Dark-Alex-17
57c1a86195
Introduce a WireApi enum (responses|chat) with an optional wire_api config field on the openai and openai-compatible clients, resolved through a single resolve_wire_api fn: codex forces the responses wire (an explicit wire_api: chat there is a hard config error), an explicit wire_api wins everywhere else, and the default stays chat for now. The openai dispatch generalizes from (RequestData, bool) to (RequestData, WireApi), and the openai-compatible client gains opt-in Responses dispatch (URL, body, extractor, and streaming forks). Both patch layers are now wire-aware: RequestData carries its wire, RequestPatch gains a responses section, and patch_request_data selects patches per wire. This deliberately FIXES a codex bug: chat-shaped model patches (e.g. o3-high's max_tokens/temperature/top_p/ reasoning_effort tweaks) previously merged wire-blind into codex responses bodies, poisoning them with chat-only keys. On the responses wire only an explicit responses sub-patch applies; chat-wire patching is byte-identical to before (pinned by a regression test).
1236 lines
44 KiB
Rust
1236 lines
44 KiB
Rust
use std::collections::HashSet;
|
|
use std::mem;
|
|
|
|
use super::*;
|
|
|
|
use crate::utils::{base64_decode, encode_uri, hex_encode, hmac_sha256, sha256, strip_think_tag};
|
|
|
|
use anyhow::{Context, Result, bail};
|
|
use aws_smithy_eventstream::frame::{DecodedFrame, MessageFrameDecoder};
|
|
use aws_smithy_eventstream::smithy::parse_response_headers;
|
|
use bytes::BytesMut;
|
|
use chrono::{DateTime, Utc};
|
|
use futures_util::StreamExt;
|
|
use indexmap::IndexMap;
|
|
use reqwest::{Client as ReqwestClient, Method, RequestBuilder};
|
|
use serde::Deserialize;
|
|
use serde_json::{Value, json};
|
|
|
|
#[derive(Debug, Clone, Deserialize)]
|
|
pub struct BedrockConfig {
|
|
pub name: Option<String>,
|
|
pub access_key_id: Option<String>,
|
|
pub secret_access_key: Option<String>,
|
|
pub region: Option<String>,
|
|
pub session_token: Option<String>,
|
|
#[serde(default)]
|
|
pub models: Vec<ModelData>,
|
|
pub prompt_cache: Option<bool>,
|
|
pub patch: Option<RequestPatch>,
|
|
pub extra: Option<ExtraConfig>,
|
|
}
|
|
|
|
impl BedrockClient {
|
|
config_get_fn!(access_key_id, get_access_key_id);
|
|
config_get_fn!(secret_access_key, get_secret_access_key);
|
|
config_get_fn!(region, get_region);
|
|
config_get_fn!(session_token, get_session_token);
|
|
|
|
create_client_config!([
|
|
("access_key_id", "AWS Access Key ID", None, true),
|
|
("secret_access_key", "AWS Secret Access Key", None, true),
|
|
("region", "AWS Region", None, false),
|
|
]);
|
|
|
|
fn chat_completions_builder(
|
|
&self,
|
|
client: &ReqwestClient,
|
|
data: ChatCompletionsData,
|
|
) -> Result<RequestBuilder> {
|
|
let access_key_id = self.get_access_key_id()?;
|
|
let secret_access_key = self.get_secret_access_key()?;
|
|
let region = self.get_region()?;
|
|
let session_token = self.get_session_token().ok();
|
|
let host = format!("bedrock-runtime.{region}.amazonaws.com");
|
|
|
|
let model_name = &self.model.real_name();
|
|
|
|
let uri = if data.stream {
|
|
format!("/model/{model_name}/converse-stream")
|
|
} else {
|
|
format!("/model/{model_name}/converse")
|
|
};
|
|
|
|
let body = build_chat_completions_body(
|
|
data,
|
|
&self.model,
|
|
self.config.prompt_cache.unwrap_or(true),
|
|
)?;
|
|
|
|
let mut request_data = RequestData::new("", body);
|
|
self.patch_request_data(&mut request_data);
|
|
let RequestData {
|
|
url: _,
|
|
headers,
|
|
body,
|
|
..
|
|
} = request_data;
|
|
|
|
let builder = aws_fetch(
|
|
client,
|
|
&AwsCredentials {
|
|
access_key_id,
|
|
secret_access_key,
|
|
region,
|
|
session_token,
|
|
},
|
|
AwsRequest {
|
|
method: Method::POST,
|
|
host,
|
|
service: "bedrock".into(),
|
|
uri,
|
|
querystring: "".into(),
|
|
headers,
|
|
body: body.to_string(),
|
|
},
|
|
)?;
|
|
|
|
Ok(builder)
|
|
}
|
|
|
|
fn embeddings_builder(
|
|
&self,
|
|
client: &ReqwestClient,
|
|
data: &EmbeddingsData,
|
|
) -> Result<RequestBuilder> {
|
|
let access_key_id = self.get_access_key_id()?;
|
|
let secret_access_key = self.get_secret_access_key()?;
|
|
let region = self.get_region()?;
|
|
let session_token = self.get_session_token().ok();
|
|
let host = format!("bedrock-runtime.{region}.amazonaws.com");
|
|
|
|
let uri = format!("/model/{}/invoke", self.model.real_name());
|
|
|
|
let input_type = match data.query {
|
|
true => "search_query",
|
|
false => "search_document",
|
|
};
|
|
|
|
let body = json!({
|
|
"texts": data.texts,
|
|
"input_type": input_type,
|
|
});
|
|
|
|
let mut request_data = RequestData::new("", body);
|
|
self.patch_request_data(&mut request_data);
|
|
let RequestData {
|
|
url: _,
|
|
headers,
|
|
body,
|
|
..
