mirror of
https://gitcode.com/JianFeeeee/ModelRouter.git
synced 2026-11-07 09:47:23 +00:00
a7a0c6adee
按前导抽出三块共通实现（都是实测逐字节相同的）：
| 钩子 | 原本重复 | 各 2.2KB |
|---|---|---|
| transform_stream_chunk | github/groq/kimicode/mistral | 2210B × 4 |
| transform_response | openai/groq/kimicode/mistral/github | 2206B × 5 |
| transform_error | deepseek/github/groq/kimicode/openai/sensenova | 292B × 6 |
适配器总行数 2409（收工前 groq 单文件 146 行 → 46 行）。
★ 两个必须记下的教训（都在本提交里踩到并修好）：
1. **抽公共代码要抽最全的实现，不是最小的**。我先抽了 transform_error 的
最小版（只认 {error:{message}}），打挂了 TestOpenAITransformErrorEnvelopes
的 6 个子用例 —— 最小版丢掉了 HTML 错误页、裸文本、扁平 {code,message}、
FastAPI {detail} 等形状。openai.lua 原有的 1580B 版本才是该进前导的那份。
抽最小实现是**静默降级**：不报错，只是少认几种错误体。
2. **抽出共性时不能丢掉"非共性但重要"的细节**。openai_response 我漏了两处：
reasoning_content 透传、以及 tool_calls 的 id 规整（safe_tool_id）。
后者尤其危险 —— 本网关是扇出路由器，一个非法 id（"bash:0"）会被回放到
每个 Claude 上游导致整条 AUTO 链失败；且**流式与非流式必须同规则**，
否则同一请求两条路径产出不同 id（实测报错：
流式 "bash_0_40248ca0" vs 非流式 "bash:0"）。现两条路径都走
__prelude.tool_id。
判据：TestPreludeRefactorIsByteIdentical —— 固定输入下逐字节比对
（含"arguments 必须解码成表""usage-only chunk 必须是 done:false"
"arguments 不得留原样 JSON 字符串"等反向断言），并覆盖错误体 6 种形状
与"非 OpenAI 形状必须返回 ok=false"。
另修 channels_pool_test：断言改为**关系**（关一个模型源里少一个）而非写死
数量 —— 兜底目录会随对齐进度变化（minimax 3→4），写死数量会把正常的目录
扩充变成测试失败。
回归：internal/{lua,provider,core,gateway} 全绿。
174 lines
7.1 KiB
Lua
174 lines
7.1 KiB
Lua
-- sensenova API format adapter
|
||
-- streaming: delta only has reasoning_content, no content field
|
||
-- non-streaming: has both content and reasoning_content
|
||
|
||
local adapter = {}
|
||
adapter.name = "sensenova"
|
||
adapter.version = "1.0.0"
|
||
adapter.endpoint = "/chat/completions"
|
||
adapter.headers = {}
|
||
|
||
-- Same as openai - strip provider-specific fields
|
||
function adapter.transform_request(raw_body)
|
||
local ok, req = pcall(json.decode, raw_body)
|
||
if not ok then return raw_body end
|
||
-- The gateway's private control fields are GONE from the body (v2): the
|
||
-- neutral `thinking` field carries that intent and each adapter maps it in
|
||
-- its own protocol terms. These upstreams have no thinking switch at all,
|
||
-- so the field is simply dropped here — which is what the nil-ing below
|
||
-- used to do, except the field being dropped is no longer the kernel's
|
||
-- private one.
|
||
req.thinking = nil
|
||
req.extra_body = nil
|
||
if req.messages then
|
||
for _, msg in ipairs(req.messages) do
|
||
msg.reasoning_content = nil
|
||
end
|
||
end
|
||
return json.encode(req)
|
||
end
|
||
|
||
-- Same as openai - extract content from response
|
||
function adapter.transform_response(raw_body)
|
||
local ok, resp = pcall(json.decode, raw_body)
|
||
if not ok or resp == nil then return raw_body end
|
||
|
||
local unified = {
|
||
content = "",
|
||
finish_reason = "",
|
||
token_usage = { prompt = 0, completion = 0, total = 0 }
|
||
}
|
||
|
||
if type(resp.usage) == "table" then
|
||
unified.token_usage.prompt = resp.usage.prompt_tokens or 0
|
||
unified.token_usage.completion = resp.usage.completion_tokens or 0
|
||
unified.token_usage.total = resp.usage.total_tokens or 0
|
||
local hit = 0
|
||
if type(resp.usage.prompt_tokens_details) == "table" and resp.usage.prompt_tokens_details.cached_tokens ~= nil then
|
||
hit = resp.usage.prompt_tokens_details.cached_tokens
|
||
unified.token_usage.prompt_tokens_details = { cached_tokens = hit }
|
||
end
|
||
if (resp.usage.prompt_cache_hit_tokens or 0) > 0 then
|
||
unified.token_usage.prompt_cache_hit_tokens = resp.usage.prompt_cache_hit_tokens
|
||
unified.token_usage.prompt_cache_miss_tokens = resp.usage.prompt_cache_miss_tokens or 0
|
||
if hit == 0 then
|
||
unified.token_usage.prompt_tokens_details = { cached_tokens = resp.usage.prompt_cache_hit_tokens }
|
||
end
|
||
end
|
||
end
|
||
|
||
if type(resp.choices) == "table" and #resp.choices > 0 then
|
||
local ch = resp.choices[1]
|
||
if type(ch.message) == "table" then
|
||
unified.content = ch.message.content or ""
|
||
if ch.message.reasoning_content then
|
||
unified.reasoning_content = ch.message.reasoning_content
|
||
elseif ch.message.reasoning then
|
||
-- sensenova-6.8-flash-lite 用 reasoning 字段而不是 reasoning_content
|
||
unified.reasoning_content = ch.message.reasoning
|
||
end
|
||
-- Tool calls MUST be forwarded. Dropping them while keeping
|
||
-- finish_reason="tool_calls" makes the client replay an assistant
|
||
-- message whose function name/arguments are empty, and sensenova
|
||
-- then rejects the next turn with
|
||
-- 400 invalid tool_call function, function/name/arguments cannot be empty
|
||
-- i.e. a tool-using conversation dies on its second request.
