From 52e29529c3d9ea9877bcbbf6ca7b1ba935214495 Mon Sep 17 00:00:00 2001 From: Evan Prodromou Date: Fri, 6 Nov 2026 12:16:54 -0500 Subject: [PATCH] feat: chat message with debug output --- src/extension.ts | 106 +++++++++++++++++++++++++++++++++++++++++------ 1 file changed, 94 insertions(+), 12 deletions(-) diff --git a/src/extension.ts b/src/extension.ts index 10531c1..8ebac82 100644 --- a/src/extension.ts +++ b/src/extension.ts @@ -1,43 +1,125 @@ import * as vscode from 'vscode'; const BASE_URL = 'https://gateway.inference.coop/v1'; +const PREFIX = 'inference-coop/'; export function activate(context: vscode.ExtensionContext) { + const out = vscode.window.createOutputChannel('Inference Coop'); + context.subscriptions.push(out); + out.appendLine('activate() called'); + const provider: vscode.LanguageModelChatProvider = { async provideLanguageModelChatInformation() { + out.appendLine('provideLanguageModelChatInformation: called'); + const key = vscode.workspace .getConfiguration('inferenceCoop') .get('apiKey'); - if (!key) { - return []; - } + out.appendLine(` key present: ${Boolean(key)}`); + if (!key) {return [];} const res = await fetch(`${BASE_URL}/models`, { headers: { Authorization: `Bearer ${key}` } }); + out.appendLine(` /models status: ${res.status}`); if (!res.ok) { - vscode.window.showErrorMessage( - `Inference Coop: model list failed (HTTP ${res.status})` - ); + out.appendLine(` body: ${await res.text()}`); return []; } - const body = (await res.json()) as { data: Array<{ id: string; mode?: string }> }; - return body.data + const body = (await res.json()) as { + data: Array<{ id: string; mode?: string }>; + }; + out.appendLine(` raw models: ${body.data.length}`); + + const models = body.data .filter(m => !m.mode || m.mode === 'chat') .map(m => ({ - id: m.id, + id: `${PREFIX}${m.id}`, name: m.id, family: 'inference-coop', version: '1.0.0', maxInputTokens: 131072, maxOutputTokens: 8192, - capabilities: { toolCalling: true, imageInput: false } + capabilities: { toolCalling: true, imageInput: false } })); + + out.appendLine(` returning ${models.length} models`); + return models; }, - async provideLanguageModelChatResponse() { - throw new Error('Not implemented yet'); + + async provideLanguageModelChatResponse(model, messages, _options, progress, token) { + out.appendLine(`provideLanguageModelChatResponse: ${model.id}`); + + const key = vscode.workspace + .getConfiguration('inferenceCoop') + .get('apiKey'); + + const body = { + model: model.id.replace(PREFIX, ''), + stream: true, + messages: messages.map(m => ({ + role: + m.role === vscode.LanguageModelChatMessageRole.Assistant + ? 'assistant' + : 'user', + content: m.content + .filter( + (p): p is vscode.LanguageModelTextPart => + p instanceof vscode.LanguageModelTextPart + ) + .map(p => p.value) + .join('\n') + })) + }; + + const res = await fetch(`${BASE_URL}/chat/completions`, { + method: 'POST', + headers: { + Authorization: `Bearer ${key}`, + 'Content-Type': 'application/json' + }, + body: JSON.stringify(body), + signal: token.isCancellationRequested ? AbortSignal.abort() : undefined + }); + out.appendLine(` /chat/completions status: ${res.status}`); + if (!res.ok || !res.body) { + throw new Error(`Inference Coop: HTTP ${res.status} — ${await res.text()}`); + } + + const reader = res.body.getReader(); + const decoder = new TextDecoder(); + let buf = ''; + let chunks = 0; + while (true) { + const { done, value } = await reader.read(); + if (done) {break;} + buf += decoder.decode(value, { stream: true }); + const frames = buf.split('\n\n'); + buf = frames.pop() ?? ''; + for (const frame of frames) { + for (const line of frame.split('\n')) { + if (!line.startsWith('data:')) {continue;} + const data = line.slice(5).trim(); + if (data === '[DONE]') { + out.appendLine(` done, ${chunks} text chunks`); + return; + } + try { + const delta = JSON.parse(data).choices?.[0]?.delta?.content; + if (delta) { + chunks++; + progress.report(new vscode.LanguageModelTextPart(delta)); + } + } catch { + /* partial frame, ignore */ + } + } + } + } + out.appendLine(` stream ended, ${chunks} text chunks`); }, + async provideTokenCount() { return 0; }