Press n or j to go to the next uncovered block, b, p or k for the previous block.
| 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 | 30x 30x | /**
* SQLite schema for Stork mail storage.
*
* Uses FTS5 for full-text search across message subjects and bodies.
*/
export const SCHEMA_VERSION = 30;
export const MIGRATIONS = [
// Version 1: Initial schema
`
CREATE TABLE IF NOT EXISTS accounts (
id INTEGER PRIMARY KEY AUTOINCREMENT,
name TEXT NOT NULL,
email TEXT NOT NULL,
imap_host TEXT NOT NULL,
imap_port INTEGER NOT NULL DEFAULT 993,
imap_tls INTEGER NOT NULL DEFAULT 1,
imap_user TEXT NOT NULL,
imap_pass TEXT NOT NULL,
smtp_host TEXT,
smtp_port INTEGER DEFAULT 587,
smtp_tls INTEGER DEFAULT 1,
smtp_user TEXT,
smtp_pass TEXT,
sync_delete_from_server INTEGER NOT NULL DEFAULT 0,
created_at TEXT NOT NULL DEFAULT (datetime('now')),
updated_at TEXT NOT NULL DEFAULT (datetime('now'))
);
CREATE TABLE IF NOT EXISTS folders (
id INTEGER PRIMARY KEY AUTOINCREMENT,
account_id INTEGER NOT NULL REFERENCES accounts(id) ON DELETE CASCADE,
path TEXT NOT NULL,
name TEXT NOT NULL,
delimiter TEXT,
flags TEXT,
uid_validity INTEGER,
uid_next INTEGER,
message_count INTEGER DEFAULT 0,
unread_count INTEGER DEFAULT 0,
last_synced_at TEXT,
UNIQUE(account_id, path)
);
CREATE TABLE IF NOT EXISTS messages (
id INTEGER PRIMARY KEY AUTOINCREMENT,
account_id INTEGER NOT NULL REFERENCES accounts(id) ON DELETE CASCADE,
folder_id INTEGER NOT NULL REFERENCES folders(id) ON DELETE CASCADE,
uid INTEGER NOT NULL,
message_id TEXT,
in_reply_to TEXT,
"references" TEXT,
subject TEXT,
from_address TEXT,
from_name TEXT,
to_addresses TEXT,
cc_addresses TEXT,
bcc_addresses TEXT,
date TEXT,
text_body TEXT,
html_body TEXT,
flags TEXT,
size INTEGER,
has_attachments INTEGER DEFAULT 0,
raw_headers TEXT,
deleted_from_server INTEGER DEFAULT 0,
created_at TEXT NOT NULL DEFAULT (datetime('now')),
UNIQUE(folder_id, uid)
);
CREATE INDEX IF NOT EXISTS idx_messages_account ON messages(account_id);
CREATE INDEX IF NOT EXISTS idx_messages_folder ON messages(folder_id);
CREATE INDEX IF NOT EXISTS idx_messages_date ON messages(date DESC);
CREATE INDEX IF NOT EXISTS idx_messages_message_id ON messages(message_id);
CREATE INDEX IF NOT EXISTS idx_messages_in_reply_to ON messages(in_reply_to);
CREATE TABLE IF NOT EXISTS attachments (
id INTEGER PRIMARY KEY AUTOINCREMENT,
message_id INTEGER NOT NULL REFERENCES messages(id) ON DELETE CASCADE,
filename TEXT,
content_type TEXT,
size INTEGER,
content_id TEXT,
data BLOB
);
CREATE INDEX IF NOT EXISTS idx_attachments_message ON attachments(message_id);
CREATE VIRTUAL TABLE IF NOT EXISTS messages_fts USING fts5(
subject,
from_address,
from_name,
to_addresses,
text_body,
content=messages,
content_rowid=id,
tokenize='porter unicode61'
);
CREATE TRIGGER IF NOT EXISTS messages_ai AFTER INSERT ON messages BEGIN
INSERT INTO messages_fts(rowid, subject, from_address, from_name, to_addresses, text_body)
VALUES (new.id, new.subject, new.from_address, new.from_name, new.to_addresses, new.text_body);
END;
CREATE TRIGGER IF NOT EXISTS messages_ad AFTER DELETE ON messages BEGIN
INSERT INTO messages_fts(messages_fts, rowid, subject, from_address, from_name, to_addresses, text_body)
VALUES ('delete', old.id, old.subject, old.from_address, old.from_name, old.to_addresses, old.text_body);
END;
CREATE TRIGGER IF NOT EXISTS messages_au AFTER UPDATE ON messages BEGIN
INSERT INTO messages_fts(messages_fts, rowid, subject, from_address, from_name, to_addresses, text_body)
VALUES ('delete', old.id, old.subject, old.from_address, old.from_name, old.to_addresses, old.text_body);
INSERT INTO messages_fts(rowid, subject, from_address, from_name, to_addresses, text_body)
VALUES (new.id, new.subject, new.from_address, new.from_name, new.to_addresses, new.text_body);
END;
CREATE TABLE IF NOT EXISTS sync_state (
id INTEGER PRIMARY KEY AUTOINCREMENT,
account_id INTEGER NOT NULL REFERENCES accounts(id) ON DELETE CASCADE,
folder_id INTEGER NOT NULL REFERENCES folders(id) ON DELETE CASCADE,
last_uid INTEGER DEFAULT 0,
last_synced_at TEXT,
UNIQUE(account_id, folder_id)
);
CREATE TABLE IF NOT EXISTS schema_version (
version INTEGER NOT NULL
);
INSERT INTO schema_version (version) VALUES (1);
`,
// Version 2: Add special_use column for folder type detection
`
ALTER TABLE folders ADD COLUMN special_use TEXT;
`,
// Version 3: Add labels system — Gmail-style labels replace folders as the
// primary organizational model. IMAP folders are still synced for tracking
// sync state, but user-facing organization is label-based. When an email
// syncs from an IMAP folder, the folder name becomes a suggested initial label.
