@parseo/shared 1.0.1 → 1.0.3

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.
Files changed (3) hide show
  1. package/dist/extract.d.ts.map +1 -1
  2. package/dist/extract.js +38 -27
  3. package/package.json +9 -2
package/dist/extract.d.ts.map CHANGED
@@ -1 +1 @@
1
- {"version":3,"file":"extract.d.ts","sourceRoot":"","sources":["../src/extract.ts"],"names":[],"mappings":"AAIA,OAAO,KAAK,EAAE,QAAQ,EAAE,QAAQ,EAAe,MAAM,YAAY,CAAC;AAWlE,wBAAsB,gBAAgB,CACpC,MAAM,EAAE,MAAM,GACb,OAAO,CAAC,QAAQ,EAAE,CAAC,CAqErB;AAED,wBAAgB,SAAS,CAAC,KAAK,EAAE,QAAQ,EAAE,GAAG,QAAQ,EAAE,CAgCvD;AA+BD,wBAAgB,YAAY,CAAC,MAAM,EAAE,MAAM,GAAG,OAAO,CAAC,QAAQ,EAAE,CAAC,CAEhE;AAED,gEAAgE;AAChE,MAAM,WAAW,UAAU;IACzB,CAAC,EAAE,MAAM,CAAC;IACV,CAAC,EAAE,MAAM,CAAC;IACV,KAAK,EAAE,MAAM,CAAC;IACd,MAAM,EAAE,MAAM,CAAC;IACf,IAAI,EAAE,MAAM,CAAC;CACd;AAED;;;;GAIG;AACH,wBAAsB,kBAAkB,CACtC,MAAM,EAAE,MAAM,EACd,KAAK,EAAE,MAAM,EAAE,EACf,IAAI,CAAC,EAAE;IAAE,OAAO,CAAC,EAAE,MAAM,CAAC;IAAC,OAAO,CAAC,EAAE,MAAM,CAAA;CAAE,GAC5C,OAAO,CAAC,UAAU,EAAE,CAAC,CAkFvB"}
1
+ {"version":3,"file":"extract.d.ts","sourceRoot":"","sources":["../src/extract.ts"],"names":[],"mappings":"AAIA,OAAO,KAAK,EAAE,QAAQ,EAAE,QAAQ,EAAe,MAAM,YAAY,CAAC;AAWlE,wBAAsB,gBAAgB,CACpC,MAAM,EAAE,MAAM,GACb,OAAO,CAAC,QAAQ,EAAE,CAAC,CAqErB;AAED,wBAAgB,SAAS,CAAC,KAAK,EAAE,QAAQ,EAAE,GAAG,QAAQ,EAAE,CAgCvD;AA+BD,wBAAgB,YAAY,CAAC,MAAM,EAAE,MAAM,GAAG,OAAO,CAAC,QAAQ,EAAE,CAAC,CAEhE;AAED,gEAAgE;AAChE,MAAM,WAAW,UAAU;IACzB,CAAC,EAAE,MAAM,CAAC;IACV,CAAC,EAAE,MAAM,CAAC;IACV,KAAK,EAAE,MAAM,CAAC;IACd,MAAM,EAAE,MAAM,CAAC;IACf,IAAI,EAAE,MAAM,CAAC;CACd;AAED;;;;GAIG;AACH,wBAAsB,kBAAkB,CACtC,MAAM,EAAE,MAAM,EACd,KAAK,EAAE,MAAM,EAAE,EACf,IAAI,CAAC,EAAE;IAAE,OAAO,CAAC,EAAE,MAAM,CAAC;IAAC,OAAO,CAAC,EAAE,MAAM,CAAA;CAAE,GAC5C,OAAO,CAAC,UAAU,EAAE,CAAC,CA8FvB"}
package/dist/extract.js CHANGED
@@ -171,37 +171,48 @@ export async function extractFilledRects(buffer, pages, opts) {
171
171
  ];
172
172
  }
173
173
  else if (op === OPS.constructPath) {
174
- const subOps = args[0];
175
- const subArgs = args[1];
176
- let argIdx = 0;
177
- for (const subOp of subOps) {
178
- if (subOp === OPS.rectangle) {
179
- const rx = subArgs[argIdx], ry = subArgs[argIdx + 1];
180
- const rw = subArgs[argIdx + 2], rh = subArgs[argIdx + 3];
181
- const absW = Math.abs(rw), absH = Math.abs(rh);
182
- if (absW >= minSize && absW <= maxSize && absH >= minSize && absH <= maxSize) {
183
- const tx = currentTransform[0] * rx + currentTransform[2] * ry + currentTransform[4];
184
- const ty = currentTransform[1] * rx + currentTransform[3] * ry + currentTransform[5];
185
- results.push({
186
- x: Math.round(tx * 100) / 100,
187
- y: Math.round((viewport.height - ty) * 100) / 100,
188
- width: Math.round(absW * 100) / 100,
189