|
|
} = request_data;
|
|
|
|
let builder = aws_fetch(
|
|
client,
|
|
&AwsCredentials {
|
|
access_key_id,
|
|
secret_access_key,
|
|
region,
|
|
session_token,
|
|
},
|
|
AwsRequest {
|
|
method: Method::POST,
|
|
host,
|
|
service: "bedrock".into(),
|
|
uri,
|
|
querystring: "".into(),
|
|
headers,
|
|
body: body.to_string(),
|
|
},
|
|
)?;
|
|
|
|
Ok(builder)
|
|
}
|
|
}
|
|
|
|
#[async_trait::async_trait]
|
|
impl Client for BedrockClient {
|
|
client_common_fns!();
|
|
|
|
async fn chat_completions_inner(
|
|
&self,
|
|
client: &ReqwestClient,
|
|
data: ChatCompletionsData,
|
|
) -> Result<ChatCompletionsOutput> {
|
|
let builder = self.chat_completions_builder(client, data)?;
|
|
chat_completions(builder).await
|
|
}
|
|
|
|
async fn chat_completions_streaming_inner(
|
|
&self,
|
|
client: &ReqwestClient,
|
|
handler: &mut SseHandler,
|
|
data: ChatCompletionsData,
|
|
) -> Result<()> {
|
|
let builder = self.chat_completions_builder(client, data)?;
|
|
chat_completions_streaming(builder, handler).await
|
|
}
|
|
|
|
async fn embeddings_inner(
|
|
&self,
|
|
client: &ReqwestClient,
|
|
data: &EmbeddingsData,
|
|
) -> Result<EmbeddingsOutput> {
|
|
let builder = self.embeddings_builder(client, data)?;
|
|
embeddings(builder).await
|
|
}
|
|
}
|
|
|
|
async fn chat_completions(builder: RequestBuilder) -> Result<ChatCompletionsOutput> {
|
|
let res = builder.send().await?;
|
|
let status = res.status();
|
|
let data: Value = res.json().await?;
|
|
|
|
if !status.is_success() {
|
|
catch_error(&data, status.as_u16())?;
|
|
}
|
|
|
|
debug!("non-stream-data: {data}");
|
|
extract_chat_completions(&data)
|
|
}
|
|
|
|
async fn chat_completions_streaming(
|
|
builder: RequestBuilder,
|
|
handler: &mut SseHandler,
|
|
) -> Result<()> {
|
|
let res = builder.send().await?;
|
|
let status = res.status();
|
|
if !status.is_success() {
|
|
let data: Value = res.json().await?;
|
|
catch_error(&data, status.as_u16())?;
|
|
bail!("Invalid response data: {data}");
|
|
}
|
|
|
|
let mut function_name = String::new();
|
|
let mut function_arguments = String::new();
|
|
let mut function_id = String::new();
|
|
let mut reasoning_state = 0;
|
|
let mut thinking_text = String::new();
|
|
let mut thinking_signature = String::new();
|
|
|
|
let mut stream = res.bytes_stream();
|
|
let mut buffer = BytesMut::new();
|
|
let mut decoder = MessageFrameDecoder::new();
|
|
while let Some(chunk) = stream.next().await {
|
|
let chunk = chunk?;
|
|
buffer.extend_from_slice(&chunk);
|
|
while let DecodedFrame::Complete(message) = decoder.decode_frame(&mut buffer)? {
|
|
let response_headers = parse_response_headers(&message)?;
|
|
let message_type = response_headers.message_type.as_str();
|
|
let smithy_type = response_headers.smithy_type.as_str();
|
|
match (message_type, smithy_type) {
|
|
("event", _) => {
|
|
let data: Value = serde_json::from_slice(message.payload())?;
|
|
debug!("stream-data: {smithy_type} {data}");
|
|
match smithy_type {
|
|
"contentBlockStart" => {
|
|
if let Some(tool_use) = data["start"]["toolUse"].as_object()
|
|
&& let (Some(id), Some(name)) = (
|
|
json_str_from_map(tool_use, "toolUseId"),
|
|
json_str_from_map(tool_use, "name"),
|
|
)
|
|
{
|
|
if !function_name.is_empty() {
|
|
if function_arguments.is_empty() {
|
|
function_arguments = String::from("{}");
|
|
}
|
|
let arguments: Value =
|
|
function_arguments.parse().with_context(|| {
|
|
format!("Tool call '{function_name}' has non-JSON arguments '{function_arguments}'")
|
|
})?;
|
|
handler.tool_call(ToolCall::new(
|
|
function_name.clone(),
|
|
arguments,
|
|
Some(function_id.clone()),
|
|
))?;
|
|
}
|
|
function_arguments.clear();
|
|
function_name = name.into();
|
|
function_id = id.into();
|
|
}
|
|
}
|
|
"contentBlockDelta" => {
|
|
if let Some(text) = data["delta"]["text"].as_str() {
|
|
handler.text(text)?;
|
|
} else if let Some(text) =
|
|
data["delta"]["reasoningContent"]["text"].as_str()
|
|
{
|
|
reasoning_state = 1;
|
|
thinking_text.push_str(text);
|
|
} else if let Some(signature) =
|
|
data["delta"]["reasoningContent"]["signature"].as_str()
|
|
{
|
|
thinking_signature.push_str(signature);
|
|
} else if let Some(redacted_data) =
|
|
data["delta"]["reasoningContent"]["redactedContent"].as_str()
|
|
{
|
|
handler.thinking_block(ThinkingBlock::RedactedThinking {
|
|
data: redacted_data.to_string(),
|
|
});
|
|
} else if let Some(input) = data["delta"]["toolUse"]["input"].as_str() {
|
|
function_arguments.push_str(input);
|
|
}
|
|
}
|
|
"contentBlockStop" => {
|
|
if reasoning_state == 1 {
