@kreuzberg/kreuzcrawl 0.1.2 → 0.2.0

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/Cargo.toml CHANGED
@@ -1,6 +1,6 @@
1
1
  [package]
2
2
  name = "kreuzcrawl-node"
3
- version = "0.1.2"
3
+ version = "0.2.0"
4
4
  edition = "2024"
5
5
  license = "Elastic-2.0"
6
6
 
package/index.d.ts CHANGED
@@ -86,6 +86,21 @@ export interface JsCitationResult {
86
86
  references?: Array<JsCitationReference>
87
87
  }
88
88
 
89
+ export interface JsContentConfig {
90
+ outputFormat?: string
91
+ preprocessingPreset?: string
92
+ removeNavigation?: boolean
93
+ removeForms?: boolean
94
+ stripTags?: Array<string>
95
+ preserveTags?: Array<string>
96
+ excludeSelectors?: Array<string>
97
+ skipImages?: boolean
98
+ maxDepth?: number
99
+ wrap?: boolean
100
+ wrapWidth?: number
101
+ includeDocumentStructure?: boolean
102
+ }
103
+
89
104
  export interface JsCookieInfo {
90
105
  name?: string
91
106
  value?: string
@@ -112,8 +127,8 @@ export interface JsCrawlConfig {
112
127
  cookiesEnabled?: boolean
113
128
  auth?: JsAuthConfig
114
129
  maxBodySize?: number
115
- mainContentOnly?: boolean
116
130
  removeTags?: Array<string>
131
+ content?: JsContentConfig
117
132
  mapLimit?: number
118
133
  mapSearch?: string
119
134
  downloadAssets?: boolean
@@ -349,7 +364,6 @@ export interface JsScrapeResult {
349
364
  isPdf?: boolean
350
365
  wasSkipped?: boolean
351
366
  detectedCharset?: string
352
- mainContentOnly?: boolean
353
367
  authHeaderSent?: boolean
354
368
  responseMeta?: JsResponseMeta
355
369
  assets?: Array<JsDownloadedAsset>
package/index.js CHANGED
@@ -77,8 +77,8 @@ function requireNative() {
77
77
  try {
78
78
  const binding = require('@kreuzberg/kreuzcrawl-android-arm64')
79
79
  const bindingPackageVersion = require('@kreuzberg/kreuzcrawl-android-arm64/package.json').version
80
- if (bindingPackageVersion !== '0.1.2' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
81
- throw new Error(`Native binding package version mismatch, expected 0.1.2 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
80
+ if (bindingPackageVersion !== '0.2.0' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
81
+ throw new Error(`Native binding package version mismatch, expected 0.2.0 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
82
82
  }
83
83
  return binding
84
84
  } catch (e) {
@@ -93,8 +93,8 @@ function requireNative() {
93
93
  try {
94
94
  const binding = require('@kreuzberg/kreuzcrawl-android-arm-eabi')
95
95
  const bindingPackageVersion = require('@kreuzberg/kreuzcrawl-android-arm-eabi/package.json').version
96
- if (bindingPackageVersion !== '0.1.2' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
97
- throw new Error(`Native binding package version mismatch, expected 0.1.2 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
96
+ if (bindingPackageVersion !== '0.2.0' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
97
+ throw new Error(`Native binding package version mismatch, expected 0.2.0 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
98
98
  }
99
99
  return binding
100
100
  } catch (e) {
@@ -114,8 +114,8 @@ function requireNative() {
114
114
  try {
115
115
  const binding = require('@kreuzberg/kreuzcrawl-win32-x64-gnu')
116
116
  const bindingPackageVersion = require('@kreuzberg/kreuzcrawl-win32-x64-gnu/package.json').version
117
- if (bindingPackageVersion !== '0.1.2' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
118
- throw new Error(`Native binding package version mismatch, expected 0.1.2 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
117
+ if (bindingPackageVersion !== '0.2.0' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
118
+ throw new Error(`Native binding package version mismatch, expected 0.2.0 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
119
119
  }
120
120
  return binding
121
121
  } catch (e) {
@@ -130,8 +130,8 @@ function requireNative() {
130
130
  try {
131
131
  const binding = require('@kreuzberg/kreuzcrawl-win32-x64-msvc')
132
132
  const bindingPackageVersion = require('@kreuzberg/kreuzcrawl-win32-x64-msvc/package.json').version
133
- if (bindingPackageVersion !== '0.1.2' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
