Merge pull request #657 from jeroenterheerdt/patch-1

Update rank-function-dax.md

Author: prmerger-automator[bot]

query-languages/dax/rank-function-dax.md

@@ -21,7 +21,7 @@ RANK ( [<ties>][, <relation>][, <orderBy>][, <blanks>][, <partitionBy>][, <match
 |ties|(Optional) Defines how to handle the ranking when two or more rows are tied. </br>If specified, the supported value is either DENSE, or SKIP. </br>If omitted:</br>- Defaults to SKIP |
 |relation|(Optional) A table expression from which the output row is returned. </br>If specified, all columns in \<orderBy> and \<partitionBy> must come from it. </br>If omitted: </br>- \<orderBy> must be explicitly specified.</br>- All \<orderBy> and \<partitionBy> columns must be fully qualified and come from a single table. </br>- Defaults to ALLSELECTED() of all columns in \<orderBy> and \<partitionBy>.|
 |orderBy|(Optional) An ORDERBY() clause containing the columns that define how each partition is sorted. </br>If omitted: </br>- \<relation> must be explicitly specified. </br>- Defaults to ordering by every column in \<relation> that is not already specified in \<partitionBy>.|
-|blanks|(Optional) An enumeration that defines how to handle blank values when sorting. </br>The supported values are:<ul><li>DEFAULT (the default value), where the behavior for numerical values is blank values are ordered between zero and negative values. The behavior for strings is blank values are ordered before all strings, including empty strings.</li><li>FIRST, blanks are always ordered on the beginning, regardless of ascending or descending sorting order.</li><li>LAST, blanks are always ordered on the end, regardless of ascending or descending sorting order. </li></ul></br>Note, when \<blanks> parameter and blanks in ORDERBY() function on individual expression are both specified, \<blanks> on individual orderBy expression takes priority for the relevant orderBy expression, and orderBy expressions without \<blanks> being specified will honor \<blanks> parameter on parent Window function.|
+|blanks|(Optional) An enumeration that defines how to handle blank values when sorting. </br>The supported values are:<ul><li>KEEP (the default value), where the behavior for numerical values is blank values are ordered between zero and negative values. The behavior for strings is blank values are ordered before all strings, including empty strings.</li><li>FIRST, blanks are always ordered on the beginning, regardless of ascending or descending sorting order.</li><li>LAST, blanks are always ordered on the end, regardless of ascending or descending sorting order. </li></ul></br>Note, when \<blanks> parameter and blanks in ORDERBY() function on individual expression are both specified, \<blanks> on individual orderBy expression takes priority for the relevant orderBy expression, and orderBy expressions without \<blanks> being specified will honor \<blanks> parameter on parent Window function.|
 |partitionBy|(Optional) A PARTITIONBY() clause containing the columns that define how \<relation> is partitioned. </br> If omitted, \<relation> is treated as a single partition. |
 |matchBy|(Optional) A MATCHBY() clause containing the columns that define how to match data and identify the current row. |