|
|
reasoning_state = 0;
|
|
handler.thinking_block(ThinkingBlock::Thinking {
|
|
thinking: mem::take(&mut thinking_text),
|
|
signature: mem::take(&mut thinking_signature),
|
|
});
|
|
}
|
|
if !function_name.is_empty() {
|
|
if function_arguments.is_empty() {
|
|
function_arguments = String::from("{}");
|
|
}
|
|
let arguments: Value = function_arguments.parse().with_context(|| {
|
|
format!("Tool call '{function_name}' has non-JSON arguments '{function_arguments}'")
|
|
})?;
|
|
handler.tool_call(ToolCall::new(
|
|
mem::take(&mut function_name),
|
|
arguments,
|
|
Some(mem::take(&mut function_id)),
|
|
))?;
|
|
function_arguments.clear();
|
|
}
|
|
}
|
|
"metadata" => {
|
|
if let Some(usage) = bedrock_parse_usage(&data["usage"]) {
|
|
debug!("token-usage: {usage:?}");
|
|
handler.usage(usage);
|
|
}
|
|
}
|
|
_ => {}
|
|
}
|
|
}
|
|
("exception", _) => {
|
|
let payload = base64_decode(message.payload())?;
|
|
let data = String::from_utf8_lossy(&payload);
|
|
|
|
bail!("Invalid response data: {data} (smithy_type: {smithy_type})")
|
|
}
|
|
_ => {
|
|
bail!(
|
|
"Unrecognized message, message_type: {message_type}, smithy_type: {smithy_type}",
|
|
);
|
|
}
|
|
}
|
|
}
|
|
}
|
|
Ok(())
|
|
}
|
|
|
|
async fn embeddings(builder: RequestBuilder) -> Result<EmbeddingsOutput> {
|
|
let res = builder.send().await?;
|
|
let status = res.status();
|
|
let data: Value = res.json().await?;
|
|
|
|
if !status.is_success() {
|
|
catch_error(&data, status.as_u16())?;
|
|
}
|
|
|
|
let res_body: EmbeddingsResBody =
|
|
serde_json::from_value(data).context("Invalid embeddings data")?;
|
|
Ok(res_body.embeddings)
|
|
}
|
|
|
|
#[derive(Deserialize)]
|
|
struct EmbeddingsResBody {
|
|
embeddings: Vec<Vec<f32>>,
|
|
}
|
|
|
|
fn build_chat_completions_body(
|
|
data: ChatCompletionsData,
|
|
model: &Model,
|
|
prompt_cache: bool,
|
|
) -> Result<Value> {
|
|
let ChatCompletionsData {
|
|
mut messages,
|
|
temperature,
|
|
top_p,
|
|
reasoning_effort,
|
|
functions,
|
|
stream: _,
|
|
} = data;
|
|
|
|
let system_message = extract_system_message(&mut messages);
|
|
|
|
let mut network_image_urls = vec![];
|
|
|
|
let messages_len = messages.len();
|
|
let messages: Vec<Value> = messages
|
|
.into_iter()
|
|
.enumerate()
|
|
.flat_map(|(i, message)| {
|
|
let Message { role, content } = message;
|
|
match content {
|
|
MessageContent::Text(text) if role.is_assistant() && i != messages_len - 1 => {
|
|
vec![json!({ "role": role, "content": [ { "text": strip_think_tag(&text) } ] })]
|
|
}
|
|
MessageContent::Text(text) => vec![json!({
|
|
"role": role,
|
|
"content": [
|
|
{
|
|
"text": text,
|
|
}
|
|
],
|
|
})],
|
|
MessageContent::Array(list) => {
|
|
let content: Vec<_> = list
|
|
.into_iter()
|
|
.map(|item| match item {
|
|
MessageContentPart::Text { text } => {
|
|
json!({"text": text})
|
|
}
|
|
MessageContentPart::ImageUrl {
|
|
image_url: ImageUrl { url },
|
|
} => {
|
|
if let Some((mime_type, data)) = url
|
|
.strip_prefix("data:")
|
|
.and_then(|v| v.split_once(";base64,"))
|
|
{
|
|
json!({
|
|
"image": {
|
|
"format": mime_type.replace("image/", ""),
|
|
"source": {
|
|
"bytes": data,
|
|
}
|
|
}
|
|
})
|
|
} else {
|
|
network_image_urls.push(url.clone());
|
|
json!({ "url": url })
|
|
}
|
|
}
|
|
})
|
|
.collect();
|
|
vec![json!({
|
|
"role": role,
|
|
"content": content,
|
|
})]
|
|
}
|
|
MessageContent::ToolCalls(MessageContentToolCalls {
|
|
tool_results,
|
|
text,
|
|
sequence,
|
|
}) => {
|
|
if !sequence {
|
|
let mut assistant_parts = vec![];
|
|
let mut user_parts = vec![];
|
|
for (index, tool_result) in tool_results.iter().enumerate() {
|
|
for block in &tool_result.thinking {
|
|
assistant_parts.push(converse_reasoning_block(block));
|
|
}
|
|
let round_text = if index == 0 && !text.is_empty() {
|
|
Some(text.as_str())
|
|
} else {
|
|
tool_result.text.as_deref()
|
|
};
|
|
if let Some(round_text) = round_text {
|
|
let round_text = strip_think_tag(round_text);
|
|
let round_text = round_text.trim();
|
|
if !round_text.is_empty() {
|
|
assistant_parts.push(json!({
|
|
"text": round_text,
|
|
}))
|
|
}
|
|
}
|
|
assistant_parts.push(json!({
|
|
"toolUse": {
|
|
"toolUseId": tool_result.call.id,
|
|
"name": tool_result.call.name,
|
|
"input": tool_result.call.arguments,
|
|
}
|
|
}));
|
|
user_parts.push(json!({
|
|
"toolResult": {
|
|
"toolUseId": tool_result.call.id,
|
|
"content": [bedrock_tool_result_content(&tool_result.output)]
|
|
}
|
|
}));
|
|
}
|
|
// Empty tool_results (reachable via deserialized sessions) must not emit an empty pair.
