add distinct/expr clauses

CHANGELOG.md

@@ -1,7 +1,7 @@
 # Changes
 
 * 2.4.next in progress
-  * Address [#504](https://github.com/seancorfield/honeysql/issues/504) by adding special syntax for ignore/respect nulls. More work will be needed to support distinct/order by/limit in BigQuery array aggregation.
+  * Address [#504](https://github.com/seancorfield/honeysql/issues/504) for BigQuery support, by adding special syntax for ignore/respect nulls, as well as new `:distinct` and `:expr` clauses to allow expressions to be qualified with SQL clauses. The latter will probably be useful for other dialects too.
 
 * 2.4.1066 -- 2023-08-27
   * Add `:select` with function call and alias example to README (PR [#502](https://github.com/seancorfield/honeysql/pull/502) [@markbastian](https://github.com/markbastian)).

README.md

@@ -52,11 +52,11 @@ section of the documentation before trying to use HoneySQL to build your own que
 From Clojure:
 <!-- {:test-doc-blocks/reader-cond :clj} -->
 ```clojure
-(refer-clojure :exclude '[filter for group-by into partition-by set update])
+(refer-clojure :exclude '[distinct filter for group-by into partition-by set update])
 (require '[honey.sql :as sql]
          ;; CAUTION: this overwrites several clojure.core fns:
          ;;
-         ;; filter, for, group-by, into, partition-by, set, and update
+         ;; distinct, filter, for, group-by, into, partition-by, set, and update
          ;;
          ;; you should generally only refer in the specific
          ;; helpers that you want to use!

doc/clause-reference.md

@@ -1002,6 +1002,19 @@ user=> (sql/format (-> (select :id
 ["SELECT id, AVG(salary) OVER () AS Average, MAX(salary) OVER () AS MaxSalary FROM employee"]
 ```
 
+## distinct, expr
+
+Related to the windowing clauses above, `:distinct` and `:expr` are
+intended to let you mix clauses with expressions, such as in BigQuery's
+`ARRAY_AGG` function:
+
+```clojure
+user=> (sql/format {:select [[[:over
+                               [[:array_agg {:distinct [:ignore-nulls :col] :order-by :x}]
+                                {:partition-by :y}]]]]})
+["SELECT ARRAY_AGG (DISTINCT col IGNORE NULLS ORDER BY x ASC) OVER (PARTITION BY y)"]
+```
+
 ## order-by
 
 `:order-by` accepts a sequence of one or more ordering

src/honey/sql.cljc

@@ -55,6 +55,7 @@
    :raw :nest :with :with-recursive :intersect :union :union-all :except :except-all
    :table
    :select :select-distinct :select-distinct-on :select-top :select-distinct-top
+   :distinct :expr
    :into :bulk-collect-into
    :insert-into :replace-into :update :delete :delete-from :truncate
    :columns :set :from :using
@@ -288,13 +289,14 @@
 
   Any ? is escaped to ??."
   [k]
-  (let [n (str/replace (name k) "?" "??")]
+  (when k
-    (if (= \' (first n))
+    (let [n (str/replace (name k) "?" "??")]
-      (let [ident   (subs n 1 (count n))
+      (if (= \' (first n))
-            ident-l (str/lower-case ident)]
+        (let [ident   (subs n 1 (count n))
-        (binding [*quoted* (when-not (contains? #{"array"} ident-l) *quoted*)]
+              ident-l (str/lower-case ident)]
-          (format-entity (keyword ident))))
+          (binding [*quoted* (when-not (contains? #{"array"} ident-l) *quoted*)]
-      (-> n (dehyphen) (upper-case)))))
+            (format-entity (keyword ident))))
+        (-> n (dehyphen) (upper-case))))))
 
 (defn- sym->kw
   "Given a symbol, produce a keyword, retaining the namespace
@@ -1361,6 +1363,8 @@
          :select-distinct-on #'format-selects-on
          :select-top      #'format-select-top
          :select-distinct-top #'format-select-top
+         :distinct        (fn [k xs] (format-selects k   [[xs]]))
+         :expr            (fn [_ xs] (format-selects nil [[xs]]))
          :into            #'format-select-into
          :bulk-collect-into #'format-select-into
          :insert-into     #'format-insert
@@ -2269,4 +2273,8 @@
                              [:transport :t] [:id :name]]
                :values [[1 "Car"] [2 "Boat"] [3 "Bike"]]}
               {:pretty true})
+  (sql/format-expr [:array_agg {:distinct [:ignore-nulls :a]  :order-by :a}])
+  (sql/format-expr [:over [[:array_agg {:expr     [:respect-nulls :a] :order-by :a
+                                        :limit    10}]
+                           {:partition-by :something}]])
   )

src/honey/sql/helpers.cljc

@@ -48,7 +48,7 @@
     bulk-collect-info [& args]
 
   (as they are for all helper functions)."
-  (:refer-clojure :exclude [filter for group-by into partition-by set update])
+  (:refer-clojure :exclude [distinct filter for group-by into partition-by set update])
   (:require [clojure.core :as c]
             [honey.sql]))
 
@@ -468,6 +468,16 @@
   [& args]
   (generic :select-distinct-top args))
 
+(defn distinct
+  "Like `select-distinct` but produces DISTINCT..."
+  [& args]
+  (generic-1 :distinct args))
+
+(defn expr
+  "Like `distinct` but produces ... (i.e., just the expression that follows)."
+  [& args]
+  (generic-1 :expr args))
+
 (defn into
   "Accepts table name, optionally followed a database name."
   {:arglists '([table] [table dbname])}