clipmem 0.2.1

macOS clipboard memory backed by SQLite and searchable from OpenClaw
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
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
mod read;
mod store;

#[cfg(test)]
mod tests;

use std::path::{Path, PathBuf};
use std::str::FromStr;
use std::time::Duration;

use anyhow::{bail, Context, Result};
use clap::ValueEnum;
use rusqlite::{Connection, OpenFlags, Row};
use serde::{Deserialize, Serialize};

use crate::model::SearchHit;

const SCHEMA: &str = include_str!("db/schema.sql");
const CURRENT_SCHEMA_VERSION: i64 = 6;
const LEGACY_PRERELEASE_COLUMNS: &[&str] = &["classification", "is_text"];

pub struct Database {
    pub(super) conn: Connection,
    pub(super) path: PathBuf,
}

#[derive(Debug, Clone, Copy, PartialEq, Eq, Serialize, Deserialize, ValueEnum)]
pub enum SearchMode {
    Auto,
    Fts,
    Literal,
}

#[derive(Debug, Clone, Copy, PartialEq, Eq, Serialize, Deserialize, ValueEnum)]
pub enum TimelineSort {
    Asc,
    Desc,
}

#[derive(Debug, Clone, Copy, PartialEq, Eq, Serialize, Deserialize, ValueEnum)]
#[serde(rename_all = "snake_case")]
pub enum RetrievalKind {
    Text,
    Html,
    Rtf,
    Url,
    File,
    Image,
    Pdf,
    Binary,
    Other,
}

#[derive(Debug, Clone, PartialEq, Eq, Serialize, Deserialize, Default)]
pub struct RetrievalFilters {
    since: Option<String>,
    until: Option<String>,
    hours: Option<u32>,
    app: Option<String>,
    bundle_id: Option<String>,
    kind: Option<RetrievalKind>,
    has_text: bool,
    has_url: bool,
    has_file_url: bool,
    has_image: bool,
    has_pdf: bool,
    min_bytes: Option<usize>,
    max_bytes: Option<usize>,
}

#[derive(Debug, Clone, Serialize)]
pub struct SearchResults {
    mode_used: SearchMode,
    hits: Vec<SearchHit>,
    has_more: bool,
}

#[derive(Debug, Clone)]
pub(crate) struct Page<T> {
    items: Vec<T>,
    has_more: bool,
}

#[derive(Debug, Clone, PartialEq)]
pub(crate) struct RecentCursorState {
    last_seen_at: String,
    snapshot_id: i64,
}

#[derive(Debug, Clone, PartialEq)]
pub(crate) struct SearchCursorState {
    mode_used: SearchMode,
    score: Option<f64>,
    last_seen_at: String,
    snapshot_id: i64,
}

#[derive(Debug, Clone, PartialEq)]
pub(crate) struct TimelineCursorState {
    observed_at: String,
    event_id: i64,
}

impl SearchResults {
    #[must_use]
    pub(crate) fn new(mode_used: SearchMode, hits: Vec<SearchHit>, has_more: bool) -> Self {
        Self {
            mode_used,
            hits,
            has_more,
        }
    }

    #[must_use]
    pub fn mode_used(&self) -> SearchMode {
        self.mode_used
    }

    #[must_use]
    pub fn hits(&self) -> &[SearchHit] {
        &self.hits
    }

    #[must_use]
    pub fn has_more(&self) -> bool {
        self.has_more
    }
}

impl<T> Page<T> {
    #[must_use]
    pub(crate) fn new(items: Vec<T>, has_more: bool) -> Self {
        Self { items, has_more }
    }

    #[must_use]
    pub(crate) fn items(&self) -> &[T] {
        &self.items
    }

    #[must_use]
    pub(crate) fn into_items(self) -> Vec<T> {
        self.items
    }

    #[must_use]
    pub(crate) fn has_more(&self) -> bool {
        self.has_more
    }
}

impl RecentCursorState {
    #[must_use]
    pub(crate) fn new(last_seen_at: String, snapshot_id: i64) -> Self {
        Self {
            last_seen_at,
            snapshot_id,
        }
    }

    #[must_use]
    pub(crate) fn last_seen_at(&self) -> &str {
        &self.last_seen_at
    }

