@clarionstack/notesia 0.0.6 → 0.0.7
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +6 -4
- package/package.json +6 -2
- package/src/commands.mjs +1 -1
- package/src/import.mjs +350 -29
package/README.md
CHANGED
|
@@ -38,7 +38,7 @@ Recursively discovers files in a directory and imports them as notes into Notesi
|
|
|
38
38
|
| Option | Description | Default |
|
|
39
39
|
|---|---|---|
|
|
40
40
|
| `<path>` | Source directory or file (required) | — |
|
|
41
|
-
| `--ext <list>` | Comma-separated file extensions | `md,txt` |
|
|
41
|
+
| `--ext <list>` | Comma-separated file extensions | `md,txt,pdf,docx,json` |
|
|
42
42
|
| `--folder <name>` | Target folder name (created if absent) | — |
|
|
43
43
|
| `--tags <list>` | Comma-separated tags for all imported notes | — |
|
|
44
44
|
| `--conflict <mode>` | `skip` or `overwrite` existing notes by title | `skip` |
|
|
@@ -47,7 +47,9 @@ Recursively discovers files in a directory and imports them as notes into Notesi
|
|
|
47
47
|
| `--quiet` | Suppress stdout output | — |
|
|
48
48
|
| `--verbose` | Print per-file progress to stderr | — |
|
|
49
49
|
|
|
50
|
-
Supported file types: `.md`, `.txt`, `.html`, `.pdf
|
|
50
|
+
Supported file types: `.md`, `.txt`, `.html`, `.pdf`, `.docx`, `.json`.
|
|
51
|
+
|
|
52
|
+
JSON imports are transformed into flattened key-path text (with truncation guards for very large arrays/strings) to improve embedding quality.
|
|
51
53
|
|
|
52
54
|
Notes exceeding 100,000 characters are automatically split into multiple parts.
|
|
53
55
|
|
|
@@ -84,8 +86,8 @@ notesia import ~/Documents/notes
|
|
|
84
86
|
# Import into a specific folder with tags
|
|
85
87
|
notesia import ./project-docs --folder "Work" --tags "project,2026"
|
|
86
88
|
|
|
87
|
-
# Import HTML and
|
|
88
|
-
notesia import ./archive --ext md,txt,html,pdf
|
|
89
|
+
# Import HTML, PDF, DOCX and JSON files too
|
|
90
|
+
notesia import ./archive --ext md,txt,html,pdf,docx,json
|
|
89
91
|
|
|
90
92
|
# Overwrite existing notes with the same title
|
|
91
93
|
notesia import ./updated-notes --conflict overwrite
|
package/package.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@clarionstack/notesia",
|
|
3
3
|
"private": false,
|
|
4
|
-
"version": "0.0.
|
|
4
|
+
"version": "0.0.7",
|
|
5
5
|
"description": "Notesia desktop bridge CLI — read, write, and bulk-import notes",
|
|
6
6
|
"type": "module",
|
|
7
7
|
"files": [
|
|
@@ -23,5 +23,9 @@
|
|
|
23
23
|
],
|
|
24
24
|
"engines": {
|
|
25
25
|
"node": ">=20"
|
|
26
|
+
},
|
|
27
|
+
"dependencies": {
|
|
28
|
+
"pdfjs-dist": "^4.0.379",
|
|
29
|
+
"jszip": "^3.10.1"
|
|
26
30
|
}
|
|
27
|
-
}
|
|
31
|
+
}
|
package/src/commands.mjs
CHANGED
|
@@ -311,7 +311,7 @@ function usageText() {
|
|
|
311
311
|
' auth status [--format text|json]',
|
|
312
312
|
'',
|
|
313
313
|
'Bulk Import:',
|
|
314
|
-
' import <path> [--ext md,txt] [--folder <name>] [--tags <list>]',
|
|
314
|
+
' import <path> [--ext md,txt,pdf,docx,json] [--folder <name>] [--tags <list>]',
|
|
315
315
|
' [--conflict skip|overwrite] [--dry-run]',
|
|
316
316
|
' [--format text|json] [--quiet] [--verbose]',
|
|
317
317
|
'',
|
package/src/import.mjs
CHANGED
|
@@ -6,13 +6,19 @@
|
|
|
6
6
|
* Dexie writes happen in the renderer via the bulk_import bridge op.