`
CREATE TABLE IF NOT EXISTS labels (
id INTEGER PRIMARY KEY AUTOINCREMENT,
account_id INTEGER NOT NULL REFERENCES accounts(id) ON DELETE CASCADE,
name TEXT NOT NULL,
color TEXT,
source TEXT NOT NULL DEFAULT 'user',
created_at TEXT NOT NULL DEFAULT (datetime('now')),
UNIQUE(account_id, name)
);
CREATE INDEX IF NOT EXISTS idx_labels_account ON labels(account_id);
CREATE TABLE IF NOT EXISTS message_labels (
message_id INTEGER NOT NULL REFERENCES messages(id) ON DELETE CASCADE,
label_id INTEGER NOT NULL REFERENCES labels(id) ON DELETE CASCADE,
created_at TEXT NOT NULL DEFAULT (datetime('now')),
PRIMARY KEY (message_id, label_id)
);
CREATE INDEX IF NOT EXISTS idx_message_labels_label ON message_labels(label_id);
`,
// Version 4: Add sync_errors table for persistent error tracking.
// Each row records a single error that occurred during sync — per-message
// parse failures, folder-level IMAP errors, etc. Errors are classified
// so the UI can distinguish retriable (transient) from permanent failures,
// and include enough context (folder, UID) for automatic retry.
`
CREATE TABLE IF NOT EXISTS sync_errors (
id INTEGER PRIMARY KEY AUTOINCREMENT,
account_id INTEGER NOT NULL REFERENCES accounts(id) ON DELETE CASCADE,
folder_path TEXT,
uid INTEGER,
error_type TEXT NOT NULL,
message TEXT NOT NULL,
retriable INTEGER NOT NULL DEFAULT 1,
resolved INTEGER NOT NULL DEFAULT 0,
retry_count INTEGER NOT NULL DEFAULT 0,
created_at TEXT NOT NULL DEFAULT (datetime('now')),
resolved_at TEXT
);
CREATE INDEX IF NOT EXISTS idx_sync_errors_account ON sync_errors(account_id);
CREATE INDEX IF NOT EXISTS idx_sync_errors_unresolved ON sync_errors(account_id, resolved);
`,
// Version 5: Add drafts table for server-side draft persistence.
// Drafts store compose state (to, cc, bcc, subject, body) so they survive
// across browser sessions and devices. Each draft belongs to an account and
// optionally references an original message (for reply/forward).
`
CREATE TABLE IF NOT EXISTS drafts (
id INTEGER PRIMARY KEY AUTOINCREMENT,
account_id INTEGER NOT NULL REFERENCES accounts(id) ON DELETE CASCADE,
to_addresses TEXT,
cc_addresses TEXT,
bcc_addresses TEXT,
subject TEXT,
text_body TEXT,
html_body TEXT,
in_reply_to TEXT,
"references" TEXT,
original_message_id INTEGER REFERENCES messages(id) ON DELETE SET NULL,
compose_mode TEXT NOT NULL DEFAULT 'new',
created_at TEXT NOT NULL DEFAULT (datetime('now')),
updated_at TEXT NOT NULL DEFAULT (datetime('now'))
);
CREATE INDEX IF NOT EXISTS idx_drafts_account ON drafts(account_id);
`,
// Version 6: Add image_trusted_senders table — persistent per-sender whitelist
// for remote image loading. When a sender is trusted, their remote images are
// loaded automatically without showing the "images blocked" banner.
// Tracking pixels are still stripped regardless of trust status.
`
CREATE TABLE IF NOT EXISTS image_trusted_senders (
id INTEGER PRIMARY KEY AUTOINCREMENT,
account_id INTEGER NOT NULL REFERENCES accounts(id) ON DELETE CASCADE,
sender_address TEXT NOT NULL,
created_at TEXT NOT NULL DEFAULT (datetime('now')),
UNIQUE(account_id, sender_address)
);
CREATE INDEX IF NOT EXISTS idx_image_trusted_senders_account ON image_trusted_senders(account_id);
CREATE INDEX IF NOT EXISTS idx_image_trusted_senders_lookup ON image_trusted_senders(account_id, sender_address);
`,
// Version 7: Add connector type columns for pluggable connector architecture.
// Accounts can now use alternative ingest/send connectors (Cloudflare Email Workers,
// AWS SES) instead of the default IMAP/SMTP. Connector-specific config is stored
// in dedicated columns. Existing accounts default to imap/smtp.
`
ALTER TABLE accounts ADD COLUMN ingest_connector_type TEXT NOT NULL DEFAULT 'imap';
ALTER TABLE accounts ADD COLUMN send_connector_type TEXT NOT NULL DEFAULT 'smtp';
ALTER TABLE accounts ADD COLUMN cf_email_webhook_secret TEXT;
ALTER TABLE accounts ADD COLUMN ses_region TEXT;
ALTER TABLE accounts ADD COLUMN ses_access_key_id TEXT;
ALTER TABLE accounts ADD COLUMN ses_secret_access_key TEXT;
`,
// Version 8: Add composite index for account+date queries.
// The separate indexes on account_id and date required a full account scan
// followed by a sort on large databases. The composite index lets SQLite
// satisfy ORDER BY date DESC for a given account in a single index scan.
`
CREATE INDEX IF NOT EXISTS idx_messages_account_date ON messages(account_id, date DESC);
`,
// Version 9: Add default_view column for per-account configurable landing view.
// Stores the view to auto-select on load: 'inbox', 'unread', 'all', or 'label:<id>'.
// Defaults to 'inbox' so existing accounts keep current behavior.
`
ALTER TABLE accounts ADD COLUMN default_view TEXT NOT NULL DEFAULT 'inbox';
`,
// Version 10: Add cached message_count and unread_count to labels table.
// The labels API endpoint previously computed these via an expensive JOIN across
// message_labels and messages on every request — on a 15 GB database this took
// 18+ seconds and blocked the synchronous event loop, starving concurrent requests.
