coding-proxy 0.2.3a5__tar.gz → 0.2.4a2__tar.gz

{coding_proxy-0.2.3a5 → coding_proxy-0.2.4a2}/CHANGELOG.md

@@ -4,7 +4,7 @@
 
 ## [Unreleased]
 
-## [v0.2.3](https://github.com/ThreeFish-AI/coding-proxy/releases/tag/v0.2.3a1) — 2026-04-15
+## [v0.2.3](https://github.com/ThreeFish-AI/coding-proxy/releases/tag/v0.2.3) — 2026-04-16
 
 - feat(dashboard): 新增实时 Web Dashboard 页面，聚合展示流量与用量统计;
 - docs(user-guide): 补充 POST /v1/messages 完整 API 参考文档;

{coding_proxy-0.2.3a5 → coding_proxy-0.2.4a2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: coding-proxy
-Version: 0.2.3a5
+Version: 0.2.4a2
 Summary: A High-Availability, Transparent, and Smart Multi-Vendor Proxy for Claude Code. Support Claude Plans, GitHub Copilot, Google Antigravity, ZAI/GLM, MiniMax, Qwen, Xiaomi, Kimi, Doubao...
 Project-URL: Source Code, https://github.com/ThreeFish-AI/coding-proxy
 Project-URL: User Guide, https://github.com/ThreeFish-AI/coding-proxy/blob/master/docs/user-guide.md
@@ -56,6 +56,10 @@ When you're deeply immersed in your coding "zone" with **Claude Code** (or any A
 
 ## 🌟 Core Features
 
+<div align="center">
+    <img src="assets/dashboard-v0.2.3.png">
+</div>
+
 - **⛓️ N-tier Chained Failover**: Autonomous descending sequence, supporting Claude's official plans, as well as Coding Plans from GitHub Copilot, Z AI, MiniMax, Alibaba Qwen, Xiaomi, Kimi, Doubao, etc.
 - **🛡️ Smart Resilience & Quota Guardians**: Every single vendor node comes fully armed with an independent **Circuit Breaker** and **Quota Guard** to proactively dodge avalanches without breaking a sweat.
 - **👻 Phantom-like Transparency**: **100% transparent** to the client! No code tweaks required. Overwrite `ANTHROPIC_BASE_URL` with a single line, and you're good to go.

{coding_proxy-0.2.3a5 → coding_proxy-0.2.4a2}/README.md

@@ -29,6 +29,10 @@ When you're deeply immersed in your coding "zone" with **Claude Code** (or any A
 
 ## 🌟 Core Features
 
+<div align="center">
+    <img src="assets/dashboard-v0.2.3.png">
+</div>
+
 - **⛓️ N-tier Chained Failover**: Autonomous descending sequence, supporting Claude's official plans, as well as Coding Plans from GitHub Copilot, Z AI, MiniMax, Alibaba Qwen, Xiaomi, Kimi, Doubao, etc.
 - **🛡️ Smart Resilience & Quota Guardians**: Every single vendor node comes fully armed with an independent **Circuit Breaker** and **Quota Guard** to proactively dodge avalanches without breaking a sweat.
 - **👻 Phantom-like Transparency**: **100% transparent** to the client! No code tweaks required. Overwrite `ANTHROPIC_BASE_URL` with a single line, and you're good to go.

{coding_proxy-0.2.3a5 → coding_proxy-0.2.4a2}/docs/zh-CN/README.md

@@ -29,6 +29,10 @@
 
 ## 🌟 核心特性 (Core Features)
 
+<div align="center">
+    <img src="../../assets/dashboard-v0.2.3.png">
+</div>
+
 - **⛓️ N-tier 链式故障转移 (Failover)**：自主降序序列，支持 Claude 官方 Plans，以及 GitHub Copilot、智谱、MiniMax、阿里千问、小米、Kimi、豆包等的 Coding Plan。
 - **🛡️ 智能弹性与容灾守卫**：每个供应商节点独立配备 **熔断器 (Circuit Breaker)** 与 **配额守卫 (Quota Guard)**，防雪崩、主动避险。
 - **👻 透明无感代理机制**：对客户端 **100% 透明**！无需修改任何代码，仅需一行配置覆盖 `ANTHROPIC_BASE_URL` 即可接入。

{coding_proxy-0.2.3a5 → coding_proxy-0.2.4a2}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "coding-proxy"
-version = "0.2.3a5"
+version = "0.2.4a2"
 description = "A High-Availability, Transparent, and Smart Multi-Vendor Proxy for Claude Code. Support Claude Plans, GitHub Copilot, Google Antigravity, ZAI/GLM, MiniMax, Qwen, Xiaomi, Kimi, Doubao..."
 readme = "README.md"
 requires-python = ">=3.12"

