Merging master in

techascent · Sep 14, 2024 · 73ca10a · 73ca10a
2 parents 1eab826 + b08346b
commit 73ca10a
Show file tree

Hide file tree

Showing 3 changed files with 31 additions and 4 deletions.
diff --git a/deps.edn b/deps.edn
@@ -1,5 +1,5 @@
 {:paths ["src" "lib"]
- :deps {techascent/tech.ml.dataset {:mvn/version "7.026"}
+ :deps {techascent/tech.ml.dataset {:mvn/version "7.032"}
         techascent/tech.ml.dataset.sql {:mvn/version "7.000-beta-52"}
         net.java.dev.jna/jna {:mvn/version "5.13.0"}}
  :aliases

diff --git a/src/tmducken/duckdb.clj b/src/tmducken/duckdb.clj
@@ -319,6 +319,10 @@ tmducken.duckdb> (get-config-options)
   (.address (dt-ffi/->pointer ptr)))
 
 
+(defn toggle-long-msb
+  [x]
+  (bit-xor x Long/MIN_VALUE))
+
 (defn insert-dataset!
   "Append this dataset using the higher performance append api of duckdb.  This is recommended
   as opposed to using sql statements or prepared statements.  That being said the schema of this
@@ -437,7 +441,7 @@ tmducken.duckdb> (get-config-options)
                                           (if uuid
                                             (do
                                               (.putLong us laddr (.getLeastSignificantBits uuid))
-                                              (.putLong us (+ laddr 8) (.getMostSignificantBits uuid)))
+                                              (.putLong us (+ laddr 8) (toggle-long-msb (.getMostSignificantBits uuid))))
                                             (do
                                               (.putLong us laddr 0)
                                               (.putLong us (+ laddr 8) 0))))))))))
@@ -553,6 +557,7 @@ tmducken.duckdb> (get-config-options)
 
 (deftype ^:private UUIDReader [])
 
+
 (defn- coldata->buffer
   [^RoaringBitmap missing ^long n-rows ^long duckdb-type ^long data-ptr]
   (case (get duckdb-ffi/duckdb-type-map duckdb-type)
@@ -618,8 +623,15 @@ tmducken.duckdb> (get-config-options)
        (reify IFnDef$LO
          (invokePrim [this idx]
            (let [laddr (+ data-ptr (* idx 16))]
-             (UUID. (.getLong us (+ laddr 8))
-                    (.getLong us laddr))))))
+             ;; typedef struct {
+             ;;     uint64_t lower;
+             ;;     int64_t upper;
+             ;; } duckdb_hugeint;
+             ;; For some reason the msb when r/w with duckdb api nee to be inverted.
+             (let [msbytes (.getLong us (+ laddr 8))
+                   msbytes (toggle-long-msb msbytes)
+                   lsbytes (.getLong us laddr)]
+               (UUID. msbytes lsbytes))))))
      0 n-rows)
 
     :DUCKDB_TYPE_VARCHAR

diff --git a/test/tmducken/duckdb_test.clj b/test/tmducken/duckdb_test.clj
@@ -223,6 +223,21 @@
               {"country" "US", "name" "Seattle", "2010" 608}]
              (ds/rows result-ds))))))
 
+(deftest uuid-test
+  (let [id (random-uuid)
+        data (ds/->dataset [{:id id}] {:dataset-name "uuid_test"})]
+    (try
+      (duckdb/drop-table! @conn* "uuid_test")
+      (catch Throwable _ nil))
+    (duckdb/create-table! @conn* data)
+    (duckdb/insert-dataset! @conn* data)
+    (let [result-ds (duckdb/sql->dataset
+                      @conn*
+                      (format "SELECT id from uuid_test where id = '%s'" id))]
+      (is (some? result-ds))
+      (is (= 1 (count (ds/rows result-ds))))
+      (is (= [{"id" id}] (ds/rows result-ds))))))
+
 (deftest show-all-tables-test
   (try
     (duckdb/drop-table! @conn* "foo")