// These columns are maintained by refreshLabelCounts() at the end of each sync
// cycle. The backfill here covers existing databases; future updates happen in sync.
`
ALTER TABLE labels ADD COLUMN message_count INTEGER NOT NULL DEFAULT 0;
ALTER TABLE labels ADD COLUMN unread_count INTEGER NOT NULL DEFAULT 0;
UPDATE labels SET message_count = (
SELECT COUNT(*) FROM message_labels WHERE label_id = labels.id
);
UPDATE labels SET unread_count = (
SELECT COUNT(*) FROM message_labels ml
JOIN messages m ON m.id = ml.message_id
WHERE ml.label_id = labels.id
AND (m.flags IS NULL OR m.flags NOT LIKE '%\\Seen%')
);
`,
// Version 11: Add cached_message_count and cached_unread_count to accounts table.
// The GET /accounts/:id/all-messages/count and GET /accounts/:id/unread-messages/count
// endpoints previously did a full messages table scan with a LIKE flags filter on every
// request. On a 15 GB database this blocked the event loop for 10-30 seconds.
// These columns are maintained by refreshIdentityCounts() at the end of each sync cycle.
// NULL means "not yet computed" — the endpoint falls back to a live query once, then
// stores the result. Subsequent requests return the cached value instantly.
`
ALTER TABLE accounts ADD COLUMN cached_message_count INTEGER DEFAULT NULL;
ALTER TABLE accounts ADD COLUMN cached_unread_count INTEGER DEFAULT NULL;
`,
// Version 12: Add pending_archive flag for crash-safe archive mode deletion.
// When archive mode is on, messages are marked pending_archive=1 immediately
// after being inserted. Phase 3 of syncFolder queries this column instead of
// relying on an in-memory list — so if the process is killed after Phase 1
// (fetch) but before Phase 3 (delete), the next sync cycle picks up and
// finishes the deletion. Cleared to 0 once deleted_from_server is set.
`
ALTER TABLE messages ADD COLUMN pending_archive INTEGER NOT NULL DEFAULT 0;
CREATE INDEX IF NOT EXISTS idx_messages_pending_archive ON messages(folder_id, pending_archive);
`,
// Version 13: Decouple inbound and outbound connectors from accounts.
//
// Previously, each account row embedded all connector config (IMAP/CF for
// ingest, SMTP/SES for send). This 1:1 coupling prevents sharing a single
// SES connector across multiple identities, or mixing Fastmail IMAP with
// AWS SES for outbound deliverability.
//
// The new model:
// - inbound_connectors: one row per ingest config (IMAP creds, CF webhook secret)
// - outbound_connectors: one row per send config (SMTP creds, SES creds)
// - accounts: identity table — holds name + email + references to connectors
//
// Migration: existing account rows are migrated 1:1 (one connector per account),
// preserving all existing behaviour. Connector rows take their name from the
// account name. The old inline columns remain on accounts for now (kept for
// backward-compat reads) but new code reads from the connector tables.
//
// _account_id is a temporary migration-aid column — it is set during the INSERT
// so we can link accounts back to their new connector rows, then cleared to NULL.
// SQLite pre-3.35 cannot DROP COLUMN, so we leave it as a permanently-NULL column.
`
CREATE TABLE IF NOT EXISTS inbound_connectors (
id INTEGER PRIMARY KEY AUTOINCREMENT,
name TEXT NOT NULL,
type TEXT NOT NULL DEFAULT 'imap',
imap_host TEXT,
imap_port INTEGER DEFAULT 993,
imap_tls INTEGER DEFAULT 1,
imap_user TEXT,
imap_pass TEXT,
cf_email_webhook_secret TEXT,
created_at TEXT NOT NULL DEFAULT (datetime('now')),
updated_at TEXT NOT NULL DEFAULT (datetime('now')),
_account_id INTEGER
);
CREATE INDEX IF NOT EXISTS idx_inbound_connectors_type ON inbound_connectors(type);
CREATE TABLE IF NOT EXISTS outbound_connectors (
id INTEGER PRIMARY KEY AUTOINCREMENT,
name TEXT NOT NULL,
type TEXT NOT NULL DEFAULT 'smtp',
smtp_host TEXT,
smtp_port INTEGER DEFAULT 587,
smtp_tls INTEGER DEFAULT 1,
smtp_user TEXT,
smtp_pass TEXT,
ses_region TEXT,
ses_access_key_id TEXT,
ses_secret_access_key TEXT,
created_at TEXT NOT NULL DEFAULT (datetime('now')),
updated_at TEXT NOT NULL DEFAULT (datetime('now')),
_account_id INTEGER
);
CREATE INDEX IF NOT EXISTS idx_outbound_connectors_type ON outbound_connectors(type);
ALTER TABLE accounts ADD COLUMN inbound_connector_id INTEGER REFERENCES inbound_connectors(id);
ALTER TABLE accounts ADD COLUMN outbound_connector_id INTEGER REFERENCES outbound_connectors(id);
INSERT INTO inbound_connectors (name, type, imap_host, imap_port, imap_tls, imap_user, imap_pass, cf_email_webhook_secret, _account_id)
SELECT name || ' (Inbound)', ingest_connector_type, imap_host, imap_port, imap_tls, imap_user, imap_pass, cf_email_webhook_secret, id
FROM accounts;
INSERT INTO outbound_connectors (name, type, smtp_host, smtp_port, smtp_tls, smtp_user, smtp_pass, ses_region, ses_access_key_id, ses_secret_access_key, _account_id)
SELECT name || ' (Outbound)', send_connector_type, smtp_host, smtp_port, smtp_tls, smtp_user, smtp_pass, ses_region, ses_access_key_id, ses_secret_access_key, id
FROM accounts;
UPDATE accounts SET
inbound_connector_id = (SELECT id FROM inbound_connectors WHERE _account_id = accounts.id),
outbound_connector_id = (SELECT id FROM outbound_connectors WHERE _account_id = accounts.id);
UPDATE inbound_connectors SET _account_id = NULL;
UPDATE outbound_connectors SET _account_id = NULL;
`,
// Version 14: Remove NOT NULL constraints from legacy IMAP columns on accounts.
