BigQuery Standard SQL returns different WEEK from timestamp









up vote
1
down vote

favorite












I am converting an old query from Legacy SQL to Standard SQL and noticed my report was off. I've traced it back to Legacy SQL and Standard SQL returning different weeks based off of a UNIX millisecond timestamp.



I was under the impression the Legacy SQL query was correct, but however, I'm curious to know the difference. It's not across all weeks, but enough to throw my report off significantly.



Here's an example:



#legacySQL
SELECT WEEK(MSEC_TO_TIMESTAMP(1470631859000)) AS sign_up_week;


Output: 33



#standardSQL
SELECT EXTRACT(WEEK FROM TIMESTAMP_MILLIS(1470631859000));


Output: 32



I've viewed the following documentation on the Legacy SQL reference:



WEEK(<timestamp>)
Returns the week of a TIMESTAMP data type as an integer between 1 and 53, inclusively.

Weeks begin on Sunday, so if January 1 is on a day other than Sunday, week 1 has fewer than 7 days and the first Sunday of the year is the first day of week 2.


And from the Standard SQL reference:



WEEK(<WEEKDAY>): Returns the week number of the date in the range [0, 53]. Weeks begin on WEEKDAY. Dates prior to the first WEEKDAY of the year are in week 0. Valid values for WEEKDAY are SUNDAY, MONDAY, TUESDAY, WEDNESDAY, THURSDAY, FRIDAY, and SATURDAY.

ISOWEEK: Returns the ISO 8601 week number of the date_expression. ISOWEEKs begin on Monday. Return values are in the range [1, 53]. The first ISOWEEK of each ISO year begins on the Monday before the first Thursday of the Gregorian calendar year.


How can I get my Standard SQL query to output the same week number as my Legacy SQL query? If not, which week number is correct? It does not appear I can get them to dovetail natively.










share|improve this question



























    up vote
    1
    down vote

    favorite












    I am converting an old query from Legacy SQL to Standard SQL and noticed my report was off. I've traced it back to Legacy SQL and Standard SQL returning different weeks based off of a UNIX millisecond timestamp.



    I was under the impression the Legacy SQL query was correct, but however, I'm curious to know the difference. It's not across all weeks, but enough to throw my report off significantly.



    Here's an example:



    #legacySQL
    SELECT WEEK(MSEC_TO_TIMESTAMP(1470631859000)) AS sign_up_week;


    Output: 33



    #standardSQL
    SELECT EXTRACT(WEEK FROM TIMESTAMP_MILLIS(1470631859000));


    Output: 32



    I've viewed the following documentation on the Legacy SQL reference:



    WEEK(<timestamp>)
    Returns the week of a TIMESTAMP data type as an integer between 1 and 53, inclusively.

    Weeks begin on Sunday, so if January 1 is on a day other than Sunday, week 1 has fewer than 7 days and the first Sunday of the year is the first day of week 2.


    And from the Standard SQL reference:



    WEEK(<WEEKDAY>): Returns the week number of the date in the range [0, 53]. Weeks begin on WEEKDAY. Dates prior to the first WEEKDAY of the year are in week 0. Valid values for WEEKDAY are SUNDAY, MONDAY, TUESDAY, WEDNESDAY, THURSDAY, FRIDAY, and SATURDAY.

    ISOWEEK: Returns the ISO 8601 week number of the date_expression. ISOWEEKs begin on Monday. Return values are in the range [1, 53]. The first ISOWEEK of each ISO year begins on the Monday before the first Thursday of the Gregorian calendar year.


    How can I get my Standard SQL query to output the same week number as my Legacy SQL query? If not, which week number is correct? It does not appear I can get them to dovetail natively.










    share|improve this question

























      up vote
      1
      down vote

      favorite









      up vote
      1
      down vote

      favorite











      I am converting an old query from Legacy SQL to Standard SQL and noticed my report was off. I've traced it back to Legacy SQL and Standard SQL returning different weeks based off of a UNIX millisecond timestamp.



      I was under the impression the Legacy SQL query was correct, but however, I'm curious to know the difference. It's not across all weeks, but enough to throw my report off significantly.