    #[must_use]
    pub(crate) fn snapshot_id(&self) -> i64 {
        self.snapshot_id
    }
}

impl SearchCursorState {
    #[must_use]
    pub(crate) fn new(
        mode_used: SearchMode,
        score: Option<f64>,
        last_seen_at: String,
        snapshot_id: i64,
    ) -> Self {
        Self {
            mode_used,
            score,
            last_seen_at,
            snapshot_id,
        }
    }

    #[must_use]
    pub(crate) fn mode_used(&self) -> SearchMode {
        self.mode_used
    }

    #[must_use]
    pub(crate) fn score(&self) -> Option<f64> {
        self.score
    }

    #[must_use]
    pub(crate) fn last_seen_at(&self) -> &str {
        &self.last_seen_at
    }

    #[must_use]
    pub(crate) fn snapshot_id(&self) -> i64 {
        self.snapshot_id
    }
}

impl SearchMode {
    #[must_use]
    pub const fn as_str(self) -> &'static str {
        match self {
            Self::Auto => "auto",
            Self::Fts => "fts",
            Self::Literal => "literal",
        }
    }
}

impl TimelineSort {
    #[must_use]
    pub const fn as_str(self) -> &'static str {
        match self {
            Self::Asc => "asc",
            Self::Desc => "desc",
        }
    }
}

impl RetrievalKind {
    #[must_use]
    pub const fn as_str(self) -> &'static str {
        match self {
            Self::Text => "text",
            Self::Html => "html",
            Self::Rtf => "rtf",
            Self::Url => "url",
            Self::File => "file",
            Self::Image => "image",
            Self::Pdf => "pdf",
            Self::Binary => "binary",
            Self::Other => "other",
        }
    }
}

impl RetrievalFilters {
    #[allow(clippy::too_many_arguments)]
    #[must_use]
    pub fn new(
        since: Option<String>,
        until: Option<String>,
        hours: Option<u32>,
        app: Option<String>,
        bundle_id: Option<String>,
        kind: Option<RetrievalKind>,
        has_text: bool,
        has_url: bool,
        has_file_url: bool,
        has_image: bool,
        has_pdf: bool,
        min_bytes: Option<usize>,
        max_bytes: Option<usize>,
    ) -> Self {
        Self {
            since,
            until,
            hours,
            app,
            bundle_id,
            kind,
            has_text,
            has_url,
            has_file_url,
            has_image,
            has_pdf,
            min_bytes,
            max_bytes,
        }
    }

    #[must_use]
    pub fn since(&self) -> Option<&str> {
        self.since.as_deref()
    }

    #[must_use]
    pub fn until(&self) -> Option<&str> {
        self.until.as_deref()
    }

    #[must_use]
    pub fn hours(&self) -> Option<u32> {
        self.hours
    }

    #[must_use]
    pub fn app(&self) -> Option<&str> {
        self.app.as_deref()
    }

    #[must_use]
    pub fn bundle_id(&self) -> Option<&str> {
        self.bundle_id.as_deref()
    }

    #[must_use]
    pub fn kind(&self) -> Option<RetrievalKind> {
        self.kind
    }

    #[must_use]
    pub fn has_text(&self) -> bool {
        self.has_text
    }

    #[must_use]
    pub fn has_url(&self) -> bool {
        self.has_url
    }

    #[must_use]
    pub fn has_file_url(&self) -> bool {
        self.has_file_url
    }

    #[must_use]
    pub fn has_image(&self) -> bool {
        self.has_image
    }

    #[must_use]
    pub fn has_pdf(&self) -> bool {
        self.has_pdf
    }

    #[must_use]
    pub fn min_bytes(&self) -> Option<usize> {
        self.min_bytes
    }

    #[must_use]
    pub fn max_bytes(&self) -> Option<usize> {
        self.max_bytes
    }
}

impl TimelineCursorState {
    #[must_use]
    pub(crate) fn new(observed_at: String, event_id: i64) -> Self {
        Self {
            observed_at,
            event_id,
        }
    }

    #[must_use]
    pub(crate) fn observed_at(&self) -> &str {
        &self.observed_at
    }