{coding_proxy-0.2.3a5 → coding_proxy-0.2.4a2}/src/coding/proxy/config/config.default.yaml

@@ -51,14 +51,14 @@ vendors:
       recovery_timeout_seconds: 300
       success_threshold: 2
     quota_guard:
-      enabled: true
-      token_budget: 45000000 # 5 小时 token 预算（根据订阅计划调整）
+      enabled: false
+      token_budget: 60000000 # 5 小时 token 预算（根据订阅计划调整）
       window_hours: 5.0
       threshold_percent: 99.0
       probe_interval_seconds: 300
     weekly_quota_guard:
-      enabled: true
-      token_budget: 500000000 # 一周 token 预算（根据订阅计划调整）
+      enabled: false
+      token_budget: 800000000 # 一周 token 预算（根据订阅计划调整）
       window_hours: 168.0 # 7 天滑动窗口
       threshold_percent: 99.0
       probe_interval_seconds: 1800 # 每 30 分钟探测一次
@@ -84,7 +84,7 @@ vendors:
 
   # Vendor 2: Google Antigravity Plans（中间层，默认禁用）
   - vendor: antigravity
-    enabled: true # 启用需配置 OAuth 凭据
+    enabled: false # 启用需配置 OAuth 凭据
     client_id: "${GOOG_CLIENT_ID}" # Google OAuth2 Client ID
     client_secret: "${GOOG_CLIENT_SECRET}" # Google OAuth2 Client Secret
     refresh_token: "${GOOG_REFRESH_TOKEN}" # Google OAuth2 Refresh Token
@@ -96,7 +96,7 @@ vendors:
       recovery_timeout_seconds: 300
       success_threshold: 2
     quota_guard:
-      enabled: true # 启用后按 Premium Requests 配额管理
+      enabled: false # 启用后按 Premium Requests 配额管理
       token_budget: 0
       window_hours: 24.0
       threshold_percent: 95.0

{coding_proxy-0.2.3a5 → coding_proxy-0.2.4a2}/src/coding/proxy/logging/stats.py

@@ -198,7 +198,7 @@ async def show_usage(
 
     if cost_totals:
         total_cost_str = " + ".join(
-            f"{cur.symbol}{amt:.4f}" for cur, amt in cost_totals.items()
+            f"{cur.symbol}{amt:.2f}" for cur, amt in cost_totals.items()
         )
     else:
         total_cost_str = "-"

{coding_proxy-0.2.3a5 → coding_proxy-0.2.4a2}/src/coding/proxy/model/pricing.py

@@ -50,8 +50,8 @@ class CostValue:
     amount: float
     currency: Currency = Currency.default()
 
-    def format(self, precision: int = 4) -> str:
-        """格式化为 ``$0.1234`` 或 ``¥0.1234``."""
+    def format(self, precision: int = 2) -> str:
+        """格式化为 ``$0.12`` 或 ``¥0.12``."""
         return f"{self.currency.symbol}{self.amount:.{precision}f}"
 
     @property

{coding_proxy-0.2.3a5 → coding_proxy-0.2.4a2}/src/coding/proxy/routing/executor.py

@@ -6,6 +6,7 @@
 
 from __future__ import annotations
 
+import copy
 import logging
 import time
 from collections.abc import AsyncIterator
@@ -222,10 +223,94 @@ class _RouteExecutor:
 