134
- throw new Error(`Native binding package version mismatch, expected 0.1.2 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
133
+ if (bindingPackageVersion !== '0.2.0' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
134
+ throw new Error(`Native binding package version mismatch, expected 0.2.0 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
135
135
  }
136
136
  return binding
137
137
  } catch (e) {
@@ -147,8 +147,8 @@ function requireNative() {
147
147
  try {
148
148
  const binding = require('@kreuzberg/kreuzcrawl-win32-ia32-msvc')
149
149
  const bindingPackageVersion = require('@kreuzberg/kreuzcrawl-win32-ia32-msvc/package.json').version
150
- if (bindingPackageVersion !== '0.1.2' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
151
- throw new Error(`Native binding package version mismatch, expected 0.1.2 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
150
+ if (bindingPackageVersion !== '0.2.0' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
151
+ throw new Error(`Native binding package version mismatch, expected 0.2.0 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
152
152
  }
153
153
  return binding
154
154
  } catch (e) {
@@ -163,8 +163,8 @@ function requireNative() {
163
163
  try {
164
164
  const binding = require('@kreuzberg/kreuzcrawl-win32-arm64-msvc')
165
165
  const bindingPackageVersion = require('@kreuzberg/kreuzcrawl-win32-arm64-msvc/package.json').version
166
- if (bindingPackageVersion !== '0.1.2' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
167
- throw new Error(`Native binding package version mismatch, expected 0.1.2 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
166
+ if (bindingPackageVersion !== '0.2.0' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
167
+ throw new Error(`Native binding package version mismatch, expected 0.2.0 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
168
168
  }
169
169
  return binding
170
170
  } catch (e) {
@@ -182,8 +182,8 @@ function requireNative() {
182
182
  try {
183
183
  const binding = require('@kreuzberg/kreuzcrawl-darwin-universal')
184
184
  const bindingPackageVersion = require('@kreuzberg/kreuzcrawl-darwin-universal/package.json').version
185
- if (bindingPackageVersion !== '0.1.2' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
186
- throw new Error(`Native binding package version mismatch, expected 0.1.2 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
185
+ if (bindingPackageVersion !== '0.2.0' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
186
+ throw new Error(`Native binding package version mismatch, expected 0.2.0 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
187
187
  }
188
188
  return binding
189
189
  } catch (e) {
@@ -198,8 +198,8 @@ function requireNative() {
198
198
  try {
199
199
  const binding = require('@kreuzberg/kreuzcrawl-darwin-x64')
200
200
  const bindingPackageVersion = require('@kreuzberg/kreuzcrawl-darwin-x64/package.json').version
201
- if (bindingPackageVersion !== '0.1.2' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
202
- throw new Error(`Native binding package version mismatch, expected 0.1.2 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
201
+ if (bindingPackageVersion !== '0.2.0' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
202
+ throw new Error(`Native binding package version mismatch, expected 0.2.0 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
203
203
  }
204
204
  return binding
205
205
  } catch (e) {
@@ -214,8 +214,8 @@ function requireNative() {
214
214
  try {
215
215
  const binding = require('@kreuzberg/kreuzcrawl-darwin-arm64')
216
216
  const bindingPackageVersion = require('@kreuzberg/kreuzcrawl-darwin-arm64/package.json').version
217
- if (bindingPackageVersion !== '0.1.2' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
218
- throw new Error(`Native binding package version mismatch, expected 0.1.2 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
217
+ if (bindingPackageVersion !== '0.2.0' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
218
+ throw new Error(`Native binding package version mismatch, expected 0.2.0 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
219