    #[must_use]
    pub(crate) fn event_id(&self) -> i64 {
        self.event_id
    }
}

impl Database {
    /// Open the archive database at `path`, creating parent directories and schema state as needed.
    ///
    /// This method also hardens parent-directory and database-file permissions after opening.
    ///
    /// # Errors
    ///
    /// Returns an error if the parent directory cannot be created, the database cannot be opened,
    /// or the connection cannot be configured and bootstrapped.
    pub fn open_or_init(path: &Path) -> Result<Self> {
        if let Some(parent) = path.parent() {
            Context::with_context(std::fs::create_dir_all(parent), || {
                format!("failed to create {}", parent.display())
            })?;
            harden_path_permissions(parent, 0o700)?;
        }

        let mut conn = open_connection(
            path,
            OpenFlags::SQLITE_OPEN_READ_WRITE
                | OpenFlags::SQLITE_OPEN_CREATE
                | OpenFlags::SQLITE_OPEN_URI
                | OpenFlags::SQLITE_OPEN_NO_MUTEX,
        )?;
        prepare_connection(&mut conn)?;
        harden_path_permissions(path, 0o600)?;

        Ok(Self {
            conn,
            path: path.to_path_buf(),
        })
    }

    /// Open an existing archive database without creating parent directories or schema state.
    ///
    /// This method also hardens parent-directory and database-file permissions after opening.
    ///
    /// # Errors
    ///
    /// Returns an error if the database file does not already exist, cannot be opened, or the
    /// connection cannot be configured.
    pub fn open_existing(path: &Path) -> Result<Self> {
        let mut conn = open_connection(
            path,
            OpenFlags::SQLITE_OPEN_READ_WRITE
                | OpenFlags::SQLITE_OPEN_URI
                | OpenFlags::SQLITE_OPEN_NO_MUTEX,
        )?;
        prepare_connection(&mut conn)?;
        if let Some(parent) = path.parent() {
            harden_path_permissions(parent, 0o700)?;
        }
        harden_path_permissions(path, 0o600)?;

        Ok(Self {
            conn,
            path: path.to_path_buf(),
        })
    }

    pub(crate) fn ensure_supported_schema_shape(&self) -> Result<()> {
        if legacy_prerelease_schema_detected(&self.conn)?
            || self
                .conn
                .prepare("SELECT kind FROM item_representations LIMIT 0")
                .is_err_and(|error| error.to_string().contains("no such column: kind"))
        {
            bail!(
                "database uses an incompatible prerelease schema; move it aside and run `clipmem setup` to initialize a fresh archive"
            );
        }
        Ok(())
    }

    #[cfg(test)]
    pub(crate) fn open_in_memory() -> Result<Self> {
        let mut conn = Connection::open_in_memory()?;
        prepare_connection(&mut conn)?;
        Ok(Self {
            conn,
            path: PathBuf::from(":memory:"),
        })
    }
}

pub(super) fn sanitise_limit(limit: usize) -> usize {
    limit.clamp(1, 250)
}

pub(super) fn collect_rows<T, I>(rows: I) -> Result<Vec<T>>
where
    I: Iterator<Item = rusqlite::Result<T>>,
{
    let mut out = Vec::new();
    for row in rows {
        out.push(row?);
    }
    Ok(out)
}

pub(super) fn usize_to_i64(value: usize) -> Result<i64> {
    anyhow::Context::context(i64::try_from(value), "value exceeds SQLite INTEGER range")
}

pub(super) fn row_usize(row: &Row<'_>, index: usize) -> rusqlite::Result<usize> {
    let value = row.get::<_, i64>(index)?;
    usize::try_from(value).map_err(|source| {
        rusqlite::Error::FromSqlConversionFailure(
            index,
            rusqlite::types::Type::Integer,
            Box::new(source),
        )
    })
}

pub(super) fn row_enum<T>(row: &Row<'_>, index: usize) -> rusqlite::Result<T>
where
    T: FromStr,
    T::Err: std::error::Error + Send + Sync + 'static,
{
    let value = row.get::<_, String>(index)?;
    value.parse().map_err(|source| {
        rusqlite::Error::FromSqlConversionFailure(
            index,
            rusqlite::types::Type::Text,
            Box::new(source),
        )
    })
}

fn open_connection(path: &Path, flags: OpenFlags) -> Result<Connection> {
    anyhow::Context::with_context(Connection::open_with_flags(path, flags), || {
        format!("failed to open {}", path.display())
    })
}

fn prepare_connection(conn: &mut Connection) -> Result<()> {
    Context::context(configure_connection(conn), "configure database connection")?;
    Context::context(prepare_schema(conn), "prepare database schema")?;
    Ok(())
}

