dobo-extra 2.4.0 → 2.5.1

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/index.js CHANGED
@@ -1,84 +1,257 @@
1
- import exportTo from './lib/export-to.js'
2
- import importFrom from './lib/import-from.js'
1
+ import path from 'path'
2
+ import { json, ndjson, csv, xlsx } from './lib/helper.js'
3
+ import { createGunzip, createGzip } from 'zlib'
4
+ import scramjet from 'scramjet'
5
+ import config from './lib/config.js'
6
+
7
+ const { DataStream } = scramjet
8
+ const exts = ['.json', '.jsonl', '.ndjson', '.csv', '.xlsx', '.tsv']
3
9
 
4
10
  /**
5
- * Plugin factory
11
+ * Plugin factory.
12
+ *
13
+ * **Never** call this function directly!!! It's only-meant to be called by the {@link https://ardhi.github.io/bajo|Bajo framework} during plugin initialization.
6
14
  *
7
15
  * @param {string} pkgName - NPM package name
8
- * @returns {class}
16
+ * @returns {DoboExtra}
9
17
  */
10
18
  async function factory (pkgName) {
11
19
  const me = this
12
20
 
13
21
  /**
14
- * DoboExtra class
22
+ * DoboExtra class definition
15
23
  *
16
- * @class
24
+ * This class provides more functionality to the Dobo plugin including:
25
+ * - new additional `ndjson` format for Bajo's configHandlers
26
+ * - data import/export from/to file system in various formats (JSON, NDJSON, CSV, TSV, XLSX)
27
+ *
28
+ * @class
17
29
  */
18
30
  class DoboExtra extends this.app.baseClass.Base {
19
31
  constructor () {
32
+ /**
33
+ * Constructor
34
+ */
20
35
  super(pkgName, me.app)
21
- this.config = {
22
- export: {
23
- maxBatch: 1000,
24
- stringify: {
25
- open: '[\n',
26
- sep: ',\n',
27
- close: '\n]\n'
28
- }
29
- },
30
- import: {
31
- maxBatch: 1000
32
- },
33
- archive: {
34
- tasks: [],
35
- checkInterval: false,
36
- runEarly: true
37
- }
38
- }
39
36
 
40
- this.selfBind(['exportTo', 'importFrom'])
37
+ /**
38
+ * Configuration object
39
+ * @type {TConfig}
40
+ */
41
+ this.config = config
41
42
  }
42
43
 
43
- init = async () => {
44
- const { buildCollections } = this.app.bajo
45
- const types = ['datetime', 'date', 'timestamp']
44
+ /**
45
+ * Import data from a file into a Dobo model
46
+ *
47
+ * @async
48
+ * @method
49
+ * @param {string} source - Source file path (absolute or relative to plugin data dir)
50
+ * @param {string|boolean} dest - Destination model name or `false`. If `false`, the data will be returned instead of being imported into a model.
51
+ * @param {object} options - Import options
52
+ * @param {boolean} [options.trashOld=true] - Whether to clear the destination model before importing
53
+ * @param {number} [options.batch=100] - Number of records to import in a single batch.