      Here's an example:



      #legacySQL
      SELECT WEEK(MSEC_TO_TIMESTAMP(1470631859000)) AS sign_up_week;


      Output: 33



      #standardSQL
      SELECT EXTRACT(WEEK FROM TIMESTAMP_MILLIS(1470631859000));


      Output: 32



      I've viewed the following documentation on the Legacy SQL reference:



      WEEK(<timestamp>)
      Returns the week of a TIMESTAMP data type as an integer between 1 and 53, inclusively.

      Weeks begin on Sunday, so if January 1 is on a day other than Sunday, week 1 has fewer than 7 days and the first Sunday of the year is the first day of week 2.


      And from the Standard SQL reference:



      WEEK(<WEEKDAY>): Returns the week number of the date in the range [0, 53]. Weeks begin on WEEKDAY. Dates prior to the first WEEKDAY of the year are in week 0. Valid values for WEEKDAY are SUNDAY, MONDAY, TUESDAY, WEDNESDAY, THURSDAY, FRIDAY, and SATURDAY.

      ISOWEEK: Returns the ISO 8601 week number of the date_expression. ISOWEEKs begin on Monday. Return values are in the range [1, 53]. The first ISOWEEK of each ISO year begins on the Monday before the first Thursday of the Gregorian calendar year.


      How can I get my Standard SQL query to output the same week number as my Legacy SQL query? If not, which week number is correct? It does not appear I can get them to dovetail natively.










      share|improve this question















      I am converting an old query from Legacy SQL to Standard SQL and noticed my report was off. I've traced it back to Legacy SQL and Standard SQL returning different weeks based off of a UNIX millisecond timestamp.



      I was under the impression the Legacy SQL query was correct, but however, I'm curious to know the difference. It's not across all weeks, but enough to throw my report off significantly.



      Here's an example:



      #legacySQL
      SELECT WEEK(MSEC_TO_TIMESTAMP(1470631859000)) AS sign_up_week;


      Output: 33



      #standardSQL
      SELECT EXTRACT(WEEK FROM TIMESTAMP_MILLIS(1470631859000));


      Output: 32



      I've viewed the following documentation on the Legacy SQL reference:



      WEEK(<timestamp>)
      Returns the week of a TIMESTAMP data type as an integer between 1 and 53, inclusively.

      Weeks begin on Sunday, so if January 1 is on a day other than Sunday, week 1 has fewer than 7 days and the first Sunday of the year is the first day of week 2.


      And from the Standard SQL reference:



      WEEK(<WEEKDAY>): Returns the week number of the date in the range [0, 53]. Weeks begin on WEEKDAY. Dates prior to the first WEEKDAY of the year are in week 0. Valid values for WEEKDAY are SUNDAY, MONDAY, TUESDAY, WEDNESDAY, THURSDAY, FRIDAY, and SATURDAY.

      ISOWEEK: Returns the ISO 8601 week number of the date_expression. ISOWEEKs begin on Monday. Return values are in the range [1, 53]. The first ISOWEEK of each ISO year begins on the Monday before the first Thursday of the Gregorian calendar year.


      How can I get my Standard SQL query to output the same week number as my Legacy SQL query? If not, which week number is correct? It does not appear I can get them to dovetail natively.







      sql google-bigquery unix-timestamp standard-sql






      share|improve this question















      share|improve this question













      share|improve this question




      share|improve this question








      edited Nov 10 at 15:31

























      asked Nov 10 at 15:18









      Sillson

      1119




      1119






















          1 Answer
          1






          active

          oldest

          votes

















          up vote
          2
          down vote













          This is a bit subtle, but legacy and standard SQL handle weeks differently. In Legacy SQL:




          Weeks begin on Sunday, so if January 1 is on a day other than Sunday, week 1 has fewer than 7 days and the first Sunday of the year is the first day of week 2.




          In Standard SQL:




          • WEEK: Returns the week number of the date in the range [0, 53]. Weeks begin with Sunday, and dates prior to the first Sunday of the year are in week 0.