#[cfg(unix)]
fn harden_path_permissions(path: &Path, mode: u32) -> Result<()> {
    use std::os::unix::fs::PermissionsExt;

    anyhow::Context::with_context(
        std::fs::set_permissions(path, PermissionsExt::from_mode(mode)),
        || format!("failed to set permissions on {}", path.display()),
    )
}

#[cfg(not(unix))]
fn harden_path_permissions(_path: &Path, _mode: u32) -> Result<()> {
    Ok(())
}

fn configure_connection(conn: &Connection) -> Result<()> {
    configure_pragma(conn, "journal_mode", "WAL")?;
    configure_pragma(conn, "synchronous", "NORMAL")?;
    configure_pragma(conn, "foreign_keys", "ON")?;
    configure_pragma(conn, "temp_store", "MEMORY")?;
    conn.busy_timeout(Duration::from_millis(1_500))
        .context("configure SQLite busy timeout")?;
    Ok(())
}

fn configure_pragma(conn: &Connection, pragma: &str, value: &str) -> Result<()> {
    Context::with_context(conn.pragma_update(None, pragma, value), || {
        format!("configure {pragma} pragma")
    })?;
    Ok(())
}

fn prepare_schema(conn: &mut Connection) -> Result<()> {
    let tx = conn
        .transaction_with_behavior(rusqlite::TransactionBehavior::Immediate)
        .context("begin schema transaction")?;

    tx.execute_batch(SCHEMA).context("apply database schema")?;

    let user_version: i64 = tx
        .query_row("PRAGMA user_version", [], |row| row.get(0))
        .context("read PRAGMA user_version")?;

    match user_version {
        0 => {
            tx.execute(
                "INSERT INTO snapshots_fts(snapshots_fts) VALUES ('rebuild')",
                [],
            )
            .context("rebuild FTS5 index")?;
            rebuild_snapshot_stats(&tx)?;
            rebuild_snapshot_projection_cache(&tx)?;
            rebuild_snapshot_event_filter_cache(&tx)?;
            rebuild_snapshot_literal_cache(&tx)?;
            rebuild_snapshot_file_url_fts(&tx)?;
            tx.pragma_update(None, "user_version", CURRENT_SCHEMA_VERSION)
                .context("set PRAGMA user_version")?;
        }
        1 => {
            if legacy_prerelease_schema_detected(&tx)? {
                bail!(
                    "database at the current user_version uses an incompatible prerelease schema; move it aside and run `clipmem setup` to initialize a fresh archive"
                );
            }
            rebuild_snapshot_stats(&tx)?;
            rebuild_snapshot_projection_cache(&tx)?;
            rebuild_snapshot_event_filter_cache(&tx)?;
            rebuild_snapshot_literal_cache(&tx)?;
            rebuild_snapshot_file_url_fts(&tx)?;
            tx.pragma_update(None, "user_version", CURRENT_SCHEMA_VERSION)
                .context("set PRAGMA user_version")?;
        }
        2 => {
            if legacy_prerelease_schema_detected(&tx)? {
                bail!(
                    "database at the current user_version uses an incompatible prerelease schema; move it aside and run `clipmem setup` to initialize a fresh archive"
                );
            }
            rebuild_snapshot_projection_cache(&tx)?;
            rebuild_snapshot_event_filter_cache(&tx)?;
            rebuild_snapshot_literal_cache(&tx)?;
            rebuild_snapshot_file_url_fts(&tx)?;
            tx.pragma_update(None, "user_version", CURRENT_SCHEMA_VERSION)
                .context("set PRAGMA user_version")?;
        }
        3 => {
            if legacy_prerelease_schema_detected(&tx)? {
                bail!(
                    "database at the current user_version uses an incompatible prerelease schema; move it aside and run `clipmem setup` to initialize a fresh archive"
                );
            }
            rebuild_snapshot_event_filter_cache(&tx)?;
            rebuild_snapshot_literal_cache(&tx)?;
            rebuild_snapshot_file_url_fts(&tx)?;
            tx.pragma_update(None, "user_version", CURRENT_SCHEMA_VERSION)
                .context("set PRAGMA user_version")?;
        }
        4 => {
            if legacy_prerelease_schema_detected(&tx)? {
                bail!(
                    "database at the current user_version uses an incompatible prerelease schema; move it aside and run `clipmem setup` to initialize a fresh archive"
                );
            }
            rebuild_snapshot_literal_cache(&tx)?;
            rebuild_snapshot_file_url_fts(&tx)?;
            tx.pragma_update(None, "user_version", CURRENT_SCHEMA_VERSION)
                .context("set PRAGMA user_version")?;
        }
        5 => {
            if legacy_prerelease_schema_detected(&tx)? {
                bail!(
                    "database at the current user_version uses an incompatible prerelease schema; move it aside and run `clipmem setup` to initialize a fresh archive"
                );
            }
            rebuild_snapshot_file_url_fts(&tx)?;
            tx.pragma_update(None, "user_version", CURRENT_SCHEMA_VERSION)
                .context("set PRAGMA user_version")?;
        }
        CURRENT_SCHEMA_VERSION => {
            if legacy_prerelease_schema_detected(&tx)? {
                bail!(
                    "database at the current user_version uses an incompatible prerelease schema; move it aside and run `clipmem setup` to initialize a fresh archive"
                );
            }
        }
        version if version > CURRENT_SCHEMA_VERSION => {
            bail!(
                "database schema version {version} is newer than supported version {}",
                CURRENT_SCHEMA_VERSION
            );
        }
        version => {
            bail!("unsupported database schema version {version}");
        }
    }