54
+ * @param {function} [options.progressFn] - Callback function to report progress
55
+ * @param {function} [options.converterFn] - Callback function to convert each record before importing
56
+ * @param {boolean} [options.useHeader=true] - Whether to use the first row as header (for CSV/TSV/XLSX)
57
+ * @param {string} [options.fileType] - File type (json, ndjson, csv, tsv, xlsx)
58
+ * @returns {Promise<object|array>} - Imported data or summary report including file path and record count affected
59
+ */
60
+ importFrom = async (source, dest, options = {}) => {
61
+ let {
62
+ trashOld = true, batch = 100, progressFn, converterFn, useHeader = true,
63
+ fileType, createOpts = {}, parserOpts = {}
64
+ } = options
65
+ const { merge } = this.app.lib._
66
+ const { fs } = this.app.lib
67
+ const { getModel } = this.app.dobo
46
68
 
47
- async function handler ({ item }) {
48
- const { join } = this.app.bajo
49
- const { getSchema } = this.app.dobo
50
- const { has } = this.app.lib._
51
- for (const f of ['source', 'destination']) {
52
- if (!has(item, f)) throw this.error('taskMustHaveModel%s', f)
53
- const key = `${f}Field`
54
- item[key] = item[key] ?? 'createdAt'
55
- const schema = getSchema(item[f])
56
- const prop = schema.properties.find(p => p.name === item[key])
57
- if (!prop) throw this.error('unknownField%s%s', item[key], item[f])
58
- if (!types.includes(prop.type)) throw this.error('isNotSupported%s%s%s%s', item[key], item[f], prop.type, join(types))
59
- }
60
- if (item.source === item.destination) throw this.error('sourceDestMustBeDifferent')
61
- item.maxAge = item.maxAge ?? 1 // in days, less then 1 is ignored
69
+ let dmodel
70
+ if (dest !== false) dmodel = getModel(dest) // make sure dest model is valid
71
+ let file
72
+ if (path.isAbsolute(source)) file = source
73
+ else {
74
+ file = `${this.app.getPluginDataDir(this.ns)}/import/${source}`
75
+ fs.ensureDirSync(path.dirname(file))
62
76
  }
77
+ if (!fs.existsSync(file)) throw this.error('sourceFileNotExists%s', file)
78
+ let ext = fileType ? `.${fileType}` : path.extname(file)
79
+ let decompress = false
80
+ if (ext === '.gz') {
81
+ ext = path.extname(path.basename(file, '.gz'))
82
+ decompress = true
83
+ }
84
+ if (!exts.includes(ext)) throw this.error('unsupportedFormat%s', ext.slice(1))
85
+ if (trashOld && dest !== false) await dmodel.clearRecord()
86
+ const reader = fs.createReadStream(file)
87
+ batch = parseInt(batch) || 100
88
+ if (batch > this.config.import.maxBatch) batch = this.config.import.maxBatch
89
+ if (batch < 0) batch = 1
90
+ let count = 0
91
+ const pipes = [reader]
92
+ if (decompress) pipes.push(createGunzip())
93
+ if (ext === '.json') pipes.push(json.parse(parserOpts))
94
+ else if (['.ndjson', '.jsonl'].includes(ext)) pipes.push(ndjson.parse(parserOpts))
95
+ else if (ext === '.csv') pipes.push(csv.parse(merge({}, { headers: useHeader }, parserOpts)))
96
+ else if (ext === '.tsv') pipes.push(csv.parse(merge({}, { headers: useHeader }, merge({}, parserOpts, { delimiter: '\t' }))))
97
+ else if (ext === '.xlsx') pipes.push(xlsx.parse(merge({}, { header: useHeader }, parserOpts)))
98
+
99
+ const stream = DataStream.pipeline(...pipes)
100
+ let batchNo = 1
101
+ const data = []
102
+ await stream
103
+ .batch(batch)
104
+ .map(async items => {
105
+ if (items.length === 0) return null
106
+ const batchStart = new Date()
107
+ for (let item of items) {
108
+ count++
109
+ item = converterFn ? await converterFn.call(this, item) : item
110
+ if (dest !== false) await dmodel.createRecord(item, createOpts)
111
+ else data.push(item)
112
+ }
113
+ if (progressFn) await progressFn.call(this, { batchNo, data: items, batchStart, batchEnd: new Date() })
114
+ batchNo++
115
+ })
116
+ .run()
63
117
 
64
- this.archivers = await buildCollections({ ns: this.ns, handler, container: 'archive.tasks', dupChecks: ['source'], useDefaultName: false })
