Files
llmx/codex-rs/tui/src/history_cell.rs

1901 lines
64 KiB
Rust
Raw Normal View History

use crate::diff_render::create_diff_summary;
use crate::exec_command::relativize_to_home;
use crate::exec_command::strip_bash_lc_and_escape;
use crate::markdown::append_markdown;
use crate::render::line_utils::line_to_static;
tui: fix approval dialog for large commands (#3087) #### Summary - Emit a “Proposed Command” history cell when an ExecApprovalRequest arrives (parity with proposed patches). - Simplify the approval dialog: show only the reason/instructions; move the command preview into history. - Make approval/abort decision history concise: - Single line snippet; if multiline, show first line + " ...". - Truncate to 80 graphemes with ellipsis for very long commands. #### Details - History - Add `new_proposed_command` to render a header and indented command preview. - Use shared `prefix_lines` helper for first/subsequent line prefixes. - Approval UI - `UserApprovalWidget` no longer renders the command in the modal; shows optional `reason` text only. - Decision history renders an inline, dimmed snippet per rules above. - Tests (snapshot-based) - Proposed/decision flow for short command. - Proposed multi-line + aborted decision snippet with “ ...”. - Very long one-line command -> truncated snippet with “…”. - Updated existing exec approval snapshots and test reasons. <img width="1053" height="704" alt="Screenshot 2025-09-03 at 11 57 35 AM" src="https://github.com/user-attachments/assets/9ed4c316-9daf-4ac1-80ff-7ae1f481dd10" /> after approving: <img width="1053" height="704" alt="Screenshot 2025-09-03 at 11 58 18 AM" src="https://github.com/user-attachments/assets/a44e243f-eb9d-42ea-87f4-171b3fb481e7" /> rejection: <img width="1053" height="207" alt="Screenshot 2025-09-03 at 11 58 45 AM" src="https://github.com/user-attachments/assets/a022664b-ae0e-4b70-a388-509208707934" /> big command: https://github.com/user-attachments/assets/2dd976e5-799f-4af7-9682-a046e66cc494
2025-09-04 16:54:53 -07:00
use crate::render::line_utils::prefix_lines;
use crate::render::line_utils::push_owned_lines;
use crate::slash_command::SlashCommand;
codex-rs: make tool calls prettier (#1211) This PR overhauls how active tool calls and completed tool calls are displayed: 1. More use of colour to indicate success/failure and distinguish between components like tool name+arguments 2. Previously, the entire `CallToolResult` was serialized to JSON and pretty-printed. Now, we extract each individual `CallToolResultContent` and print those 1. The previous solution was wasting space by unnecessarily showing details of the `CallToolResult` struct to users, without formatting the actual tool call results nicely 2. We're now able to show users more information from tool results in less space, with nicer formatting when tools return JSON results ### Before: <img width="1251" alt="Screenshot 2025-06-03 at 11 24 26" src="https://github.com/user-attachments/assets/5a58f222-219c-4c53-ace7-d887194e30cf" /> ### After: <img width="1265" alt="image" src="https://github.com/user-attachments/assets/99fe54d0-9ebe-406a-855b-7aa529b91274" /> ## Future Work 1. Integrate image tool result handling better. We should be able to display images even if they're not the first `CallToolResultContent` 2. Users should have some way to view the full version of truncated tool results 3. It would be nice to add some left padding for tool results, make it more clear that they are results. This is doable, just a little fiddly due to the way `first_visible_line` scrolling works 4. There's almost certainly a better way to format JSON than "all on 1 line with spaces to make Ratatui wrapping work". But I think that works OK for now.
2025-06-03 14:29:26 -07:00
use crate::text_formatting::format_and_truncate_tool_result;
use crate::wrapping::RtOptions;
use crate::wrapping::word_wrap_line;
use crate::wrapping::word_wrap_lines;
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
use base64::Engine;
use codex_ansi_escape::ansi_escape_line;
use codex_common::create_config_summary_entries;
use codex_common::elapsed::format_duration;
use codex_core::auth::get_auth_file;
use codex_core::auth::try_read_auth_json;
use codex_core::config::Config;
use codex_core::config_types::ReasoningSummaryFormat;
use codex_core::plan_tool::PlanItemArg;
use codex_core::plan_tool::StepStatus;
use codex_core::plan_tool::UpdatePlanArgs;
use codex_core::project_doc::discover_project_doc_paths;
use codex_core::protocol::FileChange;
use codex_core::protocol::McpInvocation;
use codex_core::protocol::SandboxPolicy;
use codex_core::protocol::SessionConfiguredEvent;
use codex_core::protocol::TokenUsage;
use codex_protocol::mcp_protocol::ConversationId;
use codex_protocol::num_format::format_with_separators;
use codex_protocol::parse_command::ParsedCommand;
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
use image::DynamicImage;
use image::ImageReader;
use itertools::Itertools;
codex-rs: make tool calls prettier (#1211) This PR overhauls how active tool calls and completed tool calls are displayed: 1. More use of colour to indicate success/failure and distinguish between components like tool name+arguments 2. Previously, the entire `CallToolResult` was serialized to JSON and pretty-printed. Now, we extract each individual `CallToolResultContent` and print those 1. The previous solution was wasting space by unnecessarily showing details of the `CallToolResult` struct to users, without formatting the actual tool call results nicely 2. We're now able to show users more information from tool results in less space, with nicer formatting when tools return JSON results ### Before: <img width="1251" alt="Screenshot 2025-06-03 at 11 24 26" src="https://github.com/user-attachments/assets/5a58f222-219c-4c53-ace7-d887194e30cf" /> ### After: <img width="1265" alt="image" src="https://github.com/user-attachments/assets/99fe54d0-9ebe-406a-855b-7aa529b91274" /> ## Future Work 1. Integrate image tool result handling better. We should be able to display images even if they're not the first `CallToolResultContent` 2. Users should have some way to view the full version of truncated tool results 3. It would be nice to add some left padding for tool results, make it more clear that they are results. This is doable, just a little fiddly due to the way `first_visible_line` scrolling works 4. There's almost certainly a better way to format JSON than "all on 1 line with spaces to make Ratatui wrapping work". But I think that works OK for now.
2025-06-03 14:29:26 -07:00
use mcp_types::EmbeddedResourceResource;
use mcp_types::ResourceLink;
use ratatui::prelude::*;
use ratatui::style::Modifier;
use ratatui::style::Style;
use ratatui::style::Styled;
use ratatui::style::Stylize;
use ratatui::widgets::Paragraph;
use ratatui::widgets::WidgetRef;
use ratatui::widgets::Wrap;
use std::collections::HashMap;
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
use std::io::Cursor;
use std::path::Path;
use std::path::PathBuf;
use std::time::Duration;
use std::time::Instant;
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
use tracing::error;
use unicode_width::UnicodeWidthStr;
#[derive(Clone, Debug)]
pub(crate) struct CommandOutput {
pub(crate) exit_code: i32,
pub(crate) stdout: String,
pub(crate) stderr: String,
pub(crate) formatted_output: String,
}
#[derive(Clone, Debug)]
pub(crate) enum PatchEventType {
ApprovalRequest,
ApplyBegin { auto_approved: bool },
}
/// Represents an event to display in the conversation history. Returns its
/// `Vec<Line<'static>>` representation to make it easier to display in a
/// scrollable list.
pub(crate) trait HistoryCell: std::fmt::Debug + Send + Sync {
fn display_lines(&self, width: u16) -> Vec<Line<'static>>;
fn transcript_lines(&self) -> Vec<Line<'static>> {
self.display_lines(u16::MAX)
}
fn desired_height(&self, width: u16) -> u16 {
Paragraph::new(Text::from(self.display_lines(width)))
.wrap(Wrap { trim: false })
.line_count(width)
.try_into()
.unwrap_or(0)
}
fn is_stream_continuation(&self) -> bool {
false
}
}
#[derive(Debug)]
pub(crate) struct UserHistoryCell {
message: String,
}
impl HistoryCell for UserHistoryCell {
fn display_lines(&self, width: u16) -> Vec<Line<'static>> {
let mut lines: Vec<Line<'static>> = Vec::new();
// Wrap the content first, then prefix each wrapped line with the marker.
