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
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
// Copyright Materialize, Inc. and contributors. All rights reserved.
//
// Use of this software is governed by the Business Source License
// included in the LICENSE file.
//
// As of the Change Date specified in that file, in accordance with
// the Business Source License, use of this software will be governed
// by the Apache License, Version 2.0.

//! Logic for the Avro representation of the CDCv2 protocol.

use mz_avro::schema::{FullName, SchemaNode};
use repr::{Diff, Row, Timestamp};
use serde_json::json;

use anyhow::anyhow;
use avro_derive::AvroDecodable;
use differential_dataflow::capture::{Message, Progress};
use mz_avro::error::{DecodeError, Error as AvroError};
use mz_avro::schema::Schema;
use mz_avro::{
    define_unexpected, ArrayAsVecDecoder, AvroDecodable, AvroDecode, AvroDeserializer, AvroRead,
    StatefulAvroDecodable,
};
use std::{cell::RefCell, rc::Rc};

use super::decode::RowWrapper;

pub fn extract_data_columns<'a>(schema: &'a Schema) -> anyhow::Result<SchemaNode<'a>> {
    let data_name = FullName::from_parts("data", Some("com.materialize.cdc"), "");
    let data_schema = &schema
        .try_lookup_name(&data_name)
        .ok_or_else(|| anyhow!("record not found: {}", data_name))?
        .piece;
    Ok(SchemaNode {
        root: &schema,
        inner: data_schema,
        name: None,
    })
}

#[derive(AvroDecodable)]
#[state_type(Rc<RefCell<Row>>, Rc<RefCell<Vec<u8>>>)]
struct MyUpdate {
    #[state_expr(Rc::clone(&self._STATE.0), Rc::clone(&self._STATE.1))]
    data: RowWrapper,
    time: Timestamp,
    diff: Diff,
}

#[derive(AvroDecodable)]
struct Count {
    time: Timestamp,
    count: usize,
}

fn make_counts_decoder() -> impl AvroDecode<Out = Vec<(Timestamp, usize)>> {
    ArrayAsVecDecoder::new(|| {
        <Count as AvroDecodable>::new_decoder().map_decoder(|ct| Ok((ct.time, ct.count)))
    })
}

#[derive(AvroDecodable)]
struct MyProgress {
    lower: Vec<Timestamp>,
    upper: Vec<Timestamp>,
    #[decoder_factory(make_counts_decoder)]
    counts: Vec<(Timestamp, usize)>,
}

impl AvroDecode for Decoder {
    type Out = Message<Row, Timestamp, Diff>;
    fn union_branch<'a, R: AvroRead, D: AvroDeserializer>(
        self,
        idx: usize,
        _n_variants: usize,
        _null_variant: Option<usize>,
        deserializer: D,
        r: &'a mut R,
    ) -> Result<Self::Out, AvroError> {
        match idx {
            0 => {
                let packer = Rc::new(RefCell::new(Row::default()));
                let buf = Rc::new(RefCell::new(vec![]));
                let d = ArrayAsVecDecoder::new(|| {
                    <MyUpdate as StatefulAvroDecodable>::new_decoder((
                        Rc::clone(&packer),
                        Rc::clone(&buf),
                    ))
                    .map_decoder(|update| Ok((update.data.0, update.time, update.diff)))
                });
                let updates = deserializer.deserialize(r, d)?;
                Ok(Message::Updates(updates))
            }
            1 => {
                let progress =
                    deserializer.deserialize(r, <MyProgress as AvroDecodable>::new_decoder())?;
                let progress = Progress {
                    lower: progress.lower,
                    upper: progress.upper,
                    counts: progress.counts,
                };
                Ok(Message::Progress(progress))
            }

            other => Err(DecodeError::Custom(format!(
                "Unrecognized union variant in CDCv2 decoder: {}",
                other
            ))
            .into()),
        }
    }
    define_unexpected! {
        record, array, map, enum_variant, scalar, decimal, bytes, string, json, uuid, fixed
    }
}

/// Collected state to decode update batches and progress statements.
#[derive(Debug)]
pub struct Decoder;