|
|
7
7
|
*/
|
|
8
8
|
|
|
9
|
-
import { createReadStream } from 'node:fs';
|
|
10
9
|
import fs from 'node:fs/promises';
|
|
11
10
|
import path from 'node:path';
|
|
12
11
|
import readline from 'node:readline';
|
|
13
12
|
import { requestBridge } from './client.mjs';
|
|
14
13
|
|
|
15
14
|
const NOTE_CHAR_LIMIT = 100_000;
|
|
15
|
+
const DEFAULT_EXTENSIONS = ['md', 'txt', 'pdf', 'docx', 'json'];
|
|
16
|
+
const JSON_MAX_DEPTH = 12;
|
|
17
|
+
const JSON_MAX_ENTRIES = 5_000;
|
|
18
|
+
const JSON_MAX_ARRAY_ITEMS = 100;
|
|
19
|
+
const JSON_MAX_STRING_LENGTH = 500;
|
|
20
|
+
const JSON_MAX_TOP_LEVEL_KEYS = 25;
|
|
21
|
+
const SIMPLE_JSON_KEY_RE = /^[A-Za-z_$][A-Za-z0-9_$]*$/;
|
|
16
22
|
|
|
17
23
|
const USAGE = [
|
|
18
24
|
'Usage: notesia import <path> [options]',
|
|
@@ -22,7 +28,7 @@ const USAGE = [
|
|
|
22
28
|
'',
|
|
23
29
|
'Options:',
|
|
24
30
|
' --path <dir> Path to directory to import',
|
|
25
|
-
|
|
31
|
+
` --ext <list> Comma-separated extensions to import (default: ${DEFAULT_EXTENSIONS.join(',')})`,
|
|
26
32
|
' --folder <name> Target folder name in Notesia',
|
|
27
33
|
' --tags <list> Comma-separated tags to apply to all imported notes',
|
|
28
34
|
' --conflict <mode> Conflict resolution: skip | overwrite (default: skip)',
|
|
@@ -47,7 +53,7 @@ export function parseImportArgs(argv) {
|
|
|
47
53
|
|
|
48
54
|
const opts = {
|
|
49
55
|
importPath: null,
|
|
50
|
-
ext: [
|
|
56
|
+
ext: [...DEFAULT_EXTENSIONS],
|
|
51
57
|
folder: null,
|
|
52
58
|
tags: [],
|
|
53
59
|
conflict: 'skip',
|
|
@@ -76,6 +82,10 @@ export function parseImportArgs(argv) {
|
|
|
76
82
|
} else if (token === '--ext') {
|
|
77
83
|
const val = args[++i];
|
|
78
84
|
if (!val || val.startsWith('--')) { process.stderr.write('Error: --ext requires a value\n\n' + USAGE); process.exit(2); }
|
|
85
|
+
opts.ext = val.split(',').map((e) => e.trim().toLowerCase()).filter(Boolean);
|
|
86
|
+
} else if (token === '--folder') {
|
|
87
|
+
const val = args[++i];
|
|
88
|
+
if (!val || val.startsWith('--')) { process.stderr.write('Error: --folder requires a value\n\n' + USAGE); process.exit(2); }
|
|
79
89
|
opts.folder = val;
|
|
80
90
|
} else if (token === '--tags') {
|
|
81
91
|
const val = args[++i];
|
|
@@ -149,9 +159,19 @@ async function discoverFiles(rootPath, extensions) {
|
|
|
149
159
|
|
|
150
160
|
// ─── Content parsing ─────────────────────────────────────────────────────────
|
|
151
161
|
|
|
152
|
-
function titleFromPath(filePath) {
|
|
153
|
-
|
|
154
|
-
|
|
162
|
+
function titleFromPath(filePath, relativePath) {
|
|
163
|
+
// Use relative path to include folder structure in title
|
|
164
|
+
// e.g., "folder1/SKILL.md" → "folder1/SKILL"
|
|
165
|
+
const baseName = path.basename(relativePath, path.extname(relativePath));
|
|
166
|
+
const folderName = path.dirname(relativePath);
|
|
167
|
+
const stem = baseName.replace(/[-_]+/g, ' ').replace(/\s+/g, ' ').trim();
|
|
168
|
+
|
|
169
|
+
// If file is in a subfolder, prepend folder to title
|
|
170
|
+
if (folderName && folderName !== '.') {
|
|
171
|
+
const folderStem = folderName.replace(/[-_]+/g, ' ').replace(/\s+/g, ' ').trim();
|
|
172
|
+
return `${folderStem}/${stem}`;
|
|
173
|
+
}
|
|
174
|
+
return stem;
|
|
155
175
|
}
|
|
156
176
|
|
|
157
177
|
function stripHtml(html) {
|
|
@@ -163,8 +183,185 @@ function stripHtml(html) {
|
|
|
163
183
|
.trim();
|
|
164
184
|
}
|
|
165
185
|
|
|
166
|
-
|
|
167
|
-
|
|
186
|
+
function decodeXmlEntities(value) {
|
|
187
|
+
if (!value) return '';
|
|
188
|
+
return value
|
|
189
|
+
.replace(/<!\[CDATA\[([\s\S]*?)\]\]>/g, '$1')
|
|
190
|
+
.replace(/&/g, '&')
|
|
191
|
+
.replace(/</g, '<')
|
|
192
|
+
.replace(/>/g, '>')
|
|
193
|
+
.replace(/"/g, '"')
|
|
194
|
+
.replace(/'/g, "'")
|
|
195
|
+
.replace(/&#x([0-9a-fA-F]+);/g, (_, hex) => String.fromCharCode(parseInt(hex, 16)))
|
|
196
|
+
.replace(/&#(\d+);/g, (_, dec) => String.fromCharCode(parseInt(dec, 10)))
|
|
197
|
+
.replace(/\u00A0/g, ' ');
|
|
198
|
+
}
|
|
199
|
+
|
|
200
|
+
function sharedLeadingWhitespace(a, b) {
|
|
201
|
+
const max = Math.min(a.length, b.length);
|
|
202
|
+
let i = 0;
|
|
203
|
+
while (i < max && a[i] === b[i]) i++;
|
|
204
|
+
return a.slice(0, i);
|
|
205
|
+
}
|
|
206
|
+
|
|
207
|
+
function normalizeFencedCodeBlocks(markdown) {
|
|
208
|
+
if (!markdown || !markdown.includes('```')) return markdown;
|
|
209
|
+
|
|
210
|
+
return markdown.replace(/```([^\n]*)\n([\s\S]*?)\n```/g, (full, lang, body) => {
|
|
211
|
+
const lines = String(body).replace(/\u00A0/g, ' ').split('\n');
|
|
212
|
+
const nonEmpty = lines.filter((line) => line.trim().length > 0);
|
|
213
|
+
if (nonEmpty.length === 0) return full;
|
|
214
|
+
|
|
215
|
+
let commonIndent = null;
|
|
216
|
+
for (const line of nonEmpty) {
|
|
217
|
+
const indent = line.match(/^[ \t]*/)?.[0] || '';
|
|
218
|
+
if (!indent) {
|
|
219
|
+
commonIndent = '';
|
|
220
|
+
break;
|
|
221
|
+
}
|
|
222
|
+
commonIndent = commonIndent === null ? indent : sharedLeadingWhitespace(commonIndent, indent);
|
|
223
|
+
if (!commonIndent) break;
|
|
224
|
+
}
|
|
225
|
+
|
|
226
|
+
if (!commonIndent) {
|
|
227
|
+
return `\`\`\`${lang}\n${lines.join('\n')}\n\`\`\``;
|
|
228
|
+
}
|
|
229
|
+
|
|
230
|
+
const dedented = lines.map((line) =>
|
|
231
|
+
line.startsWith(commonIndent) ? line.slice(commonIndent.length) : line
|
|
232
|
+
);
|
|
233
|
+
return `\`\`\`${lang}\n${dedented.join('\n')}\n\`\`\``;
|
|
234
|
+
});
|
|
235
|
+
}
|
|
236
|
+
|
|
237
|
+
function appendJsonPath(basePath, key) {
|
|
238
|
+
if (SIMPLE_JSON_KEY_RE.test(key)) {
|
|
239
|
+
return basePath === '$' ? `$.${key}` : `${basePath}.${key}`;
|
|
240
|
+
}
|
|
241
|
+
const escapedKey = String(key).replaceAll('\\', '\\\\').replaceAll('"', '\\"');
|
|
242
|
+
return `${basePath}["${escapedKey}"]`;
|
|
243
|
+
}
|
|
244
|
+
|
|
245
|
+
function formatJsonValue(value) {
|
|
246
|
+
if (typeof value === 'string') {
|
|
247
|
+
const normalized = value.replace(/\s+/g, ' ').trim();
|
|
248
|
+
if (normalized.length > JSON_MAX_STRING_LENGTH) {
|
|
249
|
+
const truncatedChars = normalized.length - JSON_MAX_STRING_LENGTH;
|
|
250
|
+
const shortened = normalized.slice(0, JSON_MAX_STRING_LENGTH);
|
|
251
|
+
return `${JSON.stringify(shortened)} [truncated ${truncatedChars} chars]`;
|
|
252
|
+
}
|
|
253
|
+
return JSON.stringify(normalized);
|
|
254
|
+
}
|
|
255
|
+
if (value === null) return 'null';
|
|
256
|
+
if (typeof value === 'number' || typeof value === 'boolean') return String(value);
|
|
257
|
+
return JSON.stringify(value);
|
|
258
|
+
}
|
|
259
|
+
|
|
260
|
+
function flattenJsonForEmbedding(root) {
|
|
261
|
+
const lines = [];
|
|
262
|
+
const notes = [];
|
|
263
|
+
const stack = [{ path: '$', value: root, depth: 0 }];
|
|
264
|
+
let hitEntryLimit = false;
|
|
265
|
+
|
|
266
|
+
while (stack.length > 0) {
|
|
267
|
+
if (lines.length >= JSON_MAX_ENTRIES) {
|
|
268
|
+
hitEntryLimit = true;
|
|
269
|
+
break;
|
|
270
|
+
}
|
|
271
|
+
|
|
272
|
+
const current = stack.pop();
|
|
273
|
+
const { path: currentPath, value, depth } = current;
|
|
274
|
+
|
|
275
|
+
if (value === null || typeof value !== 'object') {
|
|
276
|
+
lines.push(`${currentPath}: ${formatJsonValue(value)}`);
|
|
277
|
+
continue;
|
|
278
|
+
}
|
|
279
|
+
|
|
280
|
+
if (depth >= JSON_MAX_DEPTH) {
|
|
281
|
+
lines.push(`${currentPath}: [max depth ${JSON_MAX_DEPTH} reached]`);
|
|
282
|
+
continue;
|
|
283
|
+
}
|
|
284
|
+
|
|
285
|
+
if (Array.isArray(value)) {
|
|
286
|
+
lines.push(`${currentPath}: [array length=${value.length}]`);
|
|
287
|
+
const keptItems = Math.min(value.length, JSON_MAX_ARRAY_ITEMS);
|
|
288
|
+
|
|
289
|
+
for (let i = keptItems - 1; i >= 0; i -= 1) {
|
|
290
|
+
stack.push({
|
|
291
|
+
path: `${currentPath}[${i}]`,
|
|
292
|
+
value: value[i],
|
|
293
|
+
depth: depth + 1,
|
|
294
|
+
});
|
|
295
|
+
}
|
|
296
|
+
|
|
297
|
+
if (value.length > keptItems) {
|
|
298
|
+
lines.push(`${currentPath}: [omitted after item ${keptItems - 1}] (${value.length - keptItems} items omitted)`);
|
|
299
|
+
}
|
|
300
|
+
continue;
|
|
301
|
+
}
|
|
302
|
+
|
|
303
|
+
const entries = Object.entries(value);
|
|
304
|
+
if (entries.length === 0) {
|
|
305
|
+
lines.push(`${currentPath}: {}`);
|
|
306
|
+
continue;
|
|
307
|
+
}
|
|
308
|
+
|
|
309
|
+
for (let i = entries.length - 1; i >= 0; i -= 1) {
|
|
310
|
+
const [key, childValue] = entries[i];
|
|
311
|
+
stack.push({
|
|
312
|
+
path: appendJsonPath(currentPath, key),
|
|
313
|
+
value: childValue,
|
|
314
|
+
depth: depth + 1,
|
|
315
|
+
});
|
|
316
|
+
}
|
|
317
|
+
}
|
|
318
|
+
|
|
319
|
+
if (hitEntryLimit) {
|
|
320
|
+
notes.push(`[truncated] Flattened field limit reached (${JSON_MAX_ENTRIES}); remaining fields omitted.`);
|
|
321
|
+
}
|
|
322
|
+
|
|
323
|
+
return { lines, notes };
|
|
324
|
+
}
|
|
325
|
+
|
|
326
|
+
function formatJsonForEmbedding(jsonValue, sourcePath = '') {
|
|
327
|
+
const rootType = Array.isArray(jsonValue)
|
|
328
|
+
? 'array'
|
|
329
|
+
: (jsonValue === null ? 'null' : typeof jsonValue);
|
|
330
|
+
const topLevelKeys = rootType === 'object' && jsonValue
|
|
331
|
+
? Object.keys(jsonValue)
|
|
332
|
+
: [];
|
|
333
|
+
const { lines, notes } = flattenJsonForEmbedding(jsonValue);
|
|
334
|
+
|
|
335
|
+
const output = [
|
|
336
|
+
'# JSON Import',
|
|
337
|
+
'',
|
|
338
|
+
`Source: ${sourcePath || 'unknown'}`,
|
|
339
|
+
`Root type: ${rootType}`,
|
|
340
|
+
];
|
|
341
|
+
|
|
342
|
+
if (rootType === 'array') {
|
|
343
|
+
output.push(`Top-level items: ${jsonValue.length}`);
|
|
344
|
+
} else if (rootType === 'object') {
|
|
345
|
+
const visibleKeys = topLevelKeys.slice(0, JSON_MAX_TOP_LEVEL_KEYS);
|
|
346
|
+
const suffix = topLevelKeys.length > visibleKeys.length
|
|
347
|
+
? ` (+${topLevelKeys.length - visibleKeys.length} more)`
|
|
348
|
+
: '';
|
|
349
|
+
output.push(`Top-level keys (${topLevelKeys.length}): ${visibleKeys.join(', ')}${suffix}`);
|
|
350
|
+
}
|
|
351
|
+
|
|
352
|
+
output.push('', '## Flattened Fields', '');
|
|
353
|
+
output.push(...(lines.length > 0 ? lines : ['$: [empty document]']));
|
|
354
|
+
|
|
355
|
+
if (notes.length > 0) {
|
|
356
|
+
output.push('', '## Import Notes', '');
|
|
357
|
+
output.push(...notes.map((note) => `- ${note}`));
|
|
358
|
+
}
|
|
359
|
+
|
|
360
|
+
return output.join('\n').trim();
|
|
361
|
+
}
|
|
362
|
+
|
|
363
|
+
async function parseFile(absolutePath, extension, relativePath) {
|
|
364
|
+
const defaultTitle = titleFromPath(absolutePath, relativePath);
|
|
168
365
|
try {
|
|
169
366
|
if (extension === 'md' || extension === 'txt') {
|
|
170
367
|
const body = await fs.readFile(absolutePath, 'utf8');
|
|
@@ -176,23 +373,114 @@ async function parseFile(absolutePath, extension) {
|
|
|
176
373
|
const htmlTitle = titleMatch ? titleMatch[1].replace(/\s+/g, ' ').trim() : null;
|
|
177
374
|
return { title: htmlTitle || defaultTitle, body: stripHtml(raw) };
|
|
178
375
|
}
|
|
376
|
+
if (extension === 'json') {
|
|
377
|
+
const raw = await fs.readFile(absolutePath, 'utf8');
|
|
378
|
+
let parsedJson;
|
|
379
|
+
try {
|
|
380
|
+
parsedJson = JSON.parse(raw.replace(/^\uFEFF/, ''));
|
|
381
|
+
} catch (err) {
|
|
382
|
+
return { title: defaultTitle, body: '', error: `Invalid JSON: ${err.message}` };
|
|
383
|
+
}
|
|
384
|
+
return {
|
|
385
|
+
title: defaultTitle,
|
|
386
|
+
body: formatJsonForEmbedding(parsedJson, relativePath || path.basename(absolutePath)),
|
|
387
|
+
};
|
|
388
|
+
}
|
|
179
389
|
if (extension === 'pdf') {
|
|
180
|
-
//
|
|
390
|
+
// Use pdfjs-dist legacy build for Node.js PDF text extraction
|
|
181
391
|
try {
|
|
182
|
-
const
|
|
183
|
-
pdfjsLib.GlobalWorkerOptions.workerSrc = '';
|
|
392
|
+
const pdfjsLib = await import('pdfjs-dist/legacy/build/pdf.mjs');
|
|
184
393
|
const data = await fs.readFile(absolutePath);
|
|
185
|
-
const pdf = await pdfjsLib.getDocument({ data: new Uint8Array(data)
|
|
394
|
+
const pdf = await pdfjsLib.getDocument({ data: new Uint8Array(data.buffer) }).promise;
|
|
186
395
|
const pages = [];
|
|
187
396
|
for (let i = 1; i <= pdf.numPages; i++) {
|
|
188
397
|
const page = await pdf.getPage(i);
|
|
189
398
|
const content = await page.getTextContent();
|
|
190
|
-
|
|
399
|
+
|
|
400
|
+
// Extract text with better structure preservation
|
|
401
|
+
const items = content.items;
|
|
402
|
+
let pageText = '';
|
|
403
|
+
let lastY = null;
|
|
404
|
+
|
|
405
|
+
for (const item of items) {
|
|
406
|
+
if (lastY !== null && Math.abs(item.transform[5] - lastY) > 5) {
|
|
407
|
+
// Different line
|
|
408
|
+
pageText += '\n';
|
|
409
|
+
} else if (pageText !== '') {
|
|
410
|
+
// Same line - add space between items
|
|
411
|
+
pageText += ' ';
|
|
412
|
+
}
|
|
413
|
+
pageText += item.str;
|
|
414
|
+
lastY = item.transform[5];
|
|
415
|
+
}
|
|
416
|
+
|
|
417
|
+
pages.push(pageText.trim());
|
|
191
418
|
page.cleanup();
|
|
192
419
|
}
|
|
420
|
+
|
|
193
421
|
let title = defaultTitle;
|
|
194
422
|
try { const meta = await pdf.getMetadata(); if (meta?.info?.Title?.trim()) title = meta.info.Title.trim(); } catch {}
|
|
195
|
-
|
|
423
|
+
|
|
424
|
+
// Join pages with double newline and clean up
|
|
425
|
+
const body = pages.join('\n\n')
|
|
426
|
+
.replace(/\n{3,}/g, '\n\n')
|
|
427
|
+
.replace(/[ \t]+/g, ' ')
|
|
428
|
+
.trim();
|
|
429
|
+
|
|
430
|
+
return { title, body };
|
|
431
|
+
} catch (err) {
|
|
432
|
+
return { title: defaultTitle, body: '', error: err.message };
|
|
433
|
+
}
|
|
434
|
+
}
|
|
435
|
+
if (extension === 'docx') {
|
|
436
|
+
// Use JSZip for DOCX text extraction
|
|
437
|
+
try {
|
|
438
|
+
const JSZip = (await import('jszip')).default;
|
|
439
|
+
const data = await fs.readFile(absolutePath);
|
|
440
|
+
const zip = await JSZip.loadAsync(data);
|
|
441
|
+
const docXml = await zip.file('word/document.xml')?.async('string');
|
|
442
|
+
if (!docXml) {
|
|
443
|
+
return { title: defaultTitle, body: '', error: 'Invalid DOCX file: missing document.xml' };
|
|
444
|
+
}
|
|
445
|
+
|
|
446
|
+
// Extract title from docProps/core.xml if available
|
|
447
|
+
let title = defaultTitle;
|
|
448
|
+
try {
|
|
449
|
+
const coreXml = await zip.file('docProps/core.xml')?.async('string');
|
|
450
|
+
if (coreXml) {
|
|
451
|
+
const titleMatch = coreXml.match(/<dc:title[^>]*>([^<]*)<\/dc:title>/);
|
|
452
|
+
if (titleMatch && titleMatch[1]?.trim()) {
|
|
453
|
+
title = titleMatch[1].trim();
|
|
454
|
+
}
|
|
455
|
+
}
|
|
456
|
+
} catch {}
|
|
457
|
+
|
|
458
|
+
// Parse XML to extract plain text with proper line breaks
|
|
459
|
+
let text = docXml;
|
|
460
|
+
|
|
461
|
+
// Preserve CDATA text content before stripping tags.
|
|
462
|
+
text = text.replace(/<!\[CDATA\[([\s\S]*?)\]\]>/g, '$1');
|
|
463
|
+
|
|
464
|
+
// Handle paragraph spacing - use double newline for paragraphs
|
|
465
|
+
text = text
|
|
466
|
+
.replace(/<\/w:p>/g, '\n\n')
|
|
467
|
+
.replace(/<w:tab[^>]*\/>/g, '\t')
|
|
468
|
+
.replace(/<w:br[^>]*\/>/g, '\n')
|
|
469
|
+
.replace(/<w:cr[^>]*\/>/g, '\n');
|
|
470
|
+
|
|
471
|
+
// Remove all remaining XML tags
|
|
472
|
+
text = text
|
|
473
|
+
.replace(/<[^>]+>/g, '')
|
|
474
|
+
.replace(/\r\n/g, '\n')
|
|
475
|
+
.replace(/\r/g, '\n');
|
|
476
|
+
|
|
477
|
+
text = decodeXmlEntities(text)
|
|
478
|
+
.replace(/\n{3,}/g, '\n\n')
|
|
479
|
+
.trim();
|
|
480
|
+
|
|
481
|
+
text = normalizeFencedCodeBlocks(text);
|
|
482
|
+
|
|
483
|
+
return { title, body: text };
|
|
196
484
|
} catch (err) {
|
|
197
485
|
return { title: defaultTitle, body: '', error: err.message };
|
|
198
486
|
}
|
|
@@ -214,8 +502,15 @@ function splitContent(title, body) {
|
|
|
214
502
|
if (remaining.length <= NOTE_CHAR_LIMIT) { chunks.push(remaining); break; }
|
|
215
503
|
const searchStart = Math.max(0, NOTE_CHAR_LIMIT - 500);
|
|
216
504
|
const slice = remaining.slice(searchStart, NOTE_CHAR_LIMIT);
|
|
217
|
-
const
|
|
218
|
-
|
|
505
|
+
const lastParagraphBreak = slice.lastIndexOf('\n\n');
|
|
506
|
+
if (lastParagraphBreak !== -1) {
|
|
507
|
+
chunks.push(remaining.slice(0, searchStart + lastParagraphBreak + 2));
|
|
508
|
+
remaining = remaining.slice(searchStart + lastParagraphBreak + 2);
|
|
509
|
+
continue;
|
|
510
|
+
}
|
|
511
|
+
|
|
512
|
+
const lastLineBreak = slice.lastIndexOf('\n');
|
|
513
|
+
const splitAt = lastLineBreak !== -1 ? searchStart + lastLineBreak + 1 : NOTE_CHAR_LIMIT;
|
|
219
514
|
chunks.push(remaining.slice(0, splitAt));
|
|
220
515
|
remaining = remaining.slice(splitAt);
|
|
221
516
|
}
|
|
@@ -252,7 +547,7 @@ function printResult(result, opts) {
|
|
|
252
547
|
if (item.status === 'created') {
|
|
253
548
|
reason = `noteId: ${item.noteId}`;
|
|
254
549
|
} else if (item.status === 'skipped') {
|
|
255
|
-
reason = `noteId: ${item.noteId}`;
|
|
550
|
+
reason = `Note with same title already exists in target folder (noteId: ${item.noteId})`;
|
|
256
551
|
} else if (item.status === 'overwritten') {
|
|
257
552
|
reason = `noteId: ${item.noteId}`;
|
|
258
553
|
} else if (item.status === 'failed') {
|
|
@@ -311,7 +606,7 @@ export async function runImport(argv, deps = {}) {
|
|
|
311
606
|
for (const file of files) {
|
|
312
607
|
if (opts.verbose) process.stderr.write(` [parsing] ${file.relativePath}\n`);
|
|
313
608
|
|
|
314
|
-
const parsed = await parseFile(file.absolutePath, file.extension);
|
|
609
|
+
const parsed = await parseFile(file.absolutePath, file.extension, file.relativePath);
|
|
315
610
|
|
|
316
611
|
if (parsed.error && !parsed.body) {
|
|
317
612
|
if (opts.verbose) process.stderr.write(` [failed] ${file.relativePath} — ${parsed.error}\n`);
|
|
@@ -337,9 +632,9 @@ export async function runImport(argv, deps = {}) {
|
|
|
337
632
|
source_path: item.source_path,
|
|
338
633
|
_parseError: item._parseError || null,
|
|
339
634
|
})),
|
|
340
|
-
folder: opts.folder,
|
|
341
|
-
tags: opts.tags,
|
|
342
|
-
conflict: opts.conflict,
|
|
635
|
+
folder: opts.folder || null,
|
|
636
|
+
tags: opts.tags || [],
|
|
637
|
+
conflict: opts.conflict || 'skip',
|
|
343
638
|
// Always check first - we'll prompt if needed
|
|
344
639
|
create_folder_if_missing: false,
|
|
345
640
|
dry_run: opts.dryRun,
|
|
@@ -350,7 +645,14 @@ export async function runImport(argv, deps = {}) {
|
|
|
350
645
|
|
|
351
646
|
let response;
|
|
352
647
|
try {
|
|
648
|
+
if (opts.verbose) {
|
|
649
|
+
process.stderr.write(` [debug] calling bridge with ${items.length} items, folder: "${opts.folder}"\n`);
|
|
650
|
+
}
|
|
353
651
|
response = await bridge({ op: 'bulk_import', payload }, { timeoutMs });
|
|
652
|
+
process.stderr.write(` [debug] bridge response ok: ${response?.ok}, hasData: ${!!response?.data}\n`);
|
|
653
|
+
if (opts.verbose || !response?.ok) {
|
|
654
|
+
process.stderr.write(` [debug] full response: ${JSON.stringify(response)}\n`);
|
|
655
|
+
}
|
|
354
656
|
} catch (err) {
|
|
355
657
|
const code = err.code || 'BRIDGE_ERROR';
|
|
356
658
|
if (code === 'NOTESIA_APP_NOT_RUNNING') {
|
|
@@ -361,25 +663,31 @@ export async function runImport(argv, deps = {}) {
|
|
|
361
663
|
process.exit(1);
|
|
362
664
|
}
|
|
363
665
|
|
|
364
|
-
|
|
365
|
-
|
|
366
|
-
|
|
666
|
+
// Check for errors - could be at top level or nested in data
|
|
667
|
+
const hasError = !response || response.ok === false || (response.data && response.data.ok === false);
|
|
668
|
+
if (hasError) {
|
|
669
|
+
const code = response?.error?.code || response?.data?.error?.code || 'BRIDGE_ERROR';
|
|
670
|
+
const message = response?.error?.message || response?.data?.error?.message || 'Bridge request failed';
|
|
367
671
|
|
|
368
672
|
// Handle folder not found - prompt user to create (only if --create-folder is set)
|
|
673
|
+
const errorData = response?.error || response?.data?.error;
|
|
369
674
|
if (code === 'FOLDER_NOT_FOUND' && opts.folder && opts.createFolder) {
|
|
370
|
-
const folderName =
|
|
675
|
+
const folderName = errorData?.folderName || opts.folder;
|
|
371
676
|
process.stdout.write(`\nFolder "${folderName}" does not exist in Notesia.\n`);
|
|
372
677
|
const shouldCreate = await promptYesNo('Do you want to create it and continue import? [y/N] ');
|
|
373
678
|
if (shouldCreate) {
|
|
374
679
|
// Retry with create_folder_if_missing = true
|
|
375
680
|
const retryPayload = { ...payload, create_folder_if_missing: true };
|
|
376
681
|
response = await bridge({ op: 'bulk_import', payload: retryPayload }, { timeoutMs });
|
|
377
|
-
|
|
378
|
-
|
|
379
|
-
|
|
682
|
+
// Check both top-level ok and nested data.ok
|
|
683
|
+
const retryOk = response?.ok === true && (!response.data || response.data.ok !== false);
|
|
684
|
+
if (!retryOk) {
|
|
685
|
+
const retryCode = response?.error?.code || response?.data?.error?.code || 'BRIDGE_ERROR';
|
|
686
|
+
const retryMessage = response?.error?.message || response?.data?.error?.message || 'Bridge request failed';
|
|
380
687
|
process.stderr.write(`[${retryCode}] ${retryMessage}\n`);
|
|
381
688
|
process.exit(1);
|
|
382
689
|
}
|
|
690
|
+
// Success - continue with response.data as the result
|
|
383
691
|
} else {
|
|
384
692
|
process.stderr.write('Import cancelled.\n');
|
|
385
693
|
process.exit(1);
|
|
@@ -394,7 +702,20 @@ export async function runImport(argv, deps = {}) {
|
|
|
394
702
|
}
|
|
395
703
|
}
|
|
396
704
|
|
|
397
|
-
|
|
705
|
+
// Check if we have valid data in the response
|
|
706
|
+
if (!response.data) {
|
|
707
|
+
process.stderr.write(`[BRIDGE_ERROR] No data in bridge response. Raw response: ${JSON.stringify(response)}\n`);
|
|
708
|
+
process.exit(1);
|
|
709
|
+
}
|
|
710
|
+
|
|
711
|
+
const result = {
|
|
712
|
+
created: response.data?.created ?? 0,
|
|
713
|
+
skipped: response.data?.skipped ?? 0,
|
|
714
|
+
overwritten: response.data?.overwritten ?? 0,
|
|
715
|
+
failed: response.data?.failed ?? 0,
|
|
716
|
+
elapsedMs: response.data?.elapsedMs ?? (Date.now() - startMs),
|
|
717
|
+
items: response.data?.items ?? [],
|
|
718
|
+
};
|
|
398
719
|
printResult(result, { format: opts.format, quiet: opts.quiet });
|
|
399
720
|
process.exit(result.failed === 0 ? 0 : 1);
|
|
400
721
|
}
|