rivet-cli 0.7.9

Rivet: PostgreSQL/MySQL → Parquet/CSV (local, S3, GCS). Crate name rivet-cli; binary rivet.
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
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
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
//! `rivet reconcile` — partition/window reconciliation (Epic F).
//!
//! Re-runs per-partition `COUNT(*)` against the source for a chunked export and
//! compares the result with the stored per-chunk row counts. Produces a
//! [`ReconcileReport`] — matches, mismatches, and repair candidates.
//!
//! v1 supports **chunked exports with `chunk_checkpoint: true`** (so per-chunk
//! row counts and ranges are persisted). Other modes return a clear
//! "not supported in v1" error — reconcile semantics for snapshot / incremental
//! / time-window differ (see roadmap: Epic F & Epic G).

use std::collections::HashMap;
use std::path::Path;

use crate::config::{Config, ExportConfig};
use crate::error::Result;
use crate::plan::{
    ExtractionStrategy, PartitionKind, PartitionResult, ReconcileReport, ResolvedRunPlan,
    build_plan,
};
use crate::source;
use crate::state::{ChunkTaskInfo, StateStore};

use super::chunked::build_chunk_query_sql;

/// Output format for the reconcile report.
pub enum ReconcileOutputFormat {
    /// Human-readable summary printed to stdout.
    Pretty,
    /// Pretty-printed JSON written to the given path (or stdout if `None`).
    Json(Option<String>),
}

pub fn run_reconcile_command(
    config_path: &str,
    export_name: &str,
    params: Option<&HashMap<String, String>>,
    format: ReconcileOutputFormat,
) -> Result<()> {
    let config = Config::load_with_params(config_path, params)?;
    let config_dir = Path::new(config_path)
        .parent()
        .unwrap_or_else(|| Path::new("."));

    let export = config
        .exports
        .iter()
        .find(|e| e.name == export_name)
        .ok_or_else(|| anyhow::anyhow!("export '{}' not found in config", export_name))?;

    let plan = build_plan(&config, export, config_dir, false, false, false, params)?;

    let state_path = config_dir.join(".rivet_state.db");
    let state = StateStore::open(state_path.to_str().unwrap_or(".rivet_state.db"))?;

    let report = match &plan.strategy {
        ExtractionStrategy::Chunked(_) => reconcile_chunked(&plan, &state, export)?,
        ExtractionStrategy::TimeWindow { .. } => {
            anyhow::bail!(
                "reconcile: time-window mode is not supported in v1 (Epic F). \
                 Convert to chunked mode with `chunk_by_days` for partition-level reconcile."
            );
        }
        ExtractionStrategy::Snapshot
        | ExtractionStrategy::Incremental(_)
        | ExtractionStrategy::Keyset(_) => {
            anyhow::bail!(
                "reconcile: '{}' mode has no natural partitions — use `rivet run --reconcile` for a whole-export count check",
                plan.strategy.mode_label()
            );
        }
    };

    emit_report(&report, &format)?;
    Ok(())
}

/// Run a reconcile pass against the latest chunk run and return the report.
/// Exposed so `rivet repair --auto` can build a repair plan from a fresh reconcile
/// without duplicating the logic.
pub(crate) fn reconcile_chunked_fresh(
    plan: &ResolvedRunPlan,
    state: &StateStore,
) -> Result<ReconcileReport> {
    reconcile_chunked_inner(plan, state)
}

fn reconcile_chunked(
    plan: &ResolvedRunPlan,
    state: &StateStore,
    _export: &ExportConfig,
) -> Result<ReconcileReport> {
    reconcile_chunked_inner(plan, state)
}

fn reconcile_chunked_inner(plan: &ResolvedRunPlan, state: &StateStore) -> Result<ReconcileReport> {
    let (run_id, _plan_hash, _status, _updated) = state
        .get_latest_chunk_run(&plan.export_name)?
        .ok_or_else(|| {
            anyhow::anyhow!(
                "reconcile: no chunk run recorded for export '{}'. \
                 Enable `chunk_checkpoint: true` and run the export first.",
                plan.export_name
            )
        })?;

    let tasks = state.list_chunk_tasks_for_run(&run_id)?;
    if tasks.is_empty() {
        anyhow::bail!(
            "reconcile: chunk run '{}' for export '{}' has no tasks",
            run_id,
            plan.export_name
        );
    }

    let mut src = source::create_source(&plan.source)?;
    let partitions = reconcile_chunked_tasks(plan, &tasks, |chunk_query| {
        let count_sql = format!("SELECT COUNT(*) FROM ({chunk_query}) AS _rc");
        let raw = src.query_scalar(&count_sql)?;
        Ok(raw.and_then(|s| s.trim().parse::<i64>().ok()))
    })?;

    let report = ReconcileReport::new(
        plan.export_name.clone(),
        run_id.clone(),
        plan.strategy.mode_label().to_string(),
        partitions,
    );