let wrap_width = width.saturating_sub(1); // account for the ▌ prefix
let wrapped = textwrap::wrap(
&self.message,
textwrap::Options::new(wrap_width as usize)
.wrap_algorithm(textwrap::WrapAlgorithm::FirstFit), // Match textarea wrap
);
for line in wrapped {
lines.push(vec!["".cyan().dim(), line.to_string().dim()].into());
}
lines
}
fn transcript_lines(&self) -> Vec<Line<'static>> {
let mut lines: Vec<Line<'static>> = Vec::new();
lines.push("user".cyan().bold().into());
lines.extend(self.message.lines().map(|l| l.to_string().into()));
lines
}
}
#[derive(Debug)]
pub(crate) struct AgentMessageCell {
lines: Vec<Line<'static>>,
is_first_line: bool,
}
impl AgentMessageCell {
pub(crate) fn new(lines: Vec<Line<'static>>, is_first_line: bool) -> Self {
Self {
lines,
is_first_line,
}
}
}
impl HistoryCell for AgentMessageCell {
fn display_lines(&self, width: u16) -> Vec<Line<'static>> {
word_wrap_lines(
&self.lines,
RtOptions::new(width as usize)
.initial_indent(if self.is_first_line {
"> ".into()
} else {
" ".into()
})
.subsequent_indent(" ".into()),
)
}
fn transcript_lines(&self) -> Vec<Line<'static>> {
let mut out: Vec<Line<'static>> = Vec::new();
if self.is_first_line {
out.push("codex".magenta().bold().into());
}
out.extend(self.lines.clone());
out
}
fn is_stream_continuation(&self) -> bool {
!self.is_first_line
}
}
#[derive(Debug)]
pub(crate) struct PlainHistoryCell {
lines: Vec<Line<'static>>,
}
impl HistoryCell for PlainHistoryCell {
fn display_lines(&self, _width: u16) -> Vec<Line<'static>> {
self.lines.clone()
}
}
#[derive(Debug)]
pub(crate) struct TranscriptOnlyHistoryCell {
lines: Vec<Line<'static>>,
}
impl HistoryCell for TranscriptOnlyHistoryCell {
fn display_lines(&self, _width: u16) -> Vec<Line<'static>> {
Vec::new()
}
fn transcript_lines(&self) -> Vec<Line<'static>> {
self.lines.clone()
}
}
#[derive(Debug)]
pub(crate) struct PatchHistoryCell {
event_type: PatchEventType,
changes: HashMap<PathBuf, FileChange>,
cwd: PathBuf,
}
impl HistoryCell for PatchHistoryCell {
fn display_lines(&self, width: u16) -> Vec<Line<'static>> {
create_diff_summary(
&self.changes,
self.event_type.clone(),
&self.cwd,
width as usize,
)
}
}
#[derive(Debug, Clone)]
pub(crate) struct ExecCall {
pub(crate) call_id: String,
pub(crate) command: Vec<String>,
pub(crate) parsed: Vec<ParsedCommand>,
pub(crate) output: Option<CommandOutput>,
start_time: Option<Instant>,
duration: Option<Duration>,
}
#[derive(Debug)]
pub(crate) struct ExecCell {
calls: Vec<ExecCall>,
}
impl HistoryCell for ExecCell {
fn display_lines(&self, width: u16) -> Vec<Line<'static>> {
if self.is_exploring_cell() {
self.exploring_display_lines(width)
} else {
self.command_display_lines(width)
}
}
fn transcript_lines(&self) -> Vec<Line<'static>> {
let mut lines: Vec<Line<'static>> = vec![];
for call in &self.calls {
let cmd_display = strip_bash_lc_and_escape(&call.command);
for (i, part) in cmd_display.lines().enumerate() {
if i == 0 {
lines.push(vec!["$ ".magenta(), part.to_string().into()].into());
} else {
lines.push(vec![" ".into(), part.to_string().into()].into());
}
}
if let Some(output) = call.output.as_ref() {
lines.extend(output.formatted_output.lines().map(ansi_escape_line));
let duration = call
.duration
.map(format_duration)
.unwrap_or_else(|| "unknown".to_string());
let mut result: Line = if output.exit_code == 0 {
Line::from("".green().bold())
} else {
Line::from(vec![
"".red().bold(),
format!(" ({})", output.exit_code).into(),
])
};
result.push_span(format!("{duration}").dim());
lines.push(result);
}
lines.push("".into());
}
lines
}
}
impl ExecCell {
fn is_active(&self) -> bool {
self.calls.iter().any(|c| c.output.is_none())
}
fn exploring_display_lines(&self, width: u16) -> Vec<Line<'static>> {
let mut out: Vec<Line<'static>> = Vec::new();
let active_start_time = self
.calls
.iter()
.find(|c| c.output.is_none())
.and_then(|c| c.start_time);
out.push(Line::from(vec![
if self.is_active() {
// Show an animated spinner while exploring
spinner(active_start_time)
} else {
"".bold()
},
" ".into(),
if self.is_active() {
"Exploring".bold()
} else {
"Explored".bold()
},
]));
let mut calls = self.calls.clone();
let mut out_indented = Vec::new();
while !calls.is_empty() {
let mut call = calls.remove(0);
if call
.parsed
.iter()
.all(|c| matches!(c, ParsedCommand::Read { .. }))
{
while let Some(next) = calls.first() {
if next
.parsed
.iter()
.all(|c| matches!(c, ParsedCommand::Read { .. }))
{
call.parsed.extend(next.parsed.clone());
calls.remove(0);
} else {
break;
}
}
}
let call_lines: Vec<(&str, Vec<Span<'static>>)> = if call
.parsed
.iter()
.all(|c| matches!(c, ParsedCommand::Read { .. }))
{
let names: Vec<String> = call
.parsed
.iter()
.map(|c| match c {
ParsedCommand::Read { name, .. } => name.clone(),
_ => unreachable!(),
})
.unique()
.collect();
vec![(
"Read",
itertools::Itertools::intersperse(
names.into_iter().map(|n| n.into()),
", ".dim(),
)
.collect(),
)]
} else {
let mut lines = Vec::new();
for p in call.parsed {
match p {
ParsedCommand::Read { name, .. } => {
lines.push(("Read", vec![name.into()]));
}
ParsedCommand::ListFiles { cmd, path } => {
lines.push(("List", vec![path.unwrap_or(cmd).into()]));
}
ParsedCommand::Search { cmd, query, path } => {
lines.push((
"Search",
match (query, path) {
(Some(q), Some(p)) => {
vec![q.into(), " in ".dim(), p.into()]
}
(Some(q), None) => vec![q.into()],
_ => vec![cmd.into()],
},
));
}
ParsedCommand::Unknown { cmd } => {
lines.push(("Run", vec![cmd.into()]));
}
}
}
lines
};
for (title, line) in call_lines {
let line = Line::from(line);
let initial_indent = Line::from(vec![title.cyan(), " ".into()]);
let subsequent_indent = " ".repeat(initial_indent.width()).into();
let wrapped = word_wrap_line(
&line,
RtOptions::new(width as usize)
.initial_indent(initial_indent)
.subsequent_indent(subsequent_indent),
);
push_owned_lines(&wrapped, &mut out_indented);
}
}
out.extend(prefix_lines(out_indented, "".dim(), " ".into()));
out
}
fn command_display_lines(&self, width: u16) -> Vec<Line<'static>> {
use textwrap::Options as TwOptions;
let mut lines: Vec<Line<'static>> = Vec::new();
let [call] = &self.calls.as_slice() else {
panic!("Expected exactly one call in a command display cell");
};
let success = call.output.as_ref().map(|o| o.exit_code == 0);
let bullet = match success {
Some(true) => "".green().bold(),
Some(false) => "".red().bold(),
None => spinner(call.start_time),
};
let title = if self.is_active() { "Running" } else { "Ran" };
let cmd_display = strip_bash_lc_and_escape(&call.command);
// If the command fits on the same line as the header at the current width,
// show a single compact line: "• Ran <command>". Use the width of
// "• Running " (including trailing space) as the reserved prefix width.
// If the command contains newlines, always use the multi-line variant.
let reserved = "• Running ".width();
let mut body_lines: Vec<Line<'static>> = Vec::new();
let highlighted_lines = crate::render::highlight::highlight_bash_to_lines(&cmd_display);
if highlighted_lines.len() == 1
&& highlighted_lines[0].width() < (width as usize).saturating_sub(reserved)
{
let mut line = Line::from(vec![bullet, " ".into(), title.bold(), " ".into()]);
line.extend(highlighted_lines[0].clone());
lines.push(line);
} else {
lines.push(vec![bullet, " ".into(), title.bold()].into());
for hl_line in highlighted_lines.iter() {
let opts = crate::wrapping::RtOptions::new((width as usize).saturating_sub(4))
.initial_indent("".into())
.subsequent_indent(" ".into())
// Hyphenation likes to break words on hyphens, which is bad for bash scripts --because-of-flags.
.word_splitter(textwrap::WordSplitter::NoHyphenation);
let wrapped_borrowed = crate::wrapping::word_wrap_line(hl_line, opts);
body_lines.extend(wrapped_borrowed.iter().map(|l| line_to_static(l)));
}
}
if let Some(output) = call.output.as_ref()
&& output.exit_code != 0
{
let out = output_lines(Some(output), false, false, false)
.into_iter()
.join("\n");
if !out.trim().is_empty() {
// Wrap the output.
for line in out.lines() {
let wrapped = textwrap::wrap(line, TwOptions::new(width as usize - 4));
body_lines.extend(wrapped.into_iter().map(|l| Line::from(l.to_string().dim())));
}
}
}
lines.extend(prefix_lines(body_lines, "".dim(), " ".into()));
lines
}
}
impl WidgetRef for &ExecCell {
fn render_ref(&self, area: Rect, buf: &mut Buffer) {
if area.height == 0 {
return;
}
let content_area = Rect {
x: area.x,
y: area.y,
width: area.width,
height: area.height,
};
let lines = self.display_lines(area.width);
let max_rows = area.height as usize;
let rendered = if lines.len() > max_rows {
// Keep the last `max_rows` lines in original order
lines[lines.len() - max_rows..].to_vec()
} else {
lines
};
Paragraph::new(Text::from(rendered))
.wrap(Wrap { trim: false })
.render(content_area, buf);
}
}
impl ExecCell {
/// Convert an active exec cell into a failed, completed exec cell.
/// Any call without output is marked as failed with a red ✗.
pub(crate) fn into_failed(mut self) -> ExecCell {
for call in self.calls.iter_mut() {
if call.output.is_none() {
let elapsed = call
.start_time
.map(|st| st.elapsed())
.unwrap_or_else(|| Duration::from_millis(0));
call.start_time = None;
call.duration = Some(elapsed);
call.output = Some(CommandOutput {
exit_code: 1,
stdout: String::new(),
stderr: String::new(),
formatted_output: String::new(),
});
}
}
self
}
pub(crate) fn new(call: ExecCall) -> Self {
ExecCell { calls: vec![call] }
}
fn is_exploring_call(call: &ExecCall) -> bool {
!call.parsed.is_empty()
&& call.parsed.iter().all(|p| {
matches!(
p,
ParsedCommand::Read { .. }
| ParsedCommand::ListFiles { .. }
| ParsedCommand::Search { .. }
)
})
}
fn is_exploring_cell(&self) -> bool {
self.calls.iter().all(Self::is_exploring_call)
}
pub(crate) fn with_added_call(
&self,
call_id: String,
command: Vec<String>,
parsed: Vec<ParsedCommand>,
) -> Option<Self> {
let call = ExecCall {
call_id,
command,
parsed,
output: None,
start_time: Some(Instant::now()),
duration: None,
};
if self.is_exploring_cell() && Self::is_exploring_call(&call) {
Some(Self {
calls: [self.calls.clone(), vec![call]].concat(),
})
} else {
None
}
}
pub(crate) fn complete_call(
&mut self,
call_id: &str,
output: CommandOutput,
duration: Duration,
) {
if let Some(call) = self.calls.iter_mut().rev().find(|c| c.call_id == call_id) {
call.output = Some(output);
call.duration = Some(duration);
call.start_time = None;
}
}
pub(crate) fn should_flush(&self) -> bool {
!self.is_exploring_cell() && self.calls.iter().all(|c| c.output.is_some())
}
}
#[derive(Debug)]
struct CompletedMcpToolCallWithImageOutput {
_image: DynamicImage,
}
impl HistoryCell for CompletedMcpToolCallWithImageOutput {
fn display_lines(&self, _width: u16) -> Vec<Line<'static>> {
vec!["tool result (image output omitted)".into()]
}
}
const TOOL_CALL_MAX_LINES: usize = 5;
fn title_case(s: &str) -> String {
if s.is_empty() {
return String::new();
}
let mut chars = s.chars();
let first = match chars.next() {
Some(c) => c,
None => return String::new(),
};
let rest: String = chars.as_str().to_ascii_lowercase();
first.to_uppercase().collect::<String>() + &rest
}
fn pretty_provider_name(id: &str) -> String {
if id.eq_ignore_ascii_case("openai") {
"OpenAI".to_string()
} else {
title_case(id)
}
}
/// Return the emoji followed by a hair space (U+200A).
