icepick 0.4.1

Experimental Rust client for Apache Iceberg with WASM support for AWS S3 Tables and Cloudflare R2
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

//! Convert Iceberg types to Avro values

use crate::error::Result;
use crate::spec::DataFile;
use apache_avro::types::Value;

/// Convert a DataFile to an Avro Record value for manifest entry
pub fn data_file_to_avro(data_file: &DataFile) -> Result<Value> {
    // Partition is a struct/record in Iceberg spec (field 102)
    // For unpartitioned tables (partition_spec_id=0), it's an empty record
    // TODO: When partition support is added, populate fields based on partition spec
    let partition_record = Value::Record(vec![]);

    let mut fields = vec![
        ("content".to_string(), Value::Int(0)), // 0 = DATA
        (
            "file_path".to_string(),
            Value::String(data_file.file_path().to_string()),
        ),
        (
            "file_format".to_string(),
            Value::String(data_file.file_format().to_string()),
        ),
        ("partition".to_string(), partition_record),
        (
            "record_count".to_string(),
            Value::Long(data_file.record_count()),
        ),
        (
            "file_size_in_bytes".to_string(),
            Value::Long(data_file.file_size_in_bytes()),
        ),
    ];

    // Optional column_sizes (array of key-value records)
    let column_sizes = if let Some(sizes) = data_file.column_sizes() {
        let array: Vec<Value> = sizes
            .iter()
            .map(|(k, v)| {
                Value::Record(vec![
                    ("key".to_string(), Value::Int(*k)),
                    ("value".to_string(), Value::Long(*v)),
                ])
            })
            .collect();
        Value::Union(1, Box::new(Value::Array(array)))
    } else {
        Value::Union(0, Box::new(Value::Null))
    };
    fields.push(("column_sizes".to_string(), column_sizes));

    // Optional value_counts (array of key-value records)
    let value_counts = if let Some(counts) = data_file.value_counts() {
        let array: Vec<Value> = counts
            .iter()
            .map(|(k, v)| {
                Value::Record(vec![
                    ("key".to_string(), Value::Int(*k)),
                    ("value".to_string(), Value::Long(*v)),
                ])
            })
            .collect();
        Value::Union(1, Box::new(Value::Array(array)))
    } else {
        Value::Union(0, Box::new(Value::Null))
    };
    fields.push(("value_counts".to_string(), value_counts));

    // Optional null_value_counts (array of key-value records)
    let null_value_counts = if let Some(counts) = data_file.null_value_counts() {
        let array: Vec<Value> = counts
            .iter()
            .map(|(k, v)| {
                Value::Record(vec![
                    ("key".to_string(), Value::Int(*k)),
                    ("value".to_string(), Value::Long(*v)),
                ])
            })
            .collect();
        Value::Union(1, Box::new(Value::Array(array)))
    } else {
        Value::Union(0, Box::new(Value::Null))
    };
    fields.push(("null_value_counts".to_string(), null_value_counts));

    // Optional split offsets
    let split_offsets = if let Some(offsets) = data_file.split_offsets() {
        let array: Vec<Value> = offsets.iter().map(|offset| Value::Long(*offset)).collect();
        Value::Union(1, Box::new(Value::Array(array)))
    } else {
        Value::Union(0, Box::new(Value::Null))
    };
    fields.push(("split_offsets".to_string(), split_offsets));

    // Optional key metadata (encryption)
    let key_metadata = if let Some(bytes) = data_file.key_metadata() {
        Value::Union(1, Box::new(Value::Bytes(bytes.to_vec())))
    } else {
        Value::Union(0, Box::new(Value::Null))
    };
    fields.push(("key_metadata".to_string(), key_metadata));

    // Optional lower_bounds (array of key-value records)
    let lower_bounds = if let Some(bounds) = data_file.lower_bounds() {
        let array: Vec<Value> = bounds
            .iter()
            .map(|(k, v)| {
                Value::Record(vec![
                    ("key".to_string(), Value::Int(*k)),
                    ("value".to_string(), Value::Bytes(v.clone())),
                ])
            })
            .collect();
        Value::Union(1, Box::new(Value::Array(array)))
    } else {
        Value::Union(0, Box::new(Value::Null))
    };
    fields.push(("lower_bounds".to_string(), lower_bounds));

    // Optional upper_bounds (array of key-value records)
    let upper_bounds = if let Some(bounds) = data_file.upper_bounds() {
        let array: Vec<Value> = bounds
            .iter()
            .map(|(k, v)| {
                Value::Record(vec![
                    ("key".to_string(), Value::Int(*k)),
                    ("value".to_string(), Value::Bytes(v.clone())),
                ])
            })
            .collect();
        Value::Union(1, Box::new(Value::Array(array)))
    } else {
        Value::Union(0, Box::new(Value::Null))
    };
    fields.push(("upper_bounds".to_string(), upper_bounds));

    // Optional fields set to null for MVP
    fields.push((
        "equality_ids".to_string(),
        if let Some(ids) = data_file.equality_ids() {
            let values = ids.iter().map(|id| Value::Int(*id)).collect();
            Value::Union(1, Box::new(Value::Array(values)))
        } else {
            Value::Union(0, Box::new(Value::Null))
        },
    ));
    fields.push((
        "sort_order_id".to_string(),
        Value::Union(0, Box::new(Value::Null)),
    ));

    Ok(Value::Record(fields))
}