     # ── 公开执行入口 ──────────────────────────────────────
 
+    def _prepare_body_for_tier(
+        self,
+        body: dict[str, Any],
+        tier: VendorTier,
+        normalization: Any = None,
+        session_record: Any = None,
+    ) -> dict[str, Any]:
+        """为指定 tier 准备请求体，必要时应用 Anthropic 专属修复（Phase 2）.
+
+        仅当 tier 为 Anthropic 时才执行以下处理：
+        1. tool_result 重定位 + 孤儿修复（需 normalization.has_anthropic_fixes）
+        2. 条件化 thinking block 剥离（仅跨供应商场景）
+
+        确保 Zhipu 等其他 vendor 不受影响。
+        """
+        if tier.name != "anthropic":
+            return body
+
+        needs_tool_fixes = (
+            normalization is not None and normalization.has_anthropic_fixes
+        )
+        needs_thinking_strip = self._needs_thinking_strip(normalization, session_record)
+
+        if not needs_tool_fixes and not needs_thinking_strip:
+            return body
+
+        from ..server.request_normalizer import (
+            apply_anthropic_specific_fixes,
+            strip_thinking_blocks,
+        )
+
+        body_for_vendor = copy.deepcopy(body)
+
+        if needs_tool_fixes:
+            fixes = apply_anthropic_specific_fixes(
+                body_for_vendor.get("messages", []),
+                normalization.misplaced_tool_results,
+                normalization.misplaced_log_info,
+            )
+            if fixes:
+                logger.debug(
+                    "Applied Anthropic-specific fixes for tier %s: %s",
+                    tier.name,
+                    ", ".join(fixes),
+                )
+
+        if needs_thinking_strip:
+            stripped = strip_thinking_blocks(body_for_vendor)
+            if stripped:
+                logger.debug(
+                    "Stripped %d thinking block(s) for cross-vendor compatibility",
+                    stripped,
+                )
+
+        return body_for_vendor
+
+    @staticmethod
+    def _needs_thinking_strip(normalization: Any, session_record: Any) -> bool:
+        """判断是否需要剥离 thinking blocks（仅跨供应商场景）.
+
+        信号优先级：
+        1. 请求规范化信号 — 当前请求体中检测到跨供应商产物
+        2. 会话历史信号 — provider_state 中存在非 Anthropic 供应商记录
+
+        安全默认：当无法确定会话来源时（session_record 为 None），
+        回退到始终剥离，确保与 compat_session_store 未配置时的向后兼容。
+        """
+        # Signal 1: normalization 检测到跨供应商产物
+        if normalization is not None and normalization.has_cross_vendor_signals:
+            return True
+        # Signal 2: 无会话追踪能力 → 无法判断是否跨供应商 → 安全回退到剥离
+        if session_record is None:
+            return True
+        # Signal 3: 会话历史中有非 Anthropic 供应商
+        if session_record.provider_state:
+            non_anthropic = {
+                v for v in session_record.provider_state if v != "anthropic"
+            }
+            if non_anthropic:
+                return True
+        # 纯 Anthropic 会话，无跨供应商信号 → 保留 thinking blocks
+        return False
+
     async def execute_stream(
         self,
         body: dict[str, Any],
         headers: dict[str, str],
+        normalization: Any = None,
     ) -> AsyncIterator[tuple[bytes, str]]:
         """路由流式请求，按优先级尝试各层级."""
         last_idx = len(self._tiers) - 1
@@ -257,7 +342,12 @@ class _RouteExecutor:
             usage: dict[str, Any] = {}
 
             try:
-                async for chunk in tier.vendor.send_message_stream(body, headers):
+                body_for_tier = self._prepare_body_for_tier(
+                    body, tier, normalization, session_record=session_record
+                )
+                async for chunk in tier.vendor.send_message_stream(
+                    body_for_tier, headers
+                ):
                     parse_usage_from_chunk(
                         chunk,
                         usage,
@@ -389,6 +479,7 @@ class _RouteExecutor:
         self,
         body: dict[str, Any],
         headers: dict[str, str],
+        normalization: Any = None,
     ) -> VendorResponse:
         """路由非流式请求，按优先级尝试各层级."""
         last_idx = len(self._tiers) - 1
@@ -417,7 +508,10 @@ class _RouteExecutor:
                 continue
 
             try:
-                resp = await tier.vendor.send_message(body, headers)
+                body_for_tier = self._prepare_body_for_tier(
+                    body, tier, normalization, session_record=session_record
+                )
+                resp = await tier.vendor.send_message(body_for_tier, headers)
 
                 if resp.status_code < 400:
                     duration = int((time.monotonic() - start) * 1000)

{coding_proxy-0.2.3a5 → coding_proxy-0.2.4a2}/src/coding/proxy/routing/router.py

@@ -134,18 +134,24 @@ class RequestRouter:
         self,
         body: dict[str, Any],
         headers: dict[str, str],
+        normalization: Any = None,
     ) -> AsyncIterator[tuple[bytes, str]]:
         """路由流式请求，按优先级尝试各层级."""
-        async for chunk, vendor_name in self._executor.execute_stream(body, headers):
+        async for chunk, vendor_name in self._executor.execute_stream(
+            body, headers, normalization=normalization
+        ):
             yield chunk, vendor_name
 
     async def route_message(
         self,
         body: dict[str, Any],
         headers: dict[str, str],
+        normalization: Any = None,
     ) -> Any:
         """路由非流式请求，按优先级尝试各层级."""
-        return await self._executor.execute_message(body, headers)
+        return await self._executor.execute_message(
+            body, headers, normalization=normalization
+        )
 
     # ── 生命周期 ───────────────────────────────────────────