/// Using only the hair space avoids excessive padding after the emoji while
/// still providing a small visual gap across terminals.
fn padded_emoji(emoji: &str) -> String {
format!("{emoji}\u{200A}")
}
pub(crate) fn new_session_info(
config: &Config,
event: SessionConfiguredEvent,
is_first_event: bool,
) -> PlainHistoryCell {
let SessionConfiguredEvent {
model,
session_id: _,
history_log_id: _,
history_entry_count: _,
Replay EventMsgs from Response Items when resuming a session with history. (#3123) ### Overview This PR introduces the following changes: 1. Adds a unified mechanism to convert ResponseItem into EventMsg. 2. Ensures that when a session is initialized with initial history, a vector of EventMsg is sent along with the session configuration. This allows clients to re-render the UI accordingly. 3. Added integration testing ### Caveats This implementation does not send every EventMsg that was previously dispatched to clients. The excluded events fall into two categories: • “Arguably” rolled-out events Examples include tool calls and apply-patch calls. While these events are conceptually rolled out, we currently only roll out ResponseItems. These events are already being handled elsewhere and transformed into EventMsg before being sent. • Non-rolled-out events Certain events such as TurnDiff, Error, and TokenCount are not rolled out at all. ### Future Directions At present, resuming a session involves maintaining two states: • UI State Clients can replay most of the important UI from the provided EventMsg history. • Model State The model receives the complete session history to reconstruct its internal state. This design provides a solid foundation. If, in the future, more precise UI reconstruction is needed, we have two potential paths: 1. Introduce a third data structure that allows us to derive both ResponseItems and EventMsgs. 2. Clearly divide responsibilities: the core system ensures the integrity of the model state, while clients are responsible for reconstructing the UI.
2025-09-03 21:47:00 -07:00
initial_messages: _,
rollout_path: _,
} = event;
if is_first_event {
let cwd_str = match relativize_to_home(&config.cwd) {
Some(rel) if !rel.as_os_str().is_empty() => {
let sep = std::path::MAIN_SEPARATOR;
format!("~{sep}{}", rel.display())
}
Some(_) => "~".to_string(),
None => config.cwd.display().to_string(),
};
// Discover AGENTS.md files to decide whether to suggest `/init`.
let has_agents_md = discover_project_doc_paths(config)
.map(|v| !v.is_empty())
.unwrap_or(false);
let mut lines: Vec<Line<'static>> = Vec::new();
lines.push(Line::from(vec![
">_ ".dim(),
"You are using OpenAI Codex in".bold(),
format!(" {cwd_str}").dim(),
]));
lines.push(Line::from("".dim()));
lines.push(Line::from(
" To get started, describe a task or try one of these commands:".dim(),
));
lines.push(Line::from("".dim()));
if !has_agents_md {
lines.push(Line::from(vec![
" /init".bold(),
format!(" - {}", SlashCommand::Init.description()).dim(),
]));
}
lines.push(Line::from(vec![
" /status".bold(),
format!(" - {}", SlashCommand::Status.description()).dim(),
]));
lines.push(Line::from(vec![
" /approvals".bold(),
format!(" - {}", SlashCommand::Approvals.description()).dim(),
]));
lines.push(Line::from(vec![
" /model".bold(),
format!(" - {}", SlashCommand::Model.description()).dim(),
]));
PlainHistoryCell { lines }
} else if config.model == model {
PlainHistoryCell { lines: Vec::new() }
} else {
let lines = vec![
"model changed:".magenta().bold().into(),
format!("requested: {}", config.model).into(),
format!("used: {model}").into(),
];
PlainHistoryCell { lines }
}
}
pub(crate) fn new_user_prompt(message: String) -> UserHistoryCell {
UserHistoryCell { message }
}
pub(crate) fn new_user_approval_decision(lines: Vec<Line<'static>>) -> PlainHistoryCell {
PlainHistoryCell { lines }
}
feat: support the chat completions API in the Rust CLI (#862) This is a substantial PR to add support for the chat completions API, which in turn makes it possible to use non-OpenAI model providers (just like in the TypeScript CLI): * It moves a number of structs from `client.rs` to `client_common.rs` so they can be shared. * It introduces support for the chat completions API in `chat_completions.rs`. * It updates `ModelProviderInfo` so that `env_key` is `Option<String>` instead of `String` (for e.g., ollama) and adds a `wire_api` field * It updates `client.rs` to choose between `stream_responses()` and `stream_chat_completions()` based on the `wire_api` for the `ModelProviderInfo` * It updates the `exec` and TUI CLIs to no longer fail if the `OPENAI_API_KEY` environment variable is not set * It updates the TUI so that `EventMsg::Error` is displayed more prominently when it occurs, particularly now that it is important to alert users to the `CodexErr::EnvVar` variant. * `CodexErr::EnvVar` was updated to include an optional `instructions` field so we can preserve the behavior where we direct users to https://platform.openai.com if `OPENAI_API_KEY` is not set. * Cleaned up the "welcome message" in the TUI to ensure the model provider is displayed. * Updated the docs in `codex-rs/README.md`. To exercise the chat completions API from OpenAI models, I added the following to my `config.toml`: ```toml model = "gpt-4o" model_provider = "openai-chat-completions" [model_providers.openai-chat-completions] name = "OpenAI using Chat Completions" base_url = "https://api.openai.com/v1" env_key = "OPENAI_API_KEY" wire_api = "chat" ``` Though to test a non-OpenAI provider, I installed ollama with mistral locally on my Mac because ChatGPT said that would be a good match for my hardware: ```shell brew install ollama ollama serve ollama pull mistral ``` Then I added the following to my `~/.codex/config.toml`: ```toml model = "mistral" model_provider = "ollama" ``` Note this code could certainly use more test coverage, but I want to get this in so folks can start playing with it. For reference, I believe https://github.com/openai/codex/pull/247 was roughly the comparable PR on the TypeScript side.
2025-05-08 21:46:06 -07:00
pub(crate) fn new_active_exec_command(
call_id: String,
command: Vec<String>,
parsed: Vec<ParsedCommand>,
) -> ExecCell {
ExecCell::new(ExecCall {
call_id,
command,
parsed,
output: None,
start_time: Some(Instant::now()),
duration: None,
})
}
fn spinner(start_time: Option<Instant>) -> Span<'static> {
const FRAMES: &[char] = &['⠋', '⠙', '⠹', '⠸', '⠼', '⠴', '⠦', '⠧', '⠇', '⠏'];
let idx = start_time
.map(|st| ((st.elapsed().as_millis() / 100) as usize) % FRAMES.len())
.unwrap_or(0);
let ch = FRAMES[idx];
ch.to_string().into()
}
pub(crate) fn new_active_mcp_tool_call(invocation: McpInvocation) -> PlainHistoryCell {
let title_line = Line::from(vec!["tool".magenta(), " running...".dim()]);
let lines: Vec<Line> = vec![title_line, format_mcp_invocation(invocation)];
PlainHistoryCell { lines }
}
pub(crate) fn new_web_search_call(query: String) -> PlainHistoryCell {
let lines: Vec<Line<'static>> = vec![Line::from(vec![padded_emoji("🌐").into(), query.into()])];
PlainHistoryCell { lines }
}
/// If the first content is an image, return a new cell with the image.
/// TODO(rgwood-dd): Handle images properly even if they're not the first result.
fn try_new_completed_mcp_tool_call_with_image_output(
result: &Result<mcp_types::CallToolResult, String>,
) -> Option<CompletedMcpToolCallWithImageOutput> {
match result {
Ok(mcp_types::CallToolResult { content, .. }) => {
if let Some(mcp_types::ContentBlock::ImageContent(image)) = content.first() {
let raw_data = match base64::engine::general_purpose::STANDARD.decode(&image.data) {
Ok(data) => data,
Err(e) => {
error!("Failed to decode image data: {e}");
return None;
}
};
let reader = match ImageReader::new(Cursor::new(raw_data)).with_guessed_format() {
Ok(reader) => reader,
Err(e) => {
error!("Failed to guess image format: {e}");
return None;
}
};
let image = match reader.decode() {
Ok(image) => image,
Err(e) => {
error!("Image decoding failed: {e}");
return None;
}
};
Some(CompletedMcpToolCallWithImageOutput { _image: image })
} else {
None
}
}
_ => None,
}
}
pub(crate) fn new_completed_mcp_tool_call(
num_cols: usize,
invocation: McpInvocation,
duration: Duration,
success: bool,
result: Result<mcp_types::CallToolResult, String>,
) -> Box<dyn HistoryCell> {
if let Some(cell) = try_new_completed_mcp_tool_call_with_image_output(&result) {
return Box::new(cell);
}
let duration = format_duration(duration);
let status_str = if success { "success" } else { "failed" };
let title_line = Line::from(vec![
"tool".magenta(),
" ".into(),
if success {
status_str.green()
} else {
status_str.red()
},
format!(", duration: {duration}").dim(),
]);
let mut lines: Vec<Line<'static>> = Vec::new();
lines.push(title_line);
lines.push(format_mcp_invocation(invocation));
match result {
Ok(mcp_types::CallToolResult { content, .. }) => {
if !content.is_empty() {
lines.push(Line::from(""));
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
for tool_call_result in content {
let line_text = match tool_call_result {
mcp_types::ContentBlock::TextContent(text) => {
format_and_truncate_tool_result(
&text.text,
TOOL_CALL_MAX_LINES,
num_cols,
)
}
mcp_types::ContentBlock::ImageContent(_) => {
// TODO show images even if they're not the first result, will require a refactor of `CompletedMcpToolCall`
"<image content>".to_string()
}
mcp_types::ContentBlock::AudioContent(_) => "<audio content>".to_string(),
mcp_types::ContentBlock::EmbeddedResource(resource) => {
let uri = match resource.resource {
EmbeddedResourceResource::TextResourceContents(text) => text.uri,
EmbeddedResourceResource::BlobResourceContents(blob) => blob.uri,
};
format!("embedded resource: {uri}")
}
mcp_types::ContentBlock::ResourceLink(ResourceLink { uri, .. }) => {
format!("link: {uri}")
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