|
||
if type(ch.message.tool_calls) == "table" and #ch.message.tool_calls > 0 then
|
||
local tcs = {}
|
||
for _, tc in ipairs(ch.message.tool_calls) do
|
||
local fn = tc["function"] or {}
|
||
-- arguments arrives as a JSON *string* on the wire; the
|
||
-- unified shape expects a decoded object.
|
||
local args = fn.arguments
|
||
if type(args) == "string" then
|
||
local args_ok, decoded = pcall(json.decode, args)
|
||
args = args_ok and decoded or {}
|
||
elseif type(args) ~= "table" then
|
||
args = {}
|
||
end
|
||
table.insert(tcs, {
|
||
id = tc.id,
|
||
type = tc.type or "function",
|
||
name = fn.name,
|
||
arguments = args
|
||
})
|
||
end
|
||
unified.tool_calls = tcs
|
||
end
|
||
end
|
||
unified.finish_reason = ch.finish_reason or ""
|
||
end
|
||
|
||
return json.encode(unified)
|
||
end
|
||
|
||
-- Sensenova-specific stream handling
|
||
-- Upstream puts content in reasoning_content only (no content field)
|
||
-- Also sends finish_reason="" (empty string) on every chunk
|
||
function adapter.transform_stream_chunk(raw_chunk)
|
||
local ok, chunk = pcall(json.decode, raw_chunk)
|
||
if not ok then return "" end
|
||
|
||
local uses = nil
|
||
if type(chunk.usage) == "table" then
|
||
uses = {
|
||
prompt = chunk.usage.prompt_tokens or chunk.usage.prompt or 0,
|
||
completion = chunk.usage.completion_tokens or chunk.usage.completion or 0,
|
||
total = chunk.usage.total_tokens or chunk.usage.total or 0,
|
||
}
|
||
if type(chunk.usage.prompt_tokens_details) == "table" and chunk.usage.prompt_tokens_details.cached_tokens ~= nil then
|
||
uses.prompt_tokens_details = { cached_tokens = chunk.usage.prompt_tokens_details.cached_tokens }
|
||
elseif (chunk.usage.prompt_cache_hit_tokens or 0) > 0 then
|
||
uses.prompt_cache_hit_tokens = chunk.usage.prompt_cache_hit_tokens
|
||
uses.prompt_cache_miss_tokens = chunk.usage.prompt_cache_miss_tokens or 0
|
||
uses.prompt_tokens_details = { cached_tokens = chunk.usage.prompt_cache_hit_tokens }
|
||
end
|
||
end
|
||
|
||
if not chunk.choices or #chunk.choices == 0 then
|
||
if uses ~= nil then
|
||
return json.encode({ usage = uses, done = false })
|
||
end
|
||
return ""
|
||
end
|
||
|
||
local delta = chunk.choices[1].delta or {}
|
||
|
||
-- Sensenova: delta has reasoning_content (deepseek-v4-flash) or reasoning
|
||
-- (sensenova-6.8-flash-lite) but no content field
|
||
local content = delta.content or ""
|
||
if content == "" and (delta.reasoning_content or delta.reasoning) then
|
||
content = delta.reasoning_content or delta.reasoning
|
||
end
|
||
|
||
-- Sensenova: finish_reason is "" on every chunk, "stop" on last
|
||
local fr = chunk.choices[1].finish_reason
|
||
local done = (fr == "stop" or fr == "length")
|
||
|
||
local unified = {
|
||
content = content,
|
||
done = done,
|
||
}
|
||
if chunk.choices[1].finish_reason and chunk.choices[1].finish_reason ~= "" then
|
||
unified.finish_reason = chunk.choices[1].finish_reason
|
||
end
|
||
if delta.tool_calls then
|
||
unified.tool_calls = delta.tool_calls
|
||
end
|
||
if uses ~= nil then
|
||
unified.usage = uses
|
||
end
|
||
|
||
return json.encode(unified)
|
||
end
|
||
|
||
-- 错误收敛：sensenova 为 OpenAI 风格 {error:{message,...}}；
|
||
-- 配额类错误单独点出便于客户端识别重置周期。
|
||
function adapter.transform_error(status, body)
|
||
-- OpenAI 风格错误体抽消息。此前在 deepseek/github/groq/kimicode/
|
||
-- openai/sensenova 六个适配器里逐字节相同（各 293B），已抽到共享前导。
|
||
return __prelude.openai_error(status, body)
|
||
end
|
||
|
||
return adapter
|