|
|
if assistant_parts.is_empty() {
|
|
vec![]
|
|
} else {
|
|
vec![
|
|
json!({ "role": "assistant", "content": assistant_parts }),
|
|
json!({ "role": "user", "content": user_parts }),
|
|
]
|
|
}
|
|
} else {
|
|
// One pair per round: Claude can reuse tool_use IDs across API calls.
|
|
// A round boundary is detected by the presence of round text, but
|
|
// rounds where the model emitted only tool calls (no narration)
|
|
// carry no text marker. As a backstop, also split whenever a
|
|
// toolUseId would repeat within the current assistant message.
|
|
// Converse rejects duplicate toolUseIds in a single message.
|
|
let mut messages = vec![];
|
|
let mut assistant_parts: Vec<Value> = vec![];
|
|
let mut user_parts: Vec<Value> = vec![];
|
|
let mut chunk_ids: HashSet<&str> = HashSet::new();
|
|
for (index, tool_result) in tool_results.iter().enumerate() {
|
|
let id_collision = tool_result
|
|
.call
|
|
.id
|
|
.as_deref()
|
|
.is_some_and(|id| chunk_ids.contains(id));
|
|
if index > 0 && (tool_result.text.is_some() || id_collision) {
|
|
messages.push(
|
|
json!({ "role": "assistant", "content": assistant_parts }),
|
|
);
|
|
messages.push(json!({ "role": "user", "content": user_parts }));
|
|
assistant_parts = vec![];
|
|
user_parts = vec![];
|
|
chunk_ids.clear();
|
|
}
|
|
if let Some(id) = tool_result.call.id.as_deref() {
|
|
chunk_ids.insert(id);
|
|
}
|
|
for block in &tool_result.thinking {
|
|
assistant_parts.push(converse_reasoning_block(block));
|
|
}
|
|
let round_text = if index == 0 && !text.is_empty() {
|
|
Some(text.as_str())
|
|
} else {
|
|
tool_result.text.as_deref()
|
|
};
|
|
if let Some(round_text) = round_text {
|
|
let round_text = strip_think_tag(round_text);
|
|
let round_text = round_text.trim();
|
|
if !round_text.is_empty() {
|
|
assistant_parts.push(json!({
|
|
"text": round_text,
|
|
}))
|
|
}
|
|
}
|
|
assistant_parts.push(json!({
|
|
"toolUse": {
|
|
"toolUseId": tool_result.call.id,
|
|
"name": tool_result.call.name,
|
|
"input": tool_result.call.arguments,
|
|
}
|
|
}));
|
|
user_parts.push(json!({
|
|
"toolResult": {
|
|
"toolUseId": tool_result.call.id,
|
|
"content": [bedrock_tool_result_content(&tool_result.output)]
|
|
}
|
|
}));
|
|
}
|
|
if !assistant_parts.is_empty() {
|
|
messages
|
|
.push(json!({ "role": "assistant", "content": assistant_parts }));
|
|
messages.push(json!({ "role": "user", "content": user_parts }));
|
|
}
|
|
messages
|
|
}
|
|
}
|
|
}
|
|
})
|
|
.collect();
|
|
|
|
if !network_image_urls.is_empty() {
|
|
bail!(
|
|
"The model does not support network images: {:?}",
|
|
network_image_urls
|
|
);
|
|
}
|
|
|
|
let mut body = json!({
|
|
"inferenceConfig": {},
|
|
"messages": messages,
|
|
});
|
|
if let Some(v) = system_message {
|
|
body["system"] = json!([
|
|
{
|
|
"text": v,
|
|
}
|
|
])
|
|
}
|
|
|
|
if let Some(v) = model.max_tokens_param() {
|
|
body["inferenceConfig"]["maxTokens"] = v.into();
|
|
}
|
|
if let Some(v) = temperature {
|
|
body["inferenceConfig"]["temperature"] = v.into();
|
|
}
|
|
if let Some(v) = top_p {
|
|
body["inferenceConfig"]["topP"] = v.into();
|
|
}
|
|
if let Some(v) = reasoning_effort {
|
|
body["additionalModelRequestFields"] = json!({ "output_config": { "effort": v } });
|
|
}
|
|
if let Some(functions) = functions {
|
|
let tools: Vec<_> = functions
|
|
.iter()
|
|
.map(|v| {
|
|
json!({
|
|
"toolSpec": {
|
|
"name": v.name,
|
|
"description": v.description,
|
|
"inputSchema": {
|
|
"json": v.parameters,
|
|
},
|
|
}
|
|
})
|
|
})
|
|
.collect();
|
|
body["toolConfig"] = json!({
|
|
"tools": tools,
|
|
})
|
|
}
|
|
if prompt_cache && is_anthropic_model(model) {
|
|
apply_prompt_cache_points(&mut body);
|
|
}
|
|
Ok(body)
|
|
}
|
|
|
|
/// Converse rejects cachePoint blocks for models that don't support them
|
|
/// (e.g. openai.gpt-*), so caching is only applied to Anthropic models.
|
|
fn is_anthropic_model(model: &Model) -> bool {
|
|
let name = model.real_name();
|
|
name.contains("anthropic.") || name.contains("claude")
|
|
}
|
|
|
|
/// Converse counterpart of the claude client's cache_control breakpoints:
|
|
/// a cachePoint after the tools, the system prompt, and the last two user
|
|
/// messages. See `apply_prompt_cache_breakpoints` in claude.rs for why the
|
|
/// last two.
