feat(tui): independent scroll regions for conversation and tool output by ljm3790865 · Pull Request #2113 · Hmbown/CodeWhale

ljm3790865 · 2026-05-25T08:39:20Z

Summary

Split the chat area into two independent scroll regions:

Upper region: conversation/transcript (user, assistant, thinking)
Lower region: tool output (exec, exploring, patch, etc.)

Each region has its own scroll state and cache, allowing independent scrolling. Mouse wheel events are routed to the appropriate region based on cursor position.

This fixes the scroll demon issue where running tasks would cause the entire interface to scroll, blocking UI elements.

Changes

File	Change
`crates/tui/src/tui/app.rs`	Add tool_output_scroll/cache, region areas
`crates/tui/src/tui/history.rs`	Add is_conversation_cell(), is_tool_output_cell()
`crates/tui/src/tui/widgets/mod.rs`	Add ChatRegion enum, region-aware ChatWidget
`crates/tui/src/tui/ui.rs`	Split chat area into two regions
`crates/tui/src/tui/mouse_ui.rs`	Route scroll events based on cursor position

Test plan

cargo check --workspace (compiles successfully)
Run TUI and execute a task
Verify conversation scrolls independently
Verify tool output scrolls independently

?? Generated with Claude Code

…utput Split the chat area into two independent scroll regions: - Upper region: conversation/transcript (user, assistant, thinking) - Lower region: tool output (exec, exploring, patch, etc.) Each region has its own scroll state and cache, allowing independent scrolling. Mouse wheel events are routed to the appropriate region based on cursor position. This fixes the scroll demon issue where running tasks would cause the entire interface to scroll, blocking UI elements. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

gitguardian · 2026-05-25T08:39:56Z

⚠️ GitGuardian has uncovered 1 secret following the scan of your pull request.

Please consider investigating the findings and remediating the incidents. Failure to do so may lead to compromising the associated services or software components.

Since your pull request originates from a forked repository, GitGuardian is not able to associate the secrets uncovered with secret incidents on your GitGuardian dashboard.
Skipping this check run and merging your pull request will create secret incidents on your GitGuardian dashboard.

🔎 Detected hardcoded secret in your pull request

GitGuardian id	GitGuardian status	Secret	Commit	Filename
-	-	DeepSeek API Key	`c6d25e5`	pack_claude_final.js	View secret

🛠 Guidelines to remediate hardcoded secrets

Understand the implications of revoking this secret by investigating where it is used in your code.
Replace and store your secret safely. Learn here the best practices.
Revoke and rotate this secret.
If possible, rewrite git history. Rewriting git history is not a trivial act. You might completely break other contributing developers' workflow and you risk accidentally deleting legitimate data.

To avoid such incidents in the future consider

following these best practices for managing and storing secrets including API keys and other credentials
install secret detection on pre-commit to catch secret before it leaves your machine and ease remediation.

^{🦉 GitGuardian detects secrets in your source code to help developers and security teams secure the modern development process. You are seeing this because you or someone else with access to this repository has authorized GitGuardian to scan your pull request.}

gemini-code-assist

Code Review

This pull request implements a split-pane layout in the TUI to separate conversation history from tool outputs and introduces a comprehensive set of diagnostic and deployment scripts. Reviewers identified critical security risks involving hardcoded API keys in configuration and script files, as well as portability issues with absolute file paths. Technical feedback also addresses a layout bug where chat regions may overlap with the sidebar, a regression in scroll-management logic, and suggestions for refactoring duplicated code and improving UI responsiveness to terminal resizing.

gemini-code-assist · 2026-05-25T08:40:54Z

+# Active provider + DeepSeek defaults
+# ─────────────────────────────────────────────────────────────────────────────────
+provider = "deepseek"
+api_key = "sk-f1bccc35f03d4e90be027a54ef02399a"


Hardcoded API key detected. This is a critical security risk. Please remove the key and use environment variables or a secure configuration mechanism that is not committed to version control.

