Improved CLI: explicit CLI flags for compaction settings

crates/top/rerun/src/commands/rrd/merge_compact.rs

@@ -43,20 +43,53 @@ pub struct CompactCommand {
 
     #[arg(short = 'o', long = "output", value_name = "dst.(rrd|rbl)")]
     path_to_output_rrd: String,
+
+    /// What is the threshold, in bytes, after which a Chunk cannot be compacted any further?
+    ///
+    /// Overrides RERUN_CHUNK_MAX_BYTES if set.
+    #[arg(long = "max-bytes")]
+    max_bytes: Option<u64>,
+
+    /// What is the threshold, in rows, after which a Chunk cannot be compacted any further?
+    ///
+    /// Overrides RERUN_CHUNK_MAX_ROWS if set.
+    #[arg(long = "max-rows")]
+    max_rows: Option<u64>,
+
+    /// What is the threshold, in rows, after which a Chunk cannot be compacted any further?
+    ///
+    /// This specifically applies to _non_ time-sorted chunks.
+    ///
+    /// Overrides RERUN_CHUNK_MAX_ROWS_IF_UNSORTED if set.
+    #[arg(long = "max-rows-if-unsorted")]
+    max_rows_if_unsorted: Option<u64>,
 }
 
 impl CompactCommand {
     pub fn run(&self) -> anyhow::Result<()> {
         let Self {
             path_to_input_rrds,
             path_to_output_rrd,
+            max_bytes,
+            max_rows,
+            max_rows_if_unsorted,
         } = self;
 
         let mut store_config = ChunkStoreConfig::from_env().unwrap_or_default();
         // NOTE: We're doing headless processing, there's no point in running subscribers, it will just
         // (massively) slow us down.
         store_config.enable_changelog = false;
 
+        if let Some(max_bytes) = max_bytes {
+            store_config.chunk_max_bytes = *max_bytes;
+        }
+        if let Some(max_rows) = max_rows {
+            store_config.chunk_max_rows = *max_rows;
+        }
+        if let Some(max_rows_if_unsorted) = max_rows_if_unsorted {
+            store_config.chunk_max_rows_if_unsorted = *max_rows_if_unsorted;
+        }
+
         merge_and_compact(&store_config, path_to_input_rrds, path_to_output_rrd)
     }
 }

crates/top/rerun/src/commands/rrd/mod.rs

@@ -26,12 +26,18 @@ pub enum RrdCommands {
 
     /// Compacts the contents of one or more .rrd/.rbl files and writes the result to a new file.
     ///
-    /// Use the usual environment variables to control the compaction thresholds:
+    /// Uses the usual environment variables to control the compaction thresholds:
     /// `RERUN_CHUNK_MAX_ROWS`,
     /// `RERUN_CHUNK_MAX_ROWS_IF_UNSORTED`,
     /// `RERUN_CHUNK_MAX_BYTES`.
     ///
-    /// Example: `RERUN_CHUNK_MAX_ROWS=4096 RERUN_CHUNK_MAX_BYTES=1048576 rerun rrd compact /my/recordings/*.rrd -o output.rrd`
+    /// Unless explicit flags are passed, in which case they will override environment values.
+    ///
+    /// Examples:
+    ///
+    /// * `RERUN_CHUNK_MAX_ROWS=4096 RERUN_CHUNK_MAX_BYTES=1048576 rerun rrd compact /my/recordings/*.rrd -o output.rrd`
+    ///
+    /// * `rerun rrd compact --max-rows 4096 --max-bytes=1048576 /my/recordings/*.rrd -o output.rrd`
     Compact(CompactCommand),
 
     /// Merges the contents of multiple .rrd/.rbl files, and writes the result to a new file.