kevinyu98 commented on a change in pull request #28120:
[SPARK-31349][SQL][DOCS] Document built-in aggregate functions in SQL Reference
URL: https://github.com/apache/spark/pull/28120#discussion_r405704412
##########
File path: docs/sql-ref-functions-builtin-aggregate.md
##########
@@ -19,4 +19,657 @@ license: |
limitations under the License.
---
-Aggregate functions
\ No newline at end of file
+Spark SQL provides build-in aggregate functions defined in the dataset API and
SQL interface. Aggregate functions
+operate on a group of rows and return a single value.
+
+<table class="table">
+ <thead>
+ <tr><th style="width:25%">Function</th><th>Argument
Type(s)</th><th>Description</th></tr>
+ </thead>
+ <tbody>
+ <tr>
+ <td><b>{any | some | bool_or}</b>(<i>expression</i>)</td>
+ <td>boolean</td>
+ <td>Returns true if at least one value is true.</td>
+ </tr>
+ <tr>
+ <td><b>approx_count_distinct</b>(<i>expression[, relativeSD]</i>)</td>
+ <td>(long, double)</td>
+ <td>`relativeSD` is the maximum estimation error allowed. Returns the
estimated cardinality by HyperLogLog++.</td>
+ </tr>
+ <tr>
+ <td><b>{avg | mean}</b>(<i>expression</i>)</td>
+ <td>numeric or string</td>
+ <td>Returns the average of values in the input expression.</td>
+ </tr>
+ <tr>
+ <td><b>{bool_and | every}</b>(<i>expression</i>)</td>
+ <td>boolean</td>
+ <td>Returns true if all values are true.</td>
+ </tr>
+ <tr>
+ <td><b>collect_list</b>(<i>expression</i>)</td>
+ <td>any</td>
+ <td>Collects and returns a list of non-unique elements. The function is
non-deterministic because the order of collected results depends on the order
of the rows which may be non-deterministic after a shuffle.</td>
+ </tr>
+ <tr>
+ <td><b>collect_set</b>(<i>expression</i>)</td>
+ <td>any</td>
+ <td>Collects and returns a set of unique elements. The function is
non-deterministic because the order of collected results depends on the order
of the rows which may be non-deterministic after a shuffle.</td>
+ </tr>
+ <tr>
+ <td><b>corr</b>(<i>expression1, expression2</i>)</td>
+ <td>(double, double)</td>
+ <td>Returns Pearson coefficient of correlation between a set of number
pairs.</td>
+ </tr>
+ <tr>
+ <td><b>count</b>([<b>DISTINCT</b>] <i>*</i>)</td>
+ <td>none</td>
+ <td>If specified <code>DISTINCT</code>, returns the total number of
retrieved rows are unique and not null; Otherwise, returns the total number of
retrieved rows, including rows containing null.</td>
+ </tr>
+ <tr>
+ <td><b>count</b>([<b>DISTINCT</b>] <i>expression1[,
expression2</i>])</td>
+ <td>(any, any)</td>
+ <td>If specified <code>DISTINCT</code>, returns the number of rows for
which the supplied expression(s) are unique and not null; Otherwise, returns
the number of rows for which the supplied expression(s) are all not null.</td>
+ </tr>
+ <tr>
+ <td><b>count_if</b>(<i>predicate</i>)</td>
+ <td>expression that will be used for aggregation calculation</td>
+ <td>Returns the count number from the predicate evaluate to `TRUE`
values.</td>
+ </tr>
+ <tr>
+ <td><b>count_min_sketch</b>(<i>expression, eps, confidence,
seed</i>)</td>
+ <td>(integer or string or binary, double, double, integer)</td>
+ <td>`eps` and `confidence` are the double values between 0.0 and 1.0,
`seed` is a positive integer. Returns a count-min sketch of a expression with
the given `esp`, `confidence` and `seed`. The result is an array of bytes,
which can be deserialized to a `CountMinSketch` before usage. Count-min sketch
is a probabilistic data structure used for cardinality estimation using
sub-linear space.</td>
+ </tr>
+ <tr>
+ <td><b>covar_pop</b>(<i>expression1, expression2</i>)</td>
+ <td>(double, double)</td>
+ <td>Returns the population covariance of a set of number pairs.</td>
+ </tr>
+ <tr>
+ <td><b>covar_samp</b>(<i>expression1, expression2</i>)</td>
+ <td>(double, double)</td>
+ <td>Returns the sample covariance of a set of number pairs.</td>
+ </tr>
+ <tr>
+ <td><b>{first | first_value}</b>(<i>expression[, isIgnoreNull]</i>)</td>
+ <td>(any, boolean)</td>
+ <td>Returns the first value of expression for a group of rows. If
<code>isIgnoreNull</code> is true, returns only non-null values, default is
false. This function is non-deterministic.</td>
+ </tr>
+ <tr>
+ <td><b>kurtosis</b>(<i>expression</i>)</td>
+ <td>double</td>
+ <td>Returns the kurtosis value calculated from values of a group.</td>
+ </tr>
+ <tr>
+ <td><b>{last | last_value}</b>(<i>expression[, isIgnoreNull]</i>)</td>
+ <td>(any, boolean)</td>
+ <td>Returns the last value of expression for a group of rows. If
<code>isIgnoreNull</code> is true, returns only non-null values, default is
false. This function is non-deterministic.</td>
+ </tr>
+ <tr>
+ <td><b>max</b>(<i>expression</i>)</td>
+ <td>any numeric, string, datetime or arrays of these types</td>
Review comment:
changed
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