/// Construct the schema for the CDC V2 protocol.
pub fn build_schema(row_schema: serde_json::Value) -> Schema {
    let updates_schema = json!({
        "type": "array",
        "items": {
            "name" : "update",
            "type" : "record",
            "fields" : [
                {
                    "name": "data",
                    "type": row_schema,
                },
                {
                    "name" : "time",
                    "type" : "long",
                },
                {
                    "name" : "diff",
                    "type" : "long",
                },
            ],
        },
    });

    let progress_schema = json!({
        "name" : "progress",
        "type" : "record",
        "fields" : [
            {
                "name": "lower",
                "type": {
                    "type": "array",
                    "items": "long"
                }
            },
            {
                "name": "upper",
                "type": {
                    "type": "array",
                    "items": "long"
                }
            },
            {
                "name": "counts",
                "type": {
                    "type": "array",
                    "items": {
                        "type": "record",
                        "name": "counts",
                        "fields": [
                            {
                                "name": "time",
                                "type": "long",
                            },
                            {
                                "name": "count",
                                "type": "long",
                            },
                        ],
                    }
                }
            },
        ],
    });
    let message_schema = json!([updates_schema, progress_schema,]);

    Schema::parse(&message_schema).expect("schema constrution failed")
}

#[cfg(test)]
mod tests {

    use super::*;
    use crate::avro::encode_datums_as_avro;
    use crate::encode::column_names_and_types;
    use mz_avro::types::Value;
    use mz_avro::AvroDeserializer;
    use mz_avro::GeneralDeserializer;
    use repr::{ColumnName, ColumnType, RelationDesc, Row, ScalarType};

    use crate::json::build_row_schema_json;

    /// Collected state to encode update batches and progress statements.
    #[derive(Debug)]
    struct Encoder {
        columns: Vec<(ColumnName, ColumnType)>,
    }

    impl Encoder {
        /// Creates a new CDCv2 encoder from a relation description.
        pub fn new(desc: RelationDesc) -> Self {
            let columns = column_names_and_types(desc);
            Self { columns }
        }

        /// Encodes a batch of updates as an Avro value.
        pub fn encode_updates(&self, updates: &[(Row, i64, i64)]) -> Value {
            let mut enc_updates = Vec::new();
            for (data, time, diff) in updates {
                let enc_data = encode_datums_as_avro(&**data, &self.columns);
                let enc_time = Value::Long(time.clone());
                let enc_diff = Value::Long(diff.clone());
                enc_updates.push(Value::Record(vec![
                    ("data".to_string(), enc_data),
                    ("time".to_string(), enc_time),
                    ("diff".to_string(), enc_diff),
                ]));
            }
            Value::Union {
                index: 0,
                inner: Box::new(Value::Array(enc_updates)),
                n_variants: 2,
                null_variant: None,
            }
        }

        /// Encodes the contents of a progress statement as an Avro value.
        pub fn encode_progress(
            &self,
            lower: &[i64],
            upper: &[i64],
            counts: &[(i64, i64)],
        ) -> Value {
            let enc_lower = Value::Array(lower.iter().cloned().map(Value::Long).collect());
            let enc_upper = Value::Array(upper.iter().cloned().map(Value::Long).collect());
            let enc_counts = Value::Array(
                counts
                    .iter()
                    .map(|(time, count)| {
                        Value::Record(vec![
                            ("time".to_string(), Value::Long(time.clone())),
                            ("count".to_string(), Value::Long(count.clone())),
                        ])
                    })
                    .collect(),
            );
            let enc_progress = Value::Record(vec![
                ("lower".to_string(), enc_lower),
                ("upper".to_string(), enc_upper),
                ("counts".to_string(), enc_counts),
            ]);

            Value::Union {
                index: 1,
                inner: Box::new(enc_progress),
                n_variants: 2,
                null_variant: None,
            }
        }
    }

    #[test]
    fn test_roundtrip() {
        let desc = RelationDesc::empty()
            .with_column("id", ScalarType::Int64.nullable(false))
            .with_column("price", ScalarType::Float64.nullable(true));

        let encoder = Encoder::new(desc.clone());
        let row_schema =
            build_row_schema_json(&crate::encode::column_names_and_types(desc), "data");
        let schema = build_schema(row_schema);

        let values = vec![
            encoder.encode_updates(&[]),
            encoder.encode_progress(&[0], &[3], &[]),
            encoder.encode_progress(&[3], &[], &[]),
        ];
        use mz_avro::encode::encode_to_vec;
        let mut values: Vec<_> = values
            .into_iter()
            .map(|v| encode_to_vec(&v, &schema))
            .collect();

        let g = GeneralDeserializer {
            schema: schema.top_node(),
        };
        assert!(matches!(
            g.deserialize(&mut &values.remove(0)[..], Decoder).unwrap(),
            Message::Updates(_)
        ),);
        assert!(matches!(
            g.deserialize(&mut &values.remove(0)[..], Decoder).unwrap(),
            Message::Progress(_)
        ),);
        assert!(matches!(
            g.deserialize(&mut &values.remove(0)[..], Decoder).unwrap(),
            Message::Progress(_)
        ),);
    }
}