    // Epic G: a reconcile with no mismatches and no unknowns is a fresh verified boundary.
    if report.summary.mismatches == 0 && report.summary.unknown == 0 {
        let highest = tasks
            .iter()
            .filter(|t| t.status == "completed")
            .map(|t| t.chunk_index)
            .max();
        if let Some(idx) = highest
            && let Err(e) = state.record_verified_chunked(&plan.export_name, idx, &run_id)
        {
            log::warn!(
                "export '{}': verified boundary update failed: {:#}",
                plan.export_name,
                e
            );
        }
    }

    Ok(report)
}

/// Pure classification of `ChunkTaskInfo`s into `PartitionResult`s.
///
/// Abstracts the source round-trip via `count_source` so the logic can be
/// exercised in unit tests without a live database.
pub(crate) fn reconcile_chunked_tasks<F>(
    plan: &ResolvedRunPlan,
    tasks: &[ChunkTaskInfo],
    mut count_source: F,
) -> Result<Vec<PartitionResult>>
where
    F: FnMut(&str) -> Result<Option<i64>>,
{
    let cp = match &plan.strategy {
        ExtractionStrategy::Chunked(cp) => cp,
        _ => anyhow::bail!("reconcile_chunked_tasks requires Chunked strategy"),
    };

    let mut out: Vec<PartitionResult> = Vec::with_capacity(tasks.len());
    for t in tasks {
        let exported = if t.status == "completed" {
            t.rows_written
        } else {
            None
        };

        let (start, end) = match (t.start_key.parse::<i64>(), t.end_key.parse::<i64>()) {
            (Ok(s), Ok(e)) => (s, e),
            _ => {
                // Unparseable chunk keys: keep exported count but cannot re-count source.
                out.push(PartitionResult::classify(
                    PartitionKind::Chunk,
                    format!("chunk {} [{}..{}]", t.chunk_index, t.start_key, t.end_key),
                    None,
                    exported,
                ));
                continue;
            }
        };

        let chunk_query = build_chunk_query_sql(
            &plan.base_query,
            &cp.column,
            start,
            end,
            cp.dense,
            cp.by_days.is_some(),
            plan.source.source_type,
        );
        let source_count = count_source(&chunk_query)?;

        out.push(PartitionResult::classify(
            PartitionKind::Chunk,
            format!("chunk {} [{}..{}]", t.chunk_index, start, end),
            source_count,
            exported,
        ));
    }
    Ok(out)
}

fn emit_report(report: &ReconcileReport, format: &ReconcileOutputFormat) -> Result<()> {
    match format {
        ReconcileOutputFormat::Pretty => {
            print_report_pretty(report);
        }
        ReconcileOutputFormat::Json(None) => {
            println!("{}", report.to_json_pretty()?);
        }
        ReconcileOutputFormat::Json(Some(path)) => {
            let json = report.to_json_pretty()?;
            std::fs::write(path, &json)
                .map_err(|e| anyhow::anyhow!("cannot write reconcile report '{}': {}", path, e))?;
            println!("Reconcile report written to: {}", path);
        }
    }
    Ok(())
}

fn print_report_pretty(report: &ReconcileReport) {
    println!();
    println!("  Export    : {}", report.export_name);
    println!("  Run       : {}", report.run_id);
    println!("  Strategy  : {}", report.strategy);
    println!(
        "  Partitions: {} ({} match, {} mismatch, {} unknown)",
        report.summary.total_partitions,
        report.summary.matches,
        report.summary.mismatches,
        report.summary.unknown,
    );
    println!(
        "  Rows      : source {} / exported {}",
        report.summary.total_source_rows, report.summary.total_exported_rows,
    );

    let repair = report.repair_candidates();
    if repair.is_empty() {
        println!("  Status    : all partitions match");
    } else {
        println!("  Repair candidates:");
        for p in repair {
            println!("{}{}", p.identifier, format_status_note(p));
        }
    }
    println!();
}

fn format_status_note(p: &PartitionResult) -> String {
    let s = match (p.source_count, p.exported_count) {
        (Some(s), Some(e)) => format!("source={s}, exported={e}"),
        (Some(s), None) => format!("source={s}, exported=n/a"),
        (None, Some(e)) => format!("source=n/a, exported={e}"),
        (None, None) => "no counts".to_string(),
    };
    if p.note.is_empty() {
        s
    } else {
        format!("{s} ({})", p.note)
    }
}