219
  }
220
220
  return binding
221
221
  } catch (e) {
@@ -234,8 +234,8 @@ function requireNative() {
234
234
  try {
235
235
  const binding = require('@kreuzberg/kreuzcrawl-freebsd-x64')
236
236
  const bindingPackageVersion = require('@kreuzberg/kreuzcrawl-freebsd-x64/package.json').version
237
- if (bindingPackageVersion !== '0.1.2' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
238
- throw new Error(`Native binding package version mismatch, expected 0.1.2 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
237
+ if (bindingPackageVersion !== '0.2.0' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
238
+ throw new Error(`Native binding package version mismatch, expected 0.2.0 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
239
239
  }
240
240
  return binding
241
241
  } catch (e) {
@@ -250,8 +250,8 @@ function requireNative() {
250
250
  try {
251
251
  const binding = require('@kreuzberg/kreuzcrawl-freebsd-arm64')
252
252
  const bindingPackageVersion = require('@kreuzberg/kreuzcrawl-freebsd-arm64/package.json').version
253
- if (bindingPackageVersion !== '0.1.2' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
254
- throw new Error(`Native binding package version mismatch, expected 0.1.2 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
253
+ if (bindingPackageVersion !== '0.2.0' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
254
+ throw new Error(`Native binding package version mismatch, expected 0.2.0 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
255
255
  }
256
256
  return binding
257
257
  } catch (e) {
@@ -271,8 +271,8 @@ function requireNative() {
271
271
  try {
272
272
  const binding = require('@kreuzberg/kreuzcrawl-linux-x64-musl')
273
273
  const bindingPackageVersion = require('@kreuzberg/kreuzcrawl-linux-x64-musl/package.json').version
274
- if (bindingPackageVersion !== '0.1.2' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
275
- throw new Error(`Native binding package version mismatch, expected 0.1.2 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
274
+ if (bindingPackageVersion !== '0.2.0' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
275
+ throw new Error(`Native binding package version mismatch, expected 0.2.0 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
276
276
  }
277
277
  return binding
278
278
  } catch (e) {
@@ -287,8 +287,8 @@ function requireNative() {
287
287
  try {
288
288
  const binding = require('@kreuzberg/kreuzcrawl-linux-x64-gnu')
289
289
  const bindingPackageVersion = require('@kreuzberg/kreuzcrawl-linux-x64-gnu/package.json').version
290
- if (bindingPackageVersion !== '0.1.2' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
291
- throw new Error(`Native binding package version mismatch, expected 0.1.2 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
290
+ if (bindingPackageVersion !== '0.2.0' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
291
+ throw new Error(`Native binding package version mismatch, expected 0.2.0 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
292
292
  }
293
293
  return binding
294
294
  } catch (e) {
@@ -305,8 +305,8 @@ function requireNative() {
305
305
  try {
306
306
  const binding = require('@kreuzberg/kreuzcrawl-linux-arm64-musl')
307
307
  const bindingPackageVersion = require('@kreuzberg/kreuzcrawl-linux-arm64-musl/package.json').version
308
- if (bindingPackageVersion !== '0.1.2' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
309
- throw new Error(`Native binding package version mismatch, expected 0.1.2 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
308
+ if (bindingPackageVersion !== '0.2.0' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
309
+ throw new Error(`Native binding package version mismatch, expected 0.2.0 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
310
310
  }
311
311
  return binding
312
312
  } catch (e) {
@@ -321,8 +321,8 @@ function requireNative() {
321
321
  try {
322
322
  const binding = require('@kreuzberg/kreuzcrawl-linux-arm64-gnu')
323
323
  const bindingPackageVersion = require('@kreuzberg/kreuzcrawl-linux-arm64-gnu/package.json').version
324
- if (bindingPackageVersion !== '0.1.2' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