|
|
fn apply_prompt_cache_points(body: &mut Value) {
|
|
let cache_point = || json!({ "cachePoint": { "type": "default" } });
|
|
if let Some(system) = body.get_mut("system").and_then(|v| v.as_array_mut()) {
|
|
system.push(cache_point());
|
|
}
|
|
if let Some(tools) = body
|
|
.get_mut("toolConfig")
|
|
.and_then(|v| v.get_mut("tools"))
|
|
.and_then(|v| v.as_array_mut())
|
|
{
|
|
tools.push(cache_point());
|
|
}
|
|
let Some(messages) = body.get_mut("messages").and_then(|v| v.as_array_mut()) else {
|
|
return;
|
|
};
|
|
let mut remaining = 2;
|
|
for message in messages.iter_mut().rev() {
|
|
if remaining == 0 {
|
|
break;
|
|
}
|
|
if message["role"] != "user" {
|
|
continue;
|
|
}
|
|
// Converse rejects a cachePoint with no preceding block, so an empty
|
|
// content array (a tool-call pair with no results) passes the
|
|
// breakpoint to an earlier user message.
|
|
if let Some(content) = message["content"].as_array_mut()
|
|
&& !content.is_empty()
|
|
{
|
|
content.push(cache_point());
|
|
remaining -= 1;
|
|
}
|
|
}
|
|
}
|
|
|
|
fn bedrock_parse_usage(usage: &Value) -> Option<TokenUsage> {
|
|
if !usage.is_object() {
|
|
return None;
|
|
}
|
|
Some(TokenUsage {
|
|
input_tokens: usage["inputTokens"].as_u64(),
|
|
output_tokens: usage["outputTokens"].as_u64(),
|
|
cache_creation_input_tokens: usage["cacheWriteInputTokens"].as_u64(),
|
|
cache_read_input_tokens: usage["cacheReadInputTokens"].as_u64(),
|
|
})
|
|
}
|
|
|
|
fn bedrock_tool_result_content(output: &Value) -> Value {
|
|
match output {
|
|
Value::Object(_) => json!({ "json": output }),
|
|
Value::String(text) => json!({ "text": text }),
|
|
_ => json!({ "text": output.to_string() }),
|
|
}
|
|
}
|
|
|
|
/// Maps a ThinkingBlock (Anthropic tagged shape) to the Converse
|
|
/// reasoningContent union so signed blocks replay verbatim.
|
|
fn converse_reasoning_block(block: &ThinkingBlock) -> Value {
|
|
match block {
|
|
ThinkingBlock::Thinking {
|
|
thinking,
|
|
signature,
|
|
} => json!({
|
|
"reasoningContent": {
|
|
"reasoningText": {
|
|
"text": thinking,
|
|
"signature": signature,
|
|
}
|
|
}
|
|
}),
|
|
ThinkingBlock::RedactedThinking { data } => json!({
|
|
"reasoningContent": { "redactedContent": data }
|
|
}),
|
|
}
|
|
}
|
|
|
|
fn extract_chat_completions(data: &Value) -> Result<ChatCompletionsOutput> {
|
|
let mut text = String::new();
|
|
let mut tool_calls = vec![];
|
|
let mut thinking = vec![];
|
|
if let Some(array) = data["output"]["message"]["content"].as_array() {
|
|
for item in array {
|
|
if let Some(v) = item["text"].as_str() {
|
|
if !text.is_empty() {
|
|
text.push_str("\n\n");
|
|
}
|
|
text.push_str(v);
|
|
} else if let Some(reasoning_text) =
|
|
item["reasoningContent"]["reasoningText"].as_object()
|
|
{
|
|
if let Some(v) = json_str_from_map(reasoning_text, "text") {
|
|
thinking.push(ThinkingBlock::Thinking {
|
|
thinking: v.to_string(),
|
|
signature: json_str_from_map(reasoning_text, "signature")
|
|
.unwrap_or_default()
|
|
.to_string(),
|
|
});
|
|
}
|
|
} else if let Some(v) = item["reasoningContent"]["redactedContent"].as_str() {
|
|
thinking.push(ThinkingBlock::RedactedThinking {
|
|
data: v.to_string(),
|
|
});
|
|
} else if let Some(tool_use) = item["toolUse"].as_object()
|
|
&& let (Some(id), Some(name), Some(input)) = (
|
|
json_str_from_map(tool_use, "toolUseId"),
|
|
json_str_from_map(tool_use, "name"),
|
|
tool_use.get("input"),
|
|
)
|
|
{
|
|
tool_calls.push(ToolCall::new(
|
|
name.to_string(),
|
|
input.clone(),
|
|
Some(id.to_string()),
|
|
))
|
|
}
|
|
}
|
|
}
|
|
|
|
if text.is_empty() && tool_calls.is_empty() {
|
|
bail!("Invalid response data: {data}");
|
|
}
|
|
|
|
let output = ChatCompletionsOutput {
|
|
text,
|
|
tool_calls,
|
|
thinking,
|
|
usage: bedrock_parse_usage(&data["usage"]),
|
|
};
|
|
Ok(output)
|
|
}
|
|
|
|
#[derive(Debug)]
|
|
struct AwsCredentials {
|
|
access_key_id: String,
|
|
secret_access_key: String,
|
|
region: String,
|
|
session_token: Option<String>,
|
|
}
|
|
|
|
#[derive(Debug)]
|
|
struct AwsRequest {
|
|
method: Method,
|
|
host: String,
|
|
service: String,
|
|
uri: String,
|
|
querystring: String,
|
|
headers: IndexMap<String, String>,
|
|
body: String,
|
|
}
|
|
|
|
fn aws_fetch(
|
|
client: &ReqwestClient,
|
|
credentials: &AwsCredentials,
|
|
request: AwsRequest,
|
|
) -> Result<RequestBuilder> {
|
|