#[cfg(test)]
mod tests {
    use super::*;
    use crate::config::{
        CompressionType, DestinationConfig, DestinationType, FormatType, MetaColumns, SourceConfig,
        SourceType,
    };
    use crate::plan::{ChunkedPlan, ExtractionStrategy};
    use crate::state::ChunkTaskInfo;
    use crate::tuning::SourceTuning;

    fn chunked_plan() -> ResolvedRunPlan {
        ResolvedRunPlan {
            export_name: "orders".into(),
            base_query: "SELECT * FROM orders".into(),
            strategy: ExtractionStrategy::Chunked(ChunkedPlan {
                column: "id".into(),
                chunk_size: 100,
                chunk_count: None,
                parallel: 1,
                dense: false,
                by_days: None,
                checkpoint: true,
                max_attempts: 3,
            }),
            format: FormatType::Parquet,
            compression: CompressionType::Zstd,
            compression_level: None,
            max_file_size_bytes: None,
            skip_empty: false,
            meta_columns: MetaColumns::default(),
            destination: DestinationConfig {
                destination_type: DestinationType::Local,
                path: Some("./out".into()),
                ..Default::default()
            },
            quality: None,
            tuning: SourceTuning::from_config(None),
            tuning_profile_label: "balanced (default)".into(),
            validate: false,
            reconcile: false,
            resume: false,
            source: SourceConfig {
                source_type: SourceType::Postgres,
                url: Some("postgresql://localhost/test".into()),
                url_env: None,
                url_file: None,
                host: None,
                port: None,
                user: None,
                password: None,
                password_env: None,
                database: None,
                environment: None,
                tuning: None,
                tls: None,
            },
            column_overrides: Default::default(),
            schema_drift_policy: Default::default(),
            shape_drift_warn_factor: 2.0,
            parquet: None,
        }
    }

    fn task(idx: i64, start: &str, end: &str, status: &str, rows: Option<i64>) -> ChunkTaskInfo {
        ChunkTaskInfo {
            chunk_index: idx,
            start_key: start.into(),
            end_key: end.into(),
            status: status.into(),
            attempts: 1,
            last_error: None,
            rows_written: rows,
            file_name: None,
        }
    }

    #[test]
    fn matches_and_mismatches_are_classified() {
        let plan = chunked_plan();
        let tasks = vec![
            task(0, "1", "100", "completed", Some(42)),
            task(1, "101", "200", "completed", Some(30)),
        ];
        // Stub source: chunk 0 matches, chunk 1 undercounts on export side.
        let mut n = 0;
        let parts = reconcile_chunked_tasks(&plan, &tasks, |_q| {
            n += 1;
            Ok(Some(if n == 1 { 42 } else { 33 }))
        })
        .unwrap();

        assert_eq!(parts.len(), 2);
        assert_eq!(parts[0].status, crate::plan::PartitionStatus::Match);
        assert_eq!(parts[1].status, crate::plan::PartitionStatus::Mismatch);
        assert_eq!(parts[1].source_count, Some(33));
        assert_eq!(parts[1].exported_count, Some(30));
    }

    #[test]
    fn unfinished_task_is_unknown_and_does_not_hide_source_count() {
        let plan = chunked_plan();
        let tasks = vec![task(0, "1", "100", "failed", None)];
        let parts = reconcile_chunked_tasks(&plan, &tasks, |_q| Ok(Some(42))).unwrap();
        assert_eq!(parts[0].status, crate::plan::PartitionStatus::Unknown);
        assert_eq!(parts[0].source_count, Some(42));
        assert_eq!(parts[0].exported_count, None);
    }

    #[test]
    fn unparseable_chunk_keys_are_unknown_without_source_lookup() {
        let plan = chunked_plan();
        let tasks = vec![task(0, "alpha", "omega", "completed", Some(5))];
        let mut called = false;
        let parts = reconcile_chunked_tasks(&plan, &tasks, |_q| {
            called = true;
            Ok(Some(99))
        })
        .unwrap();
        assert!(
            !called,
            "reconcile must skip source count for unparseable chunk keys"
        );
        assert_eq!(parts[0].status, crate::plan::PartitionStatus::Unknown);
        assert_eq!(parts[0].exported_count, Some(5));
    }

    #[test]
    fn chunk_query_passes_through_chunked_math() {
        let plan = chunked_plan();
        let tasks = vec![task(0, "10", "20", "completed", Some(5))];
        let mut captured = String::new();
        reconcile_chunked_tasks(&plan, &tasks, |q| {
            captured = q.to_string();
            Ok(Some(5))
        })
        .unwrap();
        // Must reuse the same WHERE predicate used during extraction (ADR-0001 shape).
        assert!(captured.contains("BETWEEN 10 AND 20"), "got: {captured}");
        assert!(
            captured.contains("\"id\""),
            "identifier must be quoted: {captured}"
        );
    }
}