}
};
lines.push(Line::styled(
line_text,
Style::default().add_modifier(Modifier::DIM),
));
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
}
}
}
Err(e) => {
lines.push(vec!["Error: ".red().bold(), e.into()].into());
}
};
Box::new(PlainHistoryCell { lines })
}
pub(crate) fn new_status_output(
config: &Config,
usage: &TokenUsage,
session_id: &Option<ConversationId>,
) -> PlainHistoryCell {
let mut lines: Vec<Line<'static>> = Vec::new();
lines.push("/status".magenta().into());
let config_entries = create_config_summary_entries(config);
let lookup = |k: &str| -> String {
config_entries
.iter()
.find(|(key, _)| *key == k)
.map(|(_, v)| v.clone())
.unwrap_or_default()
};
// 📂 Workspace
lines.push(vec![padded_emoji("📂").into(), "Workspace".bold()].into());
// Path (home-relative, e.g., ~/code/project)
let cwd_str = match relativize_to_home(&config.cwd) {
Some(rel) if !rel.as_os_str().is_empty() => {
let sep = std::path::MAIN_SEPARATOR;
format!("~{sep}{}", rel.display())
}
Some(_) => "~".to_string(),
None => config.cwd.display().to_string(),
};
lines.push(vec![" • Path: ".into(), cwd_str.into()].into());
// Approval mode (as-is)
lines.push(vec![" • Approval Mode: ".into(), lookup("approval").into()].into());
// Sandbox (simplified name only)
let sandbox_name = match &config.sandbox_policy {
SandboxPolicy::DangerFullAccess => "danger-full-access",
SandboxPolicy::ReadOnly => "read-only",
SandboxPolicy::WorkspaceWrite { .. } => "workspace-write",
};
lines.push(vec![" • Sandbox: ".into(), sandbox_name.into()].into());
// AGENTS.md files discovered via core's project_doc logic
let agents_list = {
match discover_project_doc_paths(config) {
Ok(paths) => {
let mut rels: Vec<String> = Vec::new();
for p in paths {
let display = if let Some(parent) = p.parent() {
if parent == config.cwd {
"AGENTS.md".to_string()
} else {
let mut cur = config.cwd.as_path();
let mut ups = 0usize;
let mut reached = false;
while let Some(c) = cur.parent() {
if cur == parent {
reached = true;
break;
}
cur = c;
ups += 1;
}
if reached {
let up = format!("..{}", std::path::MAIN_SEPARATOR);
format!("{}AGENTS.md", up.repeat(ups))
} else if let Ok(stripped) = p.strip_prefix(&config.cwd) {
stripped.display().to_string()
} else {
p.display().to_string()
}
}
} else {
p.display().to_string()
};
rels.push(display);
}
rels
}
Err(_) => Vec::new(),
}
};
if agents_list.is_empty() {
lines.push(" • AGENTS files: (none)".into());
} else {
lines.push(vec![" • AGENTS files: ".into(), agents_list.join(", ").into()].into());
}
lines.push("".into());
// 👤 Account (only if ChatGPT tokens exist), shown under the first block
let auth_file = get_auth_file(&config.codex_home);
if let Ok(auth) = try_read_auth_json(&auth_file)
&& let Some(tokens) = auth.tokens.clone()
{
lines.push(vec![padded_emoji("👤").into(), "Account".bold()].into());
lines.push(" • Signed in with ChatGPT".into());
let info = tokens.id_token;
if let Some(email) = &info.email {
lines.push(vec![" • Login: ".into(), email.clone().into()].into());
}
match auth.openai_api_key.as_deref() {
Some(key) if !key.is_empty() => {
lines.push(" • Using API key. Run codex login to use ChatGPT plan".into());
}
_ => {
let plan_text = info
.get_chatgpt_plan_type()
.map(|s| title_case(&s))
.unwrap_or_else(|| "Unknown".to_string());
lines.push(vec![" • Plan: ".into(), plan_text.into()].into());
}
}
lines.push("".into());
}
// 🧠 Model
lines.push(vec![padded_emoji("🧠").into(), "Model".bold()].into());
lines.push(vec![" • Name: ".into(), config.model.clone().into()].into());
let provider_disp = pretty_provider_name(&config.model_provider_id);
lines.push(vec![" • Provider: ".into(), provider_disp.into()].into());
// Only show Reasoning fields if present in config summary
let reff = lookup("reasoning effort");
if !reff.is_empty() {
lines.push(vec![" • Reasoning Effort: ".into(), title_case(&reff).into()].into());
}
let rsum = lookup("reasoning summaries");
if !rsum.is_empty() {
lines.push(vec![" • Reasoning Summaries: ".into(), title_case(&rsum).into()].into());
}
lines.push("".into());
// 💻 Client
let cli_version = crate::version::CODEX_CLI_VERSION;
lines.push(vec![padded_emoji("💻").into(), "Client".bold()].into());
lines.push(vec![" • CLI Version: ".into(), cli_version.into()].into());
lines.push("".into());
// 📊 Token Usage
lines.push(vec!["📊 ".into(), "Token Usage".bold()].into());
if let Some(session_id) = session_id {
lines.push(vec![" • Session ID: ".into(), session_id.to_string().into()].into());
}
// Input: <input> [+ <cached> cached]
let mut input_line_spans: Vec<Span<'static>> = vec![
" • Input: ".into(),
format_with_separators(usage.non_cached_input()).into(),
];
if usage.cached_input_tokens > 0 {
let cached = usage.cached_input_tokens;
input_line_spans.push(format!(" (+ {cached} cached)").into());
feat: support the chat completions API in the Rust CLI (#862) This is a substantial PR to add support for the chat completions API, which in turn makes it possible to use non-OpenAI model providers (just like in the TypeScript CLI): * It moves a number of structs from `client.rs` to `client_common.rs` so they can be shared. * It introduces support for the chat completions API in `chat_completions.rs`. * It updates `ModelProviderInfo` so that `env_key` is `Option<String>` instead of `String` (for e.g., ollama) and adds a `wire_api` field * It updates `client.rs` to choose between `stream_responses()` and `stream_chat_completions()` based on the `wire_api` for the `ModelProviderInfo` * It updates the `exec` and TUI CLIs to no longer fail if the `OPENAI_API_KEY` environment variable is not set * It updates the TUI so that `EventMsg::Error` is displayed more prominently when it occurs, particularly now that it is important to alert users to the `CodexErr::EnvVar` variant. * `CodexErr::EnvVar` was updated to include an optional `instructions` field so we can preserve the behavior where we direct users to https://platform.openai.com if `OPENAI_API_KEY` is not set. * Cleaned up the "welcome message" in the TUI to ensure the model provider is displayed. * Updated the docs in `codex-rs/README.md`. To exercise the chat completions API from OpenAI models, I added the following to my `config.toml`: ```toml model = "gpt-4o" model_provider = "openai-chat-completions" [model_providers.openai-chat-completions] name = "OpenAI using Chat Completions" base_url = "https://api.openai.com/v1" env_key = "OPENAI_API_KEY" wire_api = "chat" ``` Though to test a non-OpenAI provider, I installed ollama with mistral locally on my Mac because ChatGPT said that would be a good match for my hardware: ```shell brew install ollama ollama serve ollama pull mistral ``` Then I added the following to my `~/.codex/config.toml`: ```toml model = "mistral" model_provider = "ollama" ``` Note this code could certainly use more test coverage, but I want to get this in so folks can start playing with it. For reference, I believe https://github.com/openai/codex/pull/247 was roughly the comparable PR on the TypeScript side.
2025-05-08 21:46:06 -07:00
}
lines.push(Line::from(input_line_spans));
// Output: <output>
lines.push(Line::from(vec![
" • Output: ".into(),
format_with_separators(usage.output_tokens).into(),
]));
// Total: <total>
lines.push(Line::from(vec![
" • Total: ".into(),
format_with_separators(usage.blended_total()).into(),
]));
PlainHistoryCell { lines }
}
/// Render a summary of configured MCP servers from the current `Config`.
pub(crate) fn empty_mcp_output() -> PlainHistoryCell {
let lines: Vec<Line<'static>> = vec![
"/mcp".magenta().into(),
"".into(),
vec!["🔌 ".into(), "MCP Tools".bold()].into(),
"".into(),
" • No MCP servers configured.".italic().into(),
Line::from(vec![
" See the ".into(),
"\u{1b}]8;;https://github.com/openai/codex/blob/main/docs/config.md#mcp_servers\u{7}MCP docs\u{1b}]8;;\u{7}".underlined(),
" to configure them.".into(),
])
.style(Style::default().add_modifier(Modifier::DIM)),
];
PlainHistoryCell { lines }
}
/// Render MCP tools grouped by connection using the fully-qualified tool names.
pub(crate) fn new_mcp_tools_output(
config: &Config,
tools: std::collections::HashMap<String, mcp_types::Tool>,
) -> PlainHistoryCell {
let mut lines: Vec<Line<'static>> = vec![
"/mcp".magenta().into(),
"".into(),
vec!["🔌 ".into(), "MCP Tools".bold()].into(),
"".into(),
];
if tools.is_empty() {
lines.push(" • No MCP tools available.".italic().into());
lines.push("".into());
return PlainHistoryCell { lines };
}
for (server, cfg) in config.mcp_servers.iter() {
let prefix = format!("{server}__");
let mut names: Vec<String> = tools
.keys()
.filter(|k| k.starts_with(&prefix))
.map(|k| k[prefix.len()..].to_string())
.collect();
names.sort();
lines.push(vec![" • Server: ".into(), server.clone().into()].into());
if !cfg.command.is_empty() {
let cmd_display = format!("{} {}", cfg.command, cfg.args.join(" "));
lines.push(vec![" • Command: ".into(), cmd_display.into()].into());
}
if names.is_empty() {
lines.push(" • Tools: (none)".into());
} else {
lines.push(vec![" • Tools: ".into(), names.join(", ").into()].into());
}
lines.push(Line::from(""));
}
PlainHistoryCell { lines }
}
pub(crate) fn new_error_event(message: String) -> PlainHistoryCell {
// Use a hair space (U+200A) to create a subtle, near-invisible separation
// before the text. VS16 is intentionally omitted to keep spacing tighter
// in terminals like Ghostty.
let lines: Vec<Line<'static>> =
vec![vec![padded_emoji("🖐").red().bold(), " ".into(), message.into()].into()];
PlainHistoryCell { lines }
}
2025-08-21 01:15:24 -07:00
pub(crate) fn new_stream_error_event(message: String) -> PlainHistoryCell {
let lines: Vec<Line<'static>> = vec![vec![padded_emoji("⚠️").into(), message.dim()].into()];
2025-08-21 01:15:24 -07:00
PlainHistoryCell { lines }
}
/// Render a userfriendly plan update styled like a checkbox todo list.