//
// The original v1 schema declared imap_host/imap_user/imap_pass as NOT NULL because
// every account required IMAP. After v13 (n×m connectors), accounts no longer store
// connector config directly — they reference inbound_connectors instead. New accounts
// are created without inline IMAP fields, so those NOT NULL constraints must go.
//
// SQLite cannot DROP NOT NULL via ALTER COLUMN, so we recreate the accounts table.
// All FKs in child tables reference accounts by name and will reattach automatically
// once the table is renamed back. We disable FK enforcement during the operation to
// avoid intermediate-state violations.
`
PRAGMA foreign_keys = OFF;
BEGIN;
CREATE TABLE accounts_v14 (
id INTEGER PRIMARY KEY AUTOINCREMENT,
name TEXT NOT NULL,
email TEXT NOT NULL,
imap_host TEXT,
imap_port INTEGER NOT NULL DEFAULT 993,
imap_tls INTEGER NOT NULL DEFAULT 1,
imap_user TEXT,
imap_pass TEXT,
smtp_host TEXT,
smtp_port INTEGER DEFAULT 587,
smtp_tls INTEGER DEFAULT 1,
smtp_user TEXT,
smtp_pass TEXT,
sync_delete_from_server INTEGER NOT NULL DEFAULT 0,
created_at TEXT NOT NULL DEFAULT (datetime('now')),
updated_at TEXT NOT NULL DEFAULT (datetime('now')),
ingest_connector_type TEXT NOT NULL DEFAULT 'imap',
send_connector_type TEXT NOT NULL DEFAULT 'smtp',
cf_email_webhook_secret TEXT,
ses_region TEXT,
ses_access_key_id TEXT,
ses_secret_access_key TEXT,
default_view TEXT NOT NULL DEFAULT 'inbox',
cached_message_count INTEGER DEFAULT NULL,
cached_unread_count INTEGER DEFAULT NULL,
inbound_connector_id INTEGER REFERENCES inbound_connectors(id),
outbound_connector_id INTEGER REFERENCES outbound_connectors(id)
);
INSERT INTO accounts_v14 SELECT * FROM accounts;
DROP TABLE accounts;
ALTER TABLE accounts_v14 RENAME TO accounts;
COMMIT;
PRAGMA foreign_keys = ON;
`,
// Version 15: Unify labels — remove per-account scoping.
//
// Previously, labels had an account_id FK and were scoped to a single account.
// This produced the confusing UX where label choices varied depending on which
// account was selected in the compose or sidebar.
//
// The unified model: labels are instance-level (no account FK). All accounts share
// the same label namespace. When multiple accounts had labels with the same name
// (e.g. both had "INBOX"), the rows are merged: the lowest-id row becomes canonical
// and all message_labels pointing at the duplicate rows are redirected.
//
// Implications:
// - refreshLabelCounts() now counts across all accounts (no account filter)
// - ensureLabelsForFolders() uses ON CONFLICT(name) instead of (account_id, name)
// - applyFolderLabelsToMessages() joins labels by name only (no account scope)
`
PRAGMA foreign_keys = OFF;
BEGIN;
CREATE TEMP TABLE label_canonicals AS
SELECT name, MIN(id) AS canonical_id
FROM labels
GROUP BY name;
UPDATE message_labels
SET label_id = (
SELECT lc.canonical_id
FROM labels l
JOIN label_canonicals lc ON lc.name = l.name
WHERE l.id = message_labels.label_id
)
WHERE label_id NOT IN (SELECT canonical_id FROM label_canonicals);
DELETE FROM labels WHERE id NOT IN (SELECT canonical_id FROM label_canonicals);
CREATE TABLE labels_v15 (
id INTEGER PRIMARY KEY AUTOINCREMENT,
name TEXT NOT NULL,
color TEXT,
source TEXT NOT NULL DEFAULT 'user',
message_count INTEGER NOT NULL DEFAULT 0,
unread_count INTEGER NOT NULL DEFAULT 0,
created_at TEXT NOT NULL DEFAULT (datetime('now')),
UNIQUE(name)
);
INSERT INTO labels_v15 (id, name, color, source, message_count, unread_count, created_at)
SELECT id, name, color, source, message_count, unread_count, created_at
FROM labels;
DROP TABLE labels;
ALTER TABLE labels_v15 RENAME TO labels;
DROP INDEX IF EXISTS idx_labels_account;
CREATE INDEX IF NOT EXISTS idx_message_labels_label_v15 ON message_labels(label_id);
COMMIT;
PRAGMA foreign_keys = ON;
`,
// Version 16: Move sync_delete_from_server to inbound connectors + auto-account labels.
//
// sync_delete_from_server (Mirror vs Connector mode) is fundamentally a property
// of the inbound connector — it controls whether messages are deleted from the
// IMAP server after sync. For non-IMAP connectors (e.g. Cloudflare Email Workers),
// the concept doesn't apply. Moving it to inbound_connectors aligns the setting
// with the entity it actually affects.
//
// Auto-account labels: instead of a dedicated "Accounts" sidebar section, every
// incoming message is auto-labeled with its receiving account's name (source='account').
// These labels appear in the sidebar alongside other labels, enabling multi-label
// drill-down: click "Work" to see all Work messages, then filter further by "Inbox".
`
ALTER TABLE inbound_connectors ADD COLUMN sync_delete_from_server INTEGER NOT NULL DEFAULT 0;
UPDATE inbound_connectors SET sync_delete_from_server = (
SELECT a.sync_delete_from_server FROM accounts a
WHERE a.inbound_connector_id = inbound_connectors.id
LIMIT 1
)
WHERE EXISTS (
SELECT 1 FROM accounts a WHERE a.inbound_connector_id = inbound_connectors.id
);
`,
// Version 17: Add Cloudflare R2 inbound connector support.
