xberg 1.1.5

High-performance document intelligence library for Rust. Extract text, metadata, and structured data from PDFs, Office documents, images, and 107 formats and 371 programming languages via tree-sitter code intelligence with async/sync APIs.
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113

{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "$id": "https://docs.xberg.io/schemas/preset.schema.json",
  "title": "Xberg structured-extraction preset",
  "description": "Meta-schema every preset file under crates/presets/library/ must satisfy. Defines the canonical preset format consumed by structured-extraction catalogs.",
  "type": "object",
  "additionalProperties": false,
  "required": [
    "id",
    "version",
    "schema_name",
    "description",
    "category",
    "schema",
    "system_prompt",
    "merge_mode",
    "preferred_call_mode",
    "emit_citations"
  ],
  "properties": {
    "id": {
      "type": "string",
      "pattern": "^[a-z][a-z0-9_]*$",
      "minLength": 2,
      "maxLength": 64,
      "description": "Stable, URL-safe preset identifier. Lowercase snake_case."
    },
    "version": {
      "type": "string",
      "pattern": "^v[0-9]+$",
      "description": "Preset version, monotonically increased on breaking changes."
    },
    "schema_name": {
      "type": "string",
      "minLength": 1,
      "maxLength": 128,
      "description": "Human-readable schema name (used as the LLM tool/response name)."
    },
    "description": {
      "type": "string",
      "minLength": 1,
      "maxLength": 1024
    },
    "category": {
      "type": "string",
      "enum": ["finance", "identity", "legal", "logistics", "medical", "hr", "other"]
    },
    "tags": {
      "type": "array",
      "items": {
        "type": "string",
        "pattern": "^[a-z][a-z0-9_-]*$",
        "maxLength": 64
      },
      "maxItems": 16,
      "uniqueItems": true
    },
    "schema": {
      "type": "object",
      "description": "JSON Schema (draft 2020-12) describing the structured output shape.",
      "required": ["type", "properties"],
      "properties": {
        "type": {
          "const": "object"
        },
        "properties": {
          "type": "object",
          "minProperties": 1
        }
      }
    },
    "system_prompt": {
      "type": "string",
      "minLength": 10,
      "maxLength": 8192,
      "description": "Instruction primer sent to the vision/text model."
    },
    "context_template": {
      "type": "string",
      "maxLength": 4096,
      "description": "Mustache-style template merged with caller-supplied context map."
    },
    "merge_mode": {
      "type": "string",
      "enum": ["object_merge", "array_concat", "object_first"],
      "description": "Strategy for merging per-batch model outputs across paginated calls."
    },
    "preferred_call_mode": {
      "type": "string",
      "enum": ["text_only", "vision_only", "text_plus_vision"],
      "description": "Default call mode suggested for this preset; heuristics may override."
    },
    "emit_citations": {
      "type": "boolean",
      "description": "When true, prompt asks the model to wrap each field as {value, page, bbox, confidence}."
    },
    "sample": {
      "type": "object",
      "additionalProperties": false,
      "required": ["input_path", "output_path"],
      "properties": {
        "input_path": {
          "type": "string",
          "minLength": 1
        },
        "output_path": {
          "type": "string",
          "minLength": 1
        }
      }
    }
  }
}