pub(crate) fn new_plan_update(update: UpdatePlanArgs) -> PlanUpdateCell {
let UpdatePlanArgs { explanation, plan } = update;
PlanUpdateCell { explanation, plan }
}
#[derive(Debug)]
pub(crate) struct PlanUpdateCell {
explanation: Option<String>,
plan: Vec<PlanItemArg>,
}
impl HistoryCell for PlanUpdateCell {
fn display_lines(&self, width: u16) -> Vec<Line<'static>> {
let render_note = |text: &str| -> Vec<Line<'static>> {
let wrap_width = width.saturating_sub(4).max(1) as usize;
textwrap::wrap(text, wrap_width)
.into_iter()
.map(|s| s.to_string().dim().italic().into())
.collect()
};
let render_step = |status: &StepStatus, text: &str| -> Vec<Line<'static>> {
let (box_str, step_style) = match status {
StepStatus::Completed => ("", Style::default().crossed_out().dim()),
StepStatus::InProgress => ("", Style::default().cyan().bold()),
StepStatus::Pending => ("", Style::default().dim()),
};
let wrap_width = (width as usize)
.saturating_sub(4)
.saturating_sub(box_str.width())
.max(1);
let parts = textwrap::wrap(text, wrap_width);
let step_text = parts
.into_iter()
.map(|s| s.to_string().set_style(step_style).into())
.collect();
tui: fix approval dialog for large commands (#3087) #### Summary - Emit a “Proposed Command” history cell when an ExecApprovalRequest arrives (parity with proposed patches). - Simplify the approval dialog: show only the reason/instructions; move the command preview into history. - Make approval/abort decision history concise: - Single line snippet; if multiline, show first line + " ...". - Truncate to 80 graphemes with ellipsis for very long commands. #### Details - History - Add `new_proposed_command` to render a header and indented command preview. - Use shared `prefix_lines` helper for first/subsequent line prefixes. - Approval UI - `UserApprovalWidget` no longer renders the command in the modal; shows optional `reason` text only. - Decision history renders an inline, dimmed snippet per rules above. - Tests (snapshot-based) - Proposed/decision flow for short command. - Proposed multi-line + aborted decision snippet with “ ...”. - Very long one-line command -> truncated snippet with “…”. - Updated existing exec approval snapshots and test reasons. <img width="1053" height="704" alt="Screenshot 2025-09-03 at 11 57 35 AM" src="https://github.com/user-attachments/assets/9ed4c316-9daf-4ac1-80ff-7ae1f481dd10" /> after approving: <img width="1053" height="704" alt="Screenshot 2025-09-03 at 11 58 18 AM" src="https://github.com/user-attachments/assets/a44e243f-eb9d-42ea-87f4-171b3fb481e7" /> rejection: <img width="1053" height="207" alt="Screenshot 2025-09-03 at 11 58 45 AM" src="https://github.com/user-attachments/assets/a022664b-ae0e-4b70-a388-509208707934" /> big command: https://github.com/user-attachments/assets/2dd976e5-799f-4af7-9682-a046e66cc494
2025-09-04 16:54:53 -07:00
prefix_lines(step_text, box_str.into(), " ".into())
};
let mut lines: Vec<Line<'static>> = vec![];
lines.push(vec!["".into(), "Updated Plan".bold()].into());
let mut indented_lines = vec![];
let note = self
.explanation
.as_ref()
.map(|s| s.trim())
.filter(|t| !t.is_empty());
if let Some(expl) = note {
indented_lines.extend(render_note(expl));
};
if self.plan.is_empty() {
indented_lines.push(Line::from("(no steps provided)".dim().italic()));
} else {
for PlanItemArg { step, status } in self.plan.iter() {
indented_lines.extend(render_step(status, step));
}
}
tui: fix approval dialog for large commands (#3087) #### Summary - Emit a “Proposed Command” history cell when an ExecApprovalRequest arrives (parity with proposed patches). - Simplify the approval dialog: show only the reason/instructions; move the command preview into history. - Make approval/abort decision history concise: - Single line snippet; if multiline, show first line + " ...". - Truncate to 80 graphemes with ellipsis for very long commands. #### Details - History - Add `new_proposed_command` to render a header and indented command preview. - Use shared `prefix_lines` helper for first/subsequent line prefixes. - Approval UI - `UserApprovalWidget` no longer renders the command in the modal; shows optional `reason` text only. - Decision history renders an inline, dimmed snippet per rules above. - Tests (snapshot-based) - Proposed/decision flow for short command. - Proposed multi-line + aborted decision snippet with “ ...”. - Very long one-line command -> truncated snippet with “…”. - Updated existing exec approval snapshots and test reasons. <img width="1053" height="704" alt="Screenshot 2025-09-03 at 11 57 35 AM" src="https://github.com/user-attachments/assets/9ed4c316-9daf-4ac1-80ff-7ae1f481dd10" /> after approving: <img width="1053" height="704" alt="Screenshot 2025-09-03 at 11 58 18 AM" src="https://github.com/user-attachments/assets/a44e243f-eb9d-42ea-87f4-171b3fb481e7" /> rejection: <img width="1053" height="207" alt="Screenshot 2025-09-03 at 11 58 45 AM" src="https://github.com/user-attachments/assets/a022664b-ae0e-4b70-a388-509208707934" /> big command: https://github.com/user-attachments/assets/2dd976e5-799f-4af7-9682-a046e66cc494
2025-09-04 16:54:53 -07:00
lines.extend(prefix_lines(indented_lines, "".into(), " ".into()));
lines
}
}
/// Create a new `PendingPatch` cell that lists the filelevel summary of
/// a proposed patch. The summary lines should already be formatted (e.g.
/// "A path/to/file.rs").
pub(crate) fn new_patch_event(
event_type: PatchEventType,
changes: HashMap<PathBuf, FileChange>,
cwd: &Path,
) -> PatchHistoryCell {
PatchHistoryCell {
event_type,
changes,
cwd: cwd.to_path_buf(),
}
}
pub(crate) fn new_patch_apply_failure(stderr: String) -> PlainHistoryCell {
let mut lines: Vec<Line<'static>> = Vec::new();
// Failure title
lines.push(Line::from("✘ Failed to apply patch".magenta().bold()));
if !stderr.trim().is_empty() {
lines.extend(output_lines(
Some(&CommandOutput {
exit_code: 1,
stdout: String::new(),
stderr,
formatted_output: String::new(),
}),
true,
true,
true,
));
}
PlainHistoryCell { lines }
}
tui: fix approval dialog for large commands (#3087) #### Summary - Emit a “Proposed Command” history cell when an ExecApprovalRequest arrives (parity with proposed patches). - Simplify the approval dialog: show only the reason/instructions; move the command preview into history. - Make approval/abort decision history concise: - Single line snippet; if multiline, show first line + " ...". - Truncate to 80 graphemes with ellipsis for very long commands. #### Details - History - Add `new_proposed_command` to render a header and indented command preview. - Use shared `prefix_lines` helper for first/subsequent line prefixes. - Approval UI - `UserApprovalWidget` no longer renders the command in the modal; shows optional `reason` text only. - Decision history renders an inline, dimmed snippet per rules above. - Tests (snapshot-based) - Proposed/decision flow for short command. - Proposed multi-line + aborted decision snippet with “ ...”. - Very long one-line command -> truncated snippet with “…”. - Updated existing exec approval snapshots and test reasons. <img width="1053" height="704" alt="Screenshot 2025-09-03 at 11 57 35 AM" src="https://github.com/user-attachments/assets/9ed4c316-9daf-4ac1-80ff-7ae1f481dd10" /> after approving: <img width="1053" height="704" alt="Screenshot 2025-09-03 at 11 58 18 AM" src="https://github.com/user-attachments/assets/a44e243f-eb9d-42ea-87f4-171b3fb481e7" /> rejection: <img width="1053" height="207" alt="Screenshot 2025-09-03 at 11 58 45 AM" src="https://github.com/user-attachments/assets/a022664b-ae0e-4b70-a388-509208707934" /> big command: https://github.com/user-attachments/assets/2dd976e5-799f-4af7-9682-a046e66cc494
2025-09-04 16:54:53 -07:00
/// Create a new history cell for a proposed command approval.
/// Renders a header and the command preview similar to how proposed patches
/// show a header and summary.
pub(crate) fn new_proposed_command(command: &[String]) -> PlainHistoryCell {
let cmd = strip_bash_lc_and_escape(command);
let mut lines: Vec<Line<'static>> = Vec::new();
lines.push(Line::from(vec!["".into(), "Proposed Command".bold()]));
let highlighted_lines = crate::render::highlight::highlight_bash_to_lines(&cmd);
tui: fix approval dialog for large commands (#3087) #### Summary - Emit a “Proposed Command” history cell when an ExecApprovalRequest arrives (parity with proposed patches). - Simplify the approval dialog: show only the reason/instructions; move the command preview into history. - Make approval/abort decision history concise: - Single line snippet; if multiline, show first line + " ...". - Truncate to 80 graphemes with ellipsis for very long commands. #### Details - History - Add `new_proposed_command` to render a header and indented command preview. - Use shared `prefix_lines` helper for first/subsequent line prefixes. - Approval UI - `UserApprovalWidget` no longer renders the command in the modal; shows optional `reason` text only. - Decision history renders an inline, dimmed snippet per rules above. - Tests (snapshot-based) - Proposed/decision flow for short command. - Proposed multi-line + aborted decision snippet with “ ...”. - Very long one-line command -> truncated snippet with “…”. - Updated existing exec approval snapshots and test reasons. <img width="1053" height="704" alt="Screenshot 2025-09-03 at 11 57 35 AM" src="https://github.com/user-attachments/assets/9ed4c316-9daf-4ac1-80ff-7ae1f481dd10" /> after approving: <img width="1053" height="704" alt="Screenshot 2025-09-03 at 11 58 18 AM" src="https://github.com/user-attachments/assets/a44e243f-eb9d-42ea-87f4-171b3fb481e7" /> rejection: <img width="1053" height="207" alt="Screenshot 2025-09-03 at 11 58 45 AM" src="https://github.com/user-attachments/assets/a022664b-ae0e-4b70-a388-509208707934" /> big command: https://github.com/user-attachments/assets/2dd976e5-799f-4af7-9682-a046e66cc494
2025-09-04 16:54:53 -07:00
let initial_prefix: Span<'static> = "".dim();
let subsequent_prefix: Span<'static> = " ".into();
lines.extend(prefix_lines(
highlighted_lines,
initial_prefix,
subsequent_prefix,
));
tui: fix approval dialog for large commands (#3087) #### Summary - Emit a “Proposed Command” history cell when an ExecApprovalRequest arrives (parity with proposed patches). - Simplify the approval dialog: show only the reason/instructions; move the command preview into history. - Make approval/abort decision history concise: - Single line snippet; if multiline, show first line + " ...". - Truncate to 80 graphemes with ellipsis for very long commands. #### Details - History - Add `new_proposed_command` to render a header and indented command preview. - Use shared `prefix_lines` helper for first/subsequent line prefixes. - Approval UI - `UserApprovalWidget` no longer renders the command in the modal; shows optional `reason` text only. - Decision history renders an inline, dimmed snippet per rules above. - Tests (snapshot-based) - Proposed/decision flow for short command. - Proposed multi-line + aborted decision snippet with “ ...”. - Very long one-line command -> truncated snippet with “…”. - Updated existing exec approval snapshots and test reasons. <img width="1053" height="704" alt="Screenshot 2025-09-03 at 11 57 35 AM" src="https://github.com/user-attachments/assets/9ed4c316-9daf-4ac1-80ff-7ae1f481dd10" /> after approving: <img width="1053" height="704" alt="Screenshot 2025-09-03 at 11 58 18 AM" src="https://github.com/user-attachments/assets/a44e243f-eb9d-42ea-87f4-171b3fb481e7" /> rejection: <img width="1053" height="207" alt="Screenshot 2025-09-03 at 11 58 45 AM" src="https://github.com/user-attachments/assets/a022664b-ae0e-4b70-a388-509208707934" /> big command: https://github.com/user-attachments/assets/2dd976e5-799f-4af7-9682-a046e66cc494
2025-09-04 16:54:53 -07:00
PlainHistoryCell { lines }
}
pub(crate) fn new_reasoning_block(
full_reasoning_buffer: String,
config: &Config,
) -> TranscriptOnlyHistoryCell {
let mut lines: Vec<Line<'static>> = Vec::new();
lines.push(Line::from("thinking".magenta().italic()));
append_markdown(&full_reasoning_buffer, &mut lines, config);
TranscriptOnlyHistoryCell { lines }
}
pub(crate) fn new_reasoning_summary_block(
full_reasoning_buffer: String,
config: &Config,
) -> Vec<Box<dyn HistoryCell>> {
if config.model_family.reasoning_summary_format == ReasoningSummaryFormat::Experimental {