          So, in Legacy SQL the first Sunday is week 2. In Standard SQL, the first Sunday is week 1.






          share|improve this answer




















          • Although this may be a stupid question, is there any definition that's more 'correct'? I'm hoping to remove the legacy SQL from the codebase. I believe ISOWEEK would fit my purposes better WEEK 1-53 makes more sense for a cohort report, however I'll need to justify the discrepancy in data.
            – Sillson
            Nov 10 at 16:09







          • 2




            Use ISOWEEK if that meets your needs. At least it is a defined standard and will be consistent over time and across platforms.
            – Gordon Linoff
            Nov 10 at 16:17










          Your Answer






          StackExchange.ifUsing("editor", function ()
          StackExchange.using("externalEditor", function ()
          StackExchange.using("snippets", function ()
          StackExchange.snippets.init();
          );
          );
          , "code-snippets");

          StackExchange.ready(function()
          var channelOptions =
          tags: "".split(" "),
          id: "1"
          ;
          initTagRenderer("".split(" "), "".split(" "), channelOptions);

          StackExchange.using("externalEditor", function()
          // Have to fire editor after snippets, if snippets enabled
          if (StackExchange.settings.snippets.snippetsEnabled)
          StackExchange.using("snippets", function()
          createEditor();
          );

          else
          createEditor();

          );

          function createEditor()
          StackExchange.prepareEditor(
          heartbeatType: 'answer',
          convertImagesToLinks: true,
          noModals: true,
          showLowRepImageUploadWarning: true,
          reputationToPostImages: 10,
          bindNavPrevention: true,
          postfix: "",
          imageUploader:
          brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
          contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
          allowUrls: true
          ,
          onDemand: true,
          discardSelector: ".discard-answer"
          ,immediatelyShowMarkdownHelp:true
          );



          );













           

          draft saved


          draft discarded


















          StackExchange.ready(
          function ()
          StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f53240342%2fbigquery-standard-sql-returns-different-week-from-timestamp%23new-answer', 'question_page');

          );

          Post as a guest















          Required, but never shown

























          1 Answer
          1






          active

          oldest

          votes








          1 Answer
          1






          active

          oldest

          votes









          active

          oldest

          votes






          active

          oldest

          votes








          up vote
          2
          down vote













          This is a bit subtle, but legacy and standard SQL handle weeks differently. In Legacy SQL:




          Weeks begin on Sunday, so if January 1 is on a day other than Sunday, week 1 has fewer than 7 days and the first Sunday of the year is the first day of week 2.




          In Standard SQL:




          • WEEK: Returns the week number of the date in the range [0, 53]. Weeks begin with Sunday, and dates prior to the first Sunday of the year are in week 0.



          So, in Legacy SQL the first Sunday is week 2. In Standard SQL, the first Sunday is week 1.






          share|improve this answer




















          • Although this may be a stupid question, is there any definition that's more 'correct'? I'm hoping to remove the legacy SQL from the codebase. I believe ISOWEEK would fit my purposes better WEEK 1-53 makes more sense for a cohort report, however I'll need to justify the discrepancy in data.
            – Sillson
            Nov 10 at 16:09







          • 2




            Use ISOWEEK if that meets your needs. At least it is a defined standard and will be consistent over time and across platforms.
            – Gordon Linoff
            Nov 10 at 16:17














          up vote
          2
          down vote













          This is a bit subtle, but legacy and standard SQL handle weeks differently. In Legacy SQL:




          Weeks begin on Sunday, so if January 1 is on a day other than Sunday, week 1 has fewer than 7 days and the first Sunday of the year is the first day of week 2.




          In Standard SQL:




          • WEEK: Returns the week number of the date in the range [0, 53]. Weeks begin with Sunday, and dates prior to the first Sunday of the year are in week 0.



          So, in Legacy SQL the first Sunday is week 2. In Standard SQL, the first Sunday is week 1.






          share|improve this answer




















          • Although this may be a stupid question, is there any definition that's more 'correct'? I'm hoping to remove the legacy SQL from the codebase. I believe ISOWEEK would fit my purposes better WEEK 1-53 makes more sense for a cohort report, however I'll need to justify the discrepancy in data.
            – Sillson
            Nov 10 at 16:09







          • 2




            Use ISOWEEK if that meets your needs. At least it is a defined standard and will be consistent over time and across platforms.
            – Gordon Linoff
            Nov 10 at 16:17












          up vote
          2
          down vote










          up vote
          2
          down vote









          This is a bit subtle, but legacy and standard SQL handle weeks differently. In Legacy SQL:




          Weeks begin on Sunday, so if January 1 is on a day other than Sunday, week 1 has fewer than 7 days and the first Sunday of the year is the first day of week 2.




