2011-09-14 21:52:34) Review request for hbase, Dhruba Borthakur, Michael Stack, Prakash Khemani, and Kannan Muthukkaruppan. Summary ------- FilterList.filterKeyValue does not always return the most optimal ReturnCode in both the AND and OR conditions. For example, if you have F1 AND F2, F1 returns SKIP. It immediately returns the SKIP. However, if F2 would have returned NEXT_COL or NEXT_ROW or SEEK_NEXT_USING_HINT, we would actually be able to return the more optimal ReturnCode from F2. For AND conditions, we can always pick the most restrictive return code. For OR conditions, we must always pick the least restrictive return code. This JIRA is to review the FilterList.filterKeyValue() method to try and make it more optimal and to add a new unit test which verifies the correct behavior. This addresses bug HBASE-4410 . https://issues.apache.org/jira/browse/HBASE-4410 Diffs ----- /src/main/java/org/apache/hadoop/hbase/filter/Filter.java 1170860 /src/main/java/org/apache/hadoop/hbase/filter/FilterList.java 1170860 /src/test/java/org/apache/hadoop/hbase/filter/TestFilterList.java 1170860 Diff: https://reviews.apache.org/r/1908/diff Testing ------- Adds new tests to TestFilterList. Thanks, Jonathan
        Hide
        jiraposter@reviews.apache.org added a comment -

        -----------------------------------------------------------
        This is an automatically generated e-mail. To reply, visit:
        https://reviews.apache.org/r/1908/#review1899
        -----------------------------------------------------------

        /src/main/java/org/apache/hadoop/hbase/filter/FilterList.java
        <https://reviews.apache.org/r/1908/#comment4382>

        You could break here, no?
        There will never be more restrictive filter among the other filters.

        • Lars

        On 2011-09-14 21:52:34, Jonathan Gray wrote:

        -----------------------------------------------------------

        This is an automatically generated e-mail. To reply, visit:

        https://reviews.apache.org/r/1908/

        -----------------------------------------------------------

        (Updated 2011-09-14 21:52:34)

        Review request for hbase, Dhruba Borthakur, Michael Stack, Prakash Khemani, and Kannan Muthukkaruppan.

        Summary

        -------

        FilterList.filterKeyValue does not always return the most optimal ReturnCode in both the AND and OR conditions.

        For example, if you have F1 AND F2, F1 returns SKIP. It immediately returns the SKIP. However, if F2 would have returned NEXT_COL or NEXT_ROW or SEEK_NEXT_USING_HINT, we would actually be able to return the more optimal ReturnCode from F2.

        For AND conditions, we can always pick the most restrictive return code.

        For OR conditions, we must always pick the least restrictive return code.

        This JIRA is to review the FilterList.filterKeyValue() method to try and make it more optimal and to add a new unit test which verifies the correct behavior.

        This addresses bug HBASE-4410.

        https://issues.apache.org/jira/browse/HBASE-4410

        Diffs

        -----

        /src/main/java/org/apache/hadoop/hbase/filter/Filter.java 1170860

        /src/main/java/org/apache/hadoop/hbase/filter/FilterList.java 1170860

        /src/test/java/org/apache/hadoop/hbase/filter/TestFilterList.java 1170860

        Diff: https://reviews.apache.org/r/1908/diff

        Testing

        -------

        Adds new tests to TestFilterList.