// Experimental format is following:
// ** header **
//
// reasoning summary
//
// So we need to strip header from reasoning summary
let full_reasoning_buffer = full_reasoning_buffer.trim();
if let Some(open) = full_reasoning_buffer.find("**") {
let after_open = &full_reasoning_buffer[(open + 2)..];
if let Some(close) = after_open.find("**") {
let after_close_idx = open + 2 + close + 2;
// if we don't have anything beyond `after_close_idx`
// then we don't have a summary to inject into history
if after_close_idx < full_reasoning_buffer.len() {
let header_buffer = full_reasoning_buffer[..after_close_idx].to_string();
let summary_buffer = full_reasoning_buffer[after_close_idx..].to_string();
let mut header_lines: Vec<Line<'static>> = Vec::new();
header_lines.push(Line::from("Thinking".magenta().italic()));
append_markdown(&header_buffer, &mut header_lines, config);
let mut summary_lines: Vec<Line<'static>> = Vec::new();
summary_lines.push(Line::from("Thinking".magenta().bold()));
append_markdown(&summary_buffer, &mut summary_lines, config);
return vec![
Box::new(TranscriptOnlyHistoryCell {
lines: header_lines,
}),
Box::new(AgentMessageCell::new(summary_lines, true)),
];
}
}
}
}
vec![Box::new(new_reasoning_block(full_reasoning_buffer, config))]
}
[1/3] Parse exec commands and format them more nicely in the UI (#2095) # Note for reviewers The bulk of this PR is in in the new file, `parse_command.rs`. This file is designed to be written TDD and implemented with Codex. Do not worry about reviewing the code, just review the unit tests (if you want). If any cases are missing, we'll add more tests and have Codex fix them. I think the best approach will be to land and iterate. I have some follow-ups I want to do after this lands. The next PR after this will let us merge (and dedupe) multiple sequential cells of the same such as multiple read commands. The deduping will also be important because the model often reads the same file multiple times in a row in chunks === This PR formats common commands like reading, formatting, testing, etc more nicely: It tries to extract things like file names, tests and falls back to the cmd if it doesn't. It also only shows stdout/err if the command failed. <img width="770" height="238" alt="CleanShot 2025-08-09 at 16 05 15" src="https://github.com/user-attachments/assets/0ead179a-8910-486b-aa3d-7d26264d751e" /> <img width="348" height="158" alt="CleanShot 2025-08-09 at 16 05 32" src="https://github.com/user-attachments/assets/4302681b-5e87-4ff3-85b4-0252c6c485a9" /> <img width="834" height="324" alt="CleanShot 2025-08-09 at 16 05 56 2" src="https://github.com/user-attachments/assets/09fb3517-7bd6-40f6-a126-4172106b700f" /> Part 2: https://github.com/openai/codex/pull/2097 Part 3: https://github.com/openai/codex/pull/2110
2025-08-11 11:26:15 -07:00
fn output_lines(
output: Option<&CommandOutput>,
only_err: bool,
include_angle_pipe: bool,
include_prefix: bool,
[1/3] Parse exec commands and format them more nicely in the UI (#2095) # Note for reviewers The bulk of this PR is in in the new file, `parse_command.rs`. This file is designed to be written TDD and implemented with Codex. Do not worry about reviewing the code, just review the unit tests (if you want). If any cases are missing, we'll add more tests and have Codex fix them. I think the best approach will be to land and iterate. I have some follow-ups I want to do after this lands. The next PR after this will let us merge (and dedupe) multiple sequential cells of the same such as multiple read commands. The deduping will also be important because the model often reads the same file multiple times in a row in chunks === This PR formats common commands like reading, formatting, testing, etc more nicely: It tries to extract things like file names, tests and falls back to the cmd if it doesn't. It also only shows stdout/err if the command failed. <img width="770" height="238" alt="CleanShot 2025-08-09 at 16 05 15" src="https://github.com/user-attachments/assets/0ead179a-8910-486b-aa3d-7d26264d751e" /> <img width="348" height="158" alt="CleanShot 2025-08-09 at 16 05 32" src="https://github.com/user-attachments/assets/4302681b-5e87-4ff3-85b4-0252c6c485a9" /> <img width="834" height="324" alt="CleanShot 2025-08-09 at 16 05 56 2" src="https://github.com/user-attachments/assets/09fb3517-7bd6-40f6-a126-4172106b700f" /> Part 2: https://github.com/openai/codex/pull/2097 Part 3: https://github.com/openai/codex/pull/2110
2025-08-11 11:26:15 -07:00
) -> Vec<Line<'static>> {
let CommandOutput {
exit_code,
stdout,
stderr,
..
[1/3] Parse exec commands and format them more nicely in the UI (#2095) # Note for reviewers The bulk of this PR is in in the new file, `parse_command.rs`. This file is designed to be written TDD and implemented with Codex. Do not worry about reviewing the code, just review the unit tests (if you want). If any cases are missing, we'll add more tests and have Codex fix them. I think the best approach will be to land and iterate. I have some follow-ups I want to do after this lands. The next PR after this will let us merge (and dedupe) multiple sequential cells of the same such as multiple read commands. The deduping will also be important because the model often reads the same file multiple times in a row in chunks === This PR formats common commands like reading, formatting, testing, etc more nicely: It tries to extract things like file names, tests and falls back to the cmd if it doesn't. It also only shows stdout/err if the command failed. <img width="770" height="238" alt="CleanShot 2025-08-09 at 16 05 15" src="https://github.com/user-attachments/assets/0ead179a-8910-486b-aa3d-7d26264d751e" /> <img width="348" height="158" alt="CleanShot 2025-08-09 at 16 05 32" src="https://github.com/user-attachments/assets/4302681b-5e87-4ff3-85b4-0252c6c485a9" /> <img width="834" height="324" alt="CleanShot 2025-08-09 at 16 05 56 2" src="https://github.com/user-attachments/assets/09fb3517-7bd6-40f6-a126-4172106b700f" /> Part 2: https://github.com/openai/codex/pull/2097 Part 3: https://github.com/openai/codex/pull/2110
2025-08-11 11:26:15 -07:00
} = match output {
Some(output) if only_err && output.exit_code == 0 => return vec![],
Some(output) => output,
None => return vec![],
};
let src = if *exit_code == 0 { stdout } else { stderr };
let lines: Vec<&str> = src.lines().collect();
let total = lines.len();
let limit = TOOL_CALL_MAX_LINES;
let mut out = Vec::new();
let head_end = total.min(limit);
for (i, raw) in lines[..head_end].iter().enumerate() {
let mut line = ansi_escape_line(raw);
let prefix = if !include_prefix {
""
} else if i == 0 && include_angle_pipe {
""
[1/3] Parse exec commands and format them more nicely in the UI (#2095) # Note for reviewers The bulk of this PR is in in the new file, `parse_command.rs`. This file is designed to be written TDD and implemented with Codex. Do not worry about reviewing the code, just review the unit tests (if you want). If any cases are missing, we'll add more tests and have Codex fix them. I think the best approach will be to land and iterate. I have some follow-ups I want to do after this lands. The next PR after this will let us merge (and dedupe) multiple sequential cells of the same such as multiple read commands. The deduping will also be important because the model often reads the same file multiple times in a row in chunks === This PR formats common commands like reading, formatting, testing, etc more nicely: It tries to extract things like file names, tests and falls back to the cmd if it doesn't. It also only shows stdout/err if the command failed. <img width="770" height="238" alt="CleanShot 2025-08-09 at 16 05 15" src="https://github.com/user-attachments/assets/0ead179a-8910-486b-aa3d-7d26264d751e" /> <img width="348" height="158" alt="CleanShot 2025-08-09 at 16 05 32" src="https://github.com/user-attachments/assets/4302681b-5e87-4ff3-85b4-0252c6c485a9" /> <img width="834" height="324" alt="CleanShot 2025-08-09 at 16 05 56 2" src="https://github.com/user-attachments/assets/09fb3517-7bd6-40f6-a126-4172106b700f" /> Part 2: https://github.com/openai/codex/pull/2097 Part 3: https://github.com/openai/codex/pull/2110
2025-08-11 11:26:15 -07:00
} else {
" "
};
line.spans.insert(0, prefix.into());
line.spans.iter_mut().for_each(|span| {
span.style = span.style.add_modifier(Modifier::DIM);
});
out.push(line);
}
// If we will ellipsize less than the limit, just show it.
let show_ellipsis = total > 2 * limit;
if show_ellipsis {
let omitted = total - 2 * limit;
out.push(format!("… +{omitted} lines").into());
[1/3] Parse exec commands and format them more nicely in the UI (#2095) # Note for reviewers The bulk of this PR is in in the new file, `parse_command.rs`. This file is designed to be written TDD and implemented with Codex. Do not worry about reviewing the code, just review the unit tests (if you want). If any cases are missing, we'll add more tests and have Codex fix them. I think the best approach will be to land and iterate. I have some follow-ups I want to do after this lands. The next PR after this will let us merge (and dedupe) multiple sequential cells of the same such as multiple read commands. The deduping will also be important because the model often reads the same file multiple times in a row in chunks === This PR formats common commands like reading, formatting, testing, etc more nicely: It tries to extract things like file names, tests and falls back to the cmd if it doesn't. It also only shows stdout/err if the command failed. <img width="770" height="238" alt="CleanShot 2025-08-09 at 16 05 15" src="https://github.com/user-attachments/assets/0ead179a-8910-486b-aa3d-7d26264d751e" /> <img width="348" height="158" alt="CleanShot 2025-08-09 at 16 05 32" src="https://github.com/user-attachments/assets/4302681b-5e87-4ff3-85b4-0252c6c485a9" /> <img width="834" height="324" alt="CleanShot 2025-08-09 at 16 05 56 2" src="https://github.com/user-attachments/assets/09fb3517-7bd6-40f6-a126-4172106b700f" /> Part 2: https://github.com/openai/codex/pull/2097 Part 3: https://github.com/openai/codex/pull/2110
2025-08-11 11:26:15 -07:00
}
let tail_start = if show_ellipsis {
total - limit
} else {
head_end
};
for raw in lines[tail_start..].iter() {
let mut line = ansi_escape_line(raw);
if include_prefix {
line.spans.insert(0, " ".into());
}
[1/3] Parse exec commands and format them more nicely in the UI (#2095) # Note for reviewers The bulk of this PR is in in the new file, `parse_command.rs`. This file is designed to be written TDD and implemented with Codex. Do not worry about reviewing the code, just review the unit tests (if you want). If any cases are missing, we'll add more tests and have Codex fix them. I think the best approach will be to land and iterate. I have some follow-ups I want to do after this lands. The next PR after this will let us merge (and dedupe) multiple sequential cells of the same such as multiple read commands. The deduping will also be important because the model often reads the same file multiple times in a row in chunks === This PR formats common commands like reading, formatting, testing, etc more nicely: It tries to extract things like file names, tests and falls back to the cmd if it doesn't. It also only shows stdout/err if the command failed. <img width="770" height="238" alt="CleanShot 2025-08-09 at 16 05 15" src="https://github.com/user-attachments/assets/0ead179a-8910-486b-aa3d-7d26264d751e" /> <img width="348" height="158" alt="CleanShot 2025-08-09 at 16 05 32" src="https://github.com/user-attachments/assets/4302681b-5e87-4ff3-85b4-0252c6c485a9" /> <img width="834" height="324" alt="CleanShot 2025-08-09 at 16 05 56 2" src="https://github.com/user-attachments/assets/09fb3517-7bd6-40f6-a126-4172106b700f" /> Part 2: https://github.com/openai/codex/pull/2097 Part 3: https://github.com/openai/codex/pull/2110
2025-08-11 11:26:15 -07:00
line.spans.iter_mut().for_each(|span| {
span.style = span.style.add_modifier(Modifier::DIM);
});
out.push(line);
}
out
}
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
fn format_mcp_invocation<'a>(invocation: McpInvocation) -> Line<'a> {
let args_str = invocation
.arguments
.as_ref()
.map(|v| {
// Use compact form to keep things short but readable.