          In Standard SQL:




          • WEEK: Returns the week number of the date in the range [0, 53]. Weeks begin with Sunday, and dates prior to the first Sunday of the year are in week 0.



          So, in Legacy SQL the first Sunday is week 2. In Standard SQL, the first Sunday is week 1.






          share|improve this answer












          This is a bit subtle, but legacy and standard SQL handle weeks differently. In Legacy SQL:




          Weeks begin on Sunday, so if January 1 is on a day other than Sunday, week 1 has fewer than 7 days and the first Sunday of the year is the first day of week 2.




          In Standard SQL:




          • WEEK: Returns the week number of the date in the range [0, 53]. Weeks begin with Sunday, and dates prior to the first Sunday of the year are in week 0.



          So, in Legacy SQL the first Sunday is week 2. In Standard SQL, the first Sunday is week 1.







          share|improve this answer












          share|improve this answer



          share|improve this answer










          answered Nov 10 at 15:36









          Gordon Linoff

          743k32285390




          743k32285390











          • Although this may be a stupid question, is there any definition that's more 'correct'? I'm hoping to remove the legacy SQL from the codebase. I believe ISOWEEK would fit my purposes better WEEK 1-53 makes more sense for a cohort report, however I'll need to justify the discrepancy in data.
            – Sillson
            Nov 10 at 16:09







          • 2




            Use ISOWEEK if that meets your needs. At least it is a defined standard and will be consistent over time and across platforms.
            – Gordon Linoff
            Nov 10 at 16:17
















          • Although this may be a stupid question, is there any definition that's more 'correct'? I'm hoping to remove the legacy SQL from the codebase. I believe ISOWEEK would fit my purposes better WEEK 1-53 makes more sense for a cohort report, however I'll need to justify the discrepancy in data.
            – Sillson
            Nov 10 at 16:09







          • 2




            Use ISOWEEK if that meets your needs. At least it is a defined standard and will be consistent over time and across platforms.
            – Gordon Linoff
            Nov 10 at 16:17















          Although this may be a stupid question, is there any definition that's more 'correct'? I'm hoping to remove the legacy SQL from the codebase. I believe ISOWEEK would fit my purposes better WEEK 1-53 makes more sense for a cohort report, however I'll need to justify the discrepancy in data.
          – Sillson
          Nov 10 at 16:09





          Although this may be a stupid question, is there any definition that's more 'correct'? I'm hoping to remove the legacy SQL from the codebase. I believe ISOWEEK would fit my purposes better WEEK 1-53 makes more sense for a cohort report, however I'll need to justify the discrepancy in data.
          – Sillson
          Nov 10 at 16:09





          2




          2




          Use ISOWEEK if that meets your needs. At least it is a defined standard and will be consistent over time and across platforms.
          – Gordon Linoff
          Nov 10 at 16:17




          Use ISOWEEK if that meets your needs. At least it is a defined standard and will be consistent over time and across platforms.
          – Gordon Linoff
          Nov 10 at 16:17

















           

          draft saved


          draft discarded















































           


          draft saved


          draft discarded














          StackExchange.ready(
          function ()
          StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f53240342%2fbigquery-standard-sql-returns-different-week-from-timestamp%23new-answer', 'question_page');

          );

          Post as a guest















          Required, but never shown





















































          Required, but never shown














          Required, but never shown












          Required, but never shown







          Required, but never shown

































          Required, but never shown














          Required, but never shown












          Required, but never shown







          Required, but never shown







          這個網誌中的熱門文章

          How to read a connectionString WITH PROVIDER in .NET Core?

          Node.js Script on GitHub Pages or Amazon S3

          Museum of Modern and Contemporary Art of Trento and Rovereto