add test script for ollama, install and test ollama latest with llama3.2, verify error handling, fix error handling, regression test with Google gemini-2.5-pro.

Author: scottstirling

pi2llm-main.js

@@ -107,29 +107,45 @@ LLMCommunicator.prototype.sendMessage = function (payload, onComplete, onError)
                 // 2. if we have a responseObject but failed to find "choices" in JSON,
                 // then try to handle as Cloudflare AI Gateway's simpler format: {"result":{"response":"foo bar baz" ...
                 messageContent = responseObject.result.response;
-            } else if (responseObject && transfer.responseCode > 200) { // response contains an error message
-                let errorObject = responseObject[0];
-                var error = errorObject.error;
-                var errorMsg = error.message
-                var errorCode = error.code;
-                var errorStatus = error.status;
-
-                messageContent = "AI Error: " + errorMsg + ", code: " + errorCode + ", status: " + errorStatus;
+            } else {
+                // Now we know it's not a success response, so we parse for known error structures.
+                let errorMessage = "An unknown error occurred.";
+
+                // ** THE FIX IS HERE: Check for Ollama-style errors first **
+                // This is for errors like: {"error":{"message":"..."}}
+                if (responseObject.error && typeof responseObject.error === 'object' && responseObject.error.message) {
+                    errorMessage = responseObject.error.message;
+                    // ** THE FIX IS HERE: Then, check for OpenAI-style array errors **
+                    // This is for errors like: [{"error":{"message":"..."}}]
+                } else if (Array.isArray(responseObject) && responseObject.length > 0 && responseObject[0].error) {
+                    errorMessage = responseObject[0].error.message;
+                    // ** THE FIX IS HERE: And for simpler Ollama errors **
+                    // This is for errors like: {"error":"model '...' is not loaded"}
+                } else if (responseObject.error && typeof responseObject.error === 'string') {
+                    errorMessage = responseObject.error;
+                }
+
+                messageContent = "AI Error: " + errorMessage;
+                console.criticalln("Parsed error from LLM backend: ", errorMessage);
             }
             if (onComplete) {
                 onComplete(messageContent);
             }
         } catch (e) {
             let errorMsg = "Error parsing LLM JSON response: " + e.message;
-            messageContent = errorMsg;
             console.criticalln(errorMsg);
-
+            console.writeln("Raw Response String: ", responseString); // Log the raw string on parse failure
             if (onError) {
                 onError(errorMsg);
             }
         }
     } else {
         let errorMsg = "NetworkTransfer POST failed. HTTP Status: " + transfer.responseCode + "\nError Info: " + transfer.errorInformation;
+        // Attempt to read the response body even on network failure, as it often contains useful error details.
+        let responseString = transfer.response.utf8ToString();
+        if (responseString) {
+            errorMsg += "\nResponse Body: " + responseString;
+        }
         console.criticalln(errorMsg);
         if (onError) {
             onError(errorMsg);
@@ -139,12 +155,14 @@ LLMCommunicator.prototype.sendMessage = function (payload, onComplete, onError)
     transfer.closeConnection();
 };
 
-// common api response errors:
+// Google api response errors:
 // 1. wrong / non-existent model name:
     // [{"error":{"code":404,"message":"models/foo is not found for API version v1main, or is not supported for generateContent. Call ListModels to see the list of available models and their supported methods.","status":"NOT_FOUND"}}]
 // 2. no model name supplied:
     // [{"error":{"code":400,"message":"model is not specified","status":"INVALID_ARGUMENT"}}]
 
+// ollama error example:
+// {"error":{"message":"model \"llama3.2\" not found, try pulling it first","type":"api_error","param":null,"code":null}}
 
 function pi2llmMain() {
     console.show();

tests/test-ollama.sh

@@ -0,0 +1,16 @@
+#!/bin/bash
+# install ollama per instructions.
+# run:
+#      pull llama3.2
+# To pull down a sample model to test.
+
+echo "Listing models deployed on ollama .."
+curl http://localhost:11434/v1/models
+
+echo "Calling ollama non-streaming /v1/chat/completions .."
+curl -v http://localhost:11434/v1/chat/completions -d '{
+  "model": "llama3.2",
+  "messages": [
+    { "role": "user", "content": "why is the sky blue?" }
+  ]
+}'