325
- throw new Error(`Native binding package version mismatch, expected 0.1.2 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
324
+ if (bindingPackageVersion !== '0.2.0' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
325
+ throw new Error(`Native binding package version mismatch, expected 0.2.0 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
326
326
  }
327
327
  return binding
328
328
  } catch (e) {
@@ -339,8 +339,8 @@ function requireNative() {
339
339
  try {
340
340
  const binding = require('@kreuzberg/kreuzcrawl-linux-arm-musleabihf')
341
341
  const bindingPackageVersion = require('@kreuzberg/kreuzcrawl-linux-arm-musleabihf/package.json').version
342
- if (bindingPackageVersion !== '0.1.2' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
343
- throw new Error(`Native binding package version mismatch, expected 0.1.2 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
342
+ if (bindingPackageVersion !== '0.2.0' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
343
+ throw new Error(`Native binding package version mismatch, expected 0.2.0 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
344
344
  }
345
345
  return binding
346
346
  } catch (e) {
@@ -355,8 +355,8 @@ function requireNative() {
355
355
  try {
356
356
  const binding = require('@kreuzberg/kreuzcrawl-linux-arm-gnueabihf')
357
357
  const bindingPackageVersion = require('@kreuzberg/kreuzcrawl-linux-arm-gnueabihf/package.json').version
358
- if (bindingPackageVersion !== '0.1.2' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
359
- throw new Error(`Native binding package version mismatch, expected 0.1.2 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
358
+ if (bindingPackageVersion !== '0.2.0' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
359
+ throw new Error(`Native binding package version mismatch, expected 0.2.0 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
360
360
  }
361
361
  return binding
362
362
  } catch (e) {
@@ -373,8 +373,8 @@ function requireNative() {
373
373
  try {
374
374
  const binding = require('@kreuzberg/kreuzcrawl-linux-loong64-musl')
375
375
  const bindingPackageVersion = require('@kreuzberg/kreuzcrawl-linux-loong64-musl/package.json').version
376
- if (bindingPackageVersion !== '0.1.2' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
377
- throw new Error(`Native binding package version mismatch, expected 0.1.2 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
376
+ if (bindingPackageVersion !== '0.2.0' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
377
+ throw new Error(`Native binding package version mismatch, expected 0.2.0 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
378
378
  }
379
379
  return binding
380
380
  } catch (e) {
@@ -389,8 +389,8 @@ function requireNative() {
389
389
  try {
390
390
  const binding = require('@kreuzberg/kreuzcrawl-linux-loong64-gnu')
391
391
  const bindingPackageVersion = require('@kreuzberg/kreuzcrawl-linux-loong64-gnu/package.json').version
392
- if (bindingPackageVersion !== '0.1.2' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
393
- throw new Error(`Native binding package version mismatch, expected 0.1.2 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
392
+ if (bindingPackageVersion !== '0.2.0' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
393
+ throw new Error(`Native binding package version mismatch, expected 0.2.0 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
394
394
  }
395
395
  return binding
396
396
  } catch (e) {
@@ -407,8 +407,8 @@ function requireNative() {
407
407
  try {
408
408
  const binding = require('@kreuzberg/kreuzcrawl-linux-riscv64-musl')
409
409
  const bindingPackageVersion = require('@kreuzberg/kreuzcrawl-linux-riscv64-musl/package.json').version
410
- if (bindingPackageVersion !== '0.1.2' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
411
- throw new Error(`Native binding package version mismatch, expected 0.1.2 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
410
+ if (bindingPackageVersion !== '0.2.0' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
411
