nanos in timestamp INT96 can be negative

oap-project · Mar 23, 2024 · 8704300 · 8704300
1 parent 0cb618e
commit 8704300
Show file tree

Hide file tree

Showing 2 changed files with 9 additions and 6 deletions.
diff --git a/velox/dwio/parquet/reader/PageReader.cpp b/velox/dwio/parquet/reader/PageReader.cpp
@@ -388,15 +388,15 @@ void PageReader::prepareDictionary(const PageHeader& pageHeader) {
       for (auto i = dictionary_.numValues - 1; i >= 0; --i) {
         // Convert the timestamp into seconds and nanos since the Unix epoch,
         // 00:00:00.000000 on 1 January 1970.
-        uint64_t nanos;
+        int64_t nanos;
         memcpy(
             &nanos,
             parquetValues + i * sizeof(Int96Timestamp),
-            sizeof(uint64_t));
+            sizeof(int64_t));
         int32_t days;
         memcpy(
             &days,
-            parquetValues + i * sizeof(Int96Timestamp) + sizeof(uint64_t),
+            parquetValues + i * sizeof(Int96Timestamp) + sizeof(int64_t),
             sizeof(int32_t));
 
         values[i] = Timestamp::fromDaysAndNanos(days, nanos);

diff --git a/velox/type/Timestamp.h b/velox/type/Timestamp.h
@@ -101,7 +101,7 @@ struct Timestamp {
 
   constexpr Timestamp() : seconds_(0), nanos_(0) {}
 
-  Timestamp(int64_t seconds, uint64_t nanos)
+  Timestamp(int64_t seconds, int64_t nanos)
       : seconds_(seconds), nanos_(nanos) {
     VELOX_USER_DCHECK_GE(
         seconds, kMinSeconds, "Timestamp seconds out of range");
@@ -110,16 +110,19 @@ struct Timestamp {
     VELOX_USER_DCHECK_LE(nanos, kMaxNanos, "Timestamp nanos out of range");
   }
 
-  static Timestamp fromDaysAndNanos(int32_t days, uint64_t nanos) {
+  static Timestamp fromDaysAndNanos(int32_t days, int64_t nanos) {
     static constexpr int64_t kJulianToUnixEpochDays = 2440588LL;
     static constexpr int64_t kSecondsPerDay = 86400LL;
     static constexpr int64_t kNanosPerSecond =
         Timestamp::kNanosecondsInMillisecond * Timestamp::kMillisecondsInSecond;
 
     int64_t seconds = (days - kJulianToUnixEpochDays) * kSecondsPerDay;
-    if (nanos > Timestamp::kMaxNanos) {
+    if (nanos > static_cast<int64_t>(Timestamp::kMaxNanos)) {
       seconds += nanos / kNanosPerSecond;
       nanos -= (nanos / kNanosPerSecond) * kNanosPerSecond;
+    } else if (nanos < 0) {
+      seconds += (nanos / kNanosPerSecond - 1);
+      nanos -= (nanos / kNanosPerSecond - 1) * kNanosPerSecond;
     }
 
     return Timestamp(seconds, nanos);