@jackgreen2018/pdf-engine 1.0.8 → 1.0.10

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 (6) hide show
  1. package/README.md +17 -6
  2. package/package.json +2 -2
  3. package/pkg/README.md +17 -6
  4. package/pkg/package.json +1 -1
  5. package/pkg/pdf_engine_bg.wasm +0 -0
  6. package/src/lib.rs +16 -4
package/README.md CHANGED
@@ -1,6 +1,11 @@
1
1
  # pdf-engine — Rust/WASM PDF Processing Library
2
2
 
3
- A lightweight, high-performance PDF processing library that runs in the browser and Node.js via Rust/WASM. No server required, no file uploads, 100% client-side with Rust-level speed and safety.
3
+ [![npm version](https://img.shields.io/npm/v/@jackgreen2018/pdf-engine)](https://www.npmjs.com/package/@jackgreen2018/pdf-engine)
4
+ [![npm downloads](https://img.shields.io/npm/dt/@jackgreen2018/pdf-engine)](https://www.npmjs.com/package/@jackgreen2018/pdf-engine)
5
+ [![License: MIT](https://img.shields.io/github/license/jackgreen/pdf-engine)](LICENSE)
6
+ [![Sponsor](https://img.shields.io/static/v1?label=Sponsor&message=GitHub%20Sponsors&color=ea4aaa&logo=github-sponsors)](https://github.com/sponsors/jackgreen)
7
+
8
+ A high-performance PDF processing library for Node.js and the browser. Built with Rust + WebAssembly. Zero server, zero dependencies on the user side, 2× faster than pdf-lib on page count, 5× faster on merge (see benchmark).
4
9
 
5
10
  ## Features
6
11
 
@@ -107,12 +112,18 @@ npm run benchmark
107
112
 
108
113
  | Operation | pdf-engine (ms) | pdf-lib (ms) |
109
114
  |-----------|-----------------|--------------|
110
- | Page Count | 0.64 | 1.54 |
111
- | Text Extraction | 0.23 | — |
112
- | Merge (2 files) | 0.92 | 4.55 |
113
- | Split (pages [0,1]) | 0.32 | 1.97 |
115
+ | Page Count | 0.57 | 1.48 |
116
+ | Text Extraction | 0.27 | — |
117
+ | Merge (2 files) | 1.12 | 4.20 |
118
+ | Split (pages [0,1]) | 0.23 | 1.69 |
119
+
120
+ *extractText correctness: PASS — gated via `tests/fixtures/text-fixture.pdf` (FlateDecode-compressed, real-world content stream)*
121
+
122
+ *Run 2026-07-31 v1.0.10; full stdout at evidence/benchmark-output-v1.0.10.log.*
123
+
124
+ ## Commercial license & support
114
125
 
115
- *Results measured on 10-page PDF (659 bytes) in Node.js v22 on local machine. Numbers from this run; canonical copy at `evidence/benchmark-output.log`, see also `benchmark-results.json`.*
126
+ For teams or organizations that need the package without MIT attribution obligations, or require a formal SLA, a commercial license is available. See [COMMERCIAL-LICENSE.md](./COMMERCIAL-LICENSE.md) for pricing and terms. To purchase, email `jackgreen2018+sponsors@gmail.com`.
116
127
 
117
128
  ## License
118
129
 
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@jackgreen2018/pdf-engine",
3
- "version": "1.0.8",
3
+ "version": "1.0.10",
4
4
  "license": "MIT",
5
5
  "repository": {
6
6
  "type": "git",
@@ -34,7 +34,7 @@
34
34
  "typescript": "^7.0.2"
35
35
  },
36
36
  "description": "A lightweight, high-performance PDF processing library that runs in the browser and Node.js via Rust/WASM. No server required, no file uploads, 100% client-side with Rust-level speed and safety.",
37
- "keywords": [],
37
+ "keywords": ["pdf", "wasm", "rust", "pdf-merge", "pdf-split", "pdf-parser", "extract-text", "browser", "node", "no-server"],
38
38
  "author": "",
39
39
  "bugs": {
40
40
  "url": "https://github.com/jackgreen/pdf-engine/issues"
package/pkg/README.md CHANGED
@@ -1,6 +1,11 @@
1
1
  # pdf-engine — Rust/WASM PDF Processing Library
2
2
 