    tx.commit().context("commit schema transaction")?;
    Ok(())
}

fn rebuild_snapshot_stats(conn: &Connection) -> Result<()> {
    conn.execute("DELETE FROM snapshot_stats", [])
        .context("clear snapshot stats")?;
    conn.execute_batch(
        r"
        INSERT INTO snapshot_stats (
            snapshot_id,
            capture_count,
            first_observed_at,
            last_observed_at,
            last_event_id,
            last_frontmost_app_bundle_id,
            last_frontmost_app_name
        )
        SELECT
            ce.snapshot_id,
            COUNT(*) AS capture_count,
            MIN(ce.observed_at) AS first_observed_at,
            MAX(ce.observed_at) AS last_observed_at,
            (
                SELECT latest.id
                FROM capture_events latest
                WHERE latest.snapshot_id = ce.snapshot_id
                ORDER BY latest.observed_at DESC, latest.id DESC
                LIMIT 1
            ) AS last_event_id,
            (
                SELECT latest.frontmost_app_bundle_id
                FROM capture_events latest
                WHERE latest.snapshot_id = ce.snapshot_id
                ORDER BY latest.observed_at DESC, latest.id DESC
                LIMIT 1
            ) AS last_frontmost_app_bundle_id,
            (
                SELECT latest.frontmost_app_name
                FROM capture_events latest
                WHERE latest.snapshot_id = ce.snapshot_id
                ORDER BY latest.observed_at DESC, latest.id DESC
                LIMIT 1
            ) AS last_frontmost_app_name
        FROM capture_events ce
        GROUP BY ce.snapshot_id;
        ",
    )
    .context("rebuild snapshot stats")?;
    Ok(())
}

fn rebuild_snapshot_projection_cache(conn: &Connection) -> Result<()> {
    conn.execute("DELETE FROM snapshot_projection_cache", [])
        .context("clear snapshot projection cache")?;
    conn.execute_batch(
        r"
        WITH url_values AS (
            SELECT
                snapshot_id,
                GROUP_CONCAT(text_value, char(31)) AS urls
            FROM (
                SELECT DISTINCT snapshot_id, text_value
                FROM item_representations
                WHERE kind = 'url' AND text_value IS NOT NULL AND text_value != ''
                ORDER BY text_value
            )
            GROUP BY snapshot_id
        ),
        file_url_values AS (
            SELECT
                snapshot_id,
                GROUP_CONCAT(text_value, char(31)) AS file_urls
            FROM (
                SELECT DISTINCT snapshot_id, text_value
                FROM item_representations
                WHERE kind = 'file_url' AND text_value IS NOT NULL AND text_value != ''
                ORDER BY text_value
            )
            GROUP BY snapshot_id
        )
        INSERT INTO snapshot_projection_cache (snapshot_id, urls, file_urls)
        SELECT
            s.id,
            COALESCE(uv.urls, ''),
            COALESCE(fv.file_urls, '')
        FROM snapshots s
        LEFT JOIN url_values uv ON uv.snapshot_id = s.id
        LEFT JOIN file_url_values fv ON fv.snapshot_id = s.id;
        ",
    )
    .context("rebuild snapshot projection cache")?;
    Ok(())
}