+ throw new Error(`Native binding package version mismatch, expected 0.2.0 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
412
412
  }
413
413
  return binding
414
414
  } catch (e) {
@@ -423,8 +423,8 @@ function requireNative() {
423
423
  try {
424
424
  const binding = require('@kreuzberg/kreuzcrawl-linux-riscv64-gnu')
425
425
  const bindingPackageVersion = require('@kreuzberg/kreuzcrawl-linux-riscv64-gnu/package.json').version
426
- if (bindingPackageVersion !== '0.1.2' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
427
- throw new Error(`Native binding package version mismatch, expected 0.1.2 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
426
+ if (bindingPackageVersion !== '0.2.0' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
427
+ throw new Error(`Native binding package version mismatch, expected 0.2.0 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
428
428
  }
429
429
  return binding
430
430
  } catch (e) {
@@ -440,8 +440,8 @@ function requireNative() {
440
440
  try {
441
441
  const binding = require('@kreuzberg/kreuzcrawl-linux-ppc64-gnu')
442
442
  const bindingPackageVersion = require('@kreuzberg/kreuzcrawl-linux-ppc64-gnu/package.json').version
443
- if (bindingPackageVersion !== '0.1.2' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
444
- throw new Error(`Native binding package version mismatch, expected 0.1.2 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
443
+ if (bindingPackageVersion !== '0.2.0' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
444
+ throw new Error(`Native binding package version mismatch, expected 0.2.0 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
445
445
  }
446
446
  return binding
447
447
  } catch (e) {
@@ -456,8 +456,8 @@ function requireNative() {
456
456
  try {
457
457
  const binding = require('@kreuzberg/kreuzcrawl-linux-s390x-gnu')
458
458
  const bindingPackageVersion = require('@kreuzberg/kreuzcrawl-linux-s390x-gnu/package.json').version
459
- if (bindingPackageVersion !== '0.1.2' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
460
- throw new Error(`Native binding package version mismatch, expected 0.1.2 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
459
+ if (bindingPackageVersion !== '0.2.0' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
460
+ throw new Error(`Native binding package version mismatch, expected 0.2.0 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
461
461
  }
462
462
  return binding
463
463
  } catch (e) {
@@ -476,8 +476,8 @@ function requireNative() {
476
476
  try {
477
477
  const binding = require('@kreuzberg/kreuzcrawl-openharmony-arm64')
478
478
  const bindingPackageVersion = require('@kreuzberg/kreuzcrawl-openharmony-arm64/package.json').version
479
- if (bindingPackageVersion !== '0.1.2' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
480
- throw new Error(`Native binding package version mismatch, expected 0.1.2 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
479
+ if (bindingPackageVersion !== '0.2.0' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
480
+ throw new Error(`Native binding package version mismatch, expected 0.2.0 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
481
481
  }
482
482
  return binding
483
483
  } catch (e) {
@@ -492,8 +492,8 @@ function requireNative() {
492
492
  try {
493
493
  const binding = require('@kreuzberg/kreuzcrawl-openharmony-x64')
494
494
  const bindingPackageVersion = require('@kreuzberg/kreuzcrawl-openharmony-x64/package.json').version
495
- if (bindingPackageVersion !== '0.1.2' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
496
- throw new Error(`Native binding package version mismatch, expected 0.1.2 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
495
+ if (bindingPackageVersion !== '0.2.0' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
496
+ throw new Error(`Native binding package version mismatch, expected 0.2.0 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
497
497
  }
498
498
  return binding
499
499
  } catch (e) {
@@ -508,8 +508,8 @@ function requireNative() {
508
508
  try {
509
509
  const binding = require('@kreuzberg/kreuzcrawl-openharmony-arm')
510
510
  const bindingPackageVersion = require('@kreuzberg/kreuzcrawl-openharmony-arm/package.json').version
511
- if (bindingPackageVersion !== '0.1.2' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
512
- throw new Error(`Native binding package version mismatch, expected 0.1.2 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
511
+ if (bindingPackageVersion !== '0.2.0' && process.env.NAPI_RS_ENFORCE_VERSION_CHECK && process.env.NAPI_RS_ENFORCE_VERSION_CHECK !== '0') {
512
+ throw new Error(`Native binding package version mismatch, expected 0.2.0 but got ${bindingPackageVersion}. You can reinstall dependencies to fix this issue.`)
513
513
  }
514
514
  return binding
515
515
  } catch (e) {
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@kreuzberg/kreuzcrawl-darwin-arm64",
3
- "version": "0.1.2",
3
+ "version": "0.2.0",
4
4
  "cpu": [
5
5
  "arm64"
6
6
  ],
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@kreuzberg/kreuzcrawl-linux-arm64-gnu",
3
- "version": "0.1.2",
3
+ "version": "0.2.0",
4
4
  "cpu": [
5
5
  "arm64"
6
6
  ],
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@kreuzberg/kreuzcrawl-linux-x64-gnu",
3
- "version": "0.1.2",
3
+ "version": "0.2.0",
4
4
  "cpu": [
5
5
  "x64"
6
6
  ],
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@kreuzberg/kreuzcrawl-win32-x64-msvc",
3
- "version": "0.1.2",
3
+ "version": "0.2.0",
4
4
  "cpu": [
5
5
  "x64"
6
6
  ],
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@kreuzberg/kreuzcrawl",
3
- "version": "0.1.2",
3
+ "version": "0.2.0",
4
4
  "main": "index.js",
5
5
  "types": "index.d.ts",
6
6
  "repository": {
@@ -35,9 +35,9 @@
35
35
  "@napi-rs/cli": "^3.6.2"
36
36
  },
37
37
  "optionalDependencies": {
38
- "@kreuzberg/kreuzcrawl-darwin-arm64": "0.1.2",
39
- "@kreuzberg/kreuzcrawl-linux-arm64-gnu": "0.1.2",
40
- "@kreuzberg/kreuzcrawl-linux-x64-gnu": "0.1.2",
41
- "@kreuzberg/kreuzcrawl-win32-x64-msvc": "0.1.2"
38
+ "@kreuzberg/kreuzcrawl-darwin-arm64": "0.2.0",
39
+ "@kreuzberg/kreuzcrawl-linux-arm64-gnu": "0.2.0",
40
+ "@kreuzberg/kreuzcrawl-linux-x64-gnu": "0.2.0",
41
+ "@kreuzberg/kreuzcrawl-win32-x64-msvc": "0.2.0"
42
42
  }
43
43
  }
package/src/lib.rs CHANGED
@@ -48,6 +48,34 @@ pub struct JsProxyConfig {
48
48
  pub password: Option<String>,
49
49
  }
50
50
 
51
+ #[derive(Clone, Default, serde::Serialize, serde::Deserialize)]
52
+ #[napi(object)]
53
+ pub struct JsContentConfig {
54
+ #[napi(js_name = "outputFormat")]
55
+ pub output_format: Option<String>,
56
+ #[napi(js_name = "preprocessingPreset")]
57
+ pub preprocessing_preset: Option<String>,
58
+ #[napi(js_name = "removeNavigation")]
59
+ pub remove_navigation: Option<bool>,
60
+ #[napi(js_name = "removeForms")]
61
+ pub remove_forms: Option<bool>,
62
+ #[napi(js_name = "stripTags")]
63
+ pub strip_tags: Option<Vec<String>>,
64
+ #[napi(js_name = "preserveTags")]
65
+ pub preserve_tags: Option<Vec<String>>,
66
+ #[napi(js_name = "excludeSelectors")]
67
+ pub exclude_selectors: Option<Vec<String>>,
68
+ #[napi(js_name = "skipImages")]
69
+ pub skip_images: Option<bool>,
70
+ #[napi(js_name = "maxDepth")]
71
+ pub max_depth: Option<i64>,
72
+ pub wrap: Option<bool>,
73
+ #[napi(js_name = "wrapWidth")]
74
+ pub wrap_width: Option<i64>,
75
+ #[napi(js_name = "includeDocumentStructure")]
76
+ pub include_document_structure: Option<bool>,
77
+ }
78
+
51
79
  #[derive(Clone, Default, serde::Serialize, serde::Deserialize)]
52
80
  #[napi(object)]
53
81
  pub struct JsBrowserConfig {
@@ -99,10 +127,9 @@ pub struct JsCrawlConfig {
99
127
  pub auth: Option<JsAuthConfig>,
100
128
  #[napi(js_name = "maxBodySize")]
101
129
  pub max_body_size: Option<i64>,
102
- #[napi(js_name = "mainContentOnly")]
103
- pub main_content_only: Option<bool>,
104
130
  #[napi(js_name = "removeTags")]
105
131
  pub remove_tags: Option<Vec<String>>,
132
+ pub content: Option<JsContentConfig>,
106
133
  #[napi(js_name = "mapLimit")]
107
134
  pub map_limit: Option<i64>,
108
135
  #[napi(js_name = "mapSearch")]
@@ -179,8 +206,6 @@ pub struct JsScrapeResult {
179
206
  pub was_skipped: Option<bool>,
180
207
  #[napi(js_name = "detectedCharset")]
181
208
  pub detected_charset: Option<String>,
182
- #[napi(js_name = "mainContentOnly")]
183
- pub main_content_only: Option<bool>,
184
209
  #[napi(js_name = "authHeaderSent")]
185
210
  pub auth_header_sent: Option<bool>,
186
211
  #[napi(js_name = "responseMeta")]
@@ -734,6 +759,44 @@ impl From<kreuzcrawl::ProxyConfig> for JsProxyConfig {
734
759
  }
735
760
  }
736
761
 
762
+ impl From<JsContentConfig> for kreuzcrawl::ContentConfig {
763
+ fn from(val: JsContentConfig) -> Self {
764
+ Self {
765
+ output_format: val.output_format.unwrap_or_default(),
766
+ preprocessing_preset: val.preprocessing_preset.unwrap_or_default(),
767
+ remove_navigation: val.remove_navigation.unwrap_or_default(),
768
+ remove_forms: val.remove_forms.unwrap_or_default(),
769
+ strip_tags: val.strip_tags.unwrap_or_default(),
770
+ preserve_tags: val.preserve_tags.unwrap_or_default(),
771
+ exclude_selectors: val.exclude_selectors.unwrap_or_default(),
772
+ skip_images: val.skip_images.unwrap_or_default(),
773
+ max_depth: val.max_depth.map(|v| v as usize),
774
+ wrap: val.wrap.unwrap_or_default(),
775
+ wrap_width: val.wrap_width.map(|v| v as usize).unwrap_or_default(),
776
+ include_document_structure: val.include_document_structure.unwrap_or_default(),
777
+ }
778
+ }
779
+ }
780
+
781
+ impl From<kreuzcrawl::ContentConfig> for JsContentConfig {
782
+ fn from(val: kreuzcrawl::ContentConfig) -> Self {
783
+ Self {
784
+ output_format: Some(val.output_format),
785
+ preprocessing_preset: Some(val.preprocessing_preset),
786
+ remove_navigation: Some(val.remove_navigation),
787
+ remove_forms: Some(val.remove_forms),
788
+ strip_tags: Some(val.strip_tags),
789
+ preserve_tags: Some(val.preserve_tags),
790
+ exclude_selectors: Some(val.exclude_selectors),
791
+ skip_images: Some(val.skip_images),
792
+ max_depth: val.max_depth.map(|v| v as i64),
793
+ wrap: Some(val.wrap),
794
+ wrap_width: Some(val.wrap_width as i64),
795
+ include_document_structure: Some(val.include_document_structure),
796
+ }
797
+ }
798
+ }
799
+
737
800
  #[allow(clippy::field_reassign_with_default)]
738
801
  impl From<JsBrowserConfig> for kreuzcrawl::BrowserConfig {
739
802
  fn from(val: JsBrowserConfig) -> Self {
@@ -788,8 +851,8 @@ impl From<JsCrawlConfig> for kreuzcrawl::CrawlConfig {
788
851
  __result.cookies_enabled = val.cookies_enabled.unwrap_or_default();
789
852
  __result.auth = val.auth.map(Into::into);
790
853
  __result.max_body_size = val.max_body_size.map(|v| v as usize);
791
- __result.main_content_only = val.main_content_only.unwrap_or_default();
792
854
  __result.remove_tags = val.remove_tags.unwrap_or_default();
855
+ __result.content = val.content.map(Into::into).unwrap_or_default();
793
856
  __result.map_limit = val.map_limit.map(|v| v as usize);
794
857
  __result.map_search = val.map_search;
795
858
  __result.download_assets = val.download_assets.unwrap_or_default();
@@ -833,8 +896,8 @@ impl From<kreuzcrawl::CrawlConfig> for JsCrawlConfig {
833
896
  cookies_enabled: Some(val.cookies_enabled),
834
897
  auth: val.auth.map(Into::into),
835
898
  max_body_size: val.max_body_size.map(|v| v as i64),
836
- main_content_only: Some(val.main_content_only),
837
899
  remove_tags: Some(val.remove_tags),
900
+ content: Some(val.content.into()),
838
901
  map_limit: val.map_limit.map(|v| v as i64),
839
902
  map_search: val.map_search,
840
903
  download_assets: Some(val.download_assets),
@@ -919,7 +982,6 @@ impl From<JsScrapeResult> for kreuzcrawl::ScrapeResult {
919
982
  is_pdf: val.is_pdf.unwrap_or_default(),
920
983
  was_skipped: val.was_skipped.unwrap_or_default(),
921
984
  detected_charset: val.detected_charset,
922
- main_content_only: val.main_content_only.unwrap_or_default(),
923
985
  auth_header_sent: val.auth_header_sent.unwrap_or_default(),
924
986
  response_meta: val.response_meta.map(Into::into),
925
987
  assets: val
@@ -957,7 +1019,6 @@ impl From<kreuzcrawl::ScrapeResult> for JsScrapeResult {
957
1019
  is_pdf: Some(val.is_pdf),
958
1020
  was_skipped: Some(val.was_skipped),
959
1021
  detected_charset: val.detected_charset,
960
- main_content_only: Some(val.main_content_only),
961
1022
  auth_header_sent: Some(val.auth_header_sent),
962
1023
  response_meta: val.response_meta.map(Into::into),
963
1024
  assets: Some(val.assets.into_iter().map(Into::into).collect()),