//
// Replaces the direct webhook push model with a durable queue/poll pattern:
// 1. A Cloudflare Email Worker writes each email as a JSON object to an R2 bucket.
// 2. Stork polls the bucket on a configurable interval, processes pending objects,
// and deletes them after a successful DB write.
//
// This decouples email reception from server availability — emails queue in R2 while
// stork is down or locked and are processed on the next poll after unlock/restart.
//
// R2's S3-compatible API is used for listing, downloading, and deleting objects.
// Credentials are the R2 Access Key ID / Secret Access Key (not the CF API token).
`
ALTER TABLE inbound_connectors ADD COLUMN cf_r2_account_id TEXT;
ALTER TABLE inbound_connectors ADD COLUMN cf_r2_bucket_name TEXT;
ALTER TABLE inbound_connectors ADD COLUMN cf_r2_access_key_id TEXT;
ALTER TABLE inbound_connectors ADD COLUMN cf_r2_secret_access_key TEXT;
ALTER TABLE inbound_connectors ADD COLUMN cf_r2_prefix TEXT NOT NULL DEFAULT 'pending/';
ALTER TABLE inbound_connectors ADD COLUMN cf_r2_poll_interval_ms INTEGER;
`,
// Version 18: Eliminate "accounts" concept — replace with "identities".
//
// Accounts were originally IMAP-centric (one account = one mailbox with inline
// credentials). After v13 decoupled connectors into separate tables, the accounts
// table became a thin identity layer holding only name, email, and connector
// references — but still carried ~20 legacy columns (imap_*, smtp_*, ses_*, etc.)
// that are now redundant with the connector tables.
//
// This migration:
// 1. Creates a clean `identities` table with only: id, name, email,
// inbound_connector_id, outbound_connector_id, default_view, timestamps.
// 2. Renames account_id → identity_id in all child tables: messages, folders,
// sync_state, sync_errors, drafts.
// 3. Makes image_trusted_senders global (removes account_id, uses
// UNIQUE(sender_address)) — trust decisions apply across all identities.
// 4. Recreates all affected indexes with the new column names.
// 5. Drops the accounts table.
//
// The messages FTS triggers (messages_ai, messages_ad, messages_au) are dropped
// and recreated because they reference the messages table which is recreated.
`
PRAGMA foreign_keys = OFF;
BEGIN;
-- Step 1: Create identities table from accounts (only keeping relevant columns)
CREATE TABLE identities (
id INTEGER PRIMARY KEY AUTOINCREMENT,
name TEXT NOT NULL,
email TEXT NOT NULL,
inbound_connector_id INTEGER REFERENCES inbound_connectors(id),
outbound_connector_id INTEGER REFERENCES outbound_connectors(id),
default_view TEXT NOT NULL DEFAULT 'inbox',
created_at TEXT NOT NULL DEFAULT (datetime('now')),
updated_at TEXT NOT NULL DEFAULT (datetime('now'))
);
INSERT INTO identities (id, name, email, inbound_connector_id, outbound_connector_id, default_view, created_at, updated_at)
SELECT id, name, email, inbound_connector_id, outbound_connector_id, default_view, created_at, updated_at
FROM accounts;
-- Step 2: Recreate folders with identity_id instead of account_id
CREATE TABLE folders_v18 (
id INTEGER PRIMARY KEY AUTOINCREMENT,
identity_id INTEGER NOT NULL REFERENCES identities(id) ON DELETE CASCADE,
path TEXT NOT NULL,
name TEXT NOT NULL,
delimiter TEXT,
flags TEXT,
uid_validity INTEGER,
uid_next INTEGER,
message_count INTEGER DEFAULT 0,
unread_count INTEGER DEFAULT 0,
last_synced_at TEXT,
special_use TEXT,
UNIQUE(identity_id, path)
);
INSERT INTO folders_v18 (id, identity_id, path, name, delimiter, flags, uid_validity, uid_next, message_count, unread_count, last_synced_at, special_use)
SELECT id, account_id, path, name, delimiter, flags, uid_validity, uid_next, message_count, unread_count, last_synced_at, special_use
FROM folders;
DROP TABLE folders;
ALTER TABLE folders_v18 RENAME TO folders;
-- Step 3: Drop FTS triggers before recreating messages table
DROP TRIGGER IF EXISTS messages_ai;
DROP TRIGGER IF EXISTS messages_ad;
DROP TRIGGER IF EXISTS messages_au;
-- Step 4: Recreate messages with identity_id instead of account_id
CREATE TABLE messages_v18 (
id INTEGER PRIMARY KEY AUTOINCREMENT,
identity_id INTEGER NOT NULL REFERENCES identities(id) ON DELETE CASCADE,
folder_id INTEGER NOT NULL REFERENCES folders(id) ON DELETE CASCADE,
uid INTEGER NOT NULL,
message_id TEXT,
in_reply_to TEXT,
"references" TEXT,
subject TEXT,
from_address TEXT,
from_name TEXT,
to_addresses TEXT,
cc_addresses TEXT,
bcc_addresses TEXT,
date TEXT,
text_body TEXT,
html_body TEXT,
flags TEXT,
size INTEGER,
has_attachments INTEGER DEFAULT 0,
raw_headers TEXT,
deleted_from_server INTEGER DEFAULT 0,
created_at TEXT NOT NULL DEFAULT (datetime('now')),
pending_archive INTEGER NOT NULL DEFAULT 0,
UNIQUE(folder_id, uid)
);
INSERT INTO messages_v18 (id, identity_id, folder_id, uid, message_id, in_reply_to, "references", subject, from_address, from_name, to_addresses, cc_addresses, bcc_addresses, date, text_body, html_body, flags, size, has_attachments, raw_headers, deleted_from_server, created_at, pending_archive)
SELECT id, account_id, folder_id, uid, message_id, in_reply_to, "references", subject, from_address, from_name, to_addresses, cc_addresses, bcc_addresses, date, text_body, html_body, flags, size, has_attachments, raw_headers, deleted_from_server, created_at, pending_archive
