@x1a0f3n9/dsh-llm 0.1.5-rc.2 → 0.1.5-rc.3

package/lib/index.js

@@ -155,6 +155,12 @@ const STRUCTURED_CONTEXT_OVERFLOW = new RegExp(String.raw`(?:^|[^a-z0-9])context
 const TOO_LARGE_FOR_CONTEXT = new RegExp(String.raw`\b(?:request|prompt|input|messages?)\s+(?:is\s+|are\s+)?` + String.raw`too\s+(?:large|long)\s+for\s+(?:(?:this|the)\s+)?` + String.raw`(?:model(?:'s)?\s+)?context(?:\s+window)?\b`, "i");
 /** "Exceeds" wording is safe only when its object is explicitly the model context. */
 const EXCEEDS_MODEL_CONTEXT = new RegExp(String.raw`\b(?:input|prompt|request|messages?)\b.{0,40}` + String.raw`\b(?:exceed(?:s|ed)?|overflows?|is\s+larger\s+than)\b.{0,40}` + String.raw`\b(?:the\s+)?(?:model(?:'s)?\s+)?context(?:\s+(?:length|window))?\b`, "i");
+/** xAI Grok and similar OpenAI-compatible gateways name the bound as prompt length. */
+const MAXIMUM_PROMPT_LENGTH = /\bmaximum prompt length\b|\bprompt is too long\b/i;
+/** Groq and similar gateways ask the caller to shorten the message list. */
+const REDUCE_MESSAGE_LENGTH = /\breduce the length of the messages\b/i;
+/** Chinese OpenAI-compatible gateways often return localized overflow copy. */
+const CHINESE_CONTEXT_OVERFLOW = /上下文(?:窗口|长度)|(?:提示词|输入|请求).{0,8}过长/;
 /**
 * Recognize the context-overflow wording used by OpenAI-compatible providers
 * and library adapters. Adapters pass all available provider code, type, and
@@ -163,7 +169,7 @@ const EXCEEDS_MODEL_CONTEXT = new RegExp(String.raw`\b(?:input|prompt|request|me
 * @returns true when the detail identifies a request exceeding the model context window.
 */
 function isContextWindowExceededError(detail) {
-	return STRUCTURED_CONTEXT_OVERFLOW.test(detail) || /\b(?:maximum|max)(?:\s+(?:allowed|supported))?\s+context\s+(?:length|window)\b/i.test(detail) || TOO_LARGE_FOR_CONTEXT.test(detail) || /\b(?:input|prompt|request)\s+(?:is\s+)?too\s+(?:long|large)\s+for\s+(?:this|the)\s+model\b/i.test(detail) || EXCEEDS_MODEL_CONTEXT.test(detail);
+	return STRUCTURED_CONTEXT_OVERFLOW.test(detail) || /\b(?:maximum|max)(?:\s+(?:allowed|supported))?\s+context\s+(?:length|window)\b/i.test(detail) || TOO_LARGE_FOR_CONTEXT.test(detail) || /\b(?:input|prompt|request)\s+(?:is\s+)?too\s+(?:long|large)\s+for\s+(?:this|the)\s+model\b/i.test(detail) || /\btoo long for (?:this|the) model\b/i.test(detail) || EXCEEDS_MODEL_CONTEXT.test(detail) || MAXIMUM_PROMPT_LENGTH.test(detail) || REDUCE_MESSAGE_LENGTH.test(detail) || CHINESE_CONTEXT_OVERFLOW.test(detail);
 }
 /**
 * Recognize provider wording that identifies an exhausted account quota rather

package/lib/types/error.js

@@ -51,6 +51,12 @@ const TOO_LARGE_FOR_CONTEXT = new RegExp(String.raw `\b(?:request|prompt|input|m
 const EXCEEDS_MODEL_CONTEXT = new RegExp(String.raw `\b(?:input|prompt|request|messages?)\b.{0,40}`
     + String.raw `\b(?:exceed(?:s|ed)?|overflows?|is\s+larger\s+than)\b.{0,40}`
     + String.raw `\b(?:the\s+)?(?:model(?:'s)?\s+)?context(?:\s+(?:length|window))?\b`, 'i');
+/** xAI Grok and similar OpenAI-compatible gateways name the bound as prompt length. */
+const MAXIMUM_PROMPT_LENGTH = /\bmaximum prompt length\b|\bprompt is too long\b/i;
+/** Groq and similar gateways ask the caller to shorten the message list. */
+const REDUCE_MESSAGE_LENGTH = /\breduce the length of the messages\b/i;
+/** Chinese OpenAI-compatible gateways often return localized overflow copy. */
+const CHINESE_CONTEXT_OVERFLOW = /上下文(?:窗口|长度)|(?:提示词|输入|请求).{0,8}过长/;
 /**
  * Recognize the context-overflow wording used by OpenAI-compatible providers
  * and library adapters. Adapters pass all available provider code, type, and
@@ -63,7 +69,11 @@ export function isContextWindowExceededError(detail) {
         || /\b(?:maximum|max)(?:\s+(?:allowed|supported))?\s+context\s+(?:length|window)\b/i.test(detail)
         || TOO_LARGE_FOR_CONTEXT.test(detail)
         || /\b(?:input|prompt|request)\s+(?:is\s+)?too\s+(?:long|large)\s+for\s+(?:this|the)\s+model\b/i.test(detail)
-        || EXCEEDS_MODEL_CONTEXT.test(detail);
+        || /\btoo long for (?:this|the) model\b/i.test(detail)
+        || EXCEEDS_MODEL_CONTEXT.test(detail)
+        || MAXIMUM_PROMPT_LENGTH.test(detail)
+        || REDUCE_MESSAGE_LENGTH.test(detail)
+        || CHINESE_CONTEXT_OVERFLOW.test(detail);
 }
 /**
  * Recognize provider wording that identifies an exhausted account quota rather

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@x1a0f3n9/dsh-llm",
   "description": "Provider-neutral LLM service interface for the DeepSeek Harness",
-  "version": "0.1.5-rc.2",
+  "version": "0.1.5-rc.3",
   "publishConfig": {
     "access": "public"
   },
@@ -65,16 +65,16 @@
   },
   "dependencies": {
     "zod": "^4.4.3",
-    "@x1a0f3n9/dsh-util-crypto": "^0.1.5-rc.2",
+    "@x1a0f3n9/dsh-util-crypto": "^0.1.5-rc.3",
     "@deepseek-ai/schemastery": "^3.18.2",
-    "@x1a0f3n9/dsh-timeout": "^0.1.5-rc.2",
-    "@x1a0f3n9/dsh-brand": "^0.1.5-rc.2",
-    "@x1a0f3n9/dsh-util-values": "^0.1.5-rc.2",
-    "@x1a0f3n9/dsh-typert-protocol": "^0.1.5-rc.2"
+    "@x1a0f3n9/dsh-timeout": "^0.1.5-rc.3",
+    "@x1a0f3n9/dsh-brand": "^0.1.5-rc.3",
+    "@x1a0f3n9/dsh-util-values": "^0.1.5-rc.3",
+    "@x1a0f3n9/dsh-typert-protocol": "^0.1.5-rc.3"
   },
   "devDependencies": {
-    "@x1a0f3n9/dsh-attachment": "^0.1.5-rc.2",
-    "@x1a0f3n9/dsh-invariants": "^0.1.5-rc.2",
+    "@x1a0f3n9/dsh-attachment": "^0.1.5-rc.3",
+    "@x1a0f3n9/dsh-invariants": "^0.1.5-rc.3",
     "@deepseek-ai/cordis": "^4.0.2"
   }
 }