xberg 1.1.4

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
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278

//! MCP request parameter types.
//!
//! This module defines the parameter structures for all MCP tool calls.

use rmcp::schemars;
#[cfg_attr(alef, alef(skip))]
/// Request parameters for unified extraction.
#[derive(Debug, serde::Deserialize, serde::Serialize, schemars::JsonSchema)]
pub struct ExtractParams {
    /// Unified extraction input: bytes or URI.
    #[schemars(schema_with = "extract_input_schema")]
    pub input: serde_json::Value,
    /// Extraction configuration (JSON object)
    #[serde(skip_serializing_if = "Option::is_none")]
    pub config: Option<serde_json::Value>,
    /// Password for encrypted PDFs
    #[serde(skip_serializing_if = "Option::is_none")]
    pub pdf_password: Option<String>,
    /// Wire format for the response: "json" (default) or "toon"
    #[serde(skip_serializing_if = "Option::is_none")]
    pub response_format: Option<String>,
}
#[cfg_attr(alef, alef(skip))]
/// Request parameters for unified batch extraction.
#[derive(Debug, serde::Deserialize, serde::Serialize, schemars::JsonSchema)]
pub struct ExtractBatchParams {
    /// Unified extraction inputs: bytes and/or URIs.
    #[schemars(schema_with = "extract_inputs_schema")]
    pub inputs: Vec<serde_json::Value>,
    /// Extraction configuration (JSON object)
    #[serde(skip_serializing_if = "Option::is_none")]
    pub config: Option<serde_json::Value>,
    /// Password for encrypted PDFs
    #[serde(skip_serializing_if = "Option::is_none")]
    pub pdf_password: Option<String>,
    /// Wire format for the response: "json" (default) or "toon"
    #[serde(skip_serializing_if = "Option::is_none")]
    pub response_format: Option<String>,
}

/// Request parameters for MIME type detection.
#[cfg_attr(alef, alef(skip))]
#[derive(Debug, serde::Deserialize, serde::Serialize, schemars::JsonSchema)]
pub struct DetectMimeTypeParams {
    /// Path to the file
    pub path: String,
    /// Use content-based detection (default: true)
    #[serde(default = "default_use_content")]
    pub use_content: bool,
}

fn default_use_content() -> bool {
    true
}

/// Empty parameters for tools that take no arguments.
///
/// This generates `{"type": "object", "properties": {}}` which is required by
/// the MCP specification, unlike `()` which generates `{"const": null}`.
#[derive(Debug, serde::Deserialize, serde::Serialize, schemars::JsonSchema)]
pub struct EmptyParams {}

/// Request parameters for cache warm (model download).
#[cfg_attr(alef, alef(skip))]
#[derive(Debug, serde::Deserialize, serde::Serialize, schemars::JsonSchema)]
pub struct CacheWarmParams {
    /// Download all embedding model presets
    #[serde(default)]
    pub all_embeddings: bool,
    /// Specific embedding preset name to download (e.g. "balanced", "speed", "quality")
    #[serde(skip_serializing_if = "Option::is_none")]
    pub embedding_model: Option<String>,
    /// Download the default GLiNER NER model into the standard Hugging Face cache
    #[serde(default)]
    pub ner: bool,
    /// Download every known GLiNER NER model into the standard Hugging Face cache
    #[serde(default)]
    pub all_ner_models: bool,
    /// Specific GLiNER NER model alias or catalog id to download into the standard Hugging Face cache
    #[serde(skip_serializing_if = "Option::is_none")]
    pub ner_model: Option<String>,
}

fn extract_input_schema(_generator: &mut schemars::SchemaGenerator) -> schemars::Schema {
    schemars::json_schema!({
        "type": "object",
        "required": ["kind"],
        "additionalProperties": false,
        "properties": {
            "kind": {
                "type": "string",
                "enum": ["bytes", "uri"],
                "description": "Input source kind."
            },
            "bytes": {
                "type": "array",
                "items": { "type": "integer", "minimum": 0, "maximum": 255 },
                "description": "Raw bytes for kind=bytes."
            },
            "uri": {
                "type": "string",
                "description": "Local path, file:// URI, or HTTP(S) URL for kind=uri."
            },
            "mime_type": {
                "type": "string",
                "description": "Optional MIME type hint."
            },
            "filename": {
                "type": "string",
                "description": "Optional filename hint for bytes inputs."
            },
            "config": {
                "type": "object",
                "description": "Optional per-input extraction overrides."
            }
        }
    })
}

fn extract_inputs_schema(generator: &mut schemars::SchemaGenerator) -> schemars::Schema {
    let input_schema = extract_input_schema(generator);
    schemars::json_schema!({
        "type": "array",
        "items": input_schema
    })
}

#[cfg(test)]
mod tests {
    use super::*;

    #[test]
    fn test_extract_params_defaults() {
        let json = r#"{"input": {"kind": "uri", "uri": "/test.pdf"}}"#;
        let params: ExtractParams = serde_json::from_str(json).unwrap();

        assert_eq!(params.input["uri"], "/test.pdf");
        assert_eq!(params.config, None);
    }

    #[test]
    fn test_extract_params_accepts_bytes_input() {
        let json = r#"{"input": {"kind": "bytes", "bytes": [72, 105], "mime_type": "text/plain"}}"#;
        let params: ExtractParams = serde_json::from_str(json).unwrap();

        assert_eq!(params.input["kind"], "bytes");
        assert_eq!(params.config, None);
    }

    #[test]
    fn test_extract_batch_params_defaults() {
        let json = r#"{"inputs": [{"kind": "uri", "uri": "/a.pdf"}, {"kind": "uri", "uri": "/b.pdf"}]}"#;
        let params: ExtractBatchParams = serde_json::from_str(json).unwrap();

        assert_eq!(params.inputs.len(), 2);
        assert_eq!(params.config, None);
    }

    #[test]
    fn test_detect_mime_type_params_defaults() {
        let json = r#"{"path": "/test.pdf"}"#;
        let params: DetectMimeTypeParams = serde_json::from_str(json).unwrap();

        assert_eq!(params.path, "/test.pdf");
        assert!(params.use_content);
    }

    #[test]
    fn test_detect_mime_type_params_use_content_false() {
        let json = r#"{"path": "/test.pdf", "use_content": false}"#;
        let params: DetectMimeTypeParams = serde_json::from_str(json).unwrap();

        assert!(!params.use_content);
    }

    #[test]
    fn test_extract_params_with_config() {
        let json = r#"{"input": {"kind": "uri", "uri": "/test.pdf"}, "config": {"use_cache": false}}"#;
        let params: ExtractParams = serde_json::from_str(json).unwrap();

        assert_eq!(params.input["uri"], "/test.pdf");
        assert!(params.config.is_some());
    }

    #[test]
    fn test_extract_params_serialization() {
        let params = ExtractParams {
            input: serde_json::json!({
                "kind": "uri",
                "uri": "/test.pdf",
                "mime_type": "application/pdf"
            }),
            config: Some(serde_json::json!({"use_cache": false})),
            pdf_password: None,
            response_format: None,
        };

        let json = serde_json::to_string(&params).unwrap();
        let deserialized: ExtractParams = serde_json::from_str(&json).unwrap();

        assert_eq!(params.input, deserialized.input);
        assert_eq!(params.config, deserialized.config);
    }

    #[test]
    fn test_extract_batch_params_serialization() {
        let params = ExtractBatchParams {
            inputs: vec![
                serde_json::json!({"kind": "uri", "uri": "/a.pdf"}),
                serde_json::json!({"kind": "uri", "uri": "/b.pdf"}),
            ],
            config: Some(serde_json::json!({"use_cache": true})),
            pdf_password: None,
            response_format: None,
        };

        let json = serde_json::to_string(&params).unwrap();
        let deserialized: ExtractBatchParams = serde_json::from_str(&json).unwrap();

        assert_eq!(params.inputs, deserialized.inputs);
        assert_eq!(params.config, deserialized.config);
    }

    #[test]
    fn test_detect_mime_type_params_serialization() {
        let params = DetectMimeTypeParams {
            path: "/test.pdf".to_string(),
            use_content: false,
        };

        let json = serde_json::to_string(&params).unwrap();
        let deserialized: DetectMimeTypeParams = serde_json::from_str(&json).unwrap();

        assert_eq!(params.path, deserialized.path);
        assert_eq!(params.use_content, deserialized.use_content);
    }

    #[test]
    fn test_empty_params_schema_has_type_object() {
        let schema = schemars::schema_for!(EmptyParams);
        let json = serde_json::to_value(&schema).unwrap();
        assert_eq!(json["type"], "object");
    }

    #[test]
    fn test_empty_params_deserializes_from_empty_object() {
        let params: EmptyParams = serde_json::from_str("{}").unwrap();
        let _ = params;
    }

    #[test]
    fn test_cache_warm_params_defaults() {
        let json = r#"{}"#;
        let params: CacheWarmParams = serde_json::from_str(json).unwrap();
        assert!(!params.all_embeddings);
        assert!(params.embedding_model.is_none());
        assert!(!params.ner);
        assert!(!params.all_ner_models);
        assert!(params.ner_model.is_none());
    }

    #[test]
    fn test_cache_warm_params_with_values() {
        let json = r#"{
            "all_embeddings": true,
            "embedding_model": "balanced",
            "ner": true,
            "all_ner_models": true,
            "ner_model": "gliner_small-v2.5"
        }"#;
        let params: CacheWarmParams = serde_json::from_str(json).unwrap();
        assert!(params.all_embeddings);
        assert_eq!(params.embedding_model.as_deref(), Some("balanced"));
        assert!(params.ner);
        assert!(params.all_ner_models);
        assert_eq!(params.ner_model.as_deref(), Some("gliner_small-v2.5"));
    }
}