        Thanks,

        Jonathan

        Show
        jiraposter@reviews.apache.org added a comment - ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/1908/#review1899 ----------------------------------------------------------- /src/main/java/org/apache/hadoop/hbase/filter/FilterList.java < https://reviews.apache.org/r/1908/#comment4382 > You could break here, no? There will never be more restrictive filter among the other filters. Lars On 2011-09-14 21:52:34, Jonathan Gray wrote: ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/1908/ ----------------------------------------------------------- (Updated 2011-09-14 21:52:34) Review request for hbase, Dhruba Borthakur, Michael Stack, Prakash Khemani, and Kannan Muthukkaruppan. Summary ------- FilterList.filterKeyValue does not always return the most optimal ReturnCode in both the AND and OR conditions. For example, if you have F1 AND F2, F1 returns SKIP. It immediately returns the SKIP. However, if F2 would have returned NEXT_COL or NEXT_ROW or SEEK_NEXT_USING_HINT, we would actually be able to return the more optimal ReturnCode from F2. For AND conditions, we can always pick the most restrictive return code. For OR conditions, we must always pick the least restrictive return code. This JIRA is to review the FilterList.filterKeyValue() method to try and make it more optimal and to add a new unit test which verifies the correct behavior. This addresses bug HBASE-4410 . https://issues.apache.org/jira/browse/HBASE-4410 Diffs ----- /src/main/java/org/apache/hadoop/hbase/filter/Filter.java 1170860 /src/main/java/org/apache/hadoop/hbase/filter/FilterList.java 1170860 /src/test/java/org/apache/hadoop/hbase/filter/TestFilterList.java 1170860 Diff: https://reviews.apache.org/r/1908/diff Testing ------- Adds new tests to TestFilterList. Thanks, Jonathan
        Hide
        jiraposter@reviews.apache.org added a comment -

        -----------------------------------------------------------
        This is an automatically generated e-mail. To reply, visit:
        https://reviews.apache.org/r/1908/
        -----------------------------------------------------------

        Review request for hbase, Dhruba Borthakur, Michael Stack, Prakash Khemani, and Kannan Muthukkaruppan.

        Summary
        -------

        FilterList.filterKeyValue does not always return the most optimal ReturnCode in both the AND and OR conditions.

        For example, if you have F1 AND F2, F1 returns SKIP. It immediately returns the SKIP. However, if F2 would have returned NEXT_COL or NEXT_ROW or SEEK_NEXT_USING_HINT, we would actually be able to return the more optimal ReturnCode from F2.

        For AND conditions, we can always pick the most restrictive return code.

        For OR conditions, we must always pick the least restrictive return code.

        This JIRA is to review the FilterList.filterKeyValue() method to try and make it more optimal and to add a new unit test which verifies the correct behavior.

        This addresses bug HBASE-4410.
        https://issues.apache.org/jira/browse/HBASE-4410

        Diffs


        /src/main/java/org/apache/hadoop/hbase/filter/Filter.java 1170860
        /src/main/java/org/apache/hadoop/hbase/filter/FilterList.java 1170860
        /src/test/java/org/apache/hadoop/hbase/filter/TestFilterList.java 1170860

        Diff: https://reviews.apache.org/r/1908/diff

        Testing
        -------

        Adds new tests to TestFilterList.

        Thanks,

        Jonathan

        Show
        jiraposter@reviews.apache.org added a comment - ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/1908/ ----------------------------------------------------------- Review request for hbase, Dhruba Borthakur, Michael Stack, Prakash Khemani, and Kannan Muthukkaruppan. Summary ------- FilterList.filterKeyValue does not always return the most optimal ReturnCode in both the AND and OR conditions. For example, if you have F1 AND F2, F1 returns SKIP. It immediately returns the SKIP. However, if F2 would have returned NEXT_COL or NEXT_ROW or SEEK_NEXT_USING_HINT, we would actually be able to return the more optimal ReturnCode from F2. For AND conditions, we can always pick the most restrictive return code. For OR conditions, we must always pick the least restrictive return code. This JIRA is to review the FilterList.filterKeyValue() method to try and make it more optimal and to add a new unit test which verifies the correct behavior. This addresses bug HBASE-4410 . https://issues.apache.org/jira/browse/HBASE-4410 Diffs /src/main/java/org/apache/hadoop/hbase/filter/Filter.java 1170860 /src/main/java/org/apache/hadoop/hbase/filter/FilterList.java 1170860 /src/test/java/org/apache/hadoop/hbase/filter/TestFilterList.java 1170860 Diff: https://reviews.apache.org/r/1908/diff Testing ------- Adds new tests to TestFilterList. Thanks, Jonathan
        Hide
        Jonathan Gray added a comment -

        Implements changes described in description and includes unit test. New test and existing tests are passing, kicking off full suite now.

        Show
        Jonathan Gray added a comment - Implements changes described in description and includes unit test. New test and existing tests are passing, kicking off full suite now.

          People

          • Assignee:
            Jonathan Gray
            Reporter:
            Jonathan Gray
          • Votes:
            0 Vote for this issue
            Watchers:
            2 Start watching this issue

            Dates

            • Created:
              Updated:

              Development