Compare commits
| Author | SHA1 | Date | |
|---|---|---|---|
|
|
8248ff8ac7 |
@@ -2,6 +2,34 @@
|
||||
|
||||
---
|
||||
|
||||
## v0.3.98
|
||||
|
||||
### 機能追加・バグ修正: AI TOOL翻訳の信頼性向上 + GenerateUI Modelタブ再構成(ハイライト色・折りたたみ・並び替え)+ Thinking mode/Max tokens設定
|
||||
|
||||
ユーザーから「AI TOOLタブの翻訳、英語から日本語がうまく動かない(英語のまま、完了で翻訳されない)」との報告を受けて着手。原因調査後、関連するGenerateUI Modelタブの表示改善、AI TOOLタブへのLLM生成パラメータ設定追加と、複数の要望を連続して実装した一連のセッション。
|
||||
|
||||
**1. AI TOOL翻訳の信頼性向上**: 調査の結果、翻訳処理がOllamaの`/api/generate`(system role無しの生プロンプト文字列)またはLM Studio/Lemonadeの`/v1/chat/completions`(user roleのみ、system message無し)でLLMへ丸投げしており、指示追従が弱い小型・非力なローカルモデルが「翻訳してください」という指示を無視して原文をそのまま返すことがあった。それでもHTTPレスポンス自体は正常なため、画面上は「完了」と表示されたまま実際には未翻訳、という状態が発生していた。既にChatペインで使われていた`callChat()`(system role対応のchat API)を翻訳にも流用し、system promptで「翻訳結果のみを出力し、説明・前置き・原文の繰り返しをしない」と明確に指示するよう変更。加えてモデルが前置き文("Here is the translation:"等)や引用符を付けて返す場合の後処理除去(`cleanTranslationOutput`)と、出力が原文とほぼ同一=未翻訳の疑いがある場合の警告トースト(`looksUntranslated`)を追加。SPA(`ai-tab.js`)とサイドパネルAタブ(`node_sets_menu.js`)は全く同じ不具合パターンの重複実装だったため、両方に同じ修正を適用(設定は同じ`localStorage`キー`wfm_ai_settings`を共有)。
|
||||
|
||||
**2. GenerateUI Modelタブ: ハイライト色機能**: ユーザーから「読み込んだワークフローに関係なく項目を並べているため該当する項目(Checkpointなど)の色を変更表示、デフォルトは明るい緑、設定タブに色変更設定を追加したい」との要望。既存の「RAW JSON Colors」機能(`<style>`タグ動的注入+`localStorage`保存+Resetボタン)と全く同じパターンを踏襲し、`applyModelTabActiveColor()`を新設。`renderModelTab()`で各セクション(Checkpoint/VAE/Diffusion Model/Text Encoder/ControlNet/Hypernetwork)のラベルとLoRA列見出しに、対応するノードが実際にワークフローへ含まれる場合のみ`wfm-model-label-active`クラスを付与。既存CSS`.wfm-form-group label { color: var(--wfm-text-secondary); }`の方が詳細度が高く単純なクラス指定では上書きされなかったため、注入CSS側のセレクタを`.wfm-form-group label.wfm-model-label-active`(および後続の折りたたみ対応でsummary要素向けセレクタも追加)として詳細度を確実に上回るよう実装。
|
||||
|
||||
**3. ControlNet/Hypernetwork折りたたみ化 + Text Encoder並び替え**: 「使用頻度が低い、意図的に利用するControlNet、Hypernetworkは折りたたみにし、最新モデルで主流のText EncoderをControlNetの上にしたい」とのフィードバックを受け、表示順をCheckpoint→VAE→Diffusion Model→**Text Encoder**→ControlNet→Hypernetworkに変更。ControlNet/Hypernetworkは`Image Editタブのプレースホルダ設定`(v0.3.97)で既に使われていた`<details>/<summary>` + `.wfm-settings-section`パターンをそのまま再利用し、デフォルト折りたたみ状態に。折りたたんだ状態でもハイライト色(#2の機能)は見出し文字に残るため、展開しなくても「このワークフローはControlNetを使っているか」が一目で分かる設計とした。続けて「LoRAもハイライト可能にしたい」との追加要望で、`templates/index.html`のLoRA列見出し`<span>`にid付与、`renderLoraPane()`でワークフローにLoRAノードが1件でもあればハイライトクラスをtoggleするよう対応。
|
||||
|
||||
**4. AI TOOLタブ: Thinking mode / Max tokens設定**: 「Thinking mode ON/OFF、Max tokens設定を追加したい」との要望を受け実装。Ollamaは`think`パラメータ(bool)と`options.num_predict`で標準対応しているが、LM Studio/Lemonade(OpenAI互換API)には標準的なthinking切替パラメータが存在しないため`max_tokens`のみ送信し、代わりにThinking mode OFF時は出力から`<think>...</think>`(`<thinking>...</thinking>`も)タグをクライアント側で正規表現除去する方式を採用。これはOllama自身がパラメータを無視した場合の保険としても機能し、バックエンドを問わず一貫した挙動になる。SPA(`callLLM`/`callChat`/`callVLM`)とサイドパネル(`aiCallLLM`/`aiCallChat`/`aiCallVLM`)の両方に`_applyGenOptions()`ヘルパーとタグ除去処理を追加し、Translation/Chat/TOOLS(VLM)/Wildcard生成の全呼び出し経路に反映。
|
||||
- **付随調査(見送り)**: 「Unsloth Desktopアプリのバックエンド追加は可能か」との質問を受け、公式ドキュメントを調査。OpenAI/Anthropic互換APIを同一ポート(既定`localhost:8888`)で提供する点は好条件だが、既存3バックエンドと異なり**ローカルアクセスであってもAPIキー認証が常に必須**(`Authorization: Bearer sk-unsloth-...`、無いと401)と判明。実装には新規UI(APIキー入力欄)が必要になるためユーザー判断で今回は見送り、将来の作業候補として`project_next_tasks`メモリに記録。
|
||||
|
||||
**ヘルプタブ更新**: `project_v0336_conventions`の慣習(index.html + i18n.js 3言語 + app.jsのhelpIdMapの3点セット)に従い、GenerateUI TabヘルプにModelタブの新しい表示順・折りたたみ挙動(`gen-5`更新)とハイライト色機能(新規`gen-28`)、AI TOOL TabヘルプにThinking mode/Max tokens(`ai-6`更新)、SettingsタブヘルプにModelタブハイライト色設定(新規`settings-14`)を追加。この作業中に、以前から`index.html`にだけ存在し`i18n.js`(英語含む全言語)に対応キーが用意されていなかったため実際には一度も表示されていなかった「LoRA Stack — Apply writes...」の説明文を発見・復旧(`app.js`の`el.textContent = t(key)`は未定義キーでもキー名をそのまま返してHTML側の内容を問答無用に上書きするため、i18n.js側の追加漏れは静かな表示欠落を引き起こす)。
|
||||
|
||||
**検証**: 全JSファイルについて`node --check`で構文確認。開発元git リポジトリと実行時`custom_nodes`フォルダは別実体のため([[project_dev_deploy_sync]]参照)、変更の都度`diff`で意図した差分のみであることを確認してから両フォルダへ同期、CRLF/LF不一致がないことも`file`コマンドで事前チェック。各機能追加後にユーザー自身が実機(ComfyUI再起動後)で動作確認し、フィードバックを都度反映する形で進行。
|
||||
|
||||
**How to apply**:
|
||||
1. SPA(`ai-tab.js`)とサイドパネル(`node_sets_menu.js`)のように、同じ機能が2箇所に重複実装されている場合(設定は共通の`localStorage`キーを共有していても、API呼び出しロジック自体は別ファイルにコピーされている)、片方だけ修正すると「サイドパネル版だけ直っていない」という見落としが起きる。ai-tab.js相当の修正が必要な要望が来たら、必ずnode_sets_menu.js側の対応箇所も同時に確認する。
|
||||
2. ローカルLLM APIへの指示は、user発話に埋め込んだ「〜だけ出力して」という指示より、system roleとして明示的に分離した方が小型モデルの指示追従率が上がる。それでも従わないモデルへの保険として、クライアント側での出力後処理(前置き除去、echo検知)を組み合わせるとより堅牢になる。
|
||||
3. 動的注入`<style>`タグでハイライト色のような「上書き用クラス」を追加する場合、既存CSSがelement+classなど複合セレクタで詳細度を持っていると単純なクラス指定では負ける。読み込み順に依存せず確実に上書きするには、注入側のセレクタも同等以上の詳細度(同じ複合セレクタ構造にクラスを足す)にする。
|
||||
4. 折りたたみ(`<details>`)と色ハイライトを組み合わせる設計は、「低頻度機能は隠したいが、関連情報があるかどうかは常に見えていてほしい」という要望に対する良い組み合わせになる——折りたたんだ見出し自体を色の表示場所として使う。
|
||||
5. i18n.jsの「3点セット」規約が守られていないと、HTML側にどれだけ内容を書いてもJS側の`t(key)`上書きで静かに消える。ヘルプ文言を追加・編集する際は、既存の該当キーが本当に全言語ブロックに存在するか(特に英語)を都度確認する。
|
||||
|
||||
---
|
||||
|
||||
## v0.3.97
|
||||
|
||||
### 機能追加・バグ修正: KREA-2ワークフロー対応 + Labタブ拡張(Clear/Get from Previous/Bypass/ライブ反映)+ Input Imageプレースホルダ機能
|
||||
|
||||
@@ -22,7 +22,7 @@ A comprehensive workflow, asset management, and generation UI plugin for [ComfyU
|
||||
- Built-in AI tools (translation and more)
|
||||
|
||||

|
||||

|
||||

|
||||
|
||||
## Screenshots
|
||||
|
||||
@@ -95,7 +95,8 @@ A comprehensive workflow, asset management, and generation UI plugin for [ComfyU
|
||||
- **Emphasis weight editing** (v0.3.70) — select text (or just place the cursor on a word/parenthesis block) and press Ctrl+↑/↓ to adjust `(text:weight)` by ±0.05, A1111/native-ComfyUI style; parentheses are removed automatically when the weight returns to 1.0
|
||||
- **Mask slot for inpainting** (v0.3.71) — LoadImage slots whose MASK output feeds an inpaint node (e.g. `VAEEncodeForInpaint`) show an additional Mask drop zone; when a mask is set, image + mask are composited into a single RGBA upload (mask baked into the alpha channel) matching ComfyUI's native alpha→MASK extraction — no extra nodes required
|
||||
- **Placeholder Image + Clear** (v0.3.97) — each LoadImage card gets a **Clear** button (resets that card's pending selection/preview back to empty; doesn't touch the workflow node itself) and a **Placeholder** button; a collapsible **Placeholder Image** section (collapsed by default) at the top of the Image sub-panel configures one shared default — **Color** mode (set Width/Height/Color, generates a solid-color PNG on the fly) or **Image** mode (drop one default image, reused as-is) — remembered across sessions; clicking a card's own Placeholder button immediately replaces just that card's image with the configured default, so multiple LoadImage cards can each be replaced individually with one click
|
||||
- **Model tab** — Checkpoint, VAE, LoRA, ControlNet, Diffusion Model (UNETLoader / UnetLoaderGGUF / LoaderGGUF), Text Encoder (CLIPLoader / DualCLIPLoader / ClipLoaderGGUF / DualClipLoaderGGUF — single clip: type + device; dual clip: two clip selectors + type + device), **Hypernetwork** (with Strength field) selectors with filter; Raw JSON on the right
|
||||
- **Model tab** — Checkpoint, VAE, Diffusion Model (UNETLoader / UnetLoaderGGUF / LoaderGGUF), and Text Encoder (CLIPLoader / DualCLIPLoader / ClipLoaderGGUF / DualClipLoaderGGUF — single clip: type + device; dual clip: two clip selectors + type + device) are always shown; **ControlNet** and **Hypernetwork** (with Strength field) are collapsed by default (v0.3.98) since they're used less often — click their header to expand; **LoRA** (Single/Stack tabs) has its own column; all selectors have a filter; Raw JSON on the right
|
||||
- **Highlight color** (v0.3.98) — the label of each model type above, plus the LoRA column header, turns a distinct color (default bright green, customizable in the Settings tab) when the loaded workflow actually contains that type of node; stays visible even while ControlNet/Hypernetwork are collapsed, so relevance is visible without expanding
|
||||
- **LoRA Stack Group toggle** (v0.3.92) — for `Lora Loader (LoraManager)` nodes, a **Use Stack Group** checkbox next to the Single/Stack tabs switches what the Stack tab shows and edits: OFF (the default whenever a workflow is loaded) displays and edits the LoRAs actually configured in the workflow's own Lora Manager node — correctly showing every LoRA it holds, not just the first one; ON switches to the LoRA group registered in the Models tab instead; toggling either direction clears the node's LoRA list first, then loads it from the newly selected source, and switching back to OFF restores exactly the list that was present when the workflow was loaded
|
||||
- **Settings tab** — KSampler and Latent Image side by side at 50% width each; Raw JSON on the right
|
||||
- **Always-visible Raw JSON** — edit the API-format JSON directly from any tab with syntax highlighting; Apply button reloads the workflow; built-in **search bar** (always shown) finds all matches as you type with count display (`3/12`); navigate with ↑/↓ buttons or Enter / Shift+Enter; Escape or ✕ clears; current match highlighted in orange, other matches in yellow
|
||||
@@ -224,6 +225,7 @@ An experimental batch generator: runs the workflow currently loaded in GenerateU
|
||||
- **Data Management** — export all plugin data (settings, metadata, prompts, etc.) to a single JSON file; import to restore data (useful when migrating or reinstalling); API keys are excluded from exports for security
|
||||
- **Text Size** — one slider (10–28 px) adjusts font size for all prompt and chat textareas at once: Generate UI positive/negative prompts, AI Assistant chat input, Preset prompts, Wildcard prompt and file editor, and Metadata prompt full preview; takes effect immediately and saved with Save Settings
|
||||
- **RAW JSON Colors** — customize the 6 syntax highlight colors for the Raw JSON editor in Generate UI: Default Text (base), Name/Scheduler (yellow), Title (pink), Width/Height (green), Prompt/Text (cyan), Image/File (red); changes apply immediately on color pick; Reset Defaults restores the original scheme; saved to `localStorage` under `wfm_settings.jsonColors` and applied on startup
|
||||
- **GenerateUI Model Tab Highlight** (v0.3.98) — pick the color used to mark Checkpoint/VAE/Diffusion Model/Text Encoder/ControlNet/Hypernetwork labels and the LoRA column header in GenerateUI's Model tab when the loaded workflow actually contains that type of node; default bright green; changes apply immediately, Reset Defaults restores it; saved to `localStorage` under `wfm_settings.modelTabActiveColor`
|
||||
- **Wildcard Integration** — link the WFS wildcard directory to ComfyUI-Impact-Pack's `wildcards/` directory (directory junction on Windows, symlink on other OS); existing WFS wildcard files are automatically migrated; requires [ComfyUI-Impact-Pack](https://github.com/ltdrdata/ComfyUI-Impact-Pack)
|
||||
- **G'MIC-Qt Integration** — set the path to `gmic_qt.exe` for the Image Edit Filter tool; download G'MIC-Qt Standalone from `gmic.eu/download.html` (Windows: `gmic_qt-win64.zip`), extract anywhere, then enter the full path to `gmic_qt.exe` and click Save
|
||||
- **Language** — English / Japanese / Chinese
|
||||
@@ -374,6 +376,7 @@ An experimental batch generator: runs the workflow currently loaded in GenerateU
