docs: remove duplicated words and fix two garbled words across core crates

core/data_types/src/snapshot/table.rs

@@ -233,7 +233,7 @@ impl TableSnapshotPartition {
     ///
     /// This is a no-op operation and should be rather fast.
     ///
-    /// See [`partition_key`](Self::partition_key) if need need the decoded form.
+    /// See [`partition_key`](Self::partition_key) if you need the decoded form.
     pub fn key(&self) -> &[u8] {
         &self.key
     }

core/influxdb_iox_client/src/client/namespace.rs

@@ -49,7 +49,7 @@ impl Client {
 
     /// Create a namespace
     ///
-    /// `retention_period_ns` is the the retention period in nanoseconds,
+    /// `retention_period_ns` is the retention period in nanoseconds,
     /// measured from `now()`. `None` represents infinite retention (i.e. never
     /// drop data), and 0 is also mapped to `None` on the server side.
     ///
@@ -76,7 +76,7 @@ impl Client {
 
     /// Update retention for a namespace
     ///
-    /// `retention_period_ns` is the the retention period in nanoseconds,
+    /// `retention_period_ns` is the retention period in nanoseconds,
     /// measured from `now()`. `None` represents infinite retention (i.e. never
     /// drop data), and 0 is also mapped to `None` on the server side.
     ///

core/iox_query/src/exec/context.rs

@@ -294,7 +294,7 @@ impl IOxSessionConfig {
         // memory pool from the runtime. Otherwise, it creates a new `PerQueryMemoryPool`
         // with a minimum memory budget, ensuring each query has a reserved amount of
         // memory. This prevents the query from being starved of resources and allows it
-        // it to use the central memory pool if the budget is exhausted.
+        // to use the central memory pool if the budget is exhausted.
         let memory_pool: Arc<dyn MemoryPool> = match self.per_query_mem_pool_size {
             0 => Arc::clone(&self.runtime.memory_pool),
             per_query_mem_pool_size => {

core/iox_query/src/query_log.rs

@@ -1447,7 +1447,7 @@ fn collect_ingester_metrics(plan: &dyn ExecutionPlan) -> IngesterMetrics {
 /// Collect the number of rows from an [`ExecutionPlan`].
 ///
 /// ## Note
-/// This function returns returns an [`Option`] to distinguish between
+/// This function returns an [`Option`] to distinguish between
 /// no metrics being available and no rows being returned by a query.
 ///
 /// Specifically:

core/iox_query_influxql/src/plan/planner_rewrite_expression.rs

@@ -911,7 +911,7 @@ mod test {
         assert_eq!(rewrite(expr), "boolean_field != Boolean(true)");
 
         // DataFusion does not support bitwise operations when the operands are
-        // Boolean data types. These must be case to an an Int8 in order to perform
+        // Boolean data types. These must be cast to an Int8 in order to perform
         // the operation and then cast back to Boolean
 
         let expr = binary_expr("boolean_field".as_expr(), Operator::BitwiseOr, lit(true));

core/object_store_mem_cache/src/cache_system/s3_fifo_cache/s3_fifo.rs

@@ -904,7 +904,7 @@ where
     ///
     /// This scans through the "main" queue and for every entry it either:
     ///
-    /// - **move to to back:** If the entry was used, move it to the back of the "main" queue. Decrease its usage
+    /// - **move to back:** If the entry was used, move it to the back of the "main" queue. Decrease its usage
     ///   counter by one.
     /// - **move to "ghost" set:** If the entry was NOT used, remove it from the cache and add its key to the "ghost" set.
     ///

core/parquet_file/src/storage.rs

@@ -100,7 +100,7 @@ impl std::fmt::Display for StorageId {
 
 /// Inputs required to build a [`DataSourceExec`] for one or multiple files.
 ///
-/// The files shall be grouped by [`object_store_url`](Self::object_store_url). For each each object store, you shall
+/// The files shall be grouped by [`object_store_url`](Self::object_store_url). For each object store, you shall
 /// create one [`DataSourceExec`] and put each file into its own "file group".
 ///
 /// [`DataSourceExec`]: datafusion::datasource::memory::DataSourceExec

core/predicate/src/lib.rs

@@ -251,7 +251,7 @@ impl Predicate {
     /// Sets the timestamp range
     pub fn with_range(mut self, start: i64, end: i64) -> Self {
         // Without more thought, redefining the timestamp range would
-        // lose the old range. Asser that that cannot happen.
+        // lose the old range. Assert that it cannot happen.
         assert!(
             self.range.is_none(),
             "Unexpected re-definition of timestamp range"

core/trace_exporters/src/jaeger.rs

@@ -357,7 +357,7 @@ impl std::io::Write for MessageWriter {
                     // Jaeger requires that each thrift Message is encoded in exactly one UDP
                     // packet, as this ensures it either arrives in its entirety or not at all
                     //
-                    // If for whatever reason the packet is truncated, the agent will fail to
+                    // If for whatever reason the packet is truncated, the agent will fail
                     // to decode it, likely due to a missing stop-field, and discard it
                     error!(%written, %message_len, "jaeger agent exporter failed to write message as single UDP packet");
                 }

core/trogging/src/lib.rs

@@ -601,7 +601,7 @@ ERROR foo
     }
 
     // This test checks that [`make_writer`] returns a writer that implement line buffering, which means
-    // that written data is not flushed to the underlying IO writer until a a whole line is been written
+    // that written data is not flushed to the underlying IO writer until a whole line has been written
     // to the line buffered writer, possibly by multiple calls to the `write` method.
     // (In fact, the `logfmt` layer does call `write` multiple times for each log line).
     //

influxdb3/tests/server/query.rs

@@ -1733,7 +1733,7 @@ async fn api_v1_query_group_by() {
             })
             .collect::<Vec<Value>>()
             .await;
-        // Use a snapshot to assert on the output structure. This deserializes each emitted line as
+        // Use a snapshot to assert on the output structure. This deserializes each emitted line
         // as JSON first, then combines and collects them into a Vec<Value> to serialize into a JSON
         // array for the snapshot.
         insta::with_settings!({
@@ -1835,7 +1835,7 @@ async fn api_v1_query_group_by_with_nulls() {
             })
             .collect::<Vec<Value>>()
             .await;
-        // Use a snapshot to assert on the output structure. This deserializes each emitted line as
+        // Use a snapshot to assert on the output structure. This deserializes each emitted line
         // as JSON first, then combines and collects them into a Vec<Value> to serialize into a JSON
         // array for the snapshot.
         insta::with_settings!({