@llamaindex/liteparse 2.2.0 → 2.2.1
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/README.md +29 -0
- package/liteparse.linux-x64-gnu.node +0 -0
- package/package.json +8 -8
package/README.md
CHANGED
|
@@ -90,6 +90,34 @@ for (const s of screenshots) {
|
|
|
90
90
|
}
|
|
91
91
|
```
|
|
92
92
|
|
|
93
|
+
## Document Complexity
|
|
94
|
+
|
|
95
|
+
Before committing to a full parse, check whether a document needs OCR or heavier
|
|
96
|
+
processing. `isComplex` is a cheap, text-layer-only pass that returns one entry per page
|
|
97
|
+
with a `needsOcr` verdict and the signals behind it — useful for routing documents to
|
|
98
|
+
different pipelines, rejecting ones you can't handle, or estimating cost.
|
|
99
|
+
|
|
100
|
+
```typescript
|
|
101
|
+
const parser = new LiteParse();
|
|
102
|
+
const pages = await parser.isComplex('document.pdf');
|
|
103
|
+
|
|
104
|
+
if (pages.some((p) => p.needsOcr)) {
|
|
105
|
+
// Route to the OCR-enabled pipeline
|
|
106
|
+
const result = await parser.parse('document.pdf');
|
|
107
|
+
} else {
|
|
108
|
+
// Cheap path — skip OCR entirely
|
|
109
|
+
const result = await new LiteParse({ ocrEnabled: false }).parse('document.pdf');
|
|
110
|
+
}
|
|
111
|
+
|
|
112
|
+
// Inspect why specific pages were flagged
|
|
113
|
+
for (const page of pages.filter((p) => p.needsOcr)) {
|
|
114
|
+
console.log(`Page ${page.pageNumber}: ${page.reasons.join(', ')}`);
|
|
115
|
+
}
|
|
116
|
+
```
|
|
117
|
+
|
|
118
|
+
`reasons` is one of `"scanned"`, `"no-text"`, `"sparse-text"`, `"embedded-images"`,
|
|
119
|
+
`"garbled"`, or `"vector-text"`. Raw bytes work here too.
|
|
120
|
+
|
|
93
121
|
## Supported Formats
|
|
94
122
|
|
|
95
123
|
- PDF (`.pdf`)
|
|
@@ -107,4 +135,5 @@ lit parse document.pdf
|
|
|
107
135
|
lit parse document.pdf --format json -o output.json
|
|
108
136
|
lit screenshot document.pdf -o ./screenshots
|
|
109
137
|
lit batch-parse ./input ./output
|
|
138
|
+
lit is-complex document.pdf
|
|
110
139
|
```
|
|
Binary file
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@llamaindex/liteparse",
|
|
3
|
-
"version": "2.2.
|
|
3
|
+
"version": "2.2.1",
|
|
4
4
|
"description": "Fast, lightweight PDF and document parsing with spatial text extraction",
|
|
5
5
|
"type": "module",
|
|
6
6
|
"main": "./dist/lib.js",
|
|
@@ -54,13 +54,13 @@
|
|
|
54
54
|
"typescript": "~5.9.2"
|
|
55
55
|
},
|
|
56
56
|
"optionalDependencies": {
|
|
57
|
-
"@llamaindex/liteparse-darwin-arm64": "2.2.
|
|
58
|
-
"@llamaindex/liteparse-darwin-x64": "2.2.
|
|
59
|
-
"@llamaindex/liteparse-linux-arm64-gnu": "2.2.
|
|
60
|
-
"@llamaindex/liteparse-linux-x64-gnu": "2.2.
|
|
61
|
-
"@llamaindex/liteparse-win32-arm64-msvc": "2.2.
|
|
62
|
-
"@llamaindex/liteparse-win32-x64-msvc": "2.2.
|
|
63
|
-
"@llamaindex/liteparse-linux-x64-musl": "2.2.
|
|
57
|
+
"@llamaindex/liteparse-darwin-arm64": "2.2.1",
|
|
58
|
+
"@llamaindex/liteparse-darwin-x64": "2.2.1",
|
|
59
|
+
"@llamaindex/liteparse-linux-arm64-gnu": "2.2.1",
|
|
60
|
+
"@llamaindex/liteparse-linux-x64-gnu": "2.2.1",
|
|
61
|
+
"@llamaindex/liteparse-win32-arm64-msvc": "2.2.1",
|
|
62
|
+
"@llamaindex/liteparse-win32-x64-msvc": "2.2.1",
|
|
63
|
+
"@llamaindex/liteparse-linux-x64-musl": "2.2.1"
|
|
64
64
|
},
|
|
65
65
|
"engines": {
|
|
66
66
|
"node": ">=18.0.0"
|