serde_json::to_string(v).unwrap_or_else(|_| v.to_string())
})
.unwrap_or_default();
let invocation_spans = vec![
invocation.server.clone().cyan(),
".".into(),
invocation.tool.cyan(),
"(".into(),
args_str.dim(),
")".into(),
];
invocation_spans.into()
fix: introduce ResponseInputItem::McpToolCallOutput variant (#1151) The output of an MCP server tool call can be one of several types, but to date, we treated all outputs as text by showing the serialized JSON as the "tool output" in Codex: https://github.com/openai/codex/blob/25a9949c49194d5a64de54a11bcc5b4724ac9bd5/codex-rs/mcp-types/src/lib.rs#L96-L101 This PR adds support for the `ImageContent` variant so we can now display an image output from an MCP tool call. In making this change, we introduce a new `ResponseInputItem::McpToolCallOutput` variant so that we can work with the `mcp_types::CallToolResult` directly when the function call is made to an MCP server. Though arguably the more significant change is the introduction of `HistoryCell::CompletedMcpToolCallWithImageOutput`, which is a cell that uses `ratatui_image` to render an image into the terminal. To support this, we introduce `ImageRenderCache`, cache a `ratatui_image::picker::Picker`, and `ensure_image_cache()` to cache the appropriate scaled image data and dimensions based on the current terminal size. To test, I created a minimal `package.json`: ```json { "name": "kitty-mcp", "version": "1.0.0", "type": "module", "description": "MCP that returns image of kitty", "main": "index.js", "dependencies": { "@modelcontextprotocol/sdk": "^1.12.0" } } ``` with the following `index.js` to define the MCP server: ```js #!/usr/bin/env node import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { readFile } from "node:fs/promises"; import { join } from "node:path"; const IMAGE_URI = "image://Ada.png"; const server = new McpServer({ name: "Demo", version: "1.0.0", }); server.tool( "get-cat-image", "If you need a cat image, this tool will provide one.", async () => ({ content: [ { type: "image", data: await getAdaPngBase64(), mimeType: "image/png" }, ], }) ); server.resource("Ada the Cat", IMAGE_URI, async (uri) => { const base64Image = await getAdaPngBase64(); return { contents: [ { uri: uri.href, mimeType: "image/png", blob: base64Image, }, ], }; }); async function getAdaPngBase64() { const __dirname = new URL(".", import.meta.url).pathname; // From https://github.com/benjajaja/ratatui-image/blob/9705ce2c59ec669abbce2924cbfd1f5ae22c9860/assets/Ada.png const filePath = join(__dirname, "Ada.png"); const imageData = await readFile(filePath); const base64Image = imageData.toString("base64"); return base64Image; } const transport = new StdioServerTransport(); await server.connect(transport); ``` With the local changes from this PR, I added the following to my `config.toml`: ```toml [mcp_servers.kitty] command = "node" args = ["/Users/mbolin/code/kitty-mcp/index.js"] ``` Running the TUI from source: ``` cargo run --bin codex -- --model o3 'I need a picture of a cat' ``` I get: <img width="732" alt="image" src="https://github.com/user-attachments/assets/bf80b721-9ca0-4d81-aec7-77d6899e2869" /> Now, that said, I have only tested in iTerm and there is definitely some funny business with getting an accurate character-to-pixel ratio (sometimes the `CompletedMcpToolCallWithImageOutput` thinks it needs 10 rows to render instead of 4), so there is still work to be done here.
2025-05-28 19:03:17 -07:00
}
[1/3] Parse exec commands and format them more nicely in the UI (#2095) # Note for reviewers The bulk of this PR is in in the new file, `parse_command.rs`. This file is designed to be written TDD and implemented with Codex. Do not worry about reviewing the code, just review the unit tests (if you want). If any cases are missing, we'll add more tests and have Codex fix them. I think the best approach will be to land and iterate. I have some follow-ups I want to do after this lands. The next PR after this will let us merge (and dedupe) multiple sequential cells of the same such as multiple read commands. The deduping will also be important because the model often reads the same file multiple times in a row in chunks === This PR formats common commands like reading, formatting, testing, etc more nicely: It tries to extract things like file names, tests and falls back to the cmd if it doesn't. It also only shows stdout/err if the command failed. <img width="770" height="238" alt="CleanShot 2025-08-09 at 16 05 15" src="https://github.com/user-attachments/assets/0ead179a-8910-486b-aa3d-7d26264d751e" /> <img width="348" height="158" alt="CleanShot 2025-08-09 at 16 05 32" src="https://github.com/user-attachments/assets/4302681b-5e87-4ff3-85b4-0252c6c485a9" /> <img width="834" height="324" alt="CleanShot 2025-08-09 at 16 05 56 2" src="https://github.com/user-attachments/assets/09fb3517-7bd6-40f6-a126-4172106b700f" /> Part 2: https://github.com/openai/codex/pull/2097 Part 3: https://github.com/openai/codex/pull/2110
2025-08-11 11:26:15 -07:00
#[cfg(test)]
mod tests {
use super::*;
use codex_core::config::Config;
use codex_core::config::ConfigOverrides;
use codex_core::config::ConfigToml;
fn test_config() -> Config {
Config::load_from_base_config_with_overrides(
ConfigToml::default(),
ConfigOverrides::default(),
std::env::temp_dir(),
)
.expect("config")
}
fn render_lines(lines: &[Line<'static>]) -> Vec<String> {
lines
.iter()
.map(|line| {
line.spans
.iter()
.map(|span| span.content.as_ref())
.collect::<String>()
})
.collect()
}
fn render_transcript(cell: &dyn HistoryCell) -> Vec<String> {
render_lines(&cell.transcript_lines())
}
#[test]
fn coalesces_sequential_reads_within_one_call() {
// Build one exec cell with a Search followed by two Reads
let call_id = "c1".to_string();
let mut cell = ExecCell::new(ExecCall {
call_id: call_id.clone(),
command: vec!["bash".into(), "-lc".into(), "echo".into()],
parsed: vec![
ParsedCommand::Search {
query: Some("shimmer_spans".into()),
path: None,
cmd: "rg shimmer_spans".into(),
},
ParsedCommand::Read {
name: "shimmer.rs".into(),
cmd: "cat shimmer.rs".into(),
},
ParsedCommand::Read {
name: "status_indicator_widget.rs".into(),
cmd: "cat status_indicator_widget.rs".into(),
},
],
output: None,
start_time: Some(Instant::now()),
duration: None,
});
// Mark call complete so markers are ✓
cell.complete_call(
&call_id,
CommandOutput {
exit_code: 0,
stdout: String::new(),
stderr: String::new(),
formatted_output: String::new(),
},
Duration::from_millis(1),
);
let lines = cell.display_lines(80);
let rendered = render_lines(&lines).join("\n");
insta::assert_snapshot!(rendered);
}
#[test]
fn coalesces_reads_across_multiple_calls() {
let mut cell = ExecCell::new(ExecCall {
call_id: "c1".to_string(),
command: vec!["bash".into(), "-lc".into(), "echo".into()],
parsed: vec![ParsedCommand::Search {
query: Some("shimmer_spans".into()),
path: None,
cmd: "rg shimmer_spans".into(),
}],
output: None,
start_time: Some(Instant::now()),
duration: None,
});
// Call 1: Search only
cell.complete_call(
"c1",
CommandOutput {
exit_code: 0,
stdout: String::new(),
stderr: String::new(),
formatted_output: String::new(),
},
Duration::from_millis(1),
);
// Call 2: Read A
cell = cell
.with_added_call(
"c2".into(),
vec!["bash".into(), "-lc".into(), "echo".into()],
vec![ParsedCommand::Read {
name: "shimmer.rs".into(),
cmd: "cat shimmer.rs".into(),
}],
)
.unwrap();
cell.complete_call(
"c2",
CommandOutput {
exit_code: 0,
stdout: String::new(),
stderr: String::new(),
formatted_output: String::new(),
},
Duration::from_millis(1),
);
// Call 3: Read B
cell = cell
.with_added_call(
"c3".into(),
vec!["bash".into(), "-lc".into(), "echo".into()],
vec![ParsedCommand::Read {
name: "status_indicator_widget.rs".into(),
cmd: "cat status_indicator_widget.rs".into(),
}],
)
.unwrap();
cell.complete_call(
"c3",
CommandOutput {
exit_code: 0,
stdout: String::new(),
stderr: String::new(),
formatted_output: String::new(),
},
Duration::from_millis(1),
);
let lines = cell.display_lines(80);
let rendered = render_lines(&lines).join("\n");
insta::assert_snapshot!(rendered);
}
#[test]
fn coalesced_reads_dedupe_names() {
let mut cell = ExecCell::new(ExecCall {
call_id: "c1".to_string(),
command: vec!["bash".into(), "-lc".into(), "echo".into()],
parsed: vec![
ParsedCommand::Read {
name: "auth.rs".into(),
cmd: "cat auth.rs".into(),
},
ParsedCommand::Read {
name: "auth.rs".into(),
cmd: "cat auth.rs".into(),
},
ParsedCommand::Read {
name: "shimmer.rs".into(),
cmd: "cat shimmer.rs".into(),
},
],
output: None,
start_time: Some(Instant::now()),
duration: None,
});
cell.complete_call(
"c1",
CommandOutput {
exit_code: 0,
stdout: String::new(),
stderr: String::new(),
formatted_output: String::new(),
},
Duration::from_millis(1),
);
let lines = cell.display_lines(80);
let rendered = render_lines(&lines).join("\n");
insta::assert_snapshot!(rendered);
}
#[test]
fn multiline_command_wraps_with_extra_indent_on_subsequent_lines() {
// Create a completed exec cell with a multiline command
let cmd = "set -o pipefail\ncargo test --all-features --quiet".to_string();
let call_id = "c1".to_string();
let mut cell = ExecCell::new(ExecCall {
call_id: call_id.clone(),
command: vec!["bash".into(), "-lc".into(), cmd],
parsed: Vec::new(),
output: None,
start_time: Some(Instant::now()),