118
+ return dest === false ? data : { file, count }
65
119
  }
66
120
 
67
- start = async () => {
68
- if (this.config.archive.checkInterval === false || this.config.archive.checkInterval <= 0) {
69
- this.log.warn('autoArchiveDisabled')
70
- return
121
+ /**
122
+ * Export data from a Dobo model into a file
123
+ * @async
124
+ * @method
125
+ * @param {string} source - Source model name
126
+ * @param {string} dest - Destination file path (absolute or relative to plugin data dir)
127
+ * @param {object} options - Export options
128
+ * @param {object} [options.filter={}] - Filter object to select records to export
129
+ * @param {boolean} [options.ensureDir] - Whether to create the destination directory if it doesn't exist
130
+ * @param {boolean} [options.useHeader=true] - Whether to include the header row (for CSV/TSV/XLSX)
131
+ * @param {number} [options.batch=500] - Number of records to export in a single batch
132
+ * @param {function} [options.progressFn] - Callback function to report progress
133
+ * @param {Array<string>} [options.fields] - List of fields to include in the export
134
+ * @param {object} [options.parserOpts={}] - Options for the parser (e.g., CSV delimiter)
135
+ * @returns {Promise<object>} - Export summary including file path and record count affected
136
+ */
137
+ exportTo = (source, dest, options = {}) => {
138
+ let {
139
+ filter = {}, ensureDir, useHeader = true, batch = 500,
140
+ progressFn, fields, parserOpts = {}
141
+ } = options
142
+ const { importPkg } = this.app.bajo
143
+ const { fs } = this.app.lib
144
+ const { merge, omit } = this.app.lib._
145
+ const { getModel } = this.app.dobo
146
+
147
+ const getFile = async () => {
148
+ const increment = await importPkg('bajo:add-filename-increment')
149
+ let file
150
+ if (path.isAbsolute(dest)) file = dest
151
+ else {
152
+ file = `${this.app.getPluginDataDir(this.ns)}/export/${dest}`
153
+ fs.ensureDirSync(path.dirname(file))
154
+ }
155
+ file = increment(file, { fs: true, platform: 'win32' })
156
+ const dir = path.dirname(file)
157
+ if (!fs.existsSync(dir)) {
158
+ if (ensureDir) fs.ensureDirSync(dir)
159
+ else throw this.error('dirNotExists%s', dir)
160
+ }
161
+ let compress = false
162
+ let ext = path.extname(file)
163
+ if (ext === '.gz') {
164
+ compress = true
165
+ ext = path.extname(file.slice(0, -3))
166
+ // file = file.slice(0, file.length - 3)
167
+ }
168
+ if (!exts.includes(ext)) throw this.error('unsupportedFormat%s', ext.slice(1))
169
+ return { file, ext, compress }
71
170
  }
72
- if (this.config.archive.runEarly) await this.archive()
73
- this.archiveIntv = setInterval(() => {
74
- this.archive().then().catch(err => {
75
- this.log.error('archiveError%s', err.message)
76
- })
77
- }, this.config.archive.checkInterval * 60 * 1000)
78
- }
79
171
 
80
- exportTo = exportTo
81
- importFrom = importFrom
172
+ const getData = async (options = {}) => {
173
+ const { source, filter, count, stream, progressFn, fields } = options
174
+ let cnt = count ?? 0
175
+ const { find } = this.app.lib._
176
+ const { getModel } = this.app.dobo
177
+ const { maxLimit, hardCap } = this.app.dobo.config.default.filter
178
+ filter.limit = maxLimit
179
+ let sort
180
+ const model = getModel(source)
181
+ const idField = find(model.properties, { name: 'id' }).name
182
+ for (const name of ['createdAt', 'updatedAt', 'ts', 'dt']) {
183
+ const field = find(model.properties, { name })
184
+ if (field) {
185
+ sort = field.name
186
+ break
187
+ }
188
+ }
189
+ filter.sort = `${sort ?? idField}:1`
190
+ for (;;) {
191
+ const batchStart = new Date()
192
+ const { data: rows, page } = await model.findRecord(filter, { dataOnly: false, fields, fmt: true, refs: '*', noCache: true })
193
+ const data = rows.map(item => omit(item, ['_immutable', '_fmt', '_ref']))
194
+ if (data.length === 0) break
195
+ if (cnt + data.length > hardCap) {
196
+ const sliced = data.slice(0, hardCap - cnt)
197
+ await stream.pull(sliced)
198
+ cnt += sliced.length
199
+ if (progressFn) await progressFn.call(this, { batchNo: page, data: sliced, batchStart, batchEnd: new Date() })
200
+ break
201
+ }
202
+ cnt += data.length
203
+ await stream.pull(data)
204
+ if (progressFn) await progressFn.call(this, { batchNo: page, data, batchStart, batchEnd: new Date() })
205
+ filter.page++
206
+ }
207
+ await stream.end()
208
+ return cnt
209
+ }
210
+
211
+ filter.page = 1
212
+ batch = parseInt(batch) ?? 500
213
+ if (batch > this.config.export.maxBatch) batch = this.config.export.maxBatch
214
+ if (batch < 0) batch = 1
215
+ filter.limit = batch
216
+
217
+ return new Promise((resolve, reject) => {
218
+ let count = 0
219
+ let file
220
+ let ext
221
+ let stream
222
+ let compress
223
+ let writer
224
+ getModel(source)
225
+ getFile()
226
+ .then(res => {
227
+ file = res.file
228
+ ext = res.ext
229
+ compress = res.compress
230
+ writer = fs.createWriteStream(file)
231
+ writer.on('error', err => {
232
+ reject(err)
233
+ })
234
+ writer.on('finish', () => {
235
+ resolve({ file, count })
236
+ })
237
+ stream = new DataStream()
238
+ stream = stream.flatMap(items => (items))
239
+ const pipes = []
240
+ if (ext === '.json') pipes.push(json.stringify(parserOpts))
241
+ else if (['.ndjson', '.jsonl'].includes(ext)) pipes.push(ndjson.stringify(parserOpts))
242
+ else if (ext === '.csv') pipes.push(csv.stringify(merge({}, { headers: useHeader }, parserOpts)))
243
+ else if (ext === '.tsv') pipes.push(csv.stringify(merge({}, { headers: useHeader }, merge({}, parserOpts, { delimiter: '\t' }))))
244
+ else if (ext === '.xlsx') pipes.push(xlsx.stringify(merge({}, { header: useHeader }, parserOpts)))
245
+ if (compress) pipes.push(createGzip())
246
+ DataStream.pipeline(stream, ...pipes).pipe(writer)
247
+ return getData({ source, filter, count, stream, fields, progressFn })
248
+ })
249
+ .then(cnt => {
250
+ count = cnt
251
+ })
252
+ .catch(reject)
253
+ })
254
+ }
82
255
  }
83
256
 
84
257
  return DoboExtra
package/lib/config.js ADDED
@@ -0,0 +1,28 @@
1
+ /**
2
+ * @typedef TConfig
3
+ * @memberof DoboExtra
4
+ * @type {object}
5
+ * @property {object} export - Export configuration
6
+ * @property {number} export.maxBatch - Maximum number of records to export in a single batch
7
+ * @property {object} export.stringify - Stringify options for exported data
8
+ * @property {string} export.stringify.open - Opening string for exported data
9
+ * @property {string} export.stringify.sep - Separator string for exported data
10
+ * @property {string} export.stringify.close - Closing string for exported data
11
+ * @property {object} import - Import configuration
12
+ * @property {number} import.maxBatch - Maximum number of records to import in a single batch
13
+ */
14
+ const config = {
15
+ export: {
16
+ maxBatch: 1000,
17
+ stringify: {
18
+ open: '[\n',
19
+ sep: ',\n',
20
+ close: '\n]\n'
21
+ }
22
+ },
23
+ import: {
24
+ maxBatch: 1000
25
+ }
26
+ }
27
+
28
+ export default config
package/lib/helper.js ADDED
@@ -0,0 +1,96 @@
1
+ import ndj from 'ndjson'
2
+ import fastCsv from 'fast-csv'
3
+ import xlsxparse from 'xlsx-parse-stream'
4
+ import XLSXWriteStream from '@atomictech/xlsx-write-stream'
5
+ import StreamArray from 'stream-json/streamers/StreamArray.js'
6
+ import Stringer from 'stream-json/Stringer.js'
7
+ import Disassembler from 'stream-json/Disassembler.js'
8
+ import chain from 'stream-chain'
9
+
10
+ /**
11
+ * Helper functions for Dobo Extra
12
+ * @module Helper
13
+ */
14
+
15
+ // Borrow the idea from: https://github.com/fanlia/ndjson-csv-xlsx/blob/main/index.js
16
+
17
+ const XLSXStreamer = XLSXWriteStream.default
18
+
19
+ /**
20
+ * NDJSON parser and stringifier.
21
+ *
22
+ * To be able to use this helper, do something like this:
23
+ * ```js
24
+ * const { importModule } = this.app.bajo
25
+ * const { ndjson } = await importModule('dobo-extra:/lib/helper.js', { asDefaultImport: false })
26
+ * console.log(ndjson.parse('{"a":1}\n{"b":2}')) // [{a:1},{b:2}]
27
+ * console.log(ndjson.stringify([{a:1},{b:2}])) // '{"a":1}\n{"b":2}\n'
28
+ * ```
29
+ * @type {object}
30
+ * @property {function} parse - Parse NDJSON data
31
+ * @property {function} stringify - Stringify data to NDJSON format
32
+ */
33
+ export const ndjson = {
34
+ parse: (...args) => ndj.parse(...args),
35
+ stringify: (...args) => ndj.stringify(...args)
36
+ }
37
+
38
+ /**
39
+ * CSV parser and stringifier.
40
+ *
41
+ * To be able to use this helper, do something like this:
42
+ * ```js
43
+ * const { importModule } = this.app.bajo
44
+ * const { csv } = await importModule('dobo-extra:/lib/helper.js', { asDefaultImport: false })
45
+ * ...
46
+ * ```
47
+ * @type {object}
48
+ * @property {function} parse - Parse CSV data
49
+ * @property {function} stringify - Stringify data to CSV format
50
+ */
51
+ export const csv = {
52
+ parse: (...args) => fastCsv.parse(...args),
53
+ stringify: (...args) => fastCsv.format(...args)
54
+ }
55
+
56
+ /**
57
+ * Excel XLSX parser and stringifier.
58
+ *
59
+ * To be able to use this helper, do something like this:
60
+ * ```js
61
+ * const { importModule } = this.app.bajo
62
+ * const { xlsx } = await importModule('dobo-extra:/lib/helper.js', { asDefaultImport: false })
63
+ * ...
64
+ * ```
65
+ * @type {object}
66
+ * @property {function} parse - Parse XLSX data
67
+ * @property {function} stringify - Stringify data to XLSX format
68
+ */
69
+ export const xlsx = {
70
+ parse: (...args) => xlsxparse(...args),
71
+ stringify: (...args) => new XLSXStreamer(...args)
72
+ }
73
+
74
+ /**
75
+ * JSON parser and stringifier. Use stream to handle large JSON files.
76
+ *
77
+ * To be able to use this helper, do something like this:
78
+ * ```js
79
+ * const { importModule } = this.app.bajo
80
+ * const { json } = await importModule('dobo-extra:/lib/helper.js', { asDefaultImport: false })
81
+ * ...
82
+ * ```
83
+ * @type {object}
84
+ * @property {function} parse - Parse JSON data
85
+ * @property {function} stringify - Stringify data to JSON format
86
+ */
87
+ export const json = {
88
+ parse: (...args) => chain([
89
+ StreamArray.withParser(...args),
90
+ data => data.value
91
+ ]),
92
+ stringify: (options, ...args) => chain([
93
+ new Disassembler(),
94
+ new Stringer({ ...options, makeArray: true })
95
+ ])
96
+ }
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "dobo-extra",
3
- "version": "2.4.0",
3
+ "version": "2.5.1",
4
4
  "description": "Bajo DB Extra Tools/Utility",
5
5
  "main": "index.js",
6
6
  "scripts": {
@@ -20,7 +20,7 @@
20
20
  "keywords": [
21
21
  "extra",
22
22
  "db",
23
- "driver",
23
+ "adapter",
24
24
  "bajo",
25
25
  "framework",
26
26
  "modular"
package/wiki/CHANGES.md CHANGED
@@ -1,5 +1,14 @@
1
1
  # Changes
2
2
 
3
+ ## 2026-07-25
4
+
5
+ - [2.5.1] Bug fix in `exportTo()`
6
+
7
+ ## 2026-07-13
8
+
9
+ - [2.5.0] Reorganize files
10
+ - [2.5.0] Update documentations
11
+
3
12
  ## 2026-06-12
4
13
 
5
14
  - [2.4.0] Necessary updates to ```bajo@2.18.0``` specs
package/lib/export-to.js DELETED
@@ -1,123 +0,0 @@
1
- import path from 'path'
2
- import { json, ndjson, csv, xlsx } from './ndjson-csv-xlsx.js'
3
- import { createGzip } from 'node:zlib'
4
- import scramjet from 'scramjet'
5
- import supportedExt from './io-exts.js'
6
-
7
- const { DataStream } = scramjet
8
-
9
- async function getFile (dest, ensureDir) {
10
- const { importPkg } = this.app.bajo
11
- const { fs } = this.app.lib
12
- const increment = await importPkg('bajo:add-filename-increment')
13
- let file
14
- if (path.isAbsolute(dest)) file = dest
15
- else {
16
- file = `${this.app.getPluginDataDir(this.ns)}/export/${dest}`
17
- fs.ensureDirSync(path.dirname(file))
18
- }
19
- file = increment(file, { fs: true })
20
- const dir = path.dirname(file)
21
- if (!fs.existsSync(dir)) {
22
- if (ensureDir) fs.ensureDirSync(dir)
23
- else throw this.error('dirNotExists%s', dir)
24
- }
25
- let compress = false
26
- let ext = path.extname(file)
27
- if (ext === '.gz') {
28
- compress = true
29
- ext = path.extname(file.slice(0, -3))
30
- // file = file.slice(0, file.length - 3)
31
- }
32
- if (!supportedExt.includes(ext)) throw this.error('unsupportedFormat%s', ext.slice(1))
33
- return { file, ext, compress }
34
- }
35
-
36
- async function getData ({ source, filter, count, stream, progressFn, fields }) {
37
- let cnt = count ?? 0
38
- const { find } = this.app.lib._
39
- const { recordFind, getSchema } = this.app.dobo
40
- const { maxLimit, hardCap } = this.app.dobo.config.default.filter
41
- filter.limit = maxLimit
42
- let sort
43
- const schema = getSchema(source)
44
- const idField = find(schema.properties, { name: 'id' })
45
- for (const name of ['createdAt', 'updatedAt', 'ts', 'dt']) {
46
- const field = find(schema.properties, { name })
47
- if (field) {
48
- sort = field.name
49
- break
50
- }
51
- }
52
- filter.sort = `${sort ?? idField}:1`
53
- for (;;) {
54
- const batchStart = new Date()
55
- const { data, page } = await recordFind(source, filter, { dataOnly: false, fields })
56
- if (data.length === 0) break
57
- if (cnt + data.length > hardCap) {
58
- const sliced = data.slice(0, hardCap - cnt)
59
- await stream.pull(sliced)
60
- cnt += sliced.length
61
- if (progressFn) await progressFn.call(this, { batchNo: page, data: sliced, batchStart, batchEnd: new Date() })
62
- break
63
- }
64
- cnt += data.length
65
- await stream.pull(data)
66
- if (progressFn) await progressFn.call(this, { batchNo: page, data, batchStart, batchEnd: new Date() })
67
- filter.page++
68
- }
69
- await stream.end()
70
- return cnt
71
- }
72
-
73
- function exportTo (source, dest, { filter = {}, ensureDir, useHeader = true, batch = 500, progressFn, fields } = {}, opts = {}) {
74
- const { fs } = this.app.lib
75
- const { merge } = this.app.lib._
76
- const { getInfo } = this.app.dobo
77
-
78
- filter.page = 1
79
- batch = parseInt(batch) ?? 500
80
- if (batch > this.config.export.maxBatch) batch = this.config.export.maxBatch
81
- if (batch < 0) batch = 1
82
- filter.limit = batch
83
-
84
- return new Promise((resolve, reject) => {
85
- let count = 0
86
- let file
87
- let ext
88
- let stream
89
- let compress
90
- let writer
91
- getInfo(source)
92
- getFile.call(this, dest, ensureDir)
93
- .then(res => {
94
- file = res.file
95
- ext = res.ext
96
- compress = res.compress
97
- writer = fs.createWriteStream(file)
98
- writer.on('error', err => {
99
- reject(err)
100
- })
101
- writer.on('finish', () => {
102
- resolve({ file, count })
103
- })
104
- stream = new DataStream()
105
- stream = stream.flatMap(items => (items))
106
- const pipes = []
107
- if (ext === '.json') pipes.push(json.stringify(opts))
108
- else if (['.ndjson', '.jsonl'].includes(ext)) pipes.push(ndjson.stringify(opts))
109
- else if (ext === '.csv') pipes.push(csv.stringify(merge({}, { headers: useHeader }, opts)))
110
- else if (ext === '.tsv') pipes.push(csv.stringify(merge({}, { headers: useHeader }, merge({}, opts, { delimiter: '\t' }))))
111
- else if (ext === '.xlsx') pipes.push(xlsx.stringify(merge({}, { header: useHeader }, opts)))
112
- if (compress) pipes.push(createGzip())
113
- DataStream.pipeline(stream, ...pipes).pipe(writer)
114
- return getData.call(this, { source, filter, count, stream, fields, progressFn })
115
- })
116
- .then(cnt => {
117
- count = cnt
118
- })
119
- .catch(reject)
120
- })
121
- }
122
-
123
- export default exportTo
@@ -1,65 +0,0 @@
1
- import path from 'path'
2
- import { json, ndjson, csv, xlsx } from './ndjson-csv-xlsx.js'
3
- import { createGunzip } from 'zlib'
4
- import supportedExt from './io-exts.js'
5
- import scramjet from 'scramjet'
6
-
7
- const { DataStream } = scramjet
8
-
9
- async function importFrom (source, dest, { trashOld = true, batch = 1, progressFn, converterFn, useHeader = true, fileType, createOpts = {} } = {}, opts = {}) {
10
- const { recordCreate } = this.app.dobo
11
- const { merge } = this.app.lib._
12
- const { fs } = this.app.lib
13
-
14
- if (dest !== false) this.app.dobo.getInfo(dest) // make sure dest model is valid
15
- let file
16
- if (path.isAbsolute(source)) file = source
17
- else {
18
- file = `${this.app.getPluginDataDir(this.ns)}/import/${source}`
19
- fs.ensureDirSync(path.dirname(file))
20
- }
21
- if (!fs.existsSync(file)) throw this.error('sourceFileNotExists%s', file)
22
- let ext = fileType ? `.${fileType}` : path.extname(file)
23
- let decompress = false
24
- if (ext === '.gz') {
25
- ext = path.extname(path.basename(file, '.gz'))
26
- decompress = true
27
- }
28
- if (!supportedExt.includes(ext)) throw this.error('unsupportedFormat%s', ext.slice(1))
29
- if (trashOld && dest !== false) await this.app.dobo.modelClear(dest)
30
- const reader = fs.createReadStream(file)
31
- batch = parseInt(batch) || 100
32
- if (batch > this.config.import.maxBatch) batch = this.config.import.maxBatch
33
- if (batch < 0) batch = 1
34
- let count = 0
35
- const pipes = [reader]
36
- if (decompress) pipes.push(createGunzip())
37
- if (ext === '.json') pipes.push(json.parse(opts))
38
- else if (['.ndjson', '.jsonl'].includes(ext)) pipes.push(ndjson.parse(opts))
39
- else if (ext === '.csv') pipes.push(csv.parse(merge({}, { headers: useHeader }, opts)))
40
- else if (ext === '.tsv') pipes.push(csv.parse(merge({}, { headers: useHeader }, merge({}, opts, { delimiter: '\t' }))))
41
- else if (ext === '.xlsx') pipes.push(xlsx.parse(merge({}, { header: useHeader }, opts)))
42
-
43
- const stream = DataStream.pipeline(...pipes)
44
- let batchNo = 1
45
- const data = []
46
- await stream
47
- .batch(batch)
48
- .map(async items => {
49
- if (items.length === 0) return null
50
- const batchStart = new Date()
51
- for (let item of items) {
52
- count++
53
- item = converterFn ? await converterFn.call(this, item) : item
54
- if (dest !== false) await recordCreate(dest, item, createOpts)
55
- else data.push(item)
56
- }
57
- if (progressFn) await progressFn.call(this, { batchNo, data: items, batchStart, batchEnd: new Date() })
58
- batchNo++
59
- })
60
- .run()
61
-
62
- return dest === false ? data : { file, count }
63
- }
64
-
65
- export default importFrom
package/lib/io-exts.js DELETED
@@ -1 +0,0 @@
1
- export default ['.json', '.jsonl', '.ndjson', '.csv', '.xlsx', '.tsv']