FROM messages;
DROP TABLE messages;
ALTER TABLE messages_v18 RENAME TO messages;
-- Recreate messages indexes
CREATE INDEX idx_messages_identity ON messages(identity_id);
CREATE INDEX idx_messages_folder ON messages(folder_id);
CREATE INDEX idx_messages_date ON messages(date DESC);
CREATE INDEX idx_messages_message_id ON messages(message_id);
CREATE INDEX idx_messages_in_reply_to ON messages(in_reply_to);
CREATE INDEX idx_messages_identity_date ON messages(identity_id, date DESC);
CREATE INDEX idx_messages_pending_archive ON messages(folder_id, pending_archive);
-- Recreate FTS triggers
CREATE TRIGGER messages_ai AFTER INSERT ON messages BEGIN
INSERT INTO messages_fts(rowid, subject, from_address, from_name, to_addresses, text_body)
VALUES (new.id, new.subject, new.from_address, new.from_name, new.to_addresses, new.text_body);
END;
CREATE TRIGGER messages_ad AFTER DELETE ON messages BEGIN
INSERT INTO messages_fts(messages_fts, rowid, subject, from_address, from_name, to_addresses, text_body)
VALUES ('delete', old.id, old.subject, old.from_address, old.from_name, old.to_addresses, old.text_body);
END;
CREATE TRIGGER messages_au AFTER UPDATE ON messages BEGIN
INSERT INTO messages_fts(messages_fts, rowid, subject, from_address, from_name, to_addresses, text_body)
VALUES ('delete', old.id, old.subject, old.from_address, old.from_name, old.to_addresses, old.text_body);
INSERT INTO messages_fts(rowid, subject, from_address, from_name, to_addresses, text_body)
VALUES (new.id, new.subject, new.from_address, new.from_name, new.to_addresses, new.text_body);
END;
-- Step 5: Recreate sync_state with identity_id instead of account_id
CREATE TABLE sync_state_v18 (
id INTEGER PRIMARY KEY AUTOINCREMENT,
identity_id INTEGER NOT NULL REFERENCES identities(id) ON DELETE CASCADE,
folder_id INTEGER NOT NULL REFERENCES folders(id) ON DELETE CASCADE,
last_uid INTEGER DEFAULT 0,
last_synced_at TEXT,
UNIQUE(identity_id, folder_id)
);
INSERT INTO sync_state_v18 (id, identity_id, folder_id, last_uid, last_synced_at)
SELECT id, account_id, folder_id, last_uid, last_synced_at
FROM sync_state;
DROP TABLE sync_state;
ALTER TABLE sync_state_v18 RENAME TO sync_state;
-- Step 6: Recreate sync_errors with identity_id instead of account_id
CREATE TABLE sync_errors_v18 (
id INTEGER PRIMARY KEY AUTOINCREMENT,
identity_id INTEGER NOT NULL REFERENCES identities(id) ON DELETE CASCADE,
folder_path TEXT,
uid INTEGER,
error_type TEXT NOT NULL,
message TEXT NOT NULL,
retriable INTEGER NOT NULL DEFAULT 1,
resolved INTEGER NOT NULL DEFAULT 0,
retry_count INTEGER NOT NULL DEFAULT 0,
created_at TEXT NOT NULL DEFAULT (datetime('now')),
resolved_at TEXT
);
INSERT INTO sync_errors_v18 (id, identity_id, folder_path, uid, error_type, message, retriable, resolved, retry_count, created_at, resolved_at)
SELECT id, account_id, folder_path, uid, error_type, message, retriable, resolved, retry_count, created_at, resolved_at
FROM sync_errors;
DROP TABLE sync_errors;
ALTER TABLE sync_errors_v18 RENAME TO sync_errors;
CREATE INDEX idx_sync_errors_identity ON sync_errors(identity_id);
CREATE INDEX idx_sync_errors_unresolved ON sync_errors(identity_id, resolved);
-- Step 7: Recreate drafts with identity_id instead of account_id
CREATE TABLE drafts_v18 (
id INTEGER PRIMARY KEY AUTOINCREMENT,
identity_id INTEGER NOT NULL REFERENCES identities(id) ON DELETE CASCADE,
to_addresses TEXT,
cc_addresses TEXT,
bcc_addresses TEXT,
subject TEXT,
text_body TEXT,
html_body TEXT,
in_reply_to TEXT,
"references" TEXT,
original_message_id INTEGER REFERENCES messages(id) ON DELETE SET NULL,
compose_mode TEXT NOT NULL DEFAULT 'new',
created_at TEXT NOT NULL DEFAULT (datetime('now')),
updated_at TEXT NOT NULL DEFAULT (datetime('now'))
);
INSERT INTO drafts_v18 (id, identity_id, to_addresses, cc_addresses, bcc_addresses, subject, text_body, html_body, in_reply_to, "references", original_message_id, compose_mode, created_at, updated_at)
SELECT id, account_id, to_addresses, cc_addresses, bcc_addresses, subject, text_body, html_body, in_reply_to, "references", original_message_id, compose_mode, created_at, updated_at
FROM drafts;
DROP TABLE drafts;
ALTER TABLE drafts_v18 RENAME TO drafts;
CREATE INDEX idx_drafts_identity ON drafts(identity_id);
-- Step 8: Recreate image_trusted_senders as global (no account_id)
CREATE TABLE image_trusted_senders_v18 (
id INTEGER PRIMARY KEY AUTOINCREMENT,
sender_address TEXT NOT NULL,
created_at TEXT NOT NULL DEFAULT (datetime('now')),
UNIQUE(sender_address)
);
INSERT OR IGNORE INTO image_trusted_senders_v18 (sender_address, created_at)
SELECT sender_address, MIN(created_at)
FROM image_trusted_senders
GROUP BY sender_address;
DROP TABLE image_trusted_senders;
ALTER TABLE image_trusted_senders_v18 RENAME TO image_trusted_senders;
CREATE INDEX idx_image_trusted_senders_lookup ON image_trusted_senders(sender_address);
-- Step 9: Drop the accounts table
DROP TABLE accounts;
COMMIT;
PRAGMA foreign_keys = ON;
`,
// Version 19: Identities are SEND-ONLY — remove inbound_connector_id from identities.