let AwsRequest {
|
|
method,
|
|
host,
|
|
service,
|
|
uri,
|
|
querystring,
|
|
mut headers,
|
|
body,
|
|
} = request;
|
|
let region = &credentials.region;
|
|
|
|
let endpoint = format!("https://{host}{uri}");
|
|
|
|
let now: DateTime<Utc> = Utc::now();
|
|
let amz_date = now.format("%Y%m%dT%H%M%SZ").to_string();
|
|
let date_stamp = amz_date[0..8].to_string();
|
|
headers.insert("host".into(), host.clone());
|
|
headers.insert("x-amz-date".into(), amz_date.clone());
|
|
if let Some(token) = credentials.session_token.clone() {
|
|
headers.insert("x-amz-security-token".into(), token);
|
|
}
|
|
|
|
let canonical_headers = headers
|
|
.iter()
|
|
.map(|(key, value)| format!("{key}:{value}\n"))
|
|
.collect::<Vec<_>>()
|
|
.join("");
|
|
|
|
let signed_headers = headers
|
|
.iter()
|
|
.map(|(key, _)| key.as_str())
|
|
.collect::<Vec<_>>()
|
|
.join(";");
|
|
|
|
let payload_hash = sha256(&body);
|
|
|
|
let canonical_request = format!(
|
|
"{}\n{}\n{}\n{}\n{}\n{}",
|
|
method,
|
|
encode_uri(&uri),
|
|
querystring,
|
|
canonical_headers,
|
|
signed_headers,
|
|
payload_hash
|
|
);
|
|
|
|
let algorithm = "AWS4-HMAC-SHA256";
|
|
let credential_scope = format!("{date_stamp}/{region}/{service}/aws4_request");
|
|
let string_to_sign = format!(
|
|
"{}\n{}\n{}\n{}",
|
|
algorithm,
|
|
amz_date,
|
|
credential_scope,
|
|
sha256(&canonical_request)
|
|
);
|
|
|
|
let signing_key = gen_signing_key(
|
|
&credentials.secret_access_key,
|
|
&date_stamp,
|
|
region,
|
|
&service,
|
|
);
|
|
let signature = hmac_sha256(&signing_key, &string_to_sign);
|
|
let signature = hex_encode(&signature);
|
|
|
|
let authorization_header = format!(
|
|
"{} Credential={}/{}, SignedHeaders={}, Signature={}",
|
|
algorithm, credentials.access_key_id, credential_scope, signed_headers, signature
|
|
);
|
|
|
|
headers.insert("authorization".into(), authorization_header);
|
|
|
|
debug!("Request {endpoint} {body}");
|
|
|
|
let mut request_builder = client.request(method, endpoint).body(body);
|
|
|
|
for (key, value) in &headers {
|
|
request_builder = request_builder.header(key, value);
|
|
}
|
|
|
|
Ok(request_builder)
|
|
}
|
|
|
|
fn gen_signing_key(key: &str, date_stamp: &str, region: &str, service: &str) -> Vec<u8> {
|
|
let k_date = hmac_sha256(format!("AWS4{key}").as_bytes(), date_stamp);
|
|
let k_region = hmac_sha256(&k_date, region);
|
|
let k_service = hmac_sha256(&k_region, service);
|
|
hmac_sha256(&k_service, "aws4_request")
|
|
}
|
|
|
|
#[cfg(test)]
|
|
mod tests {
|
|
use super::*;
|
|
use crate::function::{FunctionDeclaration, ToolCall, ToolResult};
|
|
|
|
fn tool_result(output: Value) -> ToolResult {
|
|
ToolResult {
|
|
call: ToolCall::new(
|
|
"fs_read".into(),
|
|
json!({"path": "x"}),
|
|
Some("tool_A".into()),
|
|
),
|
|
output,
|
|
text: None,
|
|
thinking: vec![],
|
|
}
|
|
}
|
|
|
|
fn sequence_tool_result(id: &str, text: Option<&str>) -> ToolResult {
|
|
ToolResult {
|
|
call: ToolCall::new("fs_read".into(), json!({"path": "x"}), Some(id.into())),
|
|
output: json!("ok"),
|
|
text: text.map(|t| t.to_string()),
|
|
thinking: vec![],
|
|
}
|
|
}
|
|
|
|
fn build_body_with(tool_results: Vec<ToolResult>, text: &str, sequence: bool) -> Value {
|
|
let data = ChatCompletionsData {
|
|
messages: vec![
|
|
Message::new(MessageRole::User, MessageContent::Text("hello".to_string())),
|
|
Message::new(
|
|
MessageRole::Assistant,
|
|
MessageContent::ToolCalls(MessageContentToolCalls {
|
|
tool_results,
|
|
text: text.to_string(),
|
|
sequence,
|
|
}),
|
|
),
|
|
],
|
|
temperature: None,
|
|
top_p: None,
|
|
reasoning_effort: None,
|
|
functions: None,
|
|
stream: false,
|
|
};
|
|
build_chat_completions_body(data, &Model::new("bedrock", "test"), false).unwrap()
|
|
}
|
|
|
|
fn build_body(tool_results: Vec<ToolResult>) -> Value {
|
|
build_body_with(tool_results, "", true)
|
|
}
|
|
|
|
fn tool_result_content_block(output: Value) -> Value {
|
|
let body = build_body(vec![tool_result(output)]);
|
|
body["messages"][2]["content"][0]["toolResult"]["content"][0].clone()
|
|
}
|
|
|
|
fn assert_unique_tool_use_ids_per_message(body: &Value) {
|
|
for message in body["messages"].as_array().unwrap() {
|
|
let Some(content) = message["content"].as_array() else {
|
|
continue;
|
|
};
|
|
let mut seen = HashSet::new();
|
|
for block in content {
|
|
if let Some(id) = block["toolUse"]["toolUseId"].as_str() {