|
||||
|
||||
- **4-pane layout** — Translation | Chat | TOOLS | Settings; all panes always visible simultaneously; no sub-tab switching required
|
||||
- **Translation pane** — translate text between Japanese, English, Chinese, or a custom Free language using Ollama, LM Studio, or Lemonade; language selectors with ⇄ swap button (swaps both language selectors and text content); selections saved automatically
|
||||
- **Reliability fix** (v0.3.98) — switched from a raw single-string prompt to a system-role chat request (Ollama `/api/chat`, LM Studio/Lemonade `/v1/chat/completions` with a `system` message) so small local models follow the "translation only" instruction more reliably; output is cleaned of common LLM preambles/quotes, and a warning toast appears if the model appears to have echoed the source text unchanged instead of translating it
|
||||
- **Chat pane** (v0.3.40) — multi-turn conversation with the LLM; full conversation history sent each turn for context; Enter to send, Shift+Enter for a newline; Clear button resets history; Ollama uses `/api/chat`, LM Studio and Lemonade use `/v1/chat/completions`
|
||||
- **Chat pane — image generation via Tool Calling** (v0.3.73) — if the selected model supports tool calling and the user asks for an image, a `generate_image` tool is invoked and the request runs through the same generation pipeline as the GenerateUI Generate button, with the result shown inline in the chat; an "Allow image generation" checkbox in the input row toggles this on/off (default on); the workflow used is the one currently loaded in GenerateUI, or a dedicated saved workflow configured in the Settings pane; models verified to work well for tool calling and vision: **Gemma 4:e4b**, **Qwen 3.5:9b**, **Ministral 3:3b** (Ollama)
|
||||
- **TOOLS pane (VLM)** — drop an image into the 110px drop zone, select a task (Describe Image / Create Prompt / Create Tags), and click Run to analyze with a vision model; result shown in the output area with a Copy button
|
||||
@@ -385,6 +388,7 @@ An experimental batch generator: runs the workflow currently loaded in GenerateU
|
||||
- **Settings pane** — choose backend (Ollama / LM Studio / Lemonade), set the API URL, test connection, select a model (with refresh button), and configure Free language names for translation source and destination
|
||||
- **Settings pane — Chat Image Generation** (v0.3.73) — "Use dedicated workflow" checkbox lets Chat-triggered image generation use a saved workflow of your choice instead of the one currently loaded in GenerateUI, without disturbing what's shown there (same pattern as Image Edit's Inpaint dedicated workflow)
|
||||
- **Settings pane — Chat I2I Generation** (v0.3.74) — same "Use dedicated workflow" pattern as above, but for image-to-image generation when an image is attached; independent of the text-to-image dedicated workflow setting
|
||||
- **Settings pane — Generation** (v0.3.98) — **Thinking mode** toggle (off by default; sends Ollama's `think` flag, and regardless of backend, any `<think>...</think>` block is stripped from the output when off so reasoning models don't clutter Translation/Chat/TOOLS results) and **Max tokens** (Ollama: `options.num_predict`; LM Studio/Lemonade: `max_tokens`; `0` = unlimited)
|
||||
- **Settings shared** — settings saved to `localStorage` under `wfm_ai_settings`; shared with the Library panel's AI TOOL tab so configuration is consistent across both interfaces
|
||||
- **Backend support** — Ollama (`/api/generate` for text, `/api/chat` for conversations, `/api/tags` for model list); LM Studio and Lemonade via the OpenAI-compatible API (`/v1/chat/completions`, `/v1/models`); VLM images sent as base64 (`images:[]` for Ollama, `image_url` content block for LM Studio/Lemonade)
|
||||
- **URL security** — backend URL validated via `new URL()` to enforce `http://` or `https://` scheme
|
||||
|
||||
+1
-1
@@ -1,7 +1,7 @@
|
||||
[project]
|
||||
name = "comfyui-workflow-studio"
|
||||
description = "A comprehensive workflow, asset management, and generation UI plugin for ComfyUI. Manage workflows, models, images, and prompts; batch generation and Image Feeder in the GenerateUI tab; library side panel with canvas drag & drop and AI tools."
|
||||
version = "0.3.97"
|
||||
version = "0.3.98"
|
||||
license = "MIT"
|
||||
requires-python = ">= 3.10"
|
||||
dependencies = ["jinja2"]
|
||||
|
||||
+108
-30
@@ -50,30 +50,58 @@ function fileToBase64(file) {
|
||||
});
|
||||
}
|
||||
|
||||
async function callVLM(url, backend, model, prompt, base64Image, mimeType) {
|
||||
// ============================================
|
||||
// Thinking mode / Max tokens (shared by callLLM / callChat / callVLM)
|
||||
// ============================================
|
||||
|
||||
// Ollamaは `think` (bool) と `options.num_predict` で制御できるが、
|
||||
// LM Studio/LemonadeはOpenAI互換APIのため max_tokens のみ標準対応。
|
||||
// thinking mode切替は非対応バックエンド/モデル向けに、出力からの
|
||||
// <think>タグ除去でも担保する(stripThinkingTagsを参照)。
|
||||
function _applyGenOptions(body, backend, settings) {
|
||||
const maxTokens = parseInt(settings?.maxTokens, 10);
|
||||
if (backend === "ollama") {
|
||||
body.think = !!settings?.thinkingMode;
|
||||
if (maxTokens > 0) body.options = { ...(body.options || {}), num_predict: maxTokens };
|
||||
} else if (maxTokens > 0) {
|
||||
body.max_tokens = maxTokens;
|
||||
}
|
||||
return body;
|
||||
}
|
||||
|
||||
function stripThinkingTags(text) {
|
||||
return (text || "")
|
||||
.replace(/<think>[\s\S]*?<\/think>/gi, "")
|
||||
.replace(/<thinking>[\s\S]*?<\/thinking>/gi, "")
|
||||
.trim();
|
||||
}
|
||||
|
||||
async function callVLM(url, backend, model, prompt, base64Image, mimeType, settings = {}) {
|
||||
if (backend === "ollama") {
|
||||
const res = await fetch(`${url}/api/generate`, {
|
||||
method: "POST",
|
||||
headers: { "Content-Type": "application/json" },
|
||||
body: JSON.stringify({ model, prompt, images: [base64Image], stream: false }),
|
||||
body: JSON.stringify(_applyGenOptions({ model, prompt, images: [base64Image], stream: false }, backend, settings)),
|
||||
});
|
||||
if (!res.ok) throw new Error(`HTTP ${res.status}`);
|
||||
return (await res.json()).response || "";
|
||||
const text = (await res.json()).response || "";
|
||||
return settings.thinkingMode ? text : stripThinkingTags(text);
|
||||
} else {
|
||||
const res = await fetch(`${url}/v1/chat/completions`, {
|
||||
method: "POST",
|
||||
headers: { "Content-Type": "application/json" },
|
||||
body: JSON.stringify({
|
||||
body: JSON.stringify(_applyGenOptions({
|
||||
model,
|
||||
messages: [{ role: "user", content: [
|
||||
{ type: "text", text: prompt },
|
||||
{ type: "image_url", image_url: { url: `data:${mimeType};base64,${base64Image}` } },
|
||||
]}],
|
||||
stream: false,
|
||||
}),
|
||||
}, backend, settings)),
|
||||
});
|
||||
if (!res.ok) throw new Error(`HTTP ${res.status}`);
|
||||
return (await res.json()).choices?.[0]?.message?.content || "";
|
||||
const text = (await res.json()).choices?.[0]?.message?.content || "";
|
||||
return settings.thinkingMode ? text : stripThinkingTags(text);
|
||||
}
|
||||
}
|
||||
|
||||
@@ -114,30 +142,32 @@ async function testConnection(url, backend) {
|
||||
return models.length;
|
||||
}
|
||||
|
||||
async function callLLM(url, backend, model, prompt) {
|
||||
async function callLLM(url, backend, model, prompt, settings = {}) {
|
||||
if (backend === "ollama") {
|
||||
const res = await fetch(`${url}/api/generate`, {
|
||||
method: "POST",
|
||||
headers: { "Content-Type": "application/json" },
|
||||
body: JSON.stringify({ model, prompt, stream: false }),
|
||||
body: JSON.stringify(_applyGenOptions({ model, prompt, stream: false }, backend, settings)),
|
||||
});
|
||||
if (!res.ok) throw new Error(`HTTP ${res.status}`);
|
||||
const data = await res.json();
|
||||
return data.response || "";
|
||||
const text = data.response || "";
|
||||
return settings.thinkingMode ? text : stripThinkingTags(text);
|
||||
} else {
|
||||
// LM Studio (OpenAI-compatible)
|
||||
const res = await fetch(`${url}/v1/chat/completions`, {
|
||||
method: "POST",
|
||||
headers: { "Content-Type": "application/json" },
|
||||
body: JSON.stringify({
|
||||
body: JSON.stringify(_applyGenOptions({
|
||||
model,
|
||||
messages: [{ role: "user", content: prompt }],
|
||||
stream: false,
|
||||
}),
|
||||
}, backend, settings)),
|
||||
});
|
||||
if (!res.ok) throw new Error(`HTTP ${res.status}`);
|
||||
const data = await res.json();
|
||||
return data.choices?.[0]?.message?.content || "";
|
||||
const text = data.choices?.[0]?.message?.content || "";
|
||||
return settings.thinkingMode ? text : stripThinkingTags(text);
|
||||
}
|
||||
}
|
||||
|
||||
@@ -145,11 +175,41 @@ async function callLLM(url, backend, model, prompt) {
|
||||
// Translation prompt builder
|
||||
// ============================================
|
||||
|
||||
function buildTranslationPrompt(text, srcLang, dstLang, settings) {
|
||||
const srcName = srcLang === "free" ? (settings.freeSrcLang || "Auto") : LANG_NAMES[srcLang];
|
||||
const dstName = dstLang === "free" ? (settings.freeDstLang || "English") : LANG_NAMES[dstLang];
|
||||
const fromPart = srcLang === "free" ? "" : `from ${srcName} `;
|
||||
return `Translate the following text ${fromPart}to ${dstName}. Output only the translated text, nothing else.\n\n${text}`;
|
||||
// Uses a system role (via callChat) rather than a single raw prompt string —
|
||||
// local models follow "translate only, no commentary" far more reliably
|
||||
// when it's a system instruction instead of buried inside the user text.
|
||||
function buildTranslationMessages(text, srcLang, dstLang, settings) {
|
||||
const srcName = srcLang === "free" ? (settings.freeSrcLang || "").trim() : LANG_NAMES[srcLang];
|
||||
const dstName = dstLang === "free" ? (settings.freeDstLang || "English").trim() || "English" : LANG_NAMES[dstLang];
|
||||
const fromPart = srcName ? ` from ${srcName}` : "";
|
||||
const system = `You are a professional translator. Translate the text the user sends${fromPart} into ${dstName}. `
|
||||
+ `Reply with ONLY the translated text and nothing else: no explanations, no notes, no preamble like "Here is the translation:", `
|
||||
+ `no surrounding quotes, and do not repeat or echo the original text.`;
|
||||
return [
|
||||
{ role: "system", content: system },
|
||||
{ role: "user", content: text },
|
||||
];
|
||||
}
|
||||
|
||||
// Strips common LLM wrapping/preamble artifacts that survive despite the system prompt.
|
||||
function cleanTranslationOutput(raw) {
|
||||
let s = (raw || "").trim();
|
||||
s = s.replace(/^(sure[,.]?\s*)?(here('|)s|here is)?\s*(the\s+)?translation\s*:?\s*/i, "");
|
||||
s = s.replace(/^(翻訳|译文|翻译)\s*[::]\s*/, "");
|
||||
if (s.length >= 2) {
|
||||
const first = s[0], last = s[s.length - 1];
|
||||
const pairs = { '"': '"', "'": "'", "“": "”", "「": "」" };
|
||||
if (pairs[first] === last) s = s.slice(1, -1).trim();
|
||||
}
|
||||
return s;
|
||||
}
|
||||
|
||||
// Heuristic: if the cleaned output is (near-)identical to the input, the
|
||||
// model most likely echoed the source instead of translating it.
