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
use serde::{Deserialize, Serialize};
use std::fmt::Display;
use utoipa::ToSchema;
/// PostgreSQL write mode.
///
/// Determines how the PostgreSQL output connector writes data to the target table.
#[derive(Debug, Clone, Eq, PartialEq, Deserialize, Serialize, ToSchema, Default)]
pub enum PostgresWriteMode {
/// Materialized mode: perform direct INSERT, UPDATE, and DELETE operations on the table.
/// This is the default behavior and maintains the postgres table as a materialized snapshot of the output view.
#[default]
#[serde(rename = "materialized")]
Materialized,
/// CDC (Change Data Capture) mode: write all operations as INSERT operations
/// into a Postgres table that serves as an append-only event log.
/// In this mode, inserts, updates, and deletes are all represented as new rows
/// with metadata columns describing the operation type and timestamp.
#[serde(rename = "cdc")]
Cdc,
}
impl Display for PostgresWriteMode {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
match self {
Self::Materialized => write!(f, "materialized"),
Self::Cdc => write!(f, "cdc"),
}
}
}
/// TLS/SSL configuration for PostgreSQL connectors.
#[derive(Debug, Clone, Eq, PartialEq, Deserialize, Serialize, ToSchema, Default)]
pub struct PostgresTlsConfig {
/// A sequence of CA certificates in PEM format.
pub ssl_ca_pem: Option<String>,
/// Path to a file containing a sequence of CA certificates in PEM format.
pub ssl_ca_location: Option<String>,
/// The client certificate in PEM format.
pub ssl_client_pem: Option<String>,
/// Path to the client certificate.
pub ssl_client_location: Option<String>,
/// The client certificate key in PEM format.
pub ssl_client_key: Option<String>,
/// Path to the client certificate key.
pub ssl_client_key_location: Option<String>,
/// The path to the certificate chain file.
/// The file must contain a sequence of PEM-formatted certificates,
/// the first being the leaf certificate, and the remainder forming
/// the chain of certificates up to and including the trusted root certificate.
pub ssl_certificate_chain_location: Option<String>,
/// True to enable hostname verification when using TLS. True by default.
pub verify_hostname: Option<bool>,
}
impl PostgresTlsConfig {
pub fn has_tls(&self) -> bool {
self.ssl_ca_pem.is_some() || self.ssl_ca_location.is_some()
}
}
/// Postgres input connector configuration.
#[derive(Debug, Clone, Eq, PartialEq, Deserialize, Serialize, ToSchema)]
pub struct PostgresReaderConfig {
/// Postgres URI.
/// See: <https://docs.rs/tokio-postgres/0.7.12/tokio_postgres/config/struct.Config.html>
pub uri: String,
/// Query that specifies what data to fetch from postgres.
pub query: String,
/// TLS/SSL configuration.
#[serde(flatten)]
#[schema(inline)]
pub tls: PostgresTlsConfig,
}
/// Postgres output connector configuration.
#[derive(Debug, Clone, Eq, PartialEq, Deserialize, Serialize, ToSchema)]
pub struct PostgresWriterConfig {
/// Postgres URI.
/// See: <https://docs.rs/tokio-postgres/0.7.12/tokio_postgres/config/struct.Config.html>
pub uri: String,
/// The table to write the output to.
pub table: String,
/// Write mode for the connector.
///
/// - `materialized` (default): Perform direct INSERT, UPDATE, and DELETE operations on the table.
/// - `cdc`: Write all operations as INSERT operations into an append-only event log
/// with additional metadata columns describing the operation type and timestamp.
#[serde(default)]
#[schema(default = PostgresWriteMode::default)]
pub mode: PostgresWriteMode,
/// Name of the operation metadata column in CDC mode.
///
/// Only used when `mode = "cdc"`. This column will contain:
/// - `"i"` for insert operations
/// - `"u"` for upsert operations
/// - `"d"` for delete operations
///
/// Default: `"__feldera_op"`
#[serde(default = "default_cdc_op_column")]
#[schema(default = default_cdc_op_column)]
pub cdc_op_column: String,
/// Name of the timestamp metadata column in CDC mode.
///
/// Only used when `mode = "cdc"`. This column will contain the timestamp
/// (in RFC 3339 format) when the batch of updates was output
/// by the pipeline.
///
/// Default: `"__feldera_ts"`
#[serde(default = "default_cdc_ts_column")]
#[schema(default = default_cdc_ts_column)]
pub cdc_ts_column: String,
/// TLS/SSL configuration.
#[serde(flatten)]
#[schema(inline)]
pub tls: PostgresTlsConfig,
/// The maximum number of records in a single buffer.
pub max_records_in_buffer: Option<usize>,
/// The maximum buffer size in for a single operation.
/// Note that the buffers of `INSERT`, `UPDATE` and `DELETE` queries are
/// separate.
/// Default: 1 MiB
#[schema(default = default_max_buffer_size)]
#[serde(default = "default_max_buffer_size")]
pub max_buffer_size_bytes: usize,
/// Specifies how the connector handles conflicts when executing an `INSERT`
/// into a table with a primary key. By default, an existing row with the same
/// key is overwritten. Setting this flag to `true` preserves the existing row
/// and ignores the new insert.
///
/// This setting does not affect `UPDATE` statements, which always replace the
/// value associated with the key.
///
/// This setting is not supported when `mode = "cdc"`, since all operations
/// are performed as append-only `INSERT`s into the target table.
/// Any conflict in CDC mode will result in an error.
///
/// Default: `false`
#[serde(default)]
pub on_conflict_do_nothing: bool,
/// The number of threads to use during encoding.
///
/// Default: 1
#[serde(default = "default_writer_threads")]
#[schema(default = default_writer_threads)]
pub threads: usize,
}
fn default_max_buffer_size() -> usize {
usize::pow(2, 20)
}
fn default_writer_threads() -> usize {
1
}
fn default_cdc_op_column() -> String {
"__feldera_op".to_string()
}
fn default_cdc_ts_column() -> String {
"__feldera_ts".to_string()
}
impl PostgresWriterConfig {
pub fn validate(&self) -> Result<(), String> {
match self.mode {
PostgresWriteMode::Cdc => {
if self.cdc_op_column.trim().is_empty() {
return Err("cdc_op_column cannot be empty in CDC mode".to_string());
}
if self.cdc_ts_column.trim().is_empty() {
return Err("cdc_ts_column cannot be empty in CDC mode".to_string());
}
if !self.cdc_op_column.is_ascii() {
return Err("cdc_op_column must contain only ASCII characters".to_string());
}
if !self.cdc_ts_column.is_ascii() {
return Err("cdc_ts_column must contain only ASCII characters".to_string());
}
if self.on_conflict_do_nothing {
return Err("on_conflict_do_nothing not supported in CDC mode since all operations are performed as append-only INSERTs into the target table".to_string());
}
}
PostgresWriteMode::Materialized => {
if self.cdc_ts_column != default_cdc_ts_column()
&& !self.cdc_ts_column.trim().is_empty()
{
return Err(
"cdc_ts_column must not be set when in MATERIALIZED mode".to_string()
);
}
if self.cdc_op_column != default_cdc_op_column()
&& !self.cdc_op_column.trim().is_empty()
{
return Err(
"cdc_op_column must not be set when in MATERIALIZED mode".to_string()
);
}
}
};
if self.threads == 0 {
return Err("threads must be at least 1".to_string());
}
Ok(())
}
}