|
|
assert!(
|
|
seen.insert(id.to_string()),
|
|
"duplicate toolUseId `{id}` within a single assistant message: {message}"
|
|
);
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
fn cache_fixture() -> ChatCompletionsData {
|
|
ChatCompletionsData {
|
|
messages: vec![
|
|
Message::new(MessageRole::System, MessageContent::Text("sys".to_string())),
|
|
Message::new(MessageRole::User, MessageContent::Text("hello".to_string())),
|
|
Message::new(
|
|
MessageRole::Assistant,
|
|
MessageContent::Text("hi".to_string()),
|
|
),
|
|
Message::new(MessageRole::User, MessageContent::Text("again".to_string())),
|
|
],
|
|
temperature: None,
|
|
top_p: None,
|
|
reasoning_effort: None,
|
|
functions: Some(vec![FunctionDeclaration {
|
|
name: "fs_read".to_string(),
|
|
description: "Read a file".to_string(),
|
|
parameters: Default::default(),
|
|
agent: false,
|
|
}]),
|
|
stream: false,
|
|
}
|
|
}
|
|
|
|
#[test]
|
|
fn test_bedrock_prompt_cache_inserts_cache_points_for_anthropic_models() {
|
|
let body = build_chat_completions_body(
|
|
cache_fixture(),
|
|
&Model::new("bedrock", "anthropic.claude-sonnet-4-5-20250929-v1:0"),
|
|
true,
|
|
)
|
|
.unwrap();
|
|
|
|
let cache_point = json!({ "cachePoint": { "type": "default" } });
|
|
assert_eq!(
|
|
body["system"].as_array().unwrap().last(),
|
|
Some(&cache_point),
|
|
"body: {body}"
|
|
);
|
|
assert_eq!(
|
|
body["toolConfig"]["tools"].as_array().unwrap().last(),
|
|
Some(&cache_point),
|
|
"body: {body}"
|
|
);
|
|
let messages = body["messages"].as_array().unwrap();
|
|
for idx in [0, 2] {
|
|
assert_eq!(
|
|
messages[idx]["content"].as_array().unwrap().last(),
|
|
Some(&cache_point),
|
|
"body: {body}"
|
|
);
|
|
}
|
|
assert!(
|
|
messages[1]["content"]
|
|
.as_array()
|
|
.unwrap()
|
|
.iter()
|
|
.all(|block| block.get("cachePoint").is_none()),
|
|
"assistant messages must not get cache points: {body}"
|
|
);
|
|
}
|
|
|
|
#[test]
|
|
fn test_bedrock_prompt_cache_skips_non_anthropic_models() {
|
|
let body = build_chat_completions_body(
|
|
cache_fixture(),
|
|
&Model::new("bedrock", "openai.gpt-oss-120b-1:0"),
|
|
true,
|
|
)
|
|
.unwrap();
|
|
|
|
assert!(!body.to_string().contains("cachePoint"), "body: {body}");
|
|
}
|
|
|
|
/// Converse rejects a cachePoint with no preceding block, so a user
|
|
/// message with an empty content array must be skipped and the
|
|
/// breakpoint placed on an earlier user message instead.
|
|
#[test]
|
|
fn test_bedrock_prompt_cache_skips_empty_user_content() {
|
|
let mut body = json!({
|
|
"messages": [
|
|
{ "role": "user", "content": [ { "text": "hello" } ] },
|
|
{ "role": "assistant", "content": [ { "text": "hi" } ] },
|
|
{ "role": "user", "content": [ { "text": "again" } ] },
|
|
{ "role": "assistant", "content": [ { "text": "calling" } ] },
|
|
{ "role": "user", "content": [] },
|
|
],
|
|
});
|
|
apply_prompt_cache_points(&mut body);
|
|
|
|
let cache_point = json!({ "cachePoint": { "type": "default" } });
|
|
let messages = body["messages"].as_array().unwrap();
|
|
assert!(
|
|
messages[4]["content"].as_array().unwrap().is_empty(),
|
|
"empty user content must not get a cachePoint: {body}"
|
|
);
|
|
for idx in [0, 2] {
|
|
assert_eq!(
|
|
messages[idx]["content"].as_array().unwrap().last(),
|
|
Some(&cache_point),
|
|
"body: {body}"
|
|
);
|
|
}
|
|
}
|
|
|
|
#[test]
|
|
fn sequence_splits_on_round_text() {
|
|
let body = build_body(vec![
|
|
sequence_tool_result("toolu_A", None),
|
|
sequence_tool_result("toolu_B", None),
|
|
sequence_tool_result("toolu_C", Some("running another tool")),
|
|
]);
|
|
|
|
let messages = body["messages"].as_array().unwrap();
|
|
|
|
assert_eq!(messages.len(), 5, "body: {body}");
|
|
assert_unique_tool_use_ids_per_message(&body);
|
|
}
|
|
|
|
#[test]
|
|
fn sequence_splits_on_reused_id_in_textless_round() {
|
|
let body = build_body(vec![
|
|
sequence_tool_result("toolu_A", None),
|
|
sequence_tool_result("toolu_B", None),
|
|
sequence_tool_result("toolu_A", None),
|
|
]);
|
|
|
|
let messages = body["messages"].as_array().unwrap();
|
|
|
|
assert_eq!(messages.len(), 5, "body: {body}");
|
|
assert_unique_tool_use_ids_per_message(&body);
|
|
}
|
|
|
|
#[test]
|
|
fn sequence_keeps_textless_rounds_merged_when_ids_are_unique() {
|
|
let body = build_body(vec![
|
|
sequence_tool_result("toolu_A", None),
|
|