|
||||
function looksUntranslated(input, output) {
|
||||
const normalize = (s) => s.trim().toLowerCase().replace(/[\s.,!?"'。、!?「」]/g, "");
|
||||
const a = normalize(input), b = normalize(output);
|
||||
return a.length > 0 && a === b;
|
||||
}
|
||||
|
||||
// ============================================
|
||||
@@ -217,11 +277,19 @@ function initTranslateTab() {
|
||||
outputEl.value = "";
|
||||
|
||||
try {
|
||||
const prompt = buildTranslationPrompt(text, srcLangSel.value, dstLangSel.value, settings);
|
||||
const result = await callLLM(url, backend, model, prompt);
|
||||
outputEl.value = result.trim();
|
||||
statusEl.textContent = t("aiStatusDone");
|
||||
statusEl.className = "wfm-ai-trans-status wfm-ai-status-ok";
|
||||
const messages = buildTranslationMessages(text, srcLangSel.value, dstLangSel.value, settings);
|
||||
const { content } = await callChat(url, backend, model, messages, undefined, settings);
|
||||
const cleaned = cleanTranslationOutput(content);
|
||||
outputEl.value = cleaned;
|
||||
|
||||
if (srcLangSel.value !== dstLangSel.value && looksUntranslated(text, cleaned)) {
|
||||
statusEl.textContent = t("aiStatusDone");
|
||||
statusEl.className = "wfm-ai-trans-status wfm-ai-status-error";
|
||||
showToast(t("aiToastMaybeNotTranslated"), "error");
|
||||
} else {
|
||||
statusEl.textContent = t("aiStatusDone");
|
||||
statusEl.className = "wfm-ai-trans-status wfm-ai-status-ok";
|
||||
}
|
||||
} catch (err) {
|
||||
statusEl.textContent = `${t("aiStatusConnectFail")}${err.message}`;
|
||||
statusEl.className = "wfm-ai-trans-status wfm-ai-status-error";
|
||||
@@ -288,6 +356,11 @@ function initSettingsTab() {
|
||||
const urlInput = document.getElementById("wfm-ai-backend-url");
|
||||
if (urlInput && saved.backendUrl) urlInput.value = saved.backendUrl;
|
||||
|
||||
const thinkingModeCheckbox = document.getElementById("wfm-ai-thinking-mode-checkbox");
|
||||
if (thinkingModeCheckbox) thinkingModeCheckbox.checked = !!saved.thinkingMode;
|
||||
const maxTokensInput = document.getElementById("wfm-ai-max-tokens");
|
||||
if (maxTokensInput && saved.maxTokens) maxTokensInput.value = saved.maxTokens;
|
||||
|
||||
// Switch URL to the new backend's default when the user changes backend
|
||||
backendRadios.forEach((r) => {
|
||||
r.addEventListener("change", () => {
|
||||
@@ -395,6 +468,8 @@ function initSettingsTab() {
|
||||
const chatGenDedicatedFilename = chatGenSelect?.value || "";
|
||||
const chatGenI2IDedicatedEnabled = !!chatGenI2ICheckbox?.checked;
|
||||
const chatGenI2IDedicatedFilename = chatGenI2ISelect?.value || "";
|
||||
const thinkingMode = !!thinkingModeCheckbox?.checked;
|
||||
const maxTokens = Math.max(0, parseInt(maxTokensInput?.value, 10) || 0);
|
||||
|
||||
if (url && !isValidBackendUrl(url)) {
|
||||
showToast(t("aiToastInvalidUrl"), "error");
|
||||
@@ -403,6 +478,7 @@ function initSettingsTab() {
|
||||
|
||||
saveAiSettings({
|
||||
backend, backendUrl: url, model, freeSrcLang, freeDstLang,
|
||||
thinkingMode, maxTokens,
|
||||
chatGenDedicatedEnabled, chatGenDedicatedFilename,
|
||||
chatGenI2IDedicatedEnabled, chatGenI2IDedicatedFilename,
|
||||
});
|
||||
@@ -453,10 +529,10 @@ function _formatMessagesForBackend(messages, backend) {
|
||||
});
|
||||
}
|
||||
|
||||
async function callChat(url, backend, model, messages, tools) {
|
||||
async function callChat(url, backend, model, messages, tools, settings = {}) {
|
||||
const formattedMessages = _formatMessagesForBackend(messages, backend);
|
||||
if (backend === "ollama") {
|
||||
const body = { model, messages: formattedMessages, stream: false };
|
||||
const body = _applyGenOptions({ model, messages: formattedMessages, stream: false }, backend, settings);
|
||||
if (tools) body.tools = tools;
|
||||
const res = await fetch(`${url}/api/chat`, {
|
||||
method: "POST",
|
||||
@@ -465,9 +541,10 @@ async function callChat(url, backend, model, messages, tools) {
|
||||
});
|
||||
if (!res.ok) throw new Error(`HTTP ${res.status}`);
|
||||
const message = (await res.json()).message || {};
|
||||
return { content: message.content || "", toolCalls: message.tool_calls || null };
|
||||
const content = message.content || "";
|
||||
return { content: settings.thinkingMode ? content : stripThinkingTags(content), toolCalls: message.tool_calls || null };
|
||||
} else {
|
||||
const body = { model, messages: formattedMessages, stream: false };
|
||||
const body = _applyGenOptions({ model, messages: formattedMessages, stream: false }, backend, settings);
|
||||
if (tools) { body.tools = tools; body.tool_choice = "auto"; }
|
||||
const res = await fetch(`${url}/v1/chat/completions`, {
|
||||
method: "POST",
|
||||
@@ -476,7 +553,8 @@ async function callChat(url, backend, model, messages, tools) {
|
||||
});
|
||||
if (!res.ok) throw new Error(`HTTP ${res.status}`);
|
||||
const message = (await res.json()).choices?.[0]?.message || {};
|
||||
return { content: message.content || "", toolCalls: message.tool_calls || null };
|
||||
const content = message.content || "";
|
||||
return { content: settings.thinkingMode ? content : stripThinkingTags(content), toolCalls: message.tool_calls || null };
|
||||
}
|
||||
}
|
||||
|
||||
@@ -909,7 +987,7 @@ function initChatTab() {
|
||||
const messagesToSend = skillPrompt
|
||||
? [{ role: "system", content: skillPrompt }, ...chatHistory]
|
||||
: chatHistory;
|
||||
const reply = await callChat(url, backend, model, messagesToSend, tools);
|
||||
const reply = await callChat(url, backend, model, messagesToSend, tools, settings);
|
||||
const toolCall = reply.toolCalls?.find(
|
||||
(tc) => (tc.function?.name || tc.name) === "generate_image"
|
||||
);
|
||||
@@ -1069,7 +1147,7 @@ function initVlmTab() {
|
||||
const prompt = `Generate ${count} wildcard entries for the category "${name}". Output only plain text in English, one entry per line, no numbers, no markdown, no asterisks, no bold, nothing else.`;
|
||||
|
||||
try {
|
||||
const result = await callLLM(url, backend, model, prompt);
|
||||
const result = await callLLM(url, backend, model, prompt, settings);
|
||||
resultEl.value = result.trim()
|
||||
.split("\n")
|
||||
.map(l => l.replace(/\*\*/g, "").replace(/^\*\s*/, "").replace(/^\d+\.\s*/, "").trim())
|
||||
@@ -1095,7 +1173,7 @@ function initVlmTab() {
|
||||
resultEl.value = "";
|
||||
|
||||
try {
|
||||
const result = await callVLM(url, backend, model, VLM_PROMPTS[task], _toolsImage.base64, _toolsImage.mimeType);
|
||||
const result = await callVLM(url, backend, model, VLM_PROMPTS[task], _toolsImage.base64, _toolsImage.mimeType, settings);
|
||||
resultEl.value = result.trim();
|
||||
statusEl.textContent = t("aiStatusDone");
|
||||
statusEl.className = "wfm-ai-trans-status wfm-ai-status-ok";
|
||||
|
||||
+12
-2
@@ -364,6 +364,14 @@ function applyI18nToHtml() {
|
||||
}
|
||||
const aiModelRefreshBtn = document.getElementById("wfm-ai-model-refresh-btn");
|
||||
if (aiModelRefreshBtn) aiModelRefreshBtn.textContent = t("aiSettingsRefreshBtn");
|
||||
const aiGenerationTitle = document.getElementById("wfm-ai-settings-generation-title");
|
||||
if (aiGenerationTitle) aiGenerationTitle.textContent = t("aiSettingsGenerationTitle");
|
||||
const aiThinkingModeLabel = document.getElementById("wfm-ai-thinking-mode-label");
|
||||
if (aiThinkingModeLabel) aiThinkingModeLabel.textContent = t("aiSettingsThinkingMode");
|
||||
const aiMaxTokensLabel = document.getElementById("wfm-ai-max-tokens-label");
|
||||
if (aiMaxTokensLabel) aiMaxTokensLabel.textContent = t("aiSettingsMaxTokens");
|
||||
const aiMaxTokensInput = document.getElementById("wfm-ai-max-tokens");
|
||||
if (aiMaxTokensInput) aiMaxTokensInput.placeholder = t("aiSettingsMaxTokensPlaceholder");
|
||||
const aiFreeLangTitle = document.getElementById("wfm-ai-settings-freelang-title");
|
||||
if (aiFreeLangTitle) aiFreeLangTitle.textContent = t("aiSettingsFreeLang");
|
||||
const aiFreeSrcLabel = document.getElementById("wfm-ai-free-src-label");
|
||||
@@ -446,6 +454,7 @@ function applyI18nToHtml() {
|
||||
"wfm-help-gen-22": "helpGen22", "wfm-help-gen-23": "helpGen23",
|
||||
"wfm-help-gen-24": "helpGen24", "wfm-help-gen-25": "helpGen25",
|
||||
"wfm-help-gen-26": "helpGen26", "wfm-help-gen-27": "helpGen27",
|
||||
"wfm-help-gen-28": "helpGen28",
|
||||
"wfm-help-feeder-title": "helpFeederTitle",
|
||||
"wfm-help-feeder-desc": "helpFeederDesc",
|
||||
"wfm-help-feeder-imgloop-title": "helpFeederImgloopTitle",
|
||||
@@ -495,7 +504,7 @@ function applyI18nToHtml() {
|
||||
"wfm-help-settings-7": "helpSettings7", "wfm-help-settings-8": "helpSettings8",
|
||||
"wfm-help-settings-9": "helpSettings9", "wfm-help-settings-10": "helpSettings10",
|
||||
"wfm-help-settings-11": "helpSettings11", "wfm-help-settings-12": "helpSettings12",
|
||||
"wfm-help-settings-13": "helpSettings13",
|
||||
"wfm-help-settings-13": "helpSettings13", "wfm-help-settings-14": "helpSettings14",
|
||||
"wfm-help-nodes-1": "helpNodes1", "wfm-help-nodes-2": "helpNodes2",
|
||||
"wfm-help-nodes-3": "helpNodes3", "wfm-help-nodes-4": "helpNodes4",
|
||||
"wfm-help-nodes-5": "helpNodes5", "wfm-help-nodes-6": "helpNodes6",
|
||||
@@ -756,7 +765,7 @@ import { initWorkflowTab } from "./workflow-tab.js";
|
||||
import { initNodesTab } from "./nodes-tab.js";
|
||||
import { initGenerateTab } from "./generate-tab.js";
|
||||
import { initPromptTab } from "./prompt-tab.js";
|
||||
import { initSettingsTab, applyTheme, getSavedTheme, applyTextareaFontSize, applyJsonColors } from "./settings-tab.js";
|
||||
import { initSettingsTab, applyTheme, getSavedTheme, applyTextareaFontSize, applyJsonColors, applyModelTabActiveColor } from "./settings-tab.js";
|
||||
import { initModelsTab } from "./models-tab.js";
|
||||
import { initGalleryTab } from "./gallery-tab.js";
|
||||
import { initMetadataTab } from "./metadata-tab.js";
|
||||
@@ -772,6 +781,7 @@ try {
|
||||
const _s = getSettings();
|
||||
if (_s.textareaFontSize) applyTextareaFontSize(_s.textareaFontSize);
|
||||
applyJsonColors(_s.jsonColors);
|
||||
applyModelTabActiveColor(_s.modelTabActiveColor);
|
||||
} catch {}
|
||||
|
||||
function initHelpTab() {
|
||||
|
||||
+63
-39
@@ -616,10 +616,14 @@ export const comfyEditor = {
|
||||
const el = document.getElementById(containerId);
|
||||
if (!el) return;
|
||||
|
||||
const sections = [
|
||||
// 使用頻度の高い順: Checkpoint/VAE/Diffusion Model/Text Encoder は常時表示。
|
||||
// ControlNet/Hypernetwork は意図的に使う場面が限られるため折りたたみにする。
|
||||
const mainSections = [
|
||||
{ label: "Checkpoint", key: "checkpoints", nodes: analysis.checkpoint_nodes, inputKey: "ckpt_name" },
|
||||
{ label: "VAE", key: "vaes", nodes: analysis.vae_nodes, inputKey: "vae_name" },
|
||||
{ label: "Diffusion Model", key: "diffusionModels", nodes: analysis.diffusion_model_nodes, inputKey: "unet_name" },
|
||||
];
|
||||
const collapsibleSections = [
|
||||
{ label: "ControlNet", key: "controlNets", nodes: analysis.controlnet_nodes, inputKey: "control_net_name" },
|
||||
{
|
||||
label: "Hypernetwork", key: "hypernetworks", nodes: analysis.hypernetwork_nodes, inputKey: "hypernetwork_name",
|
||||
@@ -627,46 +631,65 @@ export const comfyEditor = {
|
||||
},
|
||||
];
|
||||
|
||||