duration: None,
});
// Mark call complete so it renders as "Ran"
cell.complete_call(
&call_id,
CommandOutput {
exit_code: 0,
stdout: String::new(),
stderr: String::new(),
formatted_output: String::new(),
},
Duration::from_millis(1),
);
// Small width to force wrapping on both lines
let width: u16 = 28;
let lines = cell.display_lines(width);
let rendered = render_lines(&lines).join("\n");
insta::assert_snapshot!(rendered);
}
#[test]
fn single_line_command_compact_when_fits() {
let call_id = "c1".to_string();
let mut cell = ExecCell::new(ExecCall {
call_id: call_id.clone(),
command: vec!["echo".into(), "ok".into()],
parsed: Vec::new(),
output: None,
start_time: Some(Instant::now()),
duration: None,
});
cell.complete_call(
&call_id,
CommandOutput {
exit_code: 0,
stdout: String::new(),
stderr: String::new(),
formatted_output: String::new(),
},
Duration::from_millis(1),
);
// Wide enough that it fits inline
let lines = cell.display_lines(80);
let rendered = render_lines(&lines).join("\n");
insta::assert_snapshot!(rendered);
}
#[test]
fn single_line_command_wraps_with_four_space_continuation() {
let call_id = "c1".to_string();
let long = "a_very_long_token_without_spaces_to_force_wrapping".to_string();
let mut cell = ExecCell::new(ExecCall {
call_id: call_id.clone(),
command: vec!["bash".into(), "-lc".into(), long],
parsed: Vec::new(),
output: None,
start_time: Some(Instant::now()),
duration: None,
});
cell.complete_call(
&call_id,
CommandOutput {
exit_code: 0,
stdout: String::new(),
stderr: String::new(),
formatted_output: String::new(),
},
Duration::from_millis(1),
);
let lines = cell.display_lines(24);
let rendered = render_lines(&lines).join("\n");
insta::assert_snapshot!(rendered);
}
#[test]
fn multiline_command_without_wrap_uses_branch_then_eight_spaces() {
let call_id = "c1".to_string();
let cmd = "echo one\necho two".to_string();
let mut cell = ExecCell::new(ExecCall {
call_id: call_id.clone(),
command: vec!["bash".into(), "-lc".into(), cmd],
parsed: Vec::new(),
output: None,
start_time: Some(Instant::now()),
duration: None,
});
cell.complete_call(
&call_id,
CommandOutput {
exit_code: 0,
stdout: String::new(),
stderr: String::new(),
formatted_output: String::new(),
},
Duration::from_millis(1),
);
let lines = cell.display_lines(80);
let rendered = render_lines(&lines).join("\n");
insta::assert_snapshot!(rendered);
}
#[test]
fn multiline_command_both_lines_wrap_with_correct_prefixes() {
let call_id = "c1".to_string();
let cmd = "first_token_is_long_enough_to_wrap\nsecond_token_is_also_long_enough_to_wrap"
.to_string();
let mut cell = ExecCell::new(ExecCall {
call_id: call_id.clone(),
command: vec!["bash".into(), "-lc".into(), cmd],
parsed: Vec::new(),
output: None,
start_time: Some(Instant::now()),
duration: None,
});
cell.complete_call(
&call_id,
CommandOutput {
exit_code: 0,
stdout: String::new(),
stderr: String::new(),
formatted_output: String::new(),
},
Duration::from_millis(1),
);
let lines = cell.display_lines(28);
let rendered = render_lines(&lines).join("\n");
insta::assert_snapshot!(rendered);
}
#[test]
fn stderr_tail_more_than_five_lines_snapshot() {
// Build an exec cell with a non-zero exit and 10 lines on stderr to exercise
// the head/tail rendering and gutter prefixes.
let call_id = "c_err".to_string();
let mut cell = ExecCell::new(ExecCall {
call_id: call_id.clone(),
command: vec!["bash".into(), "-lc".into(), "seq 1 10 1>&2 && false".into()],
parsed: Vec::new(),
output: None,
start_time: Some(Instant::now()),
duration: None,
});
let stderr: String = (1..=10)
.map(|n| n.to_string())
.collect::<Vec<_>>()
.join("\n");
cell.complete_call(
&call_id,
CommandOutput {
exit_code: 1,
stdout: String::new(),
stderr,
formatted_output: String::new(),
},
Duration::from_millis(1),
);
let rendered = cell
.display_lines(80)
.iter()
.map(|l| {
l.spans
.iter()
.map(|s| s.content.as_ref())
.collect::<String>()
})
.collect::<Vec<_>>()
.join("\n");
insta::assert_snapshot!(rendered);
}
#[test]
fn ran_cell_multiline_with_stderr_snapshot() {
// Build an exec cell that completes (so it renders as "Ran") with a
// command long enough that it must render on its own line under the
// header, and include a couple of stderr lines to verify the output
// block prefixes and wrapping.
let call_id = "c_wrap_err".to_string();
let long_cmd =
"echo this_is_a_very_long_single_token_that_will_wrap_across_the_available_width";
let mut cell = ExecCell::new(ExecCall {
call_id: call_id.clone(),
command: vec!["bash".into(), "-lc".into(), long_cmd.to_string()],
parsed: Vec::new(),
output: None,
start_time: Some(Instant::now()),
duration: None,
});
let stderr = "error: first line on stderr\nerror: second line on stderr".to_string();
cell.complete_call(
&call_id,
CommandOutput {
exit_code: 1,
stdout: String::new(),
stderr,
formatted_output: String::new(),
},
Duration::from_millis(5),
);
// Narrow width to force the command to render under the header line.
let width: u16 = 28;
let rendered = cell
.display_lines(width)
.iter()
.map(|l| {
l.spans
.iter()
.map(|s| s.content.as_ref())
.collect::<String>()
})
.collect::<Vec<_>>()
.join("\n");
insta::assert_snapshot!(rendered);
}
#[test]
fn user_history_cell_wraps_and_prefixes_each_line_snapshot() {
let msg = "one two three four five six seven";
let cell = UserHistoryCell {
message: msg.to_string(),
};
// Small width to force wrapping more clearly. Effective wrap width is width-1 due to the ▌ prefix.
let width: u16 = 12;
let lines = cell.display_lines(width);
let rendered = render_lines(&lines).join("\n");
insta::assert_snapshot!(rendered);
}
#[test]
fn plan_update_with_note_and_wrapping_snapshot() {
// Long explanation forces wrapping; include long step text to verify step wrapping and alignment.
let update = UpdatePlanArgs {
explanation: Some(
"Ill update Grafana call error handling by adding retries and clearer messages when the backend is unreachable."
.to_string(),
),
plan: vec![
PlanItemArg {
step: "Investigate existing error paths and logging around HTTP timeouts".into(),
status: StepStatus::Completed,
},
PlanItemArg {
step: "Harden Grafana client error handling with retry/backoff and userfriendly messages".into(),
status: StepStatus::InProgress,
},
PlanItemArg {
step: "Add tests for transient failure scenarios and surfacing to the UI".into(),
status: StepStatus::Pending,
},
],
};
let cell = new_plan_update(update);
// Narrow width to force wrapping for both the note and steps
let lines = cell.display_lines(32);
let rendered = render_lines(&lines).join("\n");
insta::assert_snapshot!(rendered);
}
#[test]
fn plan_update_without_note_snapshot() {
let update = UpdatePlanArgs {
explanation: None,
plan: vec![
PlanItemArg {
step: "Define error taxonomy".into(),
status: StepStatus::InProgress,
},
PlanItemArg {
step: "Implement mapping to user messages".into(),
status: StepStatus::Pending,
},
],
};
let cell = new_plan_update(update);
let lines = cell.display_lines(40);
let rendered = render_lines(&lines).join("\n");
insta::assert_snapshot!(rendered);
}
#[test]
fn reasoning_summary_block_returns_reasoning_cell_when_feature_disabled() {
let mut config = test_config();
config.model_family.reasoning_summary_format = ReasoningSummaryFormat::Experimental;
let cells =
new_reasoning_summary_block("Detailed reasoning goes here.".to_string(), &config);
assert_eq!(cells.len(), 1);
let rendered = render_transcript(cells[0].as_ref());
assert_eq!(rendered, vec!["thinking", "Detailed reasoning goes here."]);
}
#[test]
fn reasoning_summary_block_falls_back_when_header_is_missing() {
let mut config = test_config();
config.model_family.reasoning_summary_format = ReasoningSummaryFormat::Experimental;
let cells = new_reasoning_summary_block(
"**High level reasoning without closing".to_string(),
&config,
);
assert_eq!(cells.len(), 1);
let rendered = render_transcript(cells[0].as_ref());
assert_eq!(
rendered,
vec!["thinking", "**High level reasoning without closing"]
);
}
#[test]
fn reasoning_summary_block_falls_back_when_summary_is_missing() {
let mut config = test_config();
config.model_family.reasoning_summary_format = ReasoningSummaryFormat::Experimental;
let cells = new_reasoning_summary_block(
"**High level reasoning without closing**".to_string(),
&config,
);
assert_eq!(cells.len(), 1);
let rendered = render_transcript(cells[0].as_ref());
assert_eq!(
rendered,
vec!["thinking", "High level reasoning without closing"]
);
let cells = new_reasoning_summary_block(
"**High level reasoning without closing**\n\n ".to_string(),
&config,
);
assert_eq!(cells.len(), 1);
let rendered = render_transcript(cells[0].as_ref());
assert_eq!(
rendered,
vec!["thinking", "High level reasoning without closing"]
);
}
#[test]
fn reasoning_summary_block_splits_header_and_summary_when_present() {
let mut config = test_config();
config.model_family.reasoning_summary_format = ReasoningSummaryFormat::Experimental;
let cells = new_reasoning_summary_block(
"**High level plan**\n\nWe should fix the bug next.".to_string(),
&config,
);
assert_eq!(cells.len(), 2);
let header_lines = render_transcript(cells[0].as_ref());
assert_eq!(header_lines, vec!["Thinking", "High level plan"]);
let summary_lines = render_transcript(cells[1].as_ref());
assert_eq!(
summary_lines,
vec!["codex", "Thinking", "We should fix the bug next."]
)
}
}