fn rebuild_snapshot_event_filter_cache(conn: &Connection) -> Result<()> {
    conn.execute("DELETE FROM snapshot_event_filter_cache", [])
        .context("clear snapshot event filter cache")?;
    conn.execute_batch(
        r"
        INSERT INTO snapshot_event_filter_cache (snapshot_id, app_names_lower, bundle_ids_lower)
        SELECT
            s.id,
            COALESCE((
                SELECT GROUP_CONCAT(app_name, char(31))
                FROM (
                    SELECT DISTINCT lower(ce.frontmost_app_name) AS app_name
                    FROM capture_events ce
                    WHERE ce.snapshot_id = s.id
                      AND ce.frontmost_app_name IS NOT NULL
                      AND ce.frontmost_app_name != ''
                    ORDER BY app_name
                )
            ), '') AS app_names_lower,
            COALESCE((
                SELECT GROUP_CONCAT(bundle_id, char(31))
                FROM (
                    SELECT DISTINCT lower(ce.frontmost_app_bundle_id) AS bundle_id
                    FROM capture_events ce
                    WHERE ce.snapshot_id = s.id
                      AND ce.frontmost_app_bundle_id IS NOT NULL
                      AND ce.frontmost_app_bundle_id != ''
                    ORDER BY bundle_id
                )
            ), '') AS bundle_ids_lower
        FROM snapshots s;
        ",
    )
    .context("rebuild snapshot event filter cache")?;
    Ok(())
}

fn rebuild_snapshot_literal_cache(conn: &Connection) -> Result<()> {
    conn.execute("DELETE FROM snapshot_literal_cache", [])
        .context("clear snapshot literal cache")?;
    conn.execute("DELETE FROM snapshots_literal_fts", [])
        .context("clear literal FTS cache")?;
    conn.execute_batch(
        r"
        INSERT INTO snapshot_literal_cache (snapshot_id, haystack)
        SELECT
            s.id,
            lower(
                COALESCE(NULLIF(s.preview_text, ''), s.search_text, '') || char(31) ||
                COALESCE(s.preview_text, '') || char(31) ||
                COALESCE(s.search_text, '') || char(31) ||
                COALESCE(sp.urls, '') || char(31) ||
                COALESCE(sp.file_urls, '') || char(31) ||
                COALESCE(ss.last_frontmost_app_name, '') || char(31) ||
                COALESCE(ss.last_frontmost_app_bundle_id, '')
            )
        FROM snapshots s
        LEFT JOIN snapshot_projection_cache sp ON sp.snapshot_id = s.id
        LEFT JOIN snapshot_stats ss ON ss.snapshot_id = s.id;
        ",
    )
    .context("rebuild snapshot literal cache")?;
    Ok(())
}

fn rebuild_snapshot_file_url_fts(conn: &Connection) -> Result<()> {
    conn.execute("INSERT INTO snapshot_file_url_fts(snapshot_file_url_fts) VALUES ('rebuild')", [])
        .context("rebuild snapshot file-url FTS")?;
    Ok(())
}

fn legacy_prerelease_schema_detected(conn: &Connection) -> Result<bool> {
    let mut stmt = conn
        .prepare("PRAGMA table_info(item_representations)")
        .context("prepare PRAGMA table_info(item_representations)")?;
    let rows = stmt
        .query_map([], |row| row.get::<_, String>(1))
        .context("read item_representations columns")?;
    let columns = collect_rows(rows).context("collect item_representations columns")?;
    if columns.is_empty() {
        return Ok(false);
    }

    let has_kind = columns.iter().any(|column| column == "kind");
    let has_legacy_marker = LEGACY_PRERELEASE_COLUMNS
        .iter()
        .any(|legacy| columns.iter().any(|column| column == legacy));
    Ok(!has_kind || has_legacy_marker)
}

#[cfg(test)]
pub(crate) fn explain_query_plan(
    conn: &Connection,
    sql: &str,
    params: &[&dyn rusqlite::ToSql],
) -> Result<Vec<String>> {
    let explain = format!("EXPLAIN QUERY PLAN {sql}");
    let mut stmt = conn
        .prepare(&explain)
        .context("prepare EXPLAIN QUERY PLAN")?;
    let rows = stmt
        .query_map(params, |row| row.get::<_, String>(3))
        .context("execute EXPLAIN QUERY PLAN")?;
    collect_rows(rows).context("collect EXPLAIN QUERY PLAN rows")
}