el.innerHTML = sections
|
||||
.map((s) => {
|
||||
const models = this.models[s.key] || [];
|
||||
const currentVal = s.nodes?.[0]?.[s.inputKey] || "";
|
||||
const isMissing = !!currentVal && !models.includes(currentVal);
|
||||
const targetOpts = s.nodes
|
||||
.map((n) => `<option value="${n.id}">ID:${n.id} (${n.title})</option>`)
|
||||
.join("");
|
||||
const extrasHtml = (s.extras || []).map((ex) => {
|
||||
const curVal = s.nodes?.[0]?.[ex.inputKey] ?? ex.defaultVal;
|
||||
return `<div style="display:flex;gap:6px;align-items:center;margin-top:4px;">
|
||||
<label style="font-size:12px;white-space:nowrap;color:var(--wfm-text-secondary);">${ex.label}</label>
|
||||
<input type="${ex.type}" class="wfm-input wfm-model-extra" id="wfm-model-${s.key}-${ex.inputKey}"
|
||||
data-key="${s.key}" data-input-key="${ex.inputKey}"
|
||||
value="${curVal}" step="${ex.step}" min="${ex.min}" max="${ex.max}"
|
||||
style="width:80px;">
|
||||
</div>`;
|
||||
}).join("");
|
||||
const renderSectionBody = (s) => {
|
||||
const models = this.models[s.key] || [];
|
||||
const currentVal = s.nodes?.[0]?.[s.inputKey] || "";
|
||||
const isMissing = !!currentVal && !models.includes(currentVal);
|
||||
const targetOpts = s.nodes
|
||||
.map((n) => `<option value="${n.id}">ID:${n.id} (${n.title})</option>`)
|
||||
.join("");
|
||||
const extrasHtml = (s.extras || []).map((ex) => {
|
||||
const curVal = s.nodes?.[0]?.[ex.inputKey] ?? ex.defaultVal;
|
||||
return `<div style="display:flex;gap:6px;align-items:center;margin-top:4px;">
|
||||
<label style="font-size:12px;white-space:nowrap;color:var(--wfm-text-secondary);">${ex.label}</label>
|
||||
<input type="${ex.type}" class="wfm-input wfm-model-extra" id="wfm-model-${s.key}-${ex.inputKey}"
|
||||
data-key="${s.key}" data-input-key="${ex.inputKey}"
|
||||
value="${curVal}" step="${ex.step}" min="${ex.min}" max="${ex.max}"
|
||||
style="width:80px;">
|
||||
</div>`;
|
||||
}).join("");
|
||||
|
||||
return `
|
||||
<div class="wfm-form-group" style="border-bottom:1px solid var(--wfm-border);padding-bottom:12px;">
|
||||
<label>${s.label}</label>
|
||||
<div class="wfm-search-wrap" style="margin-bottom:4px;width:100%;">
|
||||
<input type="text" class="wfm-input wfm-search-input wfm-model-filter" id="wfm-model-${s.key}-filter" placeholder="Filter..." data-target="wfm-model-${s.key}">
|
||||
<button type="button" class="wfm-search-clear-btn" id="wfm-model-${s.key}-filter-clear" title="Clear search">✕</button>
|
||||
</div>
|
||||
<select class="wfm-select ${isMissing ? "wfm-select-missing" : ""}" id="wfm-model-${s.key}" style="margin-bottom:4px;">
|
||||
${isMissing ? `<option value="${escapeHtml(currentVal)}" selected>⚠ ${escapeHtml(currentVal)} (${t("modelNotFound")})</option>` : ""}
|
||||
${models.map((m) => `<option value="${m}" ${m === currentVal ? "selected" : ""}>${m}</option>`).join("")}
|
||||
</select>
|
||||
${isMissing ? `<div class="wfm-model-missing-hint">${t("modelNotFoundHint")}</div>` : ""}
|
||||
${extrasHtml}
|
||||
<div style="display:flex;gap:8px;align-items:center;margin-top:4px;">
|
||||
<select class="wfm-select" id="wfm-model-${s.key}-target" style="flex:1;">${targetOpts}</select>
|
||||
<button class="wfm-btn wfm-btn-sm wfm-model-apply" data-key="${s.key}" data-input="${s.inputKey}" title="Apply (Alt+Click: Apply & Generate)">Apply</button>
|
||||
</div>
|
||||
return `
|
||||
<div class="wfm-search-wrap" style="margin-bottom:4px;width:100%;">
|
||||
<input type="text" class="wfm-input wfm-search-input wfm-model-filter" id="wfm-model-${s.key}-filter" placeholder="Filter..." data-target="wfm-model-${s.key}">
|
||||
<button type="button" class="wfm-search-clear-btn" id="wfm-model-${s.key}-filter-clear" title="Clear search">✕</button>
|
||||
</div>
|
||||
<select class="wfm-select ${isMissing ? "wfm-select-missing" : ""}" id="wfm-model-${s.key}" style="margin-bottom:4px;">
|
||||
${isMissing ? `<option value="${escapeHtml(currentVal)}" selected>⚠ ${escapeHtml(currentVal)} (${t("modelNotFound")})</option>` : ""}
|
||||
${models.map((m) => `<option value="${m}" ${m === currentVal ? "selected" : ""}>${m}</option>`).join("")}
|
||||
</select>
|
||||
${isMissing ? `<div class="wfm-model-missing-hint">${t("modelNotFoundHint")}</div>` : ""}
|
||||
${extrasHtml}
|
||||
<div style="display:flex;gap:8px;align-items:center;margin-top:4px;">
|
||||
<select class="wfm-select" id="wfm-model-${s.key}-target" style="flex:1;">${targetOpts}</select>
|
||||
<button class="wfm-btn wfm-btn-sm wfm-model-apply" data-key="${s.key}" data-input="${s.inputKey}" title="Apply (Alt+Click: Apply & Generate)">Apply</button>
|
||||
</div>
|
||||
`;
|
||||
})
|
||||
.join("") + '<div class="wfm-form-group" id="wfm-te-section" style="border-bottom:1px solid var(--wfm-border);padding-bottom:12px;"><label>Text Encoder</label></div>';
|
||||
};
|
||||
|
||||
const mainHtml = mainSections.map((s) => {
|
||||
const hasNodes = (s.nodes?.length || 0) > 0;
|
||||
return `
|
||||
<div class="wfm-form-group" style="border-bottom:1px solid var(--wfm-border);padding-bottom:12px;">
|
||||
<label class="${hasNodes ? "wfm-model-label-active" : ""}">${s.label}</label>
|
||||
${renderSectionBody(s)}
|
||||
</div>`;
|
||||
}).join("");
|
||||
|
||||
const teHtml = '<div class="wfm-form-group" id="wfm-te-section" style="border-bottom:1px solid var(--wfm-border);padding-bottom:12px;"><label>Text Encoder</label></div>';
|
||||
|
||||
const collapsibleHtml = collapsibleSections.map((s) => {
|
||||
const hasNodes = (s.nodes?.length || 0) > 0;
|
||||
return `
|
||||
<details class="wfm-settings-section">
|
||||
<summary class="wfm-settings-summary ${hasNodes ? "wfm-model-label-active" : ""}">${s.label}</summary>
|
||||
<div class="wfm-form-group" style="margin-bottom:0;">
|
||||
${renderSectionBody(s)}
|
||||
</div>
|
||||
</details>`;
|
||||
}).join("");
|
||||
|
||||
el.innerHTML = mainHtml + teHtml + collapsibleHtml;
|
||||
|
||||
// Filter inputs
|
||||
el.querySelectorAll(".wfm-model-filter").forEach((input) => {
|
||||
@@ -724,6 +747,7 @@ export const comfyEditor = {
|
||||
if (!el) return;
|
||||
|
||||
const loraNodes = analysis.lora_nodes || [];
|
||||
document.getElementById("wfm-gen-lora-col-title")?.classList.toggle("wfm-model-label-active", loraNodes.length > 0);
|
||||
const loras = this.models.loras || [];
|
||||
const defaultStackTarget = (loraNodes.find((n) => n.is_lora_manager) || loraNodes[0])?.id;
|
||||
const hasLoraManager = loraNodes.some((n) => n.is_lora_manager);
|
||||
@@ -2092,7 +2116,7 @@ async function _initTextEncoderSection(analysis, textEncoders, containerEl) {
|
||||
const deviceOpts = ["default", "cpu"].map(d => `<option value="${d}" ${d === currentDevice ? "selected" : ""}>${d}</option>`).join("");
|
||||
|
||||
section.innerHTML = `
|
||||
<label>Text Encoder</label>
|
||||
<label class="wfm-model-label-active">Text Encoder</label>
|
||||
<input type="text" class="wfm-input" id="wfm-te-filter" placeholder="Filter..." style="margin-bottom:4px;">
|
||||
<select class="wfm-select" id="wfm-te-clip1" style="margin-bottom:4px;">${clip1Opts}</select>
|
||||
${isDual ? `<select class="wfm-select" id="wfm-te-clip2" style="margin-bottom:4px;">${clip2Opts}</select>` : ""}
|
||||
|
||||
+36
-6
@@ -319,6 +319,9 @@ const LANGUAGES = {
|
||||
jsonColorCyan: "Prompts / Text",
|
||||
jsonColorRed: "Images / Files",
|
||||
jsonColorReset: "Reset to defaults",
|
||||
modelTabColorSection: "GenerateUI Model Tab Highlight",
|
||||
modelTabColorLabel: "Color for items present in the workflow",
|
||||
modelTabColorHint: "In the GenerateUI Model tab, labels for items actually present in the loaded workflow (Checkpoint, VAE, etc.) are shown in this color.",
|
||||
|
||||
// -- Prompt tab --
|
||||
tabPrompt: "Prompt",
|
||||
@@ -682,6 +685,10 @@ const LANGUAGES = {
|
||||
aiSettingsModelSection: "Model",
|
||||
aiSettingsRefreshBtn: "Refresh",
|
||||
aiSettingsModelPlaceholder: "-- Select Model --",
|
||||
aiSettingsGenerationTitle: "Generation",
|
||||
aiSettingsThinkingMode: "Thinking mode",
|
||||
aiSettingsMaxTokens: "Max tokens",
|
||||
aiSettingsMaxTokensPlaceholder: "0 = unlimited",
|
||||
aiSettingsFreeLang: "Free Language",
|
||||
aiSettingsInputLang: "Input",
|
||||
aiSettingsOutputLang: "Output",
|
||||
@@ -697,6 +704,7 @@ const LANGUAGES = {
|
||||
aiStatusConnectOk: "Connected",
|
||||
aiStatusConnectFail: "Connection failed: ",
|
||||
aiToastTransFailed: "Translation failed: ",
|
||||
aiToastMaybeNotTranslated: "The model returned the text unchanged — it may not have translated it. Try a different model.",
|
||||
aiToastNoCopyText: "No text to copy",
|
||||
aiToastCopied: "Copied",
|
||||
aiToastNoModels: "No models found",
|
||||
@@ -752,7 +760,8 @@ const LANGUAGES = {
|
||||
helpGen27: "Input tab → Image sub-panel — each LoadImage card now has three buttons: Apply (unchanged), Clear (resets that card's pending selection/preview back to empty; doesn't touch the workflow node itself, since LoadImage always needs some valid filename), and Placeholder. A Placeholder Image section — collapsed by default since it's a low-frequency feature, click to expand — sits at the top of the panel with a shared default: Color mode (set Width/Height/Color; generates a solid-color PNG on the fly) or Image mode (drop one default image, reused as-is). Clicking a card's own Placeholder button immediately replaces just that card's image with the configured default — no separate Apply click needed, and with several LoadImage cards each can be replaced individually. The Placeholder config (mode, size, color, default image) is remembered across sessions",
|
||||
helpGen3: "5-tab layout — Input (Prompt + Image), Model, Settings, Feeder, Batch — Input/Model/Settings each include a Raw JSON column for instant preview and direct editing",
|
||||
helpGen4: "Input tab: inner tabs — Prompt and Image (drag-and-drop upload); Prompt tab has Positive Prompt, Negative Prompt, and an Embeddings selector at the bottom — select a model, set Weight (default 1.0), then click Paste to insert (embedding:Name:weight) at the cursor position of the last focused textarea (defaults to Positive if neither is focused); a ↺ button next to each Apply button reverts the textarea to the node's currently applied value; select text (or just place the cursor on a word/parenthesis block) and press Ctrl+↑/↓ to adjust its emphasis weight in (text:weight) syntax by ±0.05 — the parentheses are removed automatically when the weight returns to 1.0",
|
||||
helpGen5: "Model tab: Checkpoint, VAE, LoRA (Single/Stack tabs), ControlNet, Diffusion Model (UNETLoader / UnetLoaderGGUF / LoaderGGUF), Text Encoder (CLIPLoader / DualCLIPLoader / ClipLoaderGGUF / DualClipLoaderGGUF — type and device selectors; DualCLIP shows two clip selectors), Hypernetwork (with Strength field) selectors with filter",
|
||||
helpGen5: "Model tab: Checkpoint, VAE, Diffusion Model (UNETLoader / UnetLoaderGGUF / LoaderGGUF), Text Encoder (CLIPLoader / DualCLIPLoader / ClipLoaderGGUF / DualClipLoaderGGUF — type and device selectors; DualCLIP shows two clip selectors) are always shown; ControlNet and Hypernetwork (with Strength field) are collapsed by default since they're used less often — click their header to expand; LoRA (Single/Stack tabs) has its own column to the right; all selectors have a filter; LoRA Stack — Apply writes all active LoRAs to the target node and simultaneously inserts LoRA syntax and CivitAI trigger words into the Positive prompt; re-applying updates the prompt by removing inactive LoRAs and re-adding active ones",
|
||||
helpGen28: "Model tab — Highlight color: the label of each model type (Checkpoint, VAE, Diffusion Model, Text Encoder, ControlNet, Hypernetwork) and the LoRA column header turn a distinct color (default bright green) when the loaded workflow actually contains that type of node; this stays visible even while ControlNet/Hypernetwork are collapsed, so you can tell at a glance whether they're in use without expanding them; the color is customizable in the Settings tab",
|
||||
helpGen6: "Settings tab: KSampler and Latent Image side by side, each at 50% width",
|
||||
helpGen7: "Raw JSON always visible in the right column of each tab — edit directly and Apply to reload the workflow; if the loaded workflow has Bypass or Mute nodes, a note listing them appears above the search bar (Bypass nodes are excluded with their wires reconnected around them; Mute nodes are excluded with no reconnection, matching ComfyUI itself)",
|
||||
helpGen8: "Seed control: random, fixed, increment, decrement",
|
||||
@@ -855,6 +864,7 @@ const LANGUAGES = {
|
||||
helpSettings11: "Text Size: one slider (10–28 px) adjusts font size for all prompt and chat textareas at once — Generate UI positive/negative prompts, AI Assistant chat input, Preset positive/negative, Wildcard prompt and file editor, and Metadata prompt full preview; takes effect immediately and saved with Save Settings",
|
||||
helpSettings12: "RAW JSON Colors: customize the 6 syntax highlight colors for the Raw JSON editor in Generate UI — Default Text, Name/Scheduler, Title, Width/Height, Prompt/Text, Image/File; changes apply immediately on color pick; Reset Defaults restores the original color scheme",
|
||||
helpSettings13: "CivitAI Host: choose between civitai.com (SFW only) and civitai.red (unrestricted) — determines which site opens when clicking a model link in the Models tab CivitAI panel",
|
||||
helpSettings14: "GenerateUI Model Tab Highlight: pick the color used to mark Checkpoint/VAE/Diffusion Model/Text Encoder/ControlNet/Hypernetwork labels and the LoRA column header in GenerateUI's Model tab when the loaded workflow actually contains that type of node; default bright green; changes apply immediately, Reset Defaults restores it",
|
||||
helpNodes1: "Node Browser: search by name/description/alias/tag, filter by category, package, tag, group, favorites",
|
||||
helpNodes2: "Card and Table view modes with pagination",
|
||||
helpNodes3: "Favorite nodes and add custom tags for organization",
|
||||
@@ -936,7 +946,7 @@ const LANGUAGES = {
|
||||
helpAi2: "Translation pane: translate between Japanese, English, Chinese, or Free language; ⇄ swap button swaps both language selectors and text content; selections are saved automatically",
|
||||
helpAi3: "Chat pane: multi-turn conversation with the LLM; Enter to send a message, Shift+Enter for a newline; Clear button resets the conversation history; if the model supports tool calling and asks for an image (\"generate a cat picture\", etc.) it will run generation on the workflow currently loaded in GenerateUI and show the result inline; the \"Allow image generation\" checkbox in the pane header toggles this on/off; if an image is attached (dropped in the TOOLS pane) it is sent to the LLM as vision input and, when generate_image is called, used as the base for image-to-image generation instead of text-to-image; models verified to work well for tool calling and vision: Gemma 4:e4b, Qwen 3.5:9b, Ministral 3:3b (Ollama)",
|
||||
helpAi4: "TOOLS pane — VLM: drop an image (PNG/JPG/WebP) and select a task (Describe Image, Create Prompt, Create Tags), click Run; or select \"Create wildcards\", enter a category name and count to generate plain-text entries one per line; the dropped image is also shared with the Chat pane as its attachment (shown there and used for image-to-image generation) until cleared with the ✕ button",
|
||||
helpAi6: "Settings pane: choose backend (Ollama / LM Studio / Lemonade), set API URL, test connection, select model, configure Free language names; Chat Image Generation section lets you enable a dedicated workflow for Chat-triggered text-to-image generation instead of the one currently loaded in GenerateUI; Chat I2I Generation section is the same but for image-to-image generation when an image is attached; settings are shared with the Library panel's A tab via localStorage",
|
||||
helpAi6: "Settings pane: choose backend (Ollama / LM Studio / Lemonade), set API URL, test connection, select model; Generation section — Thinking mode toggle (off by default; Ollama receives a think flag, and regardless of backend any <think>...</think> block is stripped from the output when off, so reasoning models don't clutter Translation/Chat/TOOLS results) and Max tokens (Ollama: num_predict, LM Studio/Lemonade: max_tokens; 0 = unlimited); configure Free language names; Chat Image Generation section lets you enable a dedicated workflow for Chat-triggered text-to-image generation instead of the one currently loaded in GenerateUI; Chat I2I Generation section is the same but for image-to-image generation when an image is attached; settings are shared with the Library panel's A tab via localStorage",
|
||||
helpAi7: "Chat pane — SVG: the LLM can also produce SVG graphics directly as text, with no ComfyUI workflow involved; ask it to draw something in SVG and, if the reply contains SVG code, it is rendered inline as a preview with a \"Save to Gallery\" button that writes it as a .svg file to the Gallery tab's output folder; this does not use tool calling, so it works with any model regardless of tool-calling support",
|
||||
helpAi8: "Chat pane — Skills: pick a skill from the dropdown above the message list to prepend its instructions to the conversation as a system prompt (not shown in the visible chat log); use the pencil button to open the skill manager and create, edit, or delete .md skill files, stored in user/default/Workflow-Studio/ai_skills/; a file's first line block between --- markers can define name/description shown in the dropdown; two starter skills are included by default: \"SVG Icon Generator\" (decomposes the subject into parts before drawing, for more recognizable SVG output) and \"Skill Creator\" (interviews you conversationally and drafts a new skill file, ready to save); when a reply contains a ```skill code block, a \"Save as new skill\" button appears below it that opens the manager with the draft pre-filled",
|
||||
helpAi9: "Tips for writing your own skills: give a clear frontmatter name/description so it reads well in the dropdown and tooltip; state the exact output format you want and, if there is a deliverable to extract later (like SVG or another skill file), wrap it in a distinctive fenced code block; write concrete, imperative instructions rather than vague ones - smaller local models follow explicit rules and a short worked example far better than general guidance; state hard constraints (forbidden content, security, length limits) explicitly rather than assuming they are implied; when in doubt, select the \"Skill Creator\" skill in the dropdown and describe what you want - it will interview you and draft the file for you",
|
||||
@@ -1562,6 +1572,9 @@ const LANGUAGES = {
|
||||
jsonColorCyan: "プロンプト / テキスト",
|
||||
jsonColorRed: "画像 / ファイル",
|
||||
jsonColorReset: "デフォルトに戻す",
|
||||
modelTabColorSection: "GenerateUI Modelタブ ハイライト色",
|
||||
modelTabColorLabel: "ワークフローに含まれる項目の色",
|
||||
modelTabColorHint: "GenerateUIのModelタブで、読み込んだワークフローに実際に含まれる項目(Checkpoint / VAE など)のラベルをこの色で表示します。",
|
||||
|
||||
// -- Prompt tab --
|
||||
tabPrompt: "プロンプト",
|
||||
@@ -1925,6 +1938,10 @@ const LANGUAGES = {
|
||||
aiSettingsModelSection: "モデル選択",
|
||||
aiSettingsRefreshBtn: "更新",
|
||||
aiSettingsModelPlaceholder: "-- モデルを選択 --",
|
||||
aiSettingsGenerationTitle: "生成設定",
|
||||
aiSettingsThinkingMode: "Thinking mode(思考モード)",
|
||||
aiSettingsMaxTokens: "Max tokens(最大トークン数)",
|
||||
aiSettingsMaxTokensPlaceholder: "0 = 無制限",
|
||||
aiSettingsFreeLang: "Free 言語設定",
|
||||
aiSettingsInputLang: "入力言語",
|
||||
aiSettingsOutputLang: "翻訳言語",
|
||||
@@ -1940,6 +1957,7 @@ const LANGUAGES = {
|
||||
aiStatusConnectOk: "接続成功",
|
||||
aiStatusConnectFail: "接続失敗: ",
|
||||
aiToastTransFailed: "翻訳に失敗しました: ",
|
||||
aiToastMaybeNotTranslated: "モデルが原文をそのまま返しました。翻訳されていない可能性があります。別のモデルをお試しください。",
|
||||
aiToastNoCopyText: "コピーするテキストがありません",
|
||||
aiToastCopied: "コピーしました",
|
||||
aiToastNoModels: "モデルが見つかりません",
|
||||
@@ -1995,7 +2013,8 @@ const LANGUAGES = {
|
||||
helpGen27: "Inputタブ → Imageサブパネル — 各LoadImageカードにApply(変更なし)・Clear・Placeholderの3ボタンを追加。Clearはそのカードの保留中の選択・プレビューを空に戻す(LoadImageは常に有効なファイル名が必須なため、ワークフローノード自体には触れない)。パネル上部の「Placeholder Image」セクション(使用頻度が低い機能のためデフォルトで折りたたみ、クリックで展開)で共通のデフォルトを設定できる:Colorモード(幅・高さ・色を指定し、その場で単色PNGを生成)またはImageモード(デフォルト画像を1枚ドロップし、そのまま再利用)。各カード自身のPlaceholderボタンを押すと、そのカードの画像だけを設定済みデフォルトへ即座に差し替える(別途Applyクリック不要)— LoadImageが複数ある場合も個別に差し替え可能。Placeholder設定(モード・サイズ・色・デフォルト画像)はセッションをまたいで記憶される",
|
||||
helpGen3: "5タブ構成 — Input(プロンプト+画像)、Model、Settings、Feeder、Batch — Input/Model/SettingsタブにはRaw JSON列を常時表示",
|
||||
helpGen4: "Inputタブ:Prompt・Imageの内部タブ切り替え;PromptタブにはPositive Prompt・Negative Promptに加え最下部にEmbeddingsセレクタを表示(Filter+セレクト+Weight入力+Pasteボタン)— Pasteは最後にフォーカスしたテキストエリアのカーソル位置に(embedding:Name:weight)を挿入(未フォーカス時はPositiveに追加);各Applyボタンの左隣の↺ボタンはテキストエリアをノードの現在の適用値に戻す;テキストを選択(または単語・括弧内にカーソルを置くだけでも可)してCtrl+↑/↓を押すと(text:weight)形式で強調weightを±0.05単位で調整(weightが1.0に戻ると括弧を自動解除)",
|
||||
helpGen5: "Modelタブ:Checkpoint/VAE/LoRA(Single/Stackタブ)/ControlNet/Diffusion Model(UNETLoader / UnetLoaderGGUF / LoaderGGUF対応)/Text Encoder(CLIPLoader / DualCLIPLoader / ClipLoaderGGUF / DualClipLoaderGGUF対応 — typeおよびdevice選択付き、DualCLIPは2段モデル選択)/Hypernetwork(Strength入力付き)のセレクタ(フィルタ検索付き)",
|
||||
helpGen5: "Modelタブ:Checkpoint/VAE/Diffusion Model(UNETLoader / UnetLoaderGGUF / LoaderGGUF対応)/Text Encoder(CLIPLoader / DualCLIPLoader / ClipLoaderGGUF / DualClipLoaderGGUF対応 — typeおよびdevice選択付き、DualCLIPは2段モデル選択)は常時表示;ControlNetとHypernetwork(Strength入力付き)は使用頻度が低いためデフォルトで折りたたみ — 見出しをクリックで展開;LoRA(Single/Stackタブ)は右側に独立した列を表示;すべてのセレクタにフィルタ検索付き;LoRA Stack — Applyで有効な全LoRAを対象ノードに書き込み、同時にLoRA構文とCivitAIトリガーワードをPositiveプロンプトへ挿入;再Applyでは非アクティブなLoRAを削除し、アクティブなLoRAを再追加してプロンプトを更新",
|
||||
helpGen28: "Modelタブ — ハイライト色:読み込んだワークフローに実際に該当ノードが含まれる場合、各モデル種別(Checkpoint/VAE/Diffusion Model/Text Encoder/ControlNet/Hypernetwork)のラベルとLoRA列見出しが指定色(デフォルトは明るい緑)で表示される;ControlNet/Hypernetworkが折りたたまれた状態でも色は見えるため、展開しなくても使用有無が一目で分かる;色はSettingsタブでカスタマイズ可能",
|
||||
helpGen6: "Settingsタブ:KSamplerとLatent Imageを横並び(各50%幅)",
|
||||
helpGen7: "Raw JSONを各タブの右列に常時表示 — 直接編集してApplyでワークフロー即時反映;読み込んだワークフローにBypass/Muteノードがある場合、検索バー上部に一覧が表示される(Bypassは除外され配線が再接続される、Muteは除外されるが再接続はされない点はComfyUI本体と同じ)",
|
||||
helpGen8: "シード管理:ランダム / 固定 / 増加 / 減少",
|
||||
@@ -2098,6 +2117,7 @@ const LANGUAGES = {
|
||||
helpSettings11: "テキストサイズ:スライダー1本(10〜28px)でプロンプト・チャット全テキストエリアのフォントサイズを一括調整 — GenerateUI ポジティブ/ネガティブプロンプト、AIアシスタントチャット入力、プリセット ポジティブ/ネガティブ、ワイルドカードプロンプト・ファイルエディタ、メタデータプロンプト全文プレビュー;即時反映・保存で確定",
|
||||
helpSettings12: "RAW JSON カラー:GenerateUI の Raw JSON エディタの構文ハイライト色(6色)をカスタマイズ — デフォルトテキスト・名前/スケジューラー・タイトル・幅/高さ・プロンプト/テキスト・画像/ファイル;カラーピッカーで即時反映;「デフォルトにリセット」で元の配色に戻す",
|
||||
helpSettings13: "CivitAI ホスト:civitai.com(SFWのみ)と civitai.red(制限なし)を切り替え — ModelsタブのCivitAIパネルでモデルリンクをクリックしたときに開くサイトを指定",
|
||||
helpSettings14: "GenerateUI Modelタブ ハイライト色:GenerateUIのModelタブで、読み込んだワークフローに実際に該当ノードが含まれる場合にCheckpoint/VAE/Diffusion Model/Text Encoder/ControlNet/Hypernetworkのラベルおよび LoRA 列見出しを彩色する色を指定;デフォルトは明るい緑;変更は即時反映、「デフォルトにリセット」で元に戻す",
|
||||
helpNodes1: "ノードブラウザ:名前/説明/エイリアス/タグで検索、カテゴリ・パッケージ・タグ・グループ・お気に入りフィルタ",
|
||||
helpNodes2: "カード / テーブル表示モード切り替え(ページネーション付き)",
|
||||
helpNodes3: "ノードのお気に入り登録、カスタムタグの追加で整理",
|
||||
@@ -2178,7 +2198,7 @@ const LANGUAGES = {
|
||||
helpAi2: "翻訳ペイン:Ollama、LM Studio、または Lemonade を使って日本語・英語・中国語・カスタム言語(Free)間でテキストを翻訳;⇄ ボタンで言語とテキストを同時入替;選択は自動保存",
|
||||
helpAi3: "Chat ペイン:LLM とのマルチターン会話;Enter で送信、Shift+Enter で改行;Clear ボタンで会話履歴をリセット;モデルがtool callingに対応している場合、「猫の画像を作って」のように頼むとGenerateUIに読み込み中のワークフローで生成を実行し、結果をチャット内に表示する;ペインヘッダーの「画像生成を許可する」チェックボックスでこの機能のON/OFFを切り替えられる;画像が添付されている場合(TOOLSペインにドロップした画像)はLLMにもビジョン入力として送信され、generate_image呼び出し時はT2Iの代わりにI2I(image-to-image)の元画像として使われる;tool callingとビジョンの動作を確認済みのモデル: Gemma 4:e4b、Qwen 3.5:9b、Ministral 3:3b(Ollama)",
|
||||
helpAi4: "TOOLS ペイン — VLM:画像(PNG/JPG/WebP)をドロップしてタスク(画像解説 / プロンプト作成 / タグ作成)を選んで Run;または「Create wildcards」を選んでカテゴリ名と件数を入力すると1行1エントリのワイルドカードを生成;ドロップした画像はChatペインの添付としても共有され(Chat側にも表示、I2I生成に使用)、✕ボタンでクリアするまで保持される",
|
||||
helpAi6: "Settings ペイン:バックエンド(Ollama / LM Studio / Lemonade)の選択、APIのURL設定、接続テスト、モデル選択、Free言語名の設定;Chat Image Generationセクションでは、ChatからのT2I画像生成にGenerateUIの現在の読み込みワークフローの代わりに専用ワークフローを使うよう設定できる;Chat I2I Generationセクションは同様の設定を画像添付時のI2I生成向けに提供する;設定は localStorage に保存され、Library パネルの A タブと共有される",
|
||||
helpAi6: "Settings ペイン:バックエンド(Ollama / LM Studio / Lemonade)の選択、APIのURL設定、接続テスト、モデル選択;Generationセクション — Thinking modeトグル(デフォルトOFF;OllamaにはthinkフラグとしてOFF/ONを送信、バックエンドを問わずOFF時は出力中の<think>...</think>ブロックをクライアント側で除去するため、推論モデルの思考過程がTranslation/Chat/TOOLSの結果に混入しない)とMax tokens(Ollama: num_predict、LM Studio/Lemonade: max_tokens;0で無制限);Free言語名の設定;Chat Image Generationセクションでは、ChatからのT2I画像生成にGenerateUIの現在の読み込みワークフローの代わりに専用ワークフローを使うよう設定できる;Chat I2I Generationセクションは同様の設定を画像添付時のI2I生成向けに提供する;設定は localStorage に保存され、Library パネルの A タブと共有される",
|
||||
helpAi7: "Chat ペイン — SVG:LLMはComfyUIワークフローを一切使わず、SVGグラフィックをテキストとして直接生成することもできる;「SVGで〜を描いて」のように頼み、応答にSVGコードが含まれていれば自動でインラインプレビュー表示され、「Galleryに保存」ボタンでGalleryタブの出力フォルダに.svgファイルとして保存できる;tool calling(function calling)を使用しないため、tool calling非対応のモデルでも動作する",
|
||||
helpAi8: "Chat ペイン — スキル:メッセージ一覧上部のドロップダウンでスキルを選ぶと、その指示内容が system prompt として会話の先頭に付加される(表示上のチャット履歴には含まれない);鉛筆ボタンでスキル管理パネルを開き、.mdスキルファイルを作成・編集・削除できる(保存先: user/default/Workflow-Studio/ai_skills/);ファイル先頭の---で囲まれた部分にname/descriptionを書くとドロップダウンの表示に反映される;初期状態で2つのスキルを同梱: 「SVG Icon Generator」(描く前に対象を部位分解することで認識しやすいSVGを生成)と「Skill Creator」(対話形式で質問しながら新しいスキルファイルを下書きする);応答に```skillコードブロックが含まれる場合、メッセージ下部に「スキルとして保存」ボタンが表示され、内容が入った状態で管理パネルが開く",
|
||||
helpAi9: "スキル作成のヒント:ドロップダウンやツールチップで分かりやすいよう、frontmatterのname/descriptionは明確に書く;期待する出力形式を具体的に指示し、SVGや別のスキルファイルのように後で抽出したい成果物がある場合は専用のコードフェンスで囲むよう指示する;曖昧な指示より具体的で命令形の指示の方がローカルの小型モデルには効きやすく、短い実例を1つ添えるとさらに効果的;禁止事項(不適切な内容、セキュリティ、文字数制限など)は暗黙に期待せず明示的に書く;迷ったらドロップダウンで「Skill Creator」スキルを選び、作りたいスキルの概要を伝えれば対話形式で下書きしてくれる",
|
||||
@@ -2803,6 +2823,9 @@ const LANGUAGES = {
|
||||
jsonColorCyan: "提示词 / 文本",
|
||||
jsonColorRed: "图像 / 文件",
|
||||
jsonColorReset: "恢复默认",
|
||||
modelTabColorSection: "GenerateUI Model 标签高亮颜色",
|
||||
modelTabColorLabel: "工作流中包含的项目颜色",
|
||||
modelTabColorHint: "在 GenerateUI 的 Model 标签中,当前工作流实际包含的项目(Checkpoint / VAE 等)的标签将以此颜色显示。",
|
||||
|
||||
// -- Prompt tab --
|
||||
tabPrompt: "提示词",
|
||||
@@ -3166,6 +3189,10 @@ const LANGUAGES = {
|
||||
aiSettingsModelSection: "选择模型",
|
||||
aiSettingsRefreshBtn: "刷新",
|
||||
aiSettingsModelPlaceholder: "-- 请选择模型 --",
|
||||
aiSettingsGenerationTitle: "生成设置",
|
||||
aiSettingsThinkingMode: "Thinking mode(思考模式)",
|
||||
aiSettingsMaxTokens: "Max tokens(最大令牌数)",
|
||||
aiSettingsMaxTokensPlaceholder: "0 = 无限制",
|
||||
aiSettingsFreeLang: "Free 语言设置",
|
||||
aiSettingsInputLang: "输入语言",
|
||||
aiSettingsOutputLang: "翻译语言",
|
||||
@@ -3181,6 +3208,7 @@ const LANGUAGES = {
|
||||
aiStatusConnectOk: "连接成功",
|
||||
aiStatusConnectFail: "连接失败:",
|
||||
aiToastTransFailed: "翻译失败:",
|
||||
aiToastMaybeNotTranslated: "模型原样返回了原文,可能未进行翻译。请尝试更换模型。",
|
||||
aiToastNoCopyText: "没有可复制的文字",
|
||||
aiToastCopied: "已复制",
|
||||
aiToastNoModels: "未找到模型",
|
||||
@@ -3236,7 +3264,8 @@ const LANGUAGES = {
|
||||
helpGen27: "Input标签页 → Image子面板 — 每张LoadImage卡片现在有三个按钮:Apply(不变)、Clear(将该卡片待处理的选择/预览重置为空;不会改动工作流节点本身,因为LoadImage始终需要一个有效的文件名)和Placeholder。面板顶部的「Placeholder Image」区域(属于低频功能,默认折叠,点击可展开)用于配置共享的默认值:Color模式(设置宽度/高度/颜色,即时生成纯色PNG)或Image模式(拖入一张默认图片,之后原样复用)。点击某张卡片自己的Placeholder按钮,会立即只替换该卡片的图片为已配置的默认值——无需再点Apply;当有多张LoadImage卡片时,每张都可以单独替换。Placeholder配置(模式、尺寸、颜色、默认图片)会跨会话记住",
|
||||
helpGen3: "5标签布局 — Input(提示词+图像)、Model、Settings、Feeder、Batch — Input/Model/Settings各含实时预览和直接编辑的Raw JSON列",
|
||||
helpGen4: "Input标签:含Prompt和Image子标签(拖放上传);Prompt标签显示Positive Prompt、Negative Prompt,以及底部的Embeddings选择器(过滤+下拉+权重+Paste按钮)— Paste将(embedding:Name:weight)插入最后聚焦文本框的光标位置(未聚焦时默认追加到Positive);每个Apply按钮左侧的↺按钮可将文本框恢复为节点当前已应用的值;选中文字(或将光标置于单词/括号内也可)后按Ctrl+↑/↓,即可以(text:weight)语法以±0.05为单位调整强调权重(权重恢复为1.0时自动去除括号)",
|
||||
helpGen5: "Model标签:Checkpoint/VAE/LoRA(Single/Stack标签)/ControlNet/Diffusion Model(支持UNETLoader / UnetLoaderGGUF / LoaderGGUF)/Text Encoder(支持CLIPLoader / DualCLIPLoader / ClipLoaderGGUF / DualClipLoaderGGUF — 含type和device选择器,DualCLIP显示两段模型选择)/Hypernetwork(带Strength输入)选择器(含筛选)",
|
||||
helpGen5: "Model标签:Checkpoint/VAE/Diffusion Model(支持UNETLoader / UnetLoaderGGUF / LoaderGGUF)/Text Encoder(支持CLIPLoader / DualCLIPLoader / ClipLoaderGGUF / DualClipLoaderGGUF — 含type和device选择器,DualCLIP显示两段模型选择)始终显示;ControlNet和Hypernetwork(带Strength输入)因使用频率较低默认折叠 — 点击标题展开;LoRA(Single/Stack标签)拥有独立的列;所有选择器均含筛选;LoRA Stack — 点击Apply会将所有启用的LoRA写入目标节点,并同时将LoRA语法和CivitAI触发词插入Positive提示词;再次Apply会移除未启用的LoRA并重新添加启用的LoRA以更新提示词",
|
||||
helpGen28: "Model标签 — 高亮颜色:当前工作流实际包含某类型节点时(Checkpoint/VAE/Diffusion Model/Text Encoder/ControlNet/Hypernetwork),对应标签及LoRA列标题会以指定颜色(默认亮绿色)显示;即使ControlNet/Hypernetwork处于折叠状态该颜色依然可见,无需展开即可判断是否在使用;颜色可在Settings标签中自定义",
|
||||
helpGen6: "Settings标签:KSampler与Latent Image并排显示(各占50%宽度)",
|
||||
helpGen7: "Raw JSON在每个标签右列常驻显示 — 直接编辑并Apply即时重载工作流;如果加载的工作流包含Bypass或Mute节点,搜索栏上方会显示列表(Bypass节点会被排除但其连线会重新连接;Mute节点会被排除且不重新连接,与ComfyUI本体行为一致)",
|
||||
helpGen8: "种子管理:随机/固定/递增/递减",
|
||||
@@ -3339,6 +3368,7 @@ const LANGUAGES = {
|
||||
helpSettings11: "文字大小:单个滑块(10–28 px)可一次性调整所有提示词和聊天文本框的字体大小 — GenerateUI正/负向提示词、AI助手聊天输入、预设正/负向提示、通配符提示和文件编辑器、元数据提示词全文预览;即时生效并通过「保存设置」保存",
|
||||
helpSettings12: "RAW JSON 配色:自定义GenerateUI中Raw JSON编辑器的6种语法高亮颜色 — 默认文本、名称/调度器、标题、宽度/高度、提示词/文本、图像/文件;颜色选择即时生效;「重置默认」恢复原始配色",
|
||||
helpSettings13: "CivitAI 站点:在 civitai.com(仅SFW)和 civitai.red(无限制)之间切换 — 决定在Models标签CivitAI面板中点击模型链接时打开的站点",
|
||||
helpSettings14: "GenerateUI Model标签高亮颜色:设置当前工作流实际包含对应类型节点时,GenerateUI Model标签中Checkpoint/VAE/Diffusion Model/Text Encoder/ControlNet/Hypernetwork标签及LoRA列标题的显示颜色;默认为亮绿色;更改即时生效,「重置默认」可恢复",
|
||||
helpNodes1: "节点浏览器:按名称/描述/别名/标签搜索,按分类、包、标签、组、收藏筛选",
|
||||
helpNodes2: "卡片和表格视图模式切换(带分页)",
|
||||
helpNodes3: "收藏节点,添加自定义标签进行整理",
|
||||
@@ -3420,7 +3450,7 @@ const LANGUAGES = {
|
||||
helpAi2: "翻译面板:使用 Ollama、LM Studio 或 Lemonade 在日语、英语、中文或自定义语言(Free)之间翻译;⇄ 按钮同时互换语言和文本内容;选择自动保存",
|
||||
helpAi3: "Chat 面板:与 LLM 进行多轮对话;Enter 发送消息,Shift+Enter 换行;Clear 按钮重置对话历史;如果模型支持 tool calling,当你要求生成图像(例如\"生成一张猫的图片\")时,会使用 GenerateUI 中当前加载的工作流执行生成,并在对话中显示结果;面板标题栏的\"允许生成图像\"复选框可开关此功能;如果附加了图像(在 TOOLS 面板中拖放的图像),该图像会作为视觉输入发送给 LLM,调用 generate_image 时会以该图像为基础执行图生图(I2I),而不是文生图;已验证可正常支持 tool calling 与视觉功能的模型:Gemma 4:e4b、Qwen 3.5:9b、Ministral 3:3b(Ollama)",
|
||||
helpAi4: "TOOLS 面板 — VLM:拖放图像(PNG/JPG/WebP)并选择任务(图像描述 / 提示词生成 / 标签生成),点击 Run;或选择「Create wildcards」,输入类别名称和数量生成纯文本通配符条目(每行一条);拖放的图像也会作为 Chat 面板的附件共享(在该处显示并用于图生图生成),直到点击 ✕ 按钮清除",
|
||||
helpAi6: "Settings 面板:选择后端(Ollama / LM Studio / Lemonade)、设置 API URL、测试连接、选择模型、配置 Free 语言名称;Chat 图像生成部分可让 Chat 触发的文生图使用专用工作流,而不是 GenerateUI 中当前加载的工作流;Chat I2I 生成部分提供相同设置,用于附加图像时的图生图生成;设置保存至 localStorage,并与 Library 面板的 A 标签共享",
|
||||
helpAi6: "Settings 面板:选择后端(Ollama / LM Studio / Lemonade)、设置 API URL、测试连接、选择模型;Generation 部分 — Thinking mode 开关(默认关闭;Ollama会收到think标志,无论使用哪个后端,关闭时都会从输出中去除<think>...</think>思考块,避免推理模型的思考过程混入Translation/Chat/TOOLS的结果)以及 Max tokens(Ollama: num_predict,LM Studio/Lemonade: max_tokens;0为不限);配置 Free 语言名称;Chat 图像生成部分可让 Chat 触发的文生图使用专用工作流,而不是 GenerateUI 中当前加载的工作流;Chat I2I 生成部分提供相同设置,用于附加图像时的图生图生成;设置保存至 localStorage,并与 Library 面板的 A 标签共享",
|
||||
helpAi7: "Chat 面板 — SVG:LLM 还可以直接以文本形式生成 SVG 图形,完全不涉及 ComfyUI 工作流;只需要求它用 SVG 绘制内容,如果回复中包含 SVG 代码,会自动渲染为内嵌预览,并提供\"保存到图库\"按钮,将其保存为 .svg 文件到 Gallery 标签页的输出文件夹;该功能不使用 tool calling(函数调用),因此不支持 tool calling 的模型也可以使用",
|
||||
helpAi8: "Chat 面板 — 技能:在消息列表上方的下拉菜单中选择一个技能,其内容会作为 system prompt 添加到对话开头(不会显示在可见的聊天记录中);点击铅笔按钮打开技能管理面板,可创建、编辑、删除 .md 技能文件(保存路径:user/default/Workflow-Studio/ai_skills/);文件开头 --- 之间的 name/description 字段会显示在下拉菜单中;默认内置两个技能:「SVG Icon Generator」(在绘制前先将对象拆解为多个部件,以生成更易识别的 SVG)和「Skill Creator」(通过对话向你提问并起草新的技能文件);当回复中包含```skill代码块时,消息下方会出现「保存为技能」按钮,点击后会打开管理面板并预填内容",
|
||||
helpAi9: "编写技能的小贴士:为 frontmatter 的 name/description 写清楚的内容,便于在下拉菜单和提示中阅读;明确说明期望的输出格式,如果有需要后续提取的成果物(如 SVG 或另一个技能文件),指示它用专门的代码围栏包裹;具体、祈使句式的指令比笼统的说明对本地小模型更有效,附上一个简短示例效果更佳;明确写出硬性约束(禁止内容、安全性、长度限制等),不要指望模型会自行推断;如果不确定怎么写,可以在下拉菜单中选择「Skill Creator」技能并描述你想要的技能,它会通过提问帮你起草",
|
||||
|
||||
@@ -366,6 +366,21 @@ export function applyJsonColors(colors) {
|
||||
].join("\n");
|
||||
}
|
||||
|
||||
// GenerateUI Model tab: highlight color for section labels that match the loaded workflow
|
||||
const MODEL_ACTIVE_COLOR_STYLE_ID = "wfm-model-active-color-style";
|
||||
const MODEL_ACTIVE_COLOR_DEFAULT = "#4ade80";
|
||||
|
||||
export function applyModelTabActiveColor(color) {
|
||||
const val = color || MODEL_ACTIVE_COLOR_DEFAULT;
|
||||
let el = document.getElementById(MODEL_ACTIVE_COLOR_STYLE_ID);
|
||||
if (!el) {
|
||||
el = document.createElement("style");
|
||||
el.id = MODEL_ACTIVE_COLOR_STYLE_ID;
|
||||
document.head.appendChild(el);
|
||||
}
|
||||
el.textContent = `.wfm-form-group label.wfm-model-label-active,\n.wfm-settings-summary.wfm-model-label-active,\n#wfm-gen-lora-col-title.wfm-model-label-active { color: ${val}; }`;
|
||||
}
|
||||
|
||||
export function applyTheme(themeId) {
|
||||
if (themeId) {
|
||||
document.documentElement.setAttribute("data-theme", themeId);
|
||||
@@ -577,6 +592,21 @@ export async function initSettingsTab() {
|
||||
</div>
|
||||
</details>
|
||||
|
||||
<!-- GenerateUI Model Tab Highlight Color -->
|
||||
<details class="wfm-settings-section" open>
|
||||
<summary class="wfm-settings-summary">${t("modelTabColorSection") || "GenerateUI Modelタブ ハイライト色"}</summary>
|
||||
<div class="wfm-form-group">
|
||||
<label>${t("modelTabColorLabel") || "ワークフローに含まれる項目の色"}</label>
|
||||
<div style="display:flex;align-items:center;gap:8px;">
|
||||
<input type="color" id="wfm-model-active-color" value="${settings.modelTabActiveColor || MODEL_ACTIVE_COLOR_DEFAULT}">
|
||||
<button class="wfm-btn wfm-btn-sm" id="wfm-model-active-color-reset">${t("jsonColorReset") || "デフォルトに戻す"}</button>
|
||||
</div>
|
||||
<small style="color:var(--wfm-text-secondary);font-size:11px;display:block;margin-top:4px;">
|
||||
${t("modelTabColorHint") || "GenerateUIのModelタブで、読み込んだワークフローに実際に含まれる項目(Checkpoint / VAE など)のラベルをこの色で表示します"}
|
||||
</small>
|
||||
</div>
|
||||
</details>
|
||||
|
||||
<!-- Workflow Data Folder -->
|
||||
<details class="wfm-settings-section">
|
||||
<summary class="wfm-settings-summary">${t("workflowsDir")}</summary>
|
||||
@@ -1333,6 +1363,22 @@ export async function initSettingsTab() {
|
||||
showToast(t("settingsSaved"), "success");
|
||||
});
|
||||
|
||||
// --- GenerateUI Model tab highlight color ---
|
||||
document.getElementById("wfm-model-active-color")?.addEventListener("input", (e) => {
|
||||
saveLocalSettings({ modelTabActiveColor: e.target.value });
|
||||
applyModelTabActiveColor(e.target.value);
|
||||
});
|
||||
|
||||
document.getElementById("wfm-model-active-color-reset")?.addEventListener("click", () => {
|
||||
const cur = loadLocalSettings();
|
||||
delete cur.modelTabActiveColor;
|
||||
localStorage.setItem(SETTINGS_KEY, JSON.stringify(cur));
|
||||
applyModelTabActiveColor(undefined);
|
||||
const inp = document.getElementById("wfm-model-active-color");
|
||||
if (inp) inp.value = MODEL_ACTIVE_COLOR_DEFAULT;
|
||||
showToast(t("settingsSaved"), "success");
|
||||
});
|
||||
|
||||
// Save (local settings)
|
||||
document.getElementById("wfm-settings-save")?.addEventListener("click", () => {
|
||||
const taFontSize = parseInt(document.getElementById("wfm-ta-font-size")?.value) || TEXTAREA_FONT_SIZE_DEFAULT;
|
||||
|
||||
+18
-3
@@ -350,7 +350,7 @@
|
||||
</div>
|
||||
<div class="wfm-gen-lora-col">
|
||||
<div class="wfm-gen-lora-col-header">
|
||||
<span>LoRA</span>
|
||||
<span id="wfm-gen-lora-col-title">LoRA</span>
|
||||
<button class="wfm-btn wfm-btn-xs wfm-lora-refresh-btn" id="wfm-lora-stack-refresh" title="Refresh Stack group">🔄</button>
|
||||
</div>
|
||||
<div id="wfm-gen-lora-fields" class="wfm-gen-lora-col-body">
|
||||
@@ -1330,7 +1330,8 @@
|
||||
<ul>
|
||||
<li id="wfm-help-gen-3">5-tab layout — Input (Prompt + Image), Model, Settings, Feeder, Batch — Input/Model/Settings each include a Raw JSON column for instant preview and direct editing</li>
|
||||
<li id="wfm-help-gen-4">Input tab: inner tabs — <b>Prompt</b> and <b>Image</b> (drag-and-drop upload); Prompt tab has Positive Prompt, Negative Prompt, and an <b>Embeddings</b> selector at the bottom — select a model, set Weight (default 1.0), then click <b>Paste</b> to insert <code>(embedding:Name:weight)</code> at the cursor position of the last focused textarea (defaults to Positive if neither is focused); a <b>↺</b> button next to each Apply button reverts the textarea to the node's currently applied value; select text (or just place the cursor on a word/parenthesis block) and press <b>Ctrl+↑/↓</b> to adjust its emphasis weight in <code>(text:weight)</code> syntax by ±0.05 — the parentheses are removed automatically when the weight returns to 1.0</li>
|
||||
<li id="wfm-help-gen-5">Model tab: Checkpoint, VAE, LoRA (Single/Stack tabs), ControlNet, Diffusion Model (UNETLoader / UnetLoaderGGUF / LoaderGGUF), Text Encoder (CLIPLoader / DualCLIPLoader / ClipLoaderGGUF / DualClipLoaderGGUF — type and device selectors; DualCLIP shows two clip selectors), Hypernetwork (with Strength field) selectors with filter; LoRA Stack — Apply writes all active LoRAs to the target node and simultaneously inserts LoRA syntax and CivitAI trigger words into the Positive prompt; re-applying updates the prompt by removing inactive LoRAs and re-adding active ones</li>
|
||||
<li id="wfm-help-gen-5">Model tab: Checkpoint, VAE, Diffusion Model (UNETLoader / UnetLoaderGGUF / LoaderGGUF), Text Encoder (CLIPLoader / DualCLIPLoader / ClipLoaderGGUF / DualClipLoaderGGUF — type and device selectors; DualCLIP shows two clip selectors) are always shown; ControlNet and Hypernetwork (with Strength field) are collapsed by default since they're used less often — click their header to expand; LoRA (Single/Stack tabs) has its own column to the right; all selectors have a filter; LoRA Stack — Apply writes all active LoRAs to the target node and simultaneously inserts LoRA syntax and CivitAI trigger words into the Positive prompt; re-applying updates the prompt by removing inactive LoRAs and re-adding active ones</li>
|
||||
<li id="wfm-help-gen-28">Model tab — Highlight color: the label of each model type (Checkpoint, VAE, Diffusion Model, Text Encoder, ControlNet, Hypernetwork) and the LoRA column header turn a distinct color (default bright green) when the loaded workflow actually contains that type of node; this stays visible even while ControlNet/Hypernetwork are collapsed, so you can tell at a glance whether they're in use without expanding them; the color is customizable in the Settings tab</li>
|
||||
<li id="wfm-help-gen-6">Settings tab: KSampler and Latent Image side by side, each at 50% width</li>
|
||||
<li id="wfm-help-gen-7">Raw JSON always visible in the right column of each tab — edit directly and Apply to reload the workflow; if the loaded workflow has Bypass or Mute nodes, a note listing them appears above the search bar (Bypass nodes are excluded with their wires reconnected around them; Mute nodes are excluded with no reconnection, matching ComfyUI itself)</li>
|
||||
</ul>
|
||||
@@ -1567,6 +1568,7 @@
|
||||
<ul>
|
||||
<li id="wfm-help-settings-11">Text Size: one slider (10–28 px) adjusts font size for all prompt and chat textareas at once — Generate UI positive/negative prompts, AI Assistant chat input, Preset positive/negative, Wildcard prompt and file editor, and Metadata prompt full preview; takes effect immediately and saved with Save Settings</li>
|
||||
<li id="wfm-help-settings-12">RAW JSON Colors: customize the 6 syntax highlight colors for the Raw JSON editor in Generate UI — Default Text, Name/Scheduler, Title, Width/Height, Prompt/Text, Image/File; changes apply immediately on color pick; Reset Defaults restores the original color scheme</li>
|
||||
<li id="wfm-help-settings-14">GenerateUI Model Tab Highlight: pick the color used to mark Checkpoint/VAE/Diffusion Model/Text Encoder/ControlNet/Hypernetwork labels and the LoRA column header in GenerateUI's Model tab when the loaded workflow actually contains that type of node; default bright green; changes apply immediately, Reset Defaults restores it</li>
|
||||
</ul>
|
||||
</div>
|
||||
<div class="wfm-help-card">
|
||||
@@ -1801,7 +1803,7 @@
|
||||
</div>
|
||||
<div class="wfm-help-card">
|
||||
<h3 id="wfm-help-ai-card-settings-title">Settings pane</h3>
|
||||
<p id="wfm-help-ai-6">Settings pane: choose backend (Ollama / LM Studio / Lemonade), set API URL, test connection, select model, configure Free language names; Chat Image Generation section lets you enable a dedicated workflow for Chat-triggered text-to-image generation instead of the one currently loaded in GenerateUI; Chat I2I Generation section is the same but for image-to-image generation when an image is attached; settings are shared with the Library panel's A tab via localStorage</p>
|
||||
<p id="wfm-help-ai-6">Settings pane: choose backend (Ollama / LM Studio / Lemonade), set API URL, test connection, select model; Generation section — Thinking mode toggle (off by default; Ollama receives a `think` flag, and regardless of backend any <think>...</think> block is stripped from the output when off, so reasoning models don't clutter Translation/Chat/TOOLS results) and Max tokens (Ollama: num_predict, LM Studio/Lemonade: max_tokens; 0 = unlimited); configure Free language names; Chat Image Generation section lets you enable a dedicated workflow for Chat-triggered text-to-image generation instead of the one currently loaded in GenerateUI; Chat I2I Generation section is the same but for image-to-image generation when an image is attached; settings are shared with the Library panel's A tab via localStorage</p>
|
||||
</div>
|
||||
</div>
|
||||
|
||||
@@ -2831,6 +2833,19 @@
|
||||
<button id="wfm-ai-model-refresh-btn" class="wfm-btn">Refresh</button>
|
||||
</div>
|
||||
</div>
|
||||
<div class="wfm-ai-settings-section">
|
||||
<div class="wfm-ai-settings-title" id="wfm-ai-settings-generation-title">Generation</div>
|
||||
<div class="wfm-ai-settings-row">
|
||||
<label class="wfm-ai-label" style="display:flex;align-items:center;gap:6px;cursor:pointer;">
|
||||
<input type="checkbox" id="wfm-ai-thinking-mode-checkbox">
|
||||
<span id="wfm-ai-thinking-mode-label">Thinking mode</span>
|
||||
</label>
|
||||
</div>
|
||||
<div class="wfm-ai-settings-row">
|
||||
<label class="wfm-ai-label" id="wfm-ai-max-tokens-label">Max tokens</label>
|
||||
<input type="number" id="wfm-ai-max-tokens" class="wfm-input" min="0" step="1" placeholder="0 = unlimited">
|
||||
</div>
|
||||
</div>
|
||||
<div class="wfm-ai-settings-section">
|
||||
<div class="wfm-ai-settings-title" id="wfm-ai-settings-freelang-title">Free language</div>
|
||||
<div class="wfm-ai-settings-row">
|
||||
|
||||
+101
-26
@@ -3645,22 +3645,46 @@ function saveAiCfg(patch) {
|
||||
return d;
|
||||
}
|
||||
|
||||
async function aiCallLLM(url, backend, model, prompt) {
|
||||
// Ollamaは `think` (bool) と `options.num_predict` で制御できるが、
|
||||
// LM Studio/LemonadeはOpenAI互換APIのため max_tokens のみ標準対応。
|
||||
// thinking mode切替は非対応バックエンド/モデル向けに、出力からの
|
||||
// <think>タグ除去でも担保する(aiStripThinkingTagsを参照)。
|
||||
function _aiApplyGenOptions(body, backend, settings) {
|
||||
const maxTokens = parseInt(settings?.maxTokens, 10);
|
||||
if (backend === "ollama") {
|
||||
body.think = !!settings?.thinkingMode;
|
||||
if (maxTokens > 0) body.options = { ...(body.options || {}), num_predict: maxTokens };
|
||||
} else if (maxTokens > 0) {
|
||||
body.max_tokens = maxTokens;
|
||||
}
|
||||
return body;
|
||||
}
|
||||
|
||||
function aiStripThinkingTags(text) {
|
||||
return (text || "")
|
||||
.replace(/<think>[\s\S]*?<\/think>/gi, "")
|
||||
.replace(/<thinking>[\s\S]*?<\/thinking>/gi, "")
|
||||
.trim();
|
||||
}
|
||||
|
||||
async function aiCallLLM(url, backend, model, prompt, settings = {}) {
|
||||
if (backend === "ollama") {
|
||||
const r = await fetch(`${url}/api/generate`, {
|
||||
method: "POST", headers: { "Content-Type": "application/json" },
|
||||
body: JSON.stringify({ model, prompt, stream: false }),
|
||||
body: JSON.stringify(_aiApplyGenOptions({ model, prompt, stream: false }, backend, settings)),
|
||||
});
|
||||
if (!r.ok) throw new Error(`HTTP ${r.status}`);
|
||||
return (await r.json()).response || "";
|
||||
const text = (await r.json()).response || "";
|
||||
return settings.thinkingMode ? text : aiStripThinkingTags(text);
|
||||
} else {
|
||||
const r = await fetch(`${url}/v1/chat/completions`, {
|
||||
method: "POST", headers: { "Content-Type": "application/json" },
|
||||
body: JSON.stringify({ model, messages: [{ role: "user", content: prompt }], stream: false }),
|
||||
body: JSON.stringify(_aiApplyGenOptions({ model, messages: [{ role: "user", content: prompt }], stream: false }, backend, settings)),
|
||||
});
|
||||
if (!r.ok) throw new Error(`HTTP ${r.status}`);
|
||||
const d = await r.json();
|
||||
return d.choices?.[0]?.message?.content || "";
|
||||
const text = d.choices?.[0]?.message?.content || "";
|
||||
return settings.thinkingMode ? text : aiStripThinkingTags(text);
|
||||
}
|
||||
}
|
||||
|
||||
@@ -3676,49 +3700,74 @@ function aiFileToBase64(file) {
|
||||
});
|
||||
}
|
||||
|
||||
async function aiCallVLM(url, backend, model, prompt, base64Image, mimeType) {
|
||||
async function aiCallVLM(url, backend, model, prompt, base64Image, mimeType, settings = {}) {
|
||||
if (backend === "ollama") {
|
||||
const r = await fetch(`${url}/api/generate`, {
|
||||
method: "POST", headers: { "Content-Type": "application/json" },
|
||||
body: JSON.stringify({ model, prompt, images: [base64Image], stream: false }),
|
||||
body: JSON.stringify(_aiApplyGenOptions({ model, prompt, images: [base64Image], stream: false }, backend, settings)),
|
||||
});
|
||||
if (!r.ok) throw new Error(`HTTP ${r.status}`);
|
||||
return (await r.json()).response || "";
|
||||
const text = (await r.json()).response || "";
|
||||
return settings.thinkingMode ? text : aiStripThinkingTags(text);
|
||||
} else {
|
||||
const r = await fetch(`${url}/v1/chat/completions`, {
|
||||
method: "POST", headers: { "Content-Type": "application/json" },
|
||||
body: JSON.stringify({
|
||||
body: JSON.stringify(_aiApplyGenOptions({
|
||||
model,
|
||||
messages: [{ role: "user", content: [
|
||||
{ type: "text", text: prompt },
|
||||
{ type: "image_url", image_url: { url: `data:${mimeType};base64,${base64Image}` } },
|
||||
]}],
|
||||
stream: false,
|
||||
}),
|
||||
}, backend, settings)),
|
||||
});
|
||||
if (!r.ok) throw new Error(`HTTP ${r.status}`);
|
||||
return (await r.json()).choices?.[0]?.message?.content || "";
|
||||
const text = (await r.json()).choices?.[0]?.message?.content || "";
|
||||
return settings.thinkingMode ? text : aiStripThinkingTags(text);
|
||||
}
|
||||
}
|
||||
|
||||
async function aiCallChat(url, backend, model, messages) {
|
||||
async function aiCallChat(url, backend, model, messages, settings = {}) {
|
||||
if (backend === "ollama") {
|
||||
const r = await fetch(`${url}/api/chat`, {
|
||||
method: "POST", headers: { "Content-Type": "application/json" },
|
||||
body: JSON.stringify({ model, messages, stream: false }),
|
||||
body: JSON.stringify(_aiApplyGenOptions({ model, messages, stream: false }, backend, settings)),
|
||||
});
|
||||
if (!r.ok) throw new Error(`HTTP ${r.status}`);
|
||||
return (await r.json()).message?.content || "";
|
||||
const text = (await r.json()).message?.content || "";
|
||||
return settings.thinkingMode ? text : aiStripThinkingTags(text);
|
||||
} else {
|
||||
const r = await fetch(`${url}/v1/chat/completions`, {
|
||||
method: "POST", headers: { "Content-Type": "application/json" },
|
||||
body: JSON.stringify({ model, messages, stream: false }),
|
||||
body: JSON.stringify(_aiApplyGenOptions({ model, messages, stream: false }, backend, settings)),
|
||||
});
|
||||
if (!r.ok) throw new Error(`HTTP ${r.status}`);
|
||||
return (await r.json()).choices?.[0]?.message?.content || "";
|
||||
const text = (await r.json()).choices?.[0]?.message?.content || "";
|
||||
return settings.thinkingMode ? text : aiStripThinkingTags(text);
|
||||
}
|
||||
}
|
||||
|
||||
// Strips common LLM wrapping/preamble artifacts that survive despite the system prompt.
|
||||
function cleanAiTranslationOutput(raw) {
|
||||
let s = (raw || "").trim();
|
||||
s = s.replace(/^(sure[,.]?\s*)?(here('|)s|here is)?\s*(the\s+)?translation\s*:?\s*/i, "");
|
||||
s = s.replace(/^(翻訳|译文|翻译)\s*[::]\s*/, "");
|
||||
if (s.length >= 2) {
|
||||
const first = s[0], last = s[s.length - 1];
|
||||
const pairs = { '"': '"', "'": "'", "“": "”", "「": "」" };
|
||||
if (pairs[first] === last) s = s.slice(1, -1).trim();
|
||||
}
|
||||
return s;
|
||||
}
|
||||
|
||||
// Heuristic: if the cleaned output is (near-)identical to the input, the
|
||||
// model most likely echoed the source instead of translating it.
|
||||
function aiLooksUntranslated(input, output) {
|
||||
const normalize = (s) => s.trim().toLowerCase().replace(/[\s.,!?"'。、!?「」]/g, "");
|
||||
const a = normalize(input), b = normalize(output);
|
||||
return a.length > 0 && a === b;
|
||||
}
|
||||
|
||||
async function aiFetchModels(url, backend) {
|
||||
if (backend === "ollama") {
|
||||
const r = await fetch(`${url}/api/tags`);
|
||||
@@ -3894,6 +3943,18 @@ const renderAiTab = (container) => {
|
||||
<button id="wfm-nlp-ai-model-refresh" class="wfm-nlp-ai-btn">↻</button>
|
||||
</div>
|
||||
</div>
|
||||
<div class="wfm-nlp-ai-sec">
|
||||
<div class="wfm-nlp-ai-sec-title">Generation</div>
|
||||
<div class="wfm-nlp-ai-row">
|
||||
<label class="wfm-nlp-ai-lbl" style="display:flex;align-items:center;gap:6px;cursor:pointer;">
|
||||
<input type="checkbox" id="wfm-nlp-ai-thinking-mode"> Thinking mode
|
||||
</label>
|
||||
</div>
|
||||
<div class="wfm-nlp-ai-row">
|
||||
<span class="wfm-nlp-ai-lbl">Max tokens</span>
|
||||
<input type="number" id="wfm-nlp-ai-max-tokens" class="wfm-nlp-ai-input" min="0" step="1" placeholder="0 = unlimited" style="flex:1;">
|
||||
</div>
|
||||
</div>
|
||||
<div class="wfm-nlp-ai-sec">
|
||||
<div class="wfm-nlp-ai-sec-title">Free language</div>
|
||||
<div class="wfm-nlp-ai-row">
|
||||
@@ -3947,6 +4008,11 @@ const setupAiHandlers = (container) => {
|
||||
if (freeSrcInput && cfg.freeSrcLang) freeSrcInput.value = cfg.freeSrcLang;
|
||||
if (freeDstInput && cfg.freeDstLang) freeDstInput.value = cfg.freeDstLang;
|
||||
|
||||
const thinkingModeInput = container.querySelector("#wfm-nlp-ai-thinking-mode");
|
||||
if (thinkingModeInput) thinkingModeInput.checked = !!cfg.thinkingMode;
|
||||
const maxTokensInput = container.querySelector("#wfm-nlp-ai-max-tokens");
|
||||
if (maxTokensInput && cfg.maxTokens) maxTokensInput.value = cfg.maxTokens;
|
||||
|
||||
const srcSel = container.querySelector("#wfm-nlp-ai-src");
|
||||
const dstSel = container.querySelector("#wfm-nlp-ai-dst");
|
||||
if (srcSel && cfg.srcLang) srcSel.value = cfg.srcLang;
|
||||
@@ -3978,10 +4044,13 @@ const setupAiHandlers = (container) => {
|
||||
|
||||
const srcLang = srcSel?.value || "ja";
|
||||
const dstLang = dstSel?.value || "en";
|
||||
const srcName = srcLang === "free" ? (c.freeSrcLang || "Auto") : AI_LANG_NAMES[srcLang];
|
||||
const dstName = dstLang === "free" ? (c.freeDstLang || "English") : AI_LANG_NAMES[dstLang];
|
||||
const fromPart = srcLang === "free" ? "" : `from ${srcName} `;
|
||||
const prompt = `Translate the following text ${fromPart}to ${dstName}. Output only the translated text, nothing else.\n\n${text}`;
|
||||
const srcName = srcLang === "free" ? (c.freeSrcLang || "").trim() : AI_LANG_NAMES[srcLang];
|
||||
const dstName = dstLang === "free" ? (c.freeDstLang || "English").trim() || "English" : AI_LANG_NAMES[dstLang];
|
||||
const fromPart = srcName ? ` from ${srcName}` : "";
|
||||
const systemPrompt = `You are a professional translator. Translate the text the user sends${fromPart} into ${dstName}. `
|
||||
+ `Reply with ONLY the translated text and nothing else: no explanations, no notes, no preamble like "Here is the translation:", `
|
||||
+ `no surrounding quotes, and do not repeat or echo the original text.`;
|
||||
const messages = [{ role: "system", content: systemPrompt }, { role: "user", content: text }];
|
||||
|
||||
const transBtn = container.querySelector("#wfm-nlp-ai-trans-btn");
|
||||
const statusEl = container.querySelector("#wfm-nlp-ai-status");
|
||||
@@ -3992,10 +4061,14 @@ const setupAiHandlers = (container) => {
|
||||
outputEl.value = "";
|
||||
|
||||
try {
|
||||
const result = await aiCallLLM(url, backend, model, prompt);
|
||||
outputEl.value = result.trim();
|
||||
const result = await aiCallChat(url, backend, model, messages, c);
|
||||
const cleaned = cleanAiTranslationOutput(result);
|
||||
outputEl.value = cleaned;
|
||||
statusEl.textContent = "完了";
|
||||
statusEl.className = "wfm-nlp-ai-status wfm-nlp-ai-ok";
|
||||
if (srcLang !== dstLang && aiLooksUntranslated(text, cleaned)) {
|
||||
showToast("モデルが原文をそのまま返しました。翻訳されていない可能性があります。別のモデルをお試しください。", "error");
|
||||
}
|
||||
} catch (err) {
|
||||
statusEl.textContent = `エラー: ${err.message}`;
|
||||
statusEl.className = "wfm-nlp-ai-status wfm-nlp-ai-err";
|
||||
@@ -4069,11 +4142,13 @@ const setupAiHandlers = (container) => {
|
||||
const model = container.querySelector("#wfm-nlp-ai-model")?.value || "";
|
||||
const freeSrcLang = container.querySelector("#wfm-nlp-ai-free-src")?.value?.trim() || "";
|
||||
const freeDstLang = container.querySelector("#wfm-nlp-ai-free-dst")?.value?.trim() || "";
|
||||
const thinkingMode = !!container.querySelector("#wfm-nlp-ai-thinking-mode")?.checked;
|
||||
const maxTokens = Math.max(0, parseInt(container.querySelector("#wfm-nlp-ai-max-tokens")?.value, 10) || 0);
|
||||
if (url && !isValidAiUrl(url)) {
|
||||
showToast("URLは http:// または https:// で始まる必要があります", "error");
|
||||
return;
|
||||
}
|
||||
saveAiCfg({ backend, backendUrl: url, model, freeSrcLang, freeDstLang });
|
||||
saveAiCfg({ backend, backendUrl: url, model, freeSrcLang, freeDstLang, thinkingMode, maxTokens });
|
||||
showToast("Settings saved", "success");
|
||||
});
|
||||
|
||||
@@ -4119,7 +4194,7 @@ const setupAiHandlers = (container) => {
|
||||
const messagesToSend = skillPrompt
|
||||
? [{ role: "system", content: skillPrompt }, ...chatHistory]
|
||||
: chatHistory;
|
||||
const reply = await aiCallChat(url, backend, model, messagesToSend);
|
||||
const reply = await aiCallChat(url, backend, model, messagesToSend, c);
|
||||
chatHistory.push({ role: "assistant", content: reply });
|
||||
appendChatBubble("assistant", reply);
|
||||
chatStatusEl.textContent = "";
|
||||
@@ -4351,7 +4426,7 @@ const setupAiHandlers = (container) => {
|
||||
|
||||
const prompt = `Generate ${count} wildcard entries for the category "${name}". Output only plain text in English, one entry per line, no numbers, no markdown, no asterisks, no bold, nothing else.`;
|
||||
try {
|
||||
const result = await aiCallLLM(url, backend, model, prompt);
|
||||
const result = await aiCallLLM(url, backend, model, prompt, c);
|
||||
vlmResult.value = result.trim()
|
||||
.split("\n")
|
||||
.map(l => l.replace(/\*\*/g, "").replace(/^\*\s*/, "").replace(/^\d+\.\s*/, "").trim())
|
||||
@@ -4376,7 +4451,7 @@ const setupAiHandlers = (container) => {
|
||||
vlmResult.value = "";
|
||||
|
||||
try {
|
||||
const result = await aiCallVLM(url, backend, model, VLM_PROMPTS[task], vlmImage.base64, vlmImage.mimeType);
|
||||
const result = await aiCallVLM(url, backend, model, VLM_PROMPTS[task], vlmImage.base64, vlmImage.mimeType, c);
|
||||
vlmResult.value = result.trim();
|
||||
vlmStatus.textContent = "完了";
|
||||
vlmStatus.className = "wfm-nlp-ai-status wfm-nlp-ai-ok";
|
||||
|
||||
Reference in New Issue
Block a user