From 7a53515bd81664b6dfea0158659a978b041cafa1 Mon Sep 17 00:00:00 2001
From: dario curreri <dariocurr@gmail.com>
Date: Wed, 22 Jul 2026 12:40:00 +0200
Subject: [PATCH] feat: allow configuring datafusion-cli history file location

datafusion-cli hardcoded the interactive shell's rustyline history as
the relative path `.history`, so every launch directory silently got
its own history file with no way to point it elsewhere.

Add a `--history-file <PATH>` flag and a `DATAFUSION_HISTORY_FILE`
environment variable (flag takes precedence), keeping `.history` as
the default so existing behavior is unchanged.

Closes #23796.
---
 .../examples/cli-session-context.rs           |  4 ++-
 datafusion-cli/src/exec.rs                    |  7 ++--
 datafusion-cli/src/main.rs                    | 35 +++++++++++++++++--
 docs/source/user-guide/cli/usage.md           |  2 ++
 4 files changed, 43 insertions(+), 5 deletions(-)

diff --git a/datafusion-cli/examples/cli-session-context.rs b/datafusion-cli/examples/cli-session-context.rs
index 6095072163870..fbc82923726e2 100644
--- a/datafusion-cli/examples/cli-session-context.rs
+++ b/datafusion-cli/examples/cli-session-context.rs
@@ -94,5 +94,7 @@ pub async fn main() {
         instrumented_registry: Arc::new(InstrumentedObjectStoreRegistry::new()),
     };
 
-    exec_from_repl(&my_ctx, &mut print_options).await.unwrap();
+    exec_from_repl(&my_ctx, &mut print_options, None)
+        .await
+        .unwrap();
 }
diff --git a/datafusion-cli/src/exec.rs b/datafusion-cli/src/exec.rs
index bc2c15f48debb..42f7dcb88f044 100644
--- a/datafusion-cli/src/exec.rs
+++ b/datafusion-cli/src/exec.rs
@@ -48,6 +48,7 @@ use std::collections::HashMap;
 use std::fs::File;
 use std::io::BufReader;
 use std::io::prelude::*;
+use std::path::Path;
 use tokio::signal;
 
 /// run and execute SQL statements and commands, against a context with the given print options
@@ -129,13 +130,15 @@ pub async fn exec_from_files(
 pub async fn exec_from_repl(
     ctx: &dyn CliSessionContext,
     print_options: &mut PrintOptions,
+    history_file: Option<&Path>,
 ) -> rustyline::Result<()> {
+    let history_file = history_file.unwrap_or_else(|| Path::new(".history"));
     let mut rl = Editor::new()?;
     rl.set_helper(Some(CliHelper::new(
         &ctx.task_ctx().session_config().options().sql_parser.dialect,
         print_options.color,
     )));
-    rl.load_history(".history").ok();
+    rl.load_history(history_file).ok();
 
     loop {
         match rl.readline("> ") {
@@ -210,7 +213,7 @@ pub async fn exec_from_repl(
         }
     }
 
-    rl.save_history(".history")
+    rl.save_history(history_file)
 }
 
 pub(super) async fn exec_and_print(
diff --git a/datafusion-cli/src/main.rs b/datafusion-cli/src/main.rs
index 20a2537d7c10c..00739e2951453 100644
--- a/datafusion-cli/src/main.rs
+++ b/datafusion-cli/src/main.rs
@@ -18,7 +18,7 @@
 use std::collections::HashMap;
 use std::env;
 use std::num::NonZeroUsize;
-use std::path::Path;
+use std::path::{Path, PathBuf};
 use std::process::ExitCode;
 use std::sync::{Arc, LazyLock};
 
@@ -109,6 +109,13 @@ struct Args {
     )]
     rc: Option<Vec<String>>,
 
+    #[clap(
+        long,
+        help = "Path to the file used to persist interactive shell history. Overrides DATAFUSION_HISTORY_FILE if set, default to .history",
+        value_parser(parse_valid_history_file)
+    )]
+    history_file: Option<String>,
+
     #[clap(long, value_enum, default_value_t = PrintFormat::Automatic)]
     format: PrintFormat,
 
@@ -304,12 +311,17 @@ async fn main_inner() -> Result<()> {
         }
     };
 
+    let history_file = args
+        .history_file
+        .map(PathBuf::from)
+        .or_else(|| env::var_os("DATAFUSION_HISTORY_FILE").map(PathBuf::from));
+
     if repl_mode {
         if !rc.is_empty() {
             exec::exec_from_files(&ctx, rc, &print_options).await?;
         }
         // TODO maybe we can have thiserror for cli but for now let's keep it simple
-        return exec::exec_from_repl(&ctx, &mut print_options)
+        return exec::exec_from_repl(&ctx, &mut print_options, history_file.as_deref())
             .await
             .map_err(|e| DataFusionError::External(Box::new(e)));
     }
@@ -379,6 +391,25 @@ fn parse_valid_data_dir(dir: &str) -> Result<String, String> {
     }
 }
 
+fn parse_valid_history_file(path: &str) -> Result<String, String> {
+    let path = Path::new(path);
+    if path.is_dir() {
+        return Err(format!(
+            "Invalid history file '{}': is a directory",
+            path.display()
+        ));
+    }
+    match path.parent() {
+        Some(parent) if !parent.as_os_str().is_empty() && !parent.is_dir() => {
+            Err(format!(
+                "Invalid history file '{}': parent directory does not exist",
+                path.display()
+            ))
+        }
+        _ => Ok(path.to_string_lossy().into_owned()),
+    }
+}
+
 fn parse_batch_size(size: &str) -> Result<usize, String> {
     match size.parse::<usize>() {
         Ok(size) if size > 0 => Ok(size),
diff --git a/docs/source/user-guide/cli/usage.md b/docs/source/user-guide/cli/usage.md
index 75c6698f007a5..a4fe62f17f29b 100644
--- a/docs/source/user-guide/cli/usage.md
+++ b/docs/source/user-guide/cli/usage.md
@@ -39,6 +39,8 @@ Options:
           Execute commands from file(s), then exit
   -r, --rc [<RC>...]
           Run the provided files on startup instead of ~/.datafusionrc
+      --history-file <HISTORY_FILE>
+          Path to the file used to persist interactive shell history. Overrides DATAFUSION_HISTORY_FILE if set, default to .history
       --format <FORMAT>
           [default: automatic] [possible values: csv, tsv, table, json, nd-json, automatic]
   -q, --quiet