//
// Core principle: identities hold name + email + outbound connector (for sending).
// Inbound connectors are standalone: messages, folders, sync_state, and sync_errors
// now reference inbound_connectors directly, not through identities.
//
// Changes:
// 1. identities: remove inbound_connector_id column (temp table approach)
// 2. folders: add inbound_connector_id, set identity_id = NULL on migrated rows,
// replace UNIQUE(identity_id, path) with two partial unique indexes
// 3. messages: add inbound_connector_id, set identity_id = NULL on migrated rows
// 4. sync_state: replace identity_id with inbound_connector_id
// 5. sync_errors: replace identity_id with inbound_connector_id
//
// The drafts table is unchanged: drafts are outgoing and remain identity-based.
`
PRAGMA foreign_keys = OFF;
BEGIN;
-- Step 1: Recreate identities without inbound_connector_id
-- Keep old table as identities_v19_temp for data migration
ALTER TABLE identities RENAME TO identities_v19_temp;
CREATE TABLE identities (
id INTEGER PRIMARY KEY AUTOINCREMENT,
name TEXT NOT NULL,
email TEXT NOT NULL,
outbound_connector_id INTEGER REFERENCES outbound_connectors(id),
default_view TEXT NOT NULL DEFAULT 'inbox',
created_at TEXT NOT NULL DEFAULT (datetime('now')),
updated_at TEXT NOT NULL DEFAULT (datetime('now'))
);
INSERT INTO identities (id, name, email, outbound_connector_id, default_view, created_at, updated_at)
SELECT id, name, email, outbound_connector_id, default_view, created_at, updated_at
FROM identities_v19_temp;
-- Step 2: Recreate folders with inbound_connector_id
-- Drop FTS triggers first (they reference messages which will be recreated)
DROP TRIGGER IF EXISTS messages_ai;
DROP TRIGGER IF EXISTS messages_ad;
DROP TRIGGER IF EXISTS messages_au;
CREATE TABLE folders_v19 (
id INTEGER PRIMARY KEY AUTOINCREMENT,
identity_id INTEGER REFERENCES identities(id) ON DELETE CASCADE,
inbound_connector_id INTEGER REFERENCES inbound_connectors(id) ON DELETE CASCADE,
path TEXT NOT NULL,
name TEXT NOT NULL,
delimiter TEXT,
flags TEXT,
uid_validity INTEGER,
uid_next INTEGER,
message_count INTEGER DEFAULT 0,
unread_count INTEGER DEFAULT 0,
last_synced_at TEXT,
special_use TEXT
);
-- Migrate folders: set inbound_connector_id from old identity's inbound_connector_id,
-- set identity_id = NULL (folders are for received mail, not identity-owned)
INSERT INTO folders_v19 (id, identity_id, inbound_connector_id, path, name, delimiter, flags, uid_validity, uid_next, message_count, unread_count, last_synced_at, special_use)
SELECT f.id, NULL, t.inbound_connector_id, f.path, f.name, f.delimiter, f.flags, f.uid_validity, f.uid_next, f.message_count, f.unread_count, f.last_synced_at, f.special_use
FROM folders f
JOIN identities_v19_temp t ON t.id = f.identity_id;
DROP TABLE folders;
ALTER TABLE folders_v19 RENAME TO folders;
-- Partial unique indexes replace the old UNIQUE(identity_id, path)
CREATE UNIQUE INDEX idx_folders_unique_inbound ON folders(inbound_connector_id, path) WHERE inbound_connector_id IS NOT NULL;
CREATE UNIQUE INDEX idx_folders_unique_identity ON folders(identity_id, path) WHERE identity_id IS NOT NULL;
-- Step 3: Recreate messages with inbound_connector_id
CREATE TABLE messages_v19 (
id INTEGER PRIMARY KEY AUTOINCREMENT,
identity_id INTEGER REFERENCES identities(id) ON DELETE CASCADE,
inbound_connector_id INTEGER REFERENCES inbound_connectors(id) ON DELETE CASCADE,
folder_id INTEGER NOT NULL REFERENCES folders(id) ON DELETE CASCADE,
uid INTEGER NOT NULL,
message_id TEXT,
in_reply_to TEXT,
"references" TEXT,
subject TEXT,
from_address TEXT,
from_name TEXT,
to_addresses TEXT,
cc_addresses TEXT,
bcc_addresses TEXT,
date TEXT,
text_body TEXT,
html_body TEXT,
flags TEXT,
size INTEGER,
has_attachments INTEGER DEFAULT 0,
raw_headers TEXT,
deleted_from_server INTEGER DEFAULT 0,
created_at TEXT NOT NULL DEFAULT (datetime('now')),
pending_archive INTEGER NOT NULL DEFAULT 0,
UNIQUE(folder_id, uid)
);
-- Migrate messages: set inbound_connector_id from old identity's connector, set identity_id = NULL
INSERT INTO messages_v19 (id, identity_id, inbound_connector_id, folder_id, uid, message_id, in_reply_to, "references", subject, from_address, from_name, to_addresses, cc_addresses, bcc_addresses, date, text_body, html_body, flags, size, has_attachments, raw_headers, deleted_from_server, created_at, pending_archive)
SELECT m.id, NULL, t.inbound_connector_id, m.folder_id, m.uid, m.message_id, m.in_reply_to, m."references", m.subject, m.from_address, m.from_name, m.to_addresses, m.cc_addresses, m.bcc_addresses, m.date, m.text_body, m.html_body, m.flags, m.size, m.has_attachments, m.raw_headers, m.deleted_from_server, m.created_at, m.pending_archive
FROM messages m
JOIN identities_v19_temp t ON t.id = m.identity_id;
