From 84d6848e67038f33d84d432fc09b70ab539c220c Mon Sep 17 00:00:00 2001 From: yangziyu-hhh Date: Thu, 18 Jun 2026 16:06:41 +0800 Subject: [PATCH] feat(knowledge): add document creation and import --- agent/knowledge/service.py | 124 ++++++++++++++++- channel/web/chat.html | 34 ++++- channel/web/static/css/console.css | 12 ++ channel/web/static/js/console.js | 207 ++++++++++++++++++++++++++++- channel/web/web_channel.py | 89 +++++++++++++ tests/test_knowledge_service.py | 86 ++++++++++++ tests/test_knowledge_web.py | 65 +++++++++ 7 files changed, 611 insertions(+), 6 deletions(-) diff --git a/agent/knowledge/service.py b/agent/knowledge/service.py index 80fc6748..d78ff112 100644 --- a/agent/knowledge/service.py +++ b/agent/knowledge/service.py @@ -32,6 +32,10 @@ class KnowledgeService: PROTECTED_FILES = {"index.md", "log.md"} INVALID_NAME_RE = re.compile(r'[<>:"|?*\x00-\x1f]') + IMPORT_EXTENSIONS = {".md", ".txt"} + MAX_IMPORT_FILES = 100 + MAX_IMPORT_FILE_SIZE = 10 * 1024 * 1024 + MAX_IMPORT_TOTAL_SIZE = 50 * 1024 * 1024 def __init__(self, workspace_root: str, memory_manager=None): self.workspace_root = os.path.abspath(workspace_root) @@ -100,9 +104,9 @@ class KnowledgeService: raise error[0] return result[0] if result else None - def _sync_index(self, old_paths: Iterable[str]): + def _sync_index(self, old_paths: Iterable[str], force: bool = False): old_paths = sorted(set(old_paths)) - if not old_paths: + if not old_paths and not force: return manager = self._manager() for rel_path in old_paths: @@ -110,6 +114,113 @@ class KnowledgeService: manager.mark_dirty() self._run_sync(manager.sync()) + def _sanitize_document_name(self, filename: str) -> str: + name = os.path.basename((filename or "").replace("\\", "/")).strip() + if not name: + raise ValueError("filename is required") + stem, ext = os.path.splitext(name) + if ext.lower() not in self.IMPORT_EXTENSIONS: + raise ValueError(f"unsupported file type: {ext or name}") + if not stem or stem in (".", "..") or self.INVALID_NAME_RE.search(stem): + raise ValueError("invalid filename") + safe_name = f"{stem}.md" + self._ensure_not_protected(safe_name) + return safe_name + + @staticmethod + def _decode_document_content(content) -> str: + if isinstance(content, str): + return content + if not isinstance(content, (bytes, bytearray)): + raise ValueError("document content is required") + return bytes(content).decode("utf-8-sig", errors="replace") + + def _resolve_import_destination(self, target_category: str, filename: str, + conflict_strategy: str) -> tuple: + target_rel, target_full = self._resolve_path(target_category, kind="category") + if not target_full.is_dir(): + raise FileNotFoundError(f"category not found: {target_rel}") + + safe_name = self._sanitize_document_name(filename) + destination = target_full / safe_name + rel_path = f"{target_rel}/{safe_name}" + + if destination.exists(): + if conflict_strategy == "skip": + return rel_path, destination, "skip" + if conflict_strategy == "rename": + stem = destination.stem + suffix = destination.suffix + for index in range(1, 1000): + candidate = target_full / f"{stem}-{index}{suffix}" + if not candidate.exists(): + candidate_rel = f"{target_rel}/{candidate.name}" + return candidate_rel, candidate, "write" + raise FileExistsError(f"target already exists: {rel_path}") + if conflict_strategy != "overwrite": + raise ValueError("invalid conflict strategy") + return rel_path, destination, "write" + + def create_document(self, path: str, content: str = "", overwrite: bool = False) -> dict: + rel_path, full_path = self._resolve_path(path, kind="document") + self._ensure_not_protected(rel_path) + if len((content or "").encode("utf-8")) > self.MAX_IMPORT_FILE_SIZE: + raise ValueError("file too large") + if full_path.exists() and not overwrite: + raise FileExistsError(f"target already exists: {rel_path}") + old_paths = [rel_path] if full_path.exists() else [] + full_path.parent.mkdir(parents=True, exist_ok=True) + full_path.write_text(content or "", encoding="utf-8") + self._sync_index(old_paths, force=True) + return {"path": rel_path, "created": True, "overwritten": bool(old_paths)} + + def import_documents(self, target_category: str, files: Iterable[dict], + conflict_strategy: str = "skip") -> dict: + if not isinstance(files, list): + raise ValueError("files must be a list") + if len(files) > self.MAX_IMPORT_FILES: + raise ValueError(f"too many files: max {self.MAX_IMPORT_FILES}") + results = [] + old_paths = [] + imported = skipped = failed = 0 + total_size = 0 + + for item in files: + filename = item.get("filename") if isinstance(item, dict) else None + try: + content_bytes = item.get("content") if isinstance(item, dict) else None + size = len(content_bytes.encode("utf-8")) if isinstance(content_bytes, str) else len(content_bytes or b"") + total_size += size + if total_size > self.MAX_IMPORT_TOTAL_SIZE: + raise ValueError("import batch too large") + if size > self.MAX_IMPORT_FILE_SIZE: + raise ValueError("file too large") + rel_path, destination, mode = self._resolve_import_destination( + target_category, filename, conflict_strategy + ) + if mode == "skip": + skipped += 1 + results.append({"filename": filename, "path": rel_path, "status": "skipped", + "reason": "target_exists"}) + continue + + old_exists = destination.exists() + content = self._decode_document_content(content_bytes) + destination.parent.mkdir(parents=True, exist_ok=True) + destination.write_text(content, encoding="utf-8") + if old_exists: + old_paths.append(rel_path) + imported += 1 + results.append({"filename": filename, "path": rel_path, "status": "imported", + "overwritten": old_exists}) + except Exception as exc: + failed += 1 + results.append({"filename": filename or "", "status": "failed", "reason": str(exc)}) + + if imported: + self._sync_index(old_paths, force=True) + return {"results": results, "imported": imported, "skipped": skipped, "failed": failed} + def create_category(self, path: str) -> dict: rel_path, full_path = self._resolve_path(path, kind="category") if full_path.exists(): @@ -416,6 +527,15 @@ class KnowledgeService: result = self.delete_documents(payload.get("paths") or []) elif action == "move_documents": result = self.move_documents(payload.get("paths") or [], payload.get("target_category")) + elif action == "create_document": + result = self.create_document(payload.get("path"), payload.get("content", ""), + payload.get("overwrite", False)) + elif action == "import_documents": + result = self.import_documents( + payload.get("target_category"), + payload.get("files") or [], + payload.get("conflict_strategy", "skip"), + ) else: return {"action": action, "code": 400, "message": f"unknown action: {action}", "payload": None} return {"action": action, "code": 200, "message": "success", "payload": result} diff --git a/channel/web/chat.html b/channel/web/chat.html index a77eb2a1..97fe61ce 100644 --- a/channel/web/chat.html +++ b/channel/web/chat.html @@ -845,6 +845,15 @@ class="flex items-center gap-1.5 px-3 py-1.5 rounded-lg bg-primary-500 hover:bg-primary-600 text-white text-xs font-medium cursor-pointer transition-colors"> 新建分类 + + +
+
+ + +

diff --git a/channel/web/static/css/console.css b/channel/web/static/css/console.css index a618265d..c24b3dd5 100644 --- a/channel/web/static/css/console.css +++ b/channel/web/static/css/console.css @@ -1247,6 +1247,18 @@ background: rgba(74, 190, 110, 0.1); color: #4ABE6E; } +.knowledge-import-drag-over { + outline: 2px dashed rgba(74, 190, 110, 0.55); + outline-offset: 4px; + border-radius: 14px; +} +#knowledge-dialog-card.knowledge-document-dialog { + max-width: 760px; +} +#knowledge-document-content { + min-height: 320px; + line-height: 1.55; +} /* Graph legend */ .knowledge-graph-legend { diff --git a/channel/web/static/js/console.js b/channel/web/static/js/console.js index 09f09e4d..cf8a93ef 100644 --- a/channel/web/static/js/console.js +++ b/channel/web/static/js/console.js @@ -94,6 +94,8 @@ const I18N = { knowledge_empty_guide: '在对话中发送文档、链接或主题给 Agent，它会自动整理到你的知识库中。', knowledge_go_chat: '开始对话', knowledge_new_category: '新建分类', + knowledge_new_document: '新建文档', + knowledge_import_documents: '导入文档', welcome_subtitle: '我可以帮你解答问题、管理计算机、创造和执行技能，并通过
长期记忆和知识库不断成长', example_sys_title: '系统管理', example_sys_text: '查看工作空间里有哪些文件', example_task_title: '定时任务', example_task_text: '1分钟后提醒我检查服务器', @@ -310,6 +312,8 @@ const I18N = { knowledge_empty_guide: 'Send documents, links or topics to the agent in chat, and it will automatically organize them into your knowledge base.', knowledge_go_chat: 'Start a conversation', knowledge_new_category: 'New category', + knowledge_new_document: 'New document', + knowledge_import_documents: 'Import documents', welcome_subtitle: 'I can help you answer questions, manage your computer, create and execute skills, and keep growing through
long-term memory and a personal knowledge base.', example_sys_title: 'System', example_sys_text: 'Show me the files in the workspace', example_task_title: 'Scheduler', example_task_text: 'Remind me to check the server in 5 minutes', @@ -7541,6 +7545,9 @@ let _knowledgeTreeData = []; let _knowledgeRootFiles = []; let _knowledgeCurrentFile = null; let _knowledgeGraphLoaded = false; +const KNOWLEDGE_IMPORT_MAX_FILES = 100; +const KNOWLEDGE_IMPORT_MAX_FILE_SIZE = 10 * 1024 * 1024; +const KNOWLEDGE_IMPORT_MAX_TOTAL_SIZE = 50 * 1024 * 1024; function loadKnowledgeView() { // Reset to docs tab @@ -7550,6 +7557,7 @@ function loadKnowledgeView() { fetch('/api/knowledge/list').then(r => r.json()).then(data => { if (data.status !== 'success') return; + initKnowledgeImportDropZone(); const emptyEl = document.getElementById('knowledge-empty'); const docsPanel = document.getElementById('knowledge-panel-docs'); @@ -7713,12 +7721,14 @@ function _knowledgeResultMessage(action, payload) { return action === 'create_category' ? 'Category created' : action === 'rename_category' ? 'Category renamed' : action === 'delete_category' ? 'Category deleted' : + action === 'import_documents' ? `${payload?.imported || 0} imported · ${payload?.skipped || 0} skipped · ${payload?.failed || 0} failed` : action === 'move_documents' ? `${payload?.moved || 0} document moved` : `${payload?.deleted || 0} document deleted`; } return action === 'create_category' ? '分类已创建' : action === 'rename_category' ? '分类已重命名' : action === 'delete_category' ? '分类已删除' : + action === 'import_documents' ? `导入 ${payload?.imported || 0} 个，跳过 ${payload?.skipped || 0} 个，失败 ${payload?.failed || 0} 个` : action === 'move_documents' ? `已移动 ${payload?.moved || 0} 个文档` : `已删除 ${payload?.deleted || 0} 个文档`; } @@ -7734,8 +7744,15 @@ function _knowledgeCategoryPaths(groups, parent = '') { function openKnowledgeDialog(options) { const overlay = document.getElementById('knowledge-dialog-overlay'); + const card = document.getElementById('knowledge-dialog-card'); const input = document.getElementById('knowledge-dialog-input'); const select = document.getElementById('knowledge-dialog-select'); + const textarea = document.getElementById('knowledge-dialog-textarea'); + const documentForm = document.getElementById('knowledge-document-form'); + const documentFilename = document.getElementById('knowledge-document-filename'); + const documentContent = document.getElementById('knowledge-document-content'); + const templateBtn = document.getElementById('knowledge-document-template'); + const documentPathPreview = document.getElementById('knowledge-document-path-preview'); const submit = document.getElementById('knowledge-dialog-submit'); const cancel = document.getElementById('knowledge-dialog-cancel'); document.getElementById('knowledge-dialog-title').textContent = options.title; @@ -7744,9 +7761,29 @@ function openKnowledgeDialog(options) { document.getElementById('knowledge-dialog-hint').textContent = options.hint || ''; document.getElementById('knowledge-dialog-error').classList.add('hidden'); document.getElementById('knowledge-dialog-icon').className = `fas ${options.icon || 'fa-folder'} text-emerald-500`; - input.classList.toggle('hidden', options.type === 'select'); + card.classList.toggle('knowledge-document-dialog', options.type === 'document'); + input.classList.toggle('hidden', options.type === 'select' || options.type === 'textarea' || options.type === 'document'); select.classList.toggle('hidden', options.type !== 'select'); + textarea.classList.toggle('hidden', options.type !== 'textarea'); + documentForm.classList.toggle('hidden', options.type !== 'document'); input.value = options.value || ''; + textarea.value = options.value || ''; + documentFilename.value = options.filename || ''; + documentContent.value = options.content || ''; + document.getElementById('knowledge-document-category-label').textContent = currentLang === 'zh' ? '目标分类' : 'Destination category'; + documentPathPreview.textContent = options.category + ? `knowledge/${options.category}/` + : 'knowledge/'; + documentFilename.oninput = null; + document.getElementById('knowledge-document-filename-label').textContent = currentLang === 'zh' ? '文件名' : 'Filename'; + document.getElementById('knowledge-document-content-label').textContent = currentLang === 'zh' ? 'Markdown 内容' : 'Markdown content'; + templateBtn.textContent = currentLang === 'zh' ? '插入模板' : 'Insert template'; + templateBtn.onclick = () => { + if (documentContent.value.trim()) return; + const title = (documentFilename.value || 'untitled').replace(/\.md$/i, ''); + documentContent.value = `# ${title}\n\n## 摘要\n\n\n## 关键点\n\n- \n\n## 参考\n\n`; + documentContent.focus(); + }; if (options.type === 'select') { select.innerHTML = (options.choices || []).map(value => ``).join(''); } @@ -7756,7 +7793,13 @@ function openKnowledgeDialog(options) { const close = () => overlay.classList.add('hidden'); const submitAction = async () => { - const value = (options.type === 'select' ? select.value : input.value).trim(); + const rawValue = options.type === 'select' ? select.value : + (options.type === 'textarea' ? textarea.value : + (options.type === 'document' ? { + filename: documentFilename.value.trim(), + content: documentContent.value, + } : input.value)); + const value = options.type === 'textarea' || options.type === 'document' ? rawValue : rawValue.trim(); const error = options.validate ? options.validate(value) : (!value ? (currentLang === 'zh' ? '此项不能为空' : 'This field is required') : ''); if (error) { const errorEl = document.getElementById('knowledge-dialog-error'); @@ -7774,7 +7817,7 @@ function openKnowledgeDialog(options) { overlay.onclick = event => { if (event.target === overlay) close(); }; input.onkeydown = event => { if (event.key === 'Enter') submitAction(); }; overlay.classList.remove('hidden'); - setTimeout(() => (options.type === 'select' ? select : input).focus(), 0); + setTimeout(() => (options.type === 'select' ? select : (options.type === 'textarea' ? textarea : (options.type === 'document' ? documentFilename : input))).focus(), 0); } function createKnowledgeCategory() { @@ -7788,6 +7831,164 @@ function createKnowledgeCategory() { }); } +function createKnowledgeDocument() { + const categories = _knowledgeCategoryPaths(_knowledgeTreeData); + if (!categories.length) { + _setKnowledgeStatus(currentLang === 'zh' ? '请先创建分类' : 'Create a category first', true); + return; + } + openKnowledgeDialog({ + title: currentLang === 'zh' ? '新建文档' : 'New document', + subtitle: currentLang === 'zh' ? '先选择分类，然后输入文件名' : 'Choose a category, then enter a filename', + label: currentLang === 'zh' ? '目标分类' : 'Destination category', + type: 'select', + choices: categories, + icon: 'fa-file-circle-plus', + onSubmit: category => { + openKnowledgeDocumentEditor(category); + return null; + }, + }); +} + +function openKnowledgeDocumentEditor(category) { + openKnowledgeDialog({ + title: currentLang === 'zh' ? '新建文档' : 'New document', + subtitle: currentLang === 'zh' ? `保存到 ${category}` : `Save to ${category}`, + label: '', + hint: currentLang === 'zh' ? '文件名可省略 .md 后缀；保存后会自动同步索引。' : 'The .md suffix is optional. Index sync runs after saving.', + type: 'document', + category, + filename: '', + content: '', + icon: 'fa-file-circle-plus', + validate: value => { + if (!value.filename) return currentLang === 'zh' ? '文件名不能为空' : 'Filename is required'; + if (/\.[^.]+$/i.test(value.filename) && !/\.md$/i.test(value.filename)) { + return currentLang === 'zh' ? '新建文档仅支持 .md 文件名' : 'New documents must be .md files'; + } + if (!value.content.trim()) return currentLang === 'zh' ? '内容不能为空' : 'Content is required'; + if (new Blob([value.content]).size > KNOWLEDGE_IMPORT_MAX_FILE_SIZE) { + return currentLang === 'zh' ? '内容不能超过 10MB' : 'Content cannot exceed 10MB'; + } + return ''; + }, + onSubmit: value => { + const safeName = value.filename.endsWith('.md') ? value.filename : `${value.filename}.md`; + return dispatchKnowledgeAction('create_document', { + path: `${category}/${safeName}`, + content: value.content, + overwrite: false, + }); + }, + }); +} + +function selectKnowledgeImportFiles() { + const input = document.getElementById('knowledge-import-input'); + input.value = ''; + input.onchange = () => { + if (input.files && input.files.length) openKnowledgeImportDialog(Array.from(input.files)); + }; + input.click(); +} + +function openKnowledgeImportDialog(files) { + const validationError = validateKnowledgeImportFiles(files); + if (validationError) { + _setKnowledgeStatus(validationError, true); + return; + } + const choices = _knowledgeCategoryPaths(_knowledgeTreeData); + openKnowledgeDialog({ + title: currentLang === 'zh' ? '导入文档' : 'Import documents', + subtitle: currentLang === 'zh' ? `已选择 ${files.length} 个文件` : `${files.length} file(s) selected`, + label: currentLang === 'zh' ? '目标分类' : 'Destination category', + hint: choices.length ? (currentLang === 'zh' ? '支持 Markdown 和 TXT，TXT 会转成 Markdown 文档' : 'Markdown and TXT are supported. TXT is converted to Markdown.') : + (currentLang === 'zh' ? '请先创建一个分类' : 'Create a category first'), + type: 'select', + choices, + icon: 'fa-file-arrow-up', + onSubmit: target => importKnowledgeDocuments(files, target), + }); +} + +async function importKnowledgeDocuments(files, targetCategory) { + const validationError = validateKnowledgeImportFiles(files); + if (validationError) { + _setKnowledgeStatus(validationError, true); + return null; + } + const supported = files.filter(file => /\.(md|txt)$/i.test(file.name || '')); + if (!supported.length) { + _setKnowledgeStatus(currentLang === 'zh' ? '请选择 .md 或 .txt 文件' : 'Choose .md or .txt files', true); + return null; + } + const formData = new FormData(); + formData.append('target_category', targetCategory); + formData.append('conflict_strategy', 'rename'); + supported.forEach(file => formData.append('files', file, file.name)); + _setKnowledgeStatus(currentLang === 'zh' ? '正在导入...' : 'Importing...', false, true); + try { + const response = await fetch('/api/knowledge/import', { method: 'POST', body: formData }); + const result = await response.json(); + if (result.status !== 'success') { + _setKnowledgeStatus(result.message || (currentLang === 'zh' ? '导入失败' : 'Import failed'), true); + loadKnowledgeView(); + return null; + } + _setKnowledgeStatus(_knowledgeResultMessage('import_documents', result.payload), false); + loadKnowledgeView(); + return result.payload; + } catch (error) { + _setKnowledgeStatus(currentLang === 'zh' ? '导入请求失败' : 'Import request failed', true); + return null; + } +} + +function validateKnowledgeImportFiles(files) { + if (!files || !files.length) return currentLang === 'zh' ? '请选择文件' : 'Choose files'; + if (files.length > KNOWLEDGE_IMPORT_MAX_FILES) { + return currentLang === 'zh' ? `一次最多导入 ${KNOWLEDGE_IMPORT_MAX_FILES} 个文件` : `Import at most ${KNOWLEDGE_IMPORT_MAX_FILES} files at a time`; + } + let total = 0; + for (const file of files) { + total += file.size || 0; + if ((file.size || 0) > KNOWLEDGE_IMPORT_MAX_FILE_SIZE) { + return currentLang === 'zh' ? `${file.name} 超过 10MB` : `${file.name} exceeds 10MB`; + } + } + if (total > KNOWLEDGE_IMPORT_MAX_TOTAL_SIZE) { + return currentLang === 'zh' ? '单次导入总大小不能超过 50MB' : 'Total import size cannot exceed 50MB'; + } + return ''; +} + +let _knowledgeImportDropReady = false; +function initKnowledgeImportDropZone() { + if (_knowledgeImportDropReady) return; + const panel = document.getElementById('knowledge-panel-docs'); + if (!panel) return; + _knowledgeImportDropReady = true; + ['dragenter', 'dragover'].forEach(name => { + panel.addEventListener(name, event => { + if (!event.dataTransfer || !event.dataTransfer.types.includes('Files')) return; + event.preventDefault(); + panel.classList.add('knowledge-import-drag-over'); + }); + }); + ['dragleave', 'drop'].forEach(name => { + panel.addEventListener(name, event => { + if (event.type === 'drop') { + event.preventDefault(); + const files = Array.from(event.dataTransfer?.files || []); + if (files.length) openKnowledgeImportDialog(files); + } + panel.classList.remove('knowledge-import-drag-over'); + }); + }); +} + function renameKnowledgeCategory(path) { openKnowledgeDialog({ title: currentLang === 'zh' ? '重命名分类' : 'Rename category', diff --git a/channel/web/web_channel.py b/channel/web/web_channel.py index 1d07b0e7..04a0bf70 100644 --- a/channel/web/web_channel.py +++ b/channel/web/web_channel.py @@ -181,6 +181,29 @@ def _read_uploaded_file_bytes(file_obj) -> bytes: raise TypeError(f"Unsupported uploaded content type: {type(content).__name__}") +def _read_uploaded_file_bytes_limited(file_obj, max_bytes: int) -> bytes: + """Read uploaded content and fail once it exceeds max_bytes.""" + if isinstance(file_obj, bytes): + content = file_obj + elif isinstance(file_obj, str): + content = file_obj.encode("utf-8") + elif hasattr(file_obj, "file") and hasattr(file_obj.file, "read"): + content = file_obj.file.read(max_bytes + 1) + elif hasattr(file_obj, "read"): + content = file_obj.read(max_bytes + 1) + elif hasattr(file_obj, "value"): + content = file_obj.value + else: + raise ValueError("Unable to read uploaded file content") + if isinstance(content, str): + content = content.encode("utf-8") + if not isinstance(content, bytes): + raise TypeError(f"Unsupported uploaded content type: {type(content).__name__}") + if len(content) > max_bytes: + raise ValueError("file too large") + return content + + def _raw_web_input(): """Return unprocessed multipart form data when web.py exposes rawinput.""" rawinput = getattr(getattr(web, "webapi", None), "rawinput", None) @@ -1155,6 +1178,7 @@ class WebChannel(ChatChannel): '/api/knowledge/read', 'KnowledgeReadHandler', '/api/knowledge/graph', 'KnowledgeGraphHandler', '/api/knowledge/action', 'KnowledgeActionHandler', + '/api/knowledge/import', 'KnowledgeImportHandler', '/api/scheduler', 'SchedulerHandler', '/api/sessions', 'SessionsHandler', '/api/sessions/(.*)/generate_title', 'SessionTitleHandler', @@ -4509,6 +4533,71 @@ class KnowledgeActionHandler: return json.dumps({"status": "error", "code": 500, "message": str(e), "payload": None}) +class KnowledgeImportHandler: + def POST(self): + _require_auth() + web.header('Content-Type', 'application/json; charset=utf-8') + try: + from agent.knowledge.service import KnowledgeService + content_length = int(getattr(web.ctx, "env", {}).get("CONTENT_LENGTH") or 0) + if content_length > KnowledgeService.MAX_IMPORT_TOTAL_SIZE: + return json.dumps({ + "status": "error", + "code": 413, + "message": "import batch too large", + "payload": None, + }) + params = _raw_web_input() + target_category = params.get("target_category", "") + conflict_strategy = params.get("conflict_strategy", "skip") + uploaded = _ensure_list(params.get("files")) + single = params.get("file") + if single is not None: + uploaded.append(single) + if not uploaded: + return json.dumps({"status": "error", "code": 400, "message": "No files uploaded", "payload": None}) + if len(uploaded) > KnowledgeService.MAX_IMPORT_FILES: + return json.dumps({ + "status": "error", + "code": 400, + "message": f"too many files: max {KnowledgeService.MAX_IMPORT_FILES}", + "payload": None, + }) + + files = [] + total_size = 0 + for file_obj in uploaded: + if file_obj is None: + continue + filename = getattr(file_obj, "filename", "") or getattr(file_obj, "name", "") + content = _read_uploaded_file_bytes_limited(file_obj, KnowledgeService.MAX_IMPORT_FILE_SIZE) + total_size += len(content) + if total_size > KnowledgeService.MAX_IMPORT_TOTAL_SIZE: + return json.dumps({ + "status": "error", + "code": 413, + "message": "import batch too large", + "payload": None, + }) + files.append({ + "filename": filename, + "content": content, + }) + + result = KnowledgeService(_get_workspace_root()).dispatch("import_documents", { + "target_category": target_category, + "conflict_strategy": conflict_strategy, + "files": files, + }) + return json.dumps({ + "status": "success" if result["code"] < 300 else "error", + **result, + }, ensure_ascii=False) + except Exception as e: + logger.error(f"[WebChannel] Knowledge import error: {e}", exc_info=True) + return json.dumps({"status": "error", "code": 500, "message": str(e), "payload": None}) + + class VersionHandler: def GET(self): web.header('Content-Type', 'application/json; charset=utf-8') diff --git a/tests/test_knowledge_service.py b/tests/test_knowledge_service.py index 96ef447b..d1bc9a94 100644 --- a/tests/test_knowledge_service.py +++ b/tests/test_knowledge_service.py @@ -209,3 +209,89 @@ def test_move_does_not_overwrite_target_created_concurrently(tmp_path): assert source.read_text(encoding="utf-8") == "source" assert target.read_text(encoding="utf-8") == "concurrent" assert manager.storage.deleted == [] + + +def test_create_document_writes_and_syncs(tmp_path): + svc, manager = service(tmp_path) + (tmp_path / "knowledge/notes").mkdir() + + result = svc.dispatch("create_document", { + "path": "notes/new.md", "content": "# New\nBody", + }) + + assert result["code"] == 200 + assert (tmp_path / "knowledge/notes/new.md").read_text(encoding="utf-8") == "# New\nBody" + assert manager.dirty == 1 + assert manager.synced == 1 + + +def test_import_documents_supports_md_txt_and_rename_conflicts(tmp_path): + svc, manager = service(tmp_path) + (tmp_path / "knowledge/notes").mkdir() + (tmp_path / "knowledge/notes/a.md").write_text("existing", encoding="utf-8") + + result = svc.dispatch("import_documents", { + "target_category": "notes", + "conflict_strategy": "rename", + "files": [ + {"filename": "a.md", "content": b"# A"}, + {"filename": "plain.txt", "content": "plain text"}, + ], + }) + + assert result["code"] == 200 + assert result["payload"]["imported"] == 2 + assert (tmp_path / "knowledge/notes/a-1.md").read_text(encoding="utf-8") == "# A" + assert (tmp_path / "knowledge/notes/plain.md").read_text(encoding="utf-8") == "plain text" + assert manager.storage.deleted == [] + assert manager.synced == 1 + + +def test_import_documents_skip_overwrite_and_failures(tmp_path): + svc, manager = service(tmp_path) + (tmp_path / "knowledge/notes").mkdir() + existing = tmp_path / "knowledge/notes/a.md" + existing.write_text("old", encoding="utf-8") + + skipped = svc.dispatch("import_documents", { + "target_category": "notes", + "conflict_strategy": "skip", + "files": [{"filename": "a.md", "content": b"new"}], + }) + assert skipped["payload"]["skipped"] == 1 + assert existing.read_text(encoding="utf-8") == "old" + assert manager.synced == 0 + + overwritten = svc.dispatch("import_documents", { + "target_category": "notes", + "conflict_strategy": "overwrite", + "files": [ + {"filename": "a.md", "content": b"new"}, + {"filename": "bad.pdf", "content": b"%PDF"}, + ], + }) + assert overwritten["payload"]["imported"] == 1 + assert overwritten["payload"]["failed"] == 1 + assert existing.read_text(encoding="utf-8") == "new" + assert manager.storage.deleted == ["knowledge/notes/a.md"] + assert manager.synced == 1 + + +def test_import_documents_rejects_large_files_and_batches(tmp_path): + svc, manager = service(tmp_path) + (tmp_path / "knowledge/notes").mkdir() + + too_large = svc.dispatch("import_documents", { + "target_category": "notes", + "files": [{"filename": "big.md", "content": b"x" * (svc.MAX_IMPORT_FILE_SIZE + 1)}], + }) + assert too_large["payload"]["failed"] == 1 + assert too_large["payload"]["results"][0]["reason"] == "file too large" + + too_many = svc.dispatch("import_documents", { + "target_category": "notes", + "files": [{"filename": f"{i}.md", "content": b"x"} for i in range(svc.MAX_IMPORT_FILES + 1)], + }) + assert too_many["code"] == 403 + assert "too many files" in too_many["message"] + assert manager.synced == 0 diff --git a/tests/test_knowledge_web.py b/tests/test_knowledge_web.py index 92991b98..3dd26092 100644 --- a/tests/test_knowledge_web.py +++ b/tests/test_knowledge_web.py @@ -47,15 +47,80 @@ def test_knowledge_frontend_management_contract(): js = (root / "channel/web/static/js/console.js").read_text(encoding="utf-8") assert 'id="knowledge-dialog-overlay"' in html + assert 'id="knowledge-dialog-textarea"' in html + assert 'id="knowledge-document-form"' in html + assert 'id="knowledge-document-path-preview"' in html assert "function openKnowledgeDialog(" in js assert "function _knowledgeCategoryPaths(" in js assert "dispatchKnowledgeAction('create_category'" in js + assert "dispatchKnowledgeAction('create_document'" in js assert "dispatchKnowledgeAction('rename_category'" in js assert "dispatchKnowledgeAction('delete_category'" in js assert "dispatchKnowledgeAction('delete_documents'" in js assert "dispatchKnowledgeAction('move_documents'" in js + assert 'id="knowledge-import-input"' in html + assert "function createKnowledgeDocument(" in js + assert "function openKnowledgeDocumentEditor(" in js + assert "documentPathPreview.textContent = options.category" in js + assert "options.type === 'document'" in js + assert "input.classList.toggle('hidden', options.type === 'select' || options.type === 'textarea' || options.type === 'document')" in js + assert "function selectKnowledgeImportFiles(" in js + assert "function importKnowledgeDocuments(" in js + assert "function validateKnowledgeImportFiles(" in js + assert "KNOWLEDGE_IMPORT_MAX_FILE_SIZE" in js + assert "fetch('/api/knowledge/import'" in js + assert "initKnowledgeImportDropZone()" in js knowledge_section = js[js.index("// Knowledge View"):js.index("function _hasFilterMatch")] assert "prompt(" not in knowledge_section assert "alert(" not in knowledge_section assert "if (path === 'index.md' || path === 'log.md') return '';" in knowledge_section + + +class UploadedFile: + def __init__(self, filename, content): + self.filename = filename + self.value = content + + +def test_knowledge_import_handler_delegates_to_dispatch(tmp_path): + from channel.web.web_channel import KnowledgeImportHandler + + dispatched = {"action": "import_documents", "code": 200, "message": "success", + "payload": {"imported": 2, "skipped": 0, "failed": 0}} + params = { + "target_category": "notes", + "conflict_strategy": "rename", + "files": [UploadedFile("a.md", b"# A"), UploadedFile("b.txt", b"B")], + } + + with patch("channel.web.web_channel._require_auth"), \ + patch("channel.web.web_channel.web.header"), \ + patch("channel.web.web_channel._raw_web_input", return_value=params), \ + patch("channel.web.web_channel._get_workspace_root", return_value=str(tmp_path)), \ + patch("agent.knowledge.service.KnowledgeService.dispatch", return_value=dispatched) as dispatch: + response = json.loads(KnowledgeImportHandler().POST()) + + dispatch.assert_called_once() + action, payload = dispatch.call_args.args + assert action == "import_documents" + assert payload["target_category"] == "notes" + assert payload["conflict_strategy"] == "rename" + assert [f["filename"] for f in payload["files"]] == ["a.md", "b.txt"] + assert response["status"] == "success" + assert response["payload"]["imported"] == 2 + + +def test_knowledge_import_handler_rejects_large_content_length(tmp_path): + from channel.web.web_channel import KnowledgeImportHandler + from agent.knowledge.service import KnowledgeService + + with patch("channel.web.web_channel._require_auth"), \ + patch("channel.web.web_channel.web.header"), \ + patch("channel.web.web_channel.web.ctx") as ctx: + ctx.env = {"CONTENT_LENGTH": str(KnowledgeService.MAX_IMPORT_TOTAL_SIZE + 1)} + response = json.loads(KnowledgeImportHandler().POST()) + + assert response["status"] == "error" + assert response["code"] == 413 + assert response["message"] == "import batch too large"