sequence_tool_result("toolu_B", None),
|
|
sequence_tool_result("toolu_C", None),
|
|
]);
|
|
|
|
let messages = body["messages"].as_array().unwrap();
|
|
|
|
assert_eq!(messages.len(), 3, "body: {body}");
|
|
assert_unique_tool_use_ids_per_message(&body);
|
|
}
|
|
|
|
#[test]
|
|
fn non_sequence_emits_assistant_user_pair() {
|
|
let body = build_body_with(
|
|
vec![
|
|
sequence_tool_result("toolu_A", None),
|
|
sequence_tool_result("toolu_B", None),
|
|
],
|
|
"",
|
|
false,
|
|
);
|
|
|
|
let messages = body["messages"].as_array().unwrap();
|
|
|
|
assert_eq!(messages.len(), 3, "body: {body}");
|
|
assert_eq!(messages[1]["role"], "assistant");
|
|
assert_eq!(messages[1]["content"].as_array().unwrap().len(), 2);
|
|
assert_eq!(messages[2]["role"], "user");
|
|
assert_eq!(messages[2]["content"].as_array().unwrap().len(), 2);
|
|
}
|
|
|
|
#[test]
|
|
fn non_sequence_empty_tool_results_emits_no_messages() {
|
|
let body = build_body_with(vec![], "leftover text", false);
|
|
|
|
let messages = body["messages"].as_array().unwrap();
|
|
|
|
assert_eq!(messages.len(), 1, "body: {body}");
|
|
}
|
|
|
|
#[test]
|
|
fn sequence_empty_tool_results_emits_no_messages() {
|
|
let body = build_body_with(vec![], "leftover text", true);
|
|
|
|
let messages = body["messages"].as_array().unwrap();
|
|
|
|
assert_eq!(messages.len(), 1, "body: {body}");
|
|
}
|
|
|
|
#[test]
|
|
fn test_bedrock_tool_result_object_uses_json_block() {
|
|
assert_eq!(
|
|
tool_result_content_block(json!({"output": "ok"})),
|
|
json!({"json": {"output": "ok"}})
|
|
);
|
|
assert_eq!(tool_result_content_block(json!({})), json!({"json": {}}));
|
|
}
|
|
|
|
#[test]
|
|
fn test_bedrock_tool_result_array_uses_text_block() {
|
|
assert_eq!(
|
|
tool_result_content_block(json!([{"id": 1}, {"id": 2}])),
|
|
json!({"text": "[{\"id\":1},{\"id\":2}]"})
|
|
);
|
|
}
|
|
|
|
#[test]
|
|
fn test_bedrock_tool_result_scalars_use_text_block() {
|
|
assert_eq!(
|
|
tool_result_content_block(json!("plain text")),
|
|
json!({"text": "plain text"})
|
|
);
|
|
assert_eq!(tool_result_content_block(json!(42)), json!({"text": "42"}));
|
|
assert_eq!(
|
|
tool_result_content_block(json!(null)),
|
|
json!({"text": "null"})
|
|
);
|
|
}
|
|
|
|
fn thinking_tool_result() -> ToolResult {
|
|
ToolResult {
|
|
call: ToolCall::new(
|
|
"fs_read".into(),
|
|
json!({"path": "x"}),
|
|
Some("tool_A".into()),
|
|
),
|
|
output: json!("ok"),
|
|
text: None,
|
|
thinking: vec![
|
|
ThinkingBlock::Thinking {
|
|
thinking: "let me think".into(),
|
|
signature: "sig123".into(),
|
|
},
|
|
ThinkingBlock::RedactedThinking {
|
|
data: "b64data".into(),
|
|
},
|
|
],
|
|
}
|
|
}
|
|
|
|
#[test]
|
|
fn replays_thinking_blocks_in_converse_shape() {
|
|
for sequence in [false, true] {
|
|
let body = build_body_with(vec![thinking_tool_result()], "", sequence);
|
|
|
|
let content = body["messages"][1]["content"].as_array().unwrap();
|
|
assert_eq!(
|
|
content[0],
|
|
json!({
|
|
"reasoningContent": {
|
|
"reasoningText": { "text": "let me think", "signature": "sig123" }
|
|
}
|
|
}),
|
|
"body: {body}"
|
|
);
|
|
assert_eq!(
|
|
content[1],
|
|
json!({ "reasoningContent": { "redactedContent": "b64data" } }),
|
|
"body: {body}"
|
|
);
|
|
assert!(content[2].get("toolUse").is_some(), "body: {body}");
|
|
}
|
|
}
|
|
|
|
#[test]
|
|
fn extract_chat_completions_captures_reasoning_content() {
|
|
let data = json!({
|
|
"output": {
|
|
"message": {
|
|
"content": [
|
|
{
|
|
"reasoningContent": {
|
|
"reasoningText": { "text": "let me think", "signature": "sig123" }
|
|
}
|
|
},
|
|
{ "reasoningContent": { "redactedContent": "b64data" } },
|
|
{ "text": "answer" },
|
|
{
|
|
"toolUse": {
|
|
"toolUseId": "tool_A",
|
|
"name": "fs_read",
|
|
"input": { "path": "x" }
|
|
}
|
|
}
|
|
]
|
|
}
|
|
}
|
|
});
|
|
|
|
let output = extract_chat_completions(&data).unwrap();
|
|
|
|
assert_eq!(output.text, "answer");
|
|
assert_eq!(output.tool_calls.len(), 1);
|
|
assert_eq!(output.thinking.len(), 2);
|
|
match &output.thinking[0] {
|
|
ThinkingBlock::Thinking {
|
|
thinking,
|
|
signature,
|
|
} => {
|
|
assert_eq!(thinking, "let me think");
|
|
assert_eq!(signature, "sig123");
|
|
}
|
|
other => panic!("unexpected block: {other:?}"),
|
|
}
|
|
assert!(matches!(
|
|
&output.thinking[1],
|
|
ThinkingBlock::RedactedThinking { data } if data == "b64data"
|
|
));
|
|
}
|
|
}
|