parze 0.2.2 → 0.2.3
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.
- package/README.md +6 -2
- package/dist/index.d.ts +6 -1
- package/dist/index.js +7 -1
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -35,7 +35,7 @@ const extraction2 = await client.extract(parseResult.text, {
|
|
|
35
35
|
}, parseResult.job_id);
|
|
36
36
|
|
|
37
37
|
// Get AI-suggested schema
|
|
38
|
-
const suggested = await client.suggestSchema(
|
|
38
|
+
const suggested = await client.suggestSchema(parseResult.text);
|
|
39
39
|
console.log(suggested);
|
|
40
40
|
```
|
|
41
41
|
|
|
@@ -62,12 +62,16 @@ Extract structured data from text using a schema.
|
|
|
62
62
|
- `extractionSchema` (object): Schema defining fields to extract
|
|
63
63
|
- `jobId` (string): Job ID from parse response
|
|
64
64
|
|
|
65
|
-
### `extractFile(filePath, extractionSchema)`
|
|
65
|
+
### `extractFile(filePath, extractionSchema, options?)`
|
|
66
66
|
Extract structured data from a file. Parse runs internally.
|
|
67
67
|
|
|
68
68
|
**Parameters:**
|
|
69
69
|
- `filePath` (string): Path to document file
|
|
70
70
|
- `extractionSchema` (object): Schema defining fields to extract
|
|
71
|
+
- `options` (object, optional):
|
|
72
|
+
- `preserveTables`: boolean
|
|
73
|
+
- `preserveLayout`: boolean
|
|
74
|
+
- `extractionMode`: "auto" | "ocr_only" | "llm_only" | "identity_doc"
|
|
71
75
|
|
|
72
76
|
**Returns:** Promise with extracted data and confidence scores
|
|
73
77
|
|
package/dist/index.d.ts
CHANGED
|
@@ -8,6 +8,11 @@ export interface ParseOptions {
|
|
|
8
8
|
preserveLayout?: boolean;
|
|
9
9
|
extractionMode?: 'auto' | 'ocr_only' | 'llm_only' | 'identity_doc';
|
|
10
10
|
}
|
|
11
|
+
export interface ExtractFileOptions {
|
|
12
|
+
preserveTables?: boolean;
|
|
13
|
+
preserveLayout?: boolean;
|
|
14
|
+
extractionMode?: 'auto' | 'ocr_only' | 'llm_only' | 'identity_doc';
|
|
15
|
+
}
|
|
11
16
|
export declare class ParzeClient {
|
|
12
17
|
private apiKey;
|
|
13
18
|
private baseUrl;
|
|
@@ -15,7 +20,7 @@ export declare class ParzeClient {
|
|
|
15
20
|
constructor(options: ParzeClientOptions);
|
|
16
21
|
parse(filePath: string, options?: ParseOptions): Promise<any>;
|
|
17
22
|
extract(text: string, extractionSchema: object, jobId: string): Promise<any>;
|
|
18
|
-
extractFile(filePath: string, extractionSchema: object): Promise<any>;
|
|
23
|
+
extractFile(filePath: string, extractionSchema: object, options?: ExtractFileOptions): Promise<any>;
|
|
19
24
|
suggestSchema(text: string): Promise<any>;
|
|
20
25
|
textToSchema(description: string): Promise<any>;
|
|
21
26
|
}
|
package/dist/index.js
CHANGED
|
@@ -75,10 +75,16 @@ class ParzeClient {
|
|
|
75
75
|
});
|
|
76
76
|
return response.data;
|
|
77
77
|
}
|
|
78
|
-
async extractFile(filePath, extractionSchema) {
|
|
78
|
+
async extractFile(filePath, extractionSchema, options) {
|
|
79
79
|
const form = new form_data_1.default();
|
|
80
80
|
form.append('file', fs.createReadStream(filePath));
|
|
81
81
|
form.append('extraction_schema', JSON.stringify(extractionSchema));
|
|
82
|
+
if ((options === null || options === void 0 ? void 0 : options.preserveTables) !== undefined)
|
|
83
|
+
form.append('preserve_tables', String(options.preserveTables));
|
|
84
|
+
if ((options === null || options === void 0 ? void 0 : options.preserveLayout) !== undefined)
|
|
85
|
+
form.append('preserve_layout', String(options.preserveLayout));
|
|
86
|
+
if (options === null || options === void 0 ? void 0 : options.extractionMode)
|
|
87
|
+
form.append('extraction_mode', options.extractionMode);
|
|
82
88
|
const response = await this.http.post('/extract', form, {
|
|
83
89
|
headers: form.getHeaders(),
|
|
84
90
|
});
|