- height: Math.round(absH * 100) / 100,
190
- page: pageNum,
191
- });
192
- }
193
- argIdx += 4;
174
+ // pdfjs-dist v5: args = [paintOp, [Float32Array], bounds]
175
+ // Float32Array is interleaved [subOp, coords...] where:
176
+ // 0=moveTo(x,y), 1=lineTo(x,y), 2=curveTo(6 coords), 4=closePath
177
+ const pathData = args[1]?.[0];
178
+ if (!(pathData instanceof Float32Array))
179
+ continue;
180
+ // Collect all points to detect rectangle-like shapes
181
+ const points = [];
182
+ let j = 0;
183
+ while (j < pathData.length) {
184
+ const subOp = pathData[j];
185
+ if (subOp === 0 || subOp === 1) { // moveTo / lineTo
186
+ points.push([pathData[j + 1], pathData[j + 2]]);
187
+ j += 3;
194
188
  }
195
- else if (subOp === OPS.moveTo || subOp === OPS.lineTo) {
196
- argIdx += 2;
189
+ else if (subOp === 2) { // curveTo
190
+ j += 7;
197
191
  }
198
- else if (subOp === OPS.curveTo) {
199
- argIdx += 6;
192
+ else if (subOp === 4) { // closePath
193
+ j += 1;
200
194
  }
201
- else if (subOp === OPS.curveTo2 || subOp === OPS.curveTo3) {
202
- argIdx += 4;
195
+ else {
196
+ j += 1;
197
+ }
198
+ }
199
+ if (points.length >= 2) {
200
+ const xs = points.map(p => p[0]);
201
+ const ys = points.map(p => p[1]);
202
+ const rx = Math.min(...xs), ry = Math.min(...ys);
203
+ const rw = Math.max(...xs) - rx, rh = Math.max(...ys) - ry;
204
+ const absW = Math.abs(rw), absH = Math.abs(rh);
205
+ if (absW >= minSize && absW <= maxSize && absH >= minSize && absH <= maxSize) {
206
+ const tx = currentTransform[0] * rx + currentTransform[2] * ry + currentTransform[4];
207
+ const ty = currentTransform[1] * rx + currentTransform[3] * ry + currentTransform[5];
208
+ results.push({
209
+ x: Math.round(tx * 100) / 100,
210
+ y: Math.round((viewport.height - ty) * 100) / 100,
211
+ width: Math.round(absW * 100) / 100,
212
+ height: Math.round(absH * 100) / 100,
213
+ page: pageNum,
214
+ });
203
215
  }
204
- // closePath has no args
205
216
  }
206
217
  }
207
218
  }
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@parseo/shared",
3
- "version": "1.0.1",
3
+ "version": "1.0.3",
4
4
  "type": "module",
5
5
  "main": "./dist/index.js",
6
6
  "types": "./dist/index.d.ts",
@@ -11,6 +11,13 @@
11
11
  }
12
12
  },
13
13
  "license": "MIT",
14
+ "repository": {
15
+ "type": "git",
16
+ "url": "https://github.com/PabloZimmermann/parseo",
17
+ "directory": "packages/shared"
18
+ },
19
+ "homepage": "https://github.com/PabloZimmermann/parseo#readme",
20
+ "bugs": "https://github.com/PabloZimmermann/parseo/issues",
14
21
  "publishConfig": {
15
22
  "access": "public"
16
23
  },
@@ -21,6 +28,6 @@
21
28
  "build": "tsc"
22
29
  },
23
30
  "dependencies": {
24
- "pdfjs-dist": "^4.9.155"
31
+ "pdfjs-dist": "^5.7.284"
25
32
  }
26
33
  }