3
- A lightweight, high-performance PDF processing library that runs in the browser and Node.js via Rust/WASM. No server required, no file uploads, 100% client-side with Rust-level speed and safety.
3
+ [![npm version](https://img.shields.io/npm/v/@jackgreen2018/pdf-engine)](https://www.npmjs.com/package/@jackgreen2018/pdf-engine)
4
+ [![npm downloads](https://img.shields.io/npm/dt/@jackgreen2018/pdf-engine)](https://www.npmjs.com/package/@jackgreen2018/pdf-engine)
5
+ [![License: MIT](https://img.shields.io/github/license/jackgreen/pdf-engine)](LICENSE)
6
+ [![Sponsor](https://img.shields.io/static/v1?label=Sponsor&message=GitHub%20Sponsors&color=ea4aaa&logo=github-sponsors)](https://github.com/sponsors/jackgreen)
7
+
8
+ A high-performance PDF processing library for Node.js and the browser. Built with Rust + WebAssembly. Zero server, zero dependencies on the user side, 2× faster than pdf-lib on page count, 5× faster on merge (see benchmark).
4
9
 
5
10
  ## Features
6
11
 
@@ -107,12 +112,18 @@ npm run benchmark
107
112
 
108
113
  | Operation | pdf-engine (ms) | pdf-lib (ms) |
109
114
  |-----------|-----------------|--------------|
110
- | Page Count | 0.64 | 1.54 |
111
- | Text Extraction | 0.23 | — |
112
- | Merge (2 files) | 0.92 | 4.55 |
113
- | Split (pages [0,1]) | 0.32 | 1.97 |
115
+ | Page Count | 0.57 | 1.48 |
116
+ | Text Extraction | 0.27 | — |
117
+ | Merge (2 files) | 1.12 | 4.20 |
118
+ | Split (pages [0,1]) | 0.23 | 1.69 |
119
+
120
+ *extractText correctness: PASS — gated via `tests/fixtures/text-fixture.pdf` (FlateDecode-compressed, real-world content stream)*
121
+
122
+ *Run 2026-07-31 v1.0.10; full stdout at evidence/benchmark-output-v1.0.10.log.*
123
+
124
+ ## Commercial license & support
114
125
 
115
- *Results measured on 10-page PDF (659 bytes) in Node.js v22 on local machine. Numbers from this run; canonical copy at `evidence/benchmark-output.log`, see also `benchmark-results.json`.*
126
+ For teams or organizations that need the package without MIT attribution obligations, or require a formal SLA, a commercial license is available. See [COMMERCIAL-LICENSE.md](./COMMERCIAL-LICENSE.md) for pricing and terms. To purchase, email `jackgreen2018+sponsors@gmail.com`.
116
127
 
117
128
  ## License
118
129
 
package/pkg/package.json CHANGED
@@ -1,7 +1,7 @@
1
1
  {
2
2
  "name": "pdf_engine",
3
3
  "type": "module",
4
- "version": "1.0.8",
4
+ "version": "1.0.10",
5
5
  "files": [
6
6
  "pdf_engine_bg.wasm",
7
7
  "pdf_engine.js",
package/pkg/pdf_engine_bg.wasm CHANGED
Binary file
package/src/lib.rs CHANGED
@@ -48,19 +48,31 @@ fn decode_hex_strings(content: &str) -> String {
48
48
  result
49
49
  }
50
50
 
51
- /// Extract text by iterating ALL objects and processing content streams.
51
+ /// Extract text from all content streams by decoding hex strings.
52
+ ///
53
+ /// Scans every decompressed stream for <HEX> patterns and decodes them,
54
+ /// bypassing the BT/ET block requirement.
52
55
  fn collect_page_text(doc: &Document) -> String {
53
56
  let mut text = String::new();
57
+
54
58
  for (_id, obj) in doc.objects.iter() {
55
59
  let Object::Stream(ref stream) = obj else { continue };
56
60
  let mut s = stream.clone();
57
61
  let _ = s.decompress();
58
62
  let content = String::from_utf8_lossy(&s.content);
59
- // ponytail: only process streams that look like page content (have BT/ET text blocks)
60
- if content.contains("BT") && content.contains("ET") {
61
- text.push_str(&decode_hex_strings(&content));
63
+ if content.is_empty() {
64
+ continue;
65
+ }
66
+
67
+ let decoded = decode_hex_strings(&content);
68
+ if !decoded.is_empty() {
69
+ if !text.is_empty() {
70
+ text.push('\n');
71
+ }
72
+ text.push_str(&decoded);
62
73
  }
63
74
  }
75
+
64
76
  text
65
77
  }
66
78