gemini-code-assist · 2026-05-25T08:40:54Z

+# DeepSeek Platform (DeepSeek平台配置)
+# ─────────────────────────────────────────────────────────────────────────────────
+[providers.deepseek]
+api_key = "sk-f1bccc35f03d4e90be027a54ef02399a"


Hardcoded API key detected in the provider-specific configuration section.

gemini-code-assist · 2026-05-25T08:40:54Z

+$workDir = "$env:USERPROFILE\DeepSeek Tui"
+
+# 设置环境
+$env:DEEPSEEK_API_KEY = "sk-f1bccc35f03d4e90be027a54ef02399a"


Hardcoded API key detected in the PowerShell script. Secrets should be passed via environment variables or retrieved from a secure store at runtime.

gemini-code-assist · 2026-05-25T08:40:54Z

+const envContent=[
+'# Claude Code -> DeepSeek API',
+'ANTHROPIC_BASE_URL=https://api.deepseek.com/v1',
+'ANTHROPIC_API_KEY=sk-f1bccc35f03d4e90be027a54ef02399a',


Hardcoded API key detected in the environment variable string.

gemini-code-assist · 2026-05-25T08:40:54Z

+'Write-Host "[4/5] Setting up DeepSeek API..."',
+'Copy-Item "$s\\.env" $u -Force',
+'[Environment]::SetEnvironmentVariable("ANTHROPIC_BASE_URL","https://api.deepseek.com/v1","User")',
+'[Environment]::SetEnvironmentVariable("ANTHROPIC_API_KEY","sk-f1bccc35f03d4e90be027a54ef02399a","User")',


Hardcoded API key detected in the PowerShell command string within the JavaScript file.

gemini-code-assist · 2026-05-25T08:40:54Z

+    let conversation_area = chat_layout[0];
+    let tool_output_area = chat_layout[1];


The conversation_area and tool_output_area are calculated here based on the full width of chunks[1]. However, later in the function (line 5819), a sidebar might be added which reduces the available width for the chat area. This will cause the chat widgets to overlap with the sidebar because they are rendered using these stale, full-width areas. These regions should be calculated from the finalized chat_area after the sidebar logic has been applied.

gemini-code-assist · 2026-05-25T08:40:54Z

            if update.delta_lines != 0 {
-                app.user_scrolled_during_stream = true;
                app.needs_redraw = true;
            }
        }
        MouseEventKind::ScrollDown => {
            let update = app.viewport.mouse_scroll.on_scroll(ScrollDirection::Down);
-            app.viewport.pending_scroll_delta = app
-                .viewport
-                .pending_scroll_delta
-                .saturating_add(update.delta_lines);
+            let pending_delta = if mouse_hits_tool_output_area(app, mouse) {
+                &mut app.viewport.pending_tool_scroll_delta
+            } else {
+                &mut app.viewport.pending_scroll_delta
+            };
+            *pending_delta = pending_delta.saturating_add(update.delta_lines);
            if update.delta_lines != 0 {
-                app.user_scrolled_during_stream = true;
                app.needs_redraw = true;


The removal of app.user_scrolled_during_stream = true; appears to be a regression. This flag is typically used to disable auto-scrolling to the bottom when new content arrives while the user is manually inspecting previous history. While the goal is to fix the 'scroll demon', removing this flag entirely will cause the UI to jump to the bottom whenever the assistant or a tool produces output, even if the user is actively scrolling. This logic should likely be made region-aware instead of being deleted.

gemini-code-assist · 2026-05-25T08:40:54Z

        .split(size);

+    // Split chat area into conversation (upper) and tool output (lower) regions
+    const TOOL_OUTPUT_DEFAULT_HEIGHT: u16 = 10;


Using a fixed height of 10 for the tool output region (TOOL_OUTPUT_DEFAULT_HEIGHT) can lead to poor UX on small terminal windows. If the total available height is small, the conversation region might become invisible or unusable. Consider making this height dynamic (e.g., a percentage of the total height) or allowing the user to resize it.

gemini-code-assist · 2026-05-25T08:40:54Z

+fn mouse_hits_tool_output_area(app: &App, mouse: MouseEvent) -> bool {
+    if let Some(area) = app.viewport.tool_output_area {
+        return mouse.row >= area.y && mouse.row < area.y.saturating_add(area.height)
+            && mouse.column >= area.x && mouse.column < area.x.saturating_add(area.width);
+    }
+    false
+}
+
+/// Check if mouse is over the conversation area (upper region).
+fn mouse_hits_conversation_area(app: &App, mouse: MouseEvent) -> bool {
+    if let Some(area) = app.viewport.conversation_area {
+        return mouse.row >= area.y && mouse.row < area.y.saturating_add(area.height)
+            && mouse.column >= area.x && mouse.column < area.x.saturating_add(area.width);
+    }
+    false
+}


The logic for checking if a mouse event hits a specific area is duplicated. Additionally, mouse_hits_conversation_area is defined but never used in the current changes. Consider refactoring the area check into a helper function.

Suggested change

fn mouse_hits_tool_output_area(app: &App, mouse: MouseEvent) -> bool {

if let Some(area) = app.viewport.tool_output_area {

return mouse.row >= area.y && mouse.row < area.y.saturating_add(area.height)

&& mouse.column >= area.x && mouse.column < area.x.saturating_add(area.width);

}

false

}

/// Check if mouse is over the conversation area (upper region).

fn mouse_hits_conversation_area(app: &App, mouse: MouseEvent) -> bool {

if let Some(area) = app.viewport.conversation_area {

return mouse.row >= area.y && mouse.row < area.y.saturating_add(area.height)

&& mouse.column >= area.x && mouse.column < area.x.saturating_add(area.width);

}

false

}

fn mouse_hits_area(area: Rect, mouse: MouseEvent) -> bool {

mouse.row >= area.y && mouse.row < area.y.saturating_add(area.height)

&& mouse.column >= area.x && mouse.column < area.x.saturating_add(area.width)

}

fn mouse_hits_tool_output_area(app: &App, mouse: MouseEvent) -> bool {

app.viewport.tool_output_area.map_or(false, |area| mouse_hits_area(area, mouse))

}

gemini-code-assist · 2026-05-25T08:40:54Z

+set MSYS_NO_PATHCONV=1
+set MSYS2_ARG_CONV_EXCL=*
+set PATH=C:\msys64\mingw64\bin;%PATH%
+cd /d "C:\Users\ljm37\DeepSeek Tui"


Hardcoded absolute path detected (C:\Users\ljm37\...). This script will fail on any other machine. Use relative paths or environment variables like %USERPROFILE%.

cd /d "%USERPROFILE%\DeepSeek Tui"

Hmbown · 2026-05-26T10:58:41Z

The GitGuardian failure is likely detecting what looks like an API key pattern in the test fixtures. This is usually a false positive — GitGuardian scans for secret patterns in code. Could you check if there's a base64-encoded string or key-like pattern in the PR diff that's triggering it? If it's a test-only placeholder, we can add a exclusion or mark it as a false positive.

G1 Agent and others added 2 commits May 25, 2026 16:36

chore: update

c6d25e5

gemini-code-assist Bot reviewed May 25, 2026

View reviewed changes

		let conversation_area = chat_layout[0];
		let tool_output_area = chat_layout[1];

Conversation

ljm3790865 commented May 25, 2026

Summary

Changes

Test plan

Uh oh!

gitguardian Bot commented May 25, 2026

⚠️ GitGuardian has uncovered 1 secret following the scan of your pull request.

Uh oh!

gemini-code-assist Bot left a comment

Choose a reason for hiding this comment

Code Review

Uh oh!

gemini-code-assist Bot May 25, 2026

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist Bot May 25, 2026

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist Bot May 25, 2026

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist Bot May 25, 2026

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist Bot May 25, 2026

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist Bot May 25, 2026

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist Bot May 25, 2026

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist Bot May 25, 2026

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist Bot May 25, 2026

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist Bot May 25, 2026

Choose a reason for hiding this comment

Uh oh!

Hmbown commented May 26, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants