@moxn/kb-migrate 0.3.0 → 0.4.0

package/dist/sources/notion-api.js

@@ -0,0 +1,196 @@
+/**
+ * Thin HTTP client for Notion API with rate limiting.
+ *
+ * No @notionhq/client dependency — raw REST calls with retry on 429.
+ * Designed for the migration use case: read-only traversal of workspaces.
+ */
+const NOTION_API_BASE = 'https://api.notion.com/v1';
+const NOTION_API_VERSION = '2022-06-28';
+/** Minimum delay between API calls (~3 req/sec limit) */
+const MIN_REQUEST_INTERVAL_MS = 350;
+/** Max retries on 429 */
+const MAX_RETRIES = 5;
+// Notion color → hex mapping for tag colors
+const NOTION_COLOR_MAP = {
+    default: '#6B7280',
+    gray: '#6B7280',
+    brown: '#92400E',
+    orange: '#F97316',
+    yellow: '#EAB308',
+    green: '#22C55E',
+    blue: '#3B82F6',
+    purple: '#8B5CF6',
+    pink: '#EC4899',
+    red: '#EF4444',
+};
+export function notionColorToHex(color) {
+    return NOTION_COLOR_MAP[color] ?? NOTION_COLOR_MAP.default;
+}
+// ============================================
+// Client
+// ============================================
+export class NotionApiClient {
+    token;
+    lastRequestTime = 0;
+    constructor(token) {
+        this.token = token;
+    }
+    /** Test that the token is valid. Throws if not. */
+    async validateToken() {
+        // GET /v1/users/me works as a token validation check
+        await this.request('GET', '/users/me');
+    }
+    /** List all workspace users. */
+    async listUsers() {
+        const users = [];
+        let cursor = null;
+        do {
+            const params = new URLSearchParams();
+            if (cursor)
+                params.set('start_cursor', cursor);
+            params.set('page_size', '100');
+            const response = await this.request('GET', `/users?${params}`);
+            users.push(...response.results);
+            cursor = response.has_more ? response.next_cursor : null;
+        } while (cursor);
+        return users;
+    }
+    /**
+     * Search for all pages in the workspace.
+     * Sorted by last_edited_time descending for incremental sync support.
+     */
+    async searchPages(options) {
+        const pages = [];
+        let cursor = null;
+        do {
+            const body = {
+                filter: { value: 'page', property: 'object' },
+                page_size: 100,
+            };
+            if (options?.sortByLastEdited) {
+                body.sort = { timestamp: 'last_edited_time', direction: 'descending' };
+            }
+            if (cursor)
+                body.start_cursor = cursor;
+            const response = await this.request('POST', '/search', body);
+            pages.push(...response.results);
+            cursor = response.has_more ? response.next_cursor : null;
+        } while (cursor);
+        return pages;
+    }
+    /**
+     * Search for all databases in the workspace.
+     */
+    async searchDatabases() {
+        const databases = [];
+        let cursor = null;
+        do {
+            const body = {
+                filter: { value: 'database', property: 'object' },
+                page_size: 100,
+            };
+            if (cursor)
+                body.start_cursor = cursor;
+            const response = await this.request('POST', '/search', body);
+            databases.push(...response.results);
+            cursor = response.has_more ? response.next_cursor : null;
+        } while (cursor);
+        return databases;
+    }
+    /** Get a database schema by ID. */
+    async getDatabase(databaseId) {
+        return this.request('GET', `/databases/${databaseId}`);
+    }
+    /**
+     * Query all entries (pages) in a database.
+     */
+    async queryDatabase(databaseId) {
+        const pages = [];
+        let cursor = null;
+        do {
+            const body = { page_size: 100 };
+            if (cursor)
+                body.start_cursor = cursor;
+            const response = await this.request('POST', `/databases/${databaseId}/query`, body);
+            pages.push(...response.results);
+            cursor = response.has_more ? response.next_cursor : null;
+        } while (cursor);
+        return pages;
+    }
+    /**
+     * Get all child blocks of a block (page or another block).
+     * Automatically paginates through all results.
+     */
+    async getBlockChildren(blockId) {
+        const blocks = [];
+        let cursor = null;
+        do {
+            const params = new URLSearchParams({ page_size: '100' });
+            if (cursor)
+                params.set('start_cursor', cursor);
+            const response = await this.request('GET', `/blocks/${blockId}/children?${params}`);
+            blocks.push(...response.results);
+            cursor = response.has_more ? response.next_cursor : null;
+        } while (cursor);
+        return blocks;
+    }
+    /** Get a single page by ID. */
+    async getPage(pageId) {
+        return this.request('GET', `/pages/${pageId}`);
+    }
+    // ============================================
+    // Internal: HTTP with rate limiting + retry
+    // ============================================
+    async request(method, path, body) {
+        await this.rateLimit();
+        let lastError = null;
+        for (let attempt = 0; attempt <= MAX_RETRIES; attempt++) {
+            const headers = {
+                Authorization: `Bearer ${this.token}`,
+                'Notion-Version': NOTION_API_VERSION,
+            };
+            if (body)
+                headers['Content-Type'] = 'application/json';
+            const response = await fetch(`${NOTION_API_BASE}${path}`, {
+                method,
+                headers,
+                body: body ? JSON.stringify(body) : undefined,
+            });
+            if (response.ok) {
+                return (await response.json());
+            }
+            if (response.status === 429) {
+                const retryAfter = parseInt(response.headers.get('Retry-After') || '1', 10);
+                const delay = Math.max(retryAfter * 1000, 1000) * (attempt + 1);
+                console.warn(`Rate limited. Retrying in ${delay}ms (attempt ${attempt + 1}/${MAX_RETRIES})`);
+                await sleep(delay);
+                continue;
+            }
+            const errorBody = await response.text();
+            lastError = new Error(`Notion API error ${response.status}: ${errorBody}`);
+            // Don't retry on client errors (except 429 handled above)
+            if (response.status >= 400 && response.status < 500) {
+                throw lastError;
+            }
+            // Retry on server errors
+            if (attempt < MAX_RETRIES) {
+                const delay = 1000 * (attempt + 1);
+                console.warn(`Server error ${response.status}. Retrying in ${delay}ms`);
+                await sleep(delay);
+                continue;
+            }
+        }
+        throw lastError ?? new Error('Request failed after all retries');
+    }
+    async rateLimit() {
+        const now = Date.now();
+        const elapsed = now - this.lastRequestTime;
+        if (elapsed < MIN_REQUEST_INTERVAL_MS) {
+            await sleep(MIN_REQUEST_INTERVAL_MS - elapsed);
+        }
+        this.lastRequestTime = Date.now();
+    }
+}
+function sleep(ms) {
+    return new Promise((resolve) => setTimeout(resolve, ms));
+}

package/dist/sources/notion-blocks.d.ts

@@ -0,0 +1,30 @@
+/**
+ * Converts Notion blocks to Moxn KB ContentBlocks and sections.
+ *
+ * Section boundaries are created at H2 headings.
+ * Rich text is converted to markdown.
+ */
+import type { SectionInput } from '../types.js';
+import type { NotionBlock, NotionRichText, NotionApiClient, NotionPage } from './notion-api.js';
+/** Map of page IDs to their KB paths (for cross-link resolution). */
+export type PagePathMap = Map<string, string>;
+/**
+ * Convert a page's blocks into Moxn sections.
+ *
+ * H2 headings create section boundaries.
+ * Content before the first H2 goes into an "Introduction" section.
+ */
+export declare function blocksToSections(blocks: NotionBlock[], client: NotionApiClient, pagePathMap: PagePathMap, options?: {
+    /** Track synced block IDs to detect cycles. */
+    visitedSyncedBlocks?: Set<string>;
+}): Promise<SectionInput[]>;
+/** Convert rich text array to markdown string. */
+export declare function richTextToMarkdown(richText: NotionRichText[]): string;
+/** Convert rich text array to plain text (no markdown formatting). */
+export declare function richTextToPlain(richText: NotionRichText[]): string;
+/** Normalize Notion ID: remove dashes for consistent lookup. */
+export declare function normalizeId(id: string): string;
+/**
+ * Extract the page title from Notion page properties.
+ */
+export declare function getPageTitle(page: NotionPage): string;