DROP TABLE messages;
ALTER TABLE messages_v19 RENAME TO messages;
-- Recreate messages indexes
CREATE INDEX idx_messages_inbound_connector ON messages(inbound_connector_id);
CREATE INDEX idx_messages_identity ON messages(identity_id);
CREATE INDEX idx_messages_folder ON messages(folder_id);
CREATE INDEX idx_messages_date ON messages(date DESC);
CREATE INDEX idx_messages_message_id ON messages(message_id);
CREATE INDEX idx_messages_in_reply_to ON messages(in_reply_to);
CREATE INDEX idx_messages_inbound_connector_date ON messages(inbound_connector_id, date DESC);
CREATE INDEX idx_messages_pending_archive ON messages(folder_id, pending_archive);
-- Recreate FTS triggers
CREATE TRIGGER messages_ai AFTER INSERT ON messages BEGIN
INSERT INTO messages_fts(rowid, subject, from_address, from_name, to_addresses, text_body)
VALUES (new.id, new.subject, new.from_address, new.from_name, new.to_addresses, new.text_body);
END;
CREATE TRIGGER messages_ad AFTER DELETE ON messages BEGIN
INSERT INTO messages_fts(messages_fts, rowid, subject, from_address, from_name, to_addresses, text_body)
VALUES ('delete', old.id, old.subject, old.from_address, old.from_name, old.to_addresses, old.text_body);
END;
CREATE TRIGGER messages_au AFTER UPDATE ON messages BEGIN
INSERT INTO messages_fts(messages_fts, rowid, subject, from_address, from_name, to_addresses, text_body)
VALUES ('delete', old.id, old.subject, old.from_address, old.from_name, old.to_addresses, old.text_body);
INSERT INTO messages_fts(rowid, subject, from_address, from_name, to_addresses, text_body)
VALUES (new.id, new.subject, new.from_address, new.from_name, new.to_addresses, new.text_body);
END;
-- Step 4: Recreate sync_state with inbound_connector_id instead of identity_id
CREATE TABLE sync_state_v19 (
id INTEGER PRIMARY KEY AUTOINCREMENT,
inbound_connector_id INTEGER NOT NULL REFERENCES inbound_connectors(id) ON DELETE CASCADE,
folder_id INTEGER NOT NULL REFERENCES folders(id) ON DELETE CASCADE,
last_uid INTEGER DEFAULT 0,
last_synced_at TEXT,
UNIQUE(inbound_connector_id, folder_id)
);
INSERT INTO sync_state_v19 (id, inbound_connector_id, folder_id, last_uid, last_synced_at)
SELECT ss.id, t.inbound_connector_id, ss.folder_id, ss.last_uid, ss.last_synced_at
FROM sync_state ss
JOIN identities_v19_temp t ON t.id = ss.identity_id;
DROP TABLE sync_state;
ALTER TABLE sync_state_v19 RENAME TO sync_state;
-- Step 5: Recreate sync_errors with inbound_connector_id instead of identity_id
CREATE TABLE sync_errors_v19 (
id INTEGER PRIMARY KEY AUTOINCREMENT,
inbound_connector_id INTEGER NOT NULL REFERENCES inbound_connectors(id) ON DELETE CASCADE,
folder_path TEXT,
uid INTEGER,
error_type TEXT NOT NULL,
message TEXT NOT NULL,
retriable INTEGER NOT NULL DEFAULT 1,
resolved INTEGER NOT NULL DEFAULT 0,
retry_count INTEGER NOT NULL DEFAULT 0,
created_at TEXT NOT NULL DEFAULT (datetime('now')),
resolved_at TEXT
);
INSERT INTO sync_errors_v19 (id, inbound_connector_id, folder_path, uid, error_type, message, retriable, resolved, retry_count, created_at, resolved_at)
SELECT se.id, t.inbound_connector_id, se.folder_path, se.uid, se.error_type, se.message, se.retriable, se.resolved, se.retry_count, se.created_at, se.resolved_at
FROM sync_errors se
JOIN identities_v19_temp t ON t.id = se.identity_id;
DROP TABLE sync_errors;
ALTER TABLE sync_errors_v19 RENAME TO sync_errors;
CREATE INDEX idx_sync_errors_inbound_connector ON sync_errors(inbound_connector_id);
CREATE INDEX idx_sync_errors_unresolved ON sync_errors(inbound_connector_id, resolved);
-- Step 6: Recreate drafts to fix FK reference (RENAME above updated it to identities_v19_temp)
CREATE TABLE drafts_v19 (
id INTEGER PRIMARY KEY AUTOINCREMENT,
identity_id INTEGER REFERENCES identities(id) ON DELETE CASCADE,
to_addresses TEXT,
cc_addresses TEXT,
bcc_addresses TEXT,
subject TEXT,
text_body TEXT,
html_body TEXT,
in_reply_to TEXT,
"references" TEXT,
original_message_id INTEGER REFERENCES messages(id) ON DELETE SET NULL,
compose_mode TEXT NOT NULL DEFAULT 'new',
created_at TEXT NOT NULL DEFAULT (datetime('now')),
updated_at TEXT NOT NULL DEFAULT (datetime('now'))
);
INSERT INTO drafts_v19 (id, identity_id, to_addresses, cc_addresses, bcc_addresses, subject, text_body, html_body, in_reply_to, "references", original_message_id, compose_mode, created_at, updated_at)
SELECT id, identity_id, to_addresses, cc_addresses, bcc_addresses, subject, text_body, html_body, in_reply_to, "references", original_message_id, compose_mode, created_at, updated_at
FROM drafts;
DROP TABLE drafts;
ALTER TABLE drafts_v19 RENAME TO drafts;
CREATE INDEX idx_drafts_identity ON drafts(identity_id);
-- Drop the temp table
DROP TABLE identities_v19_temp;
COMMIT;
PRAGMA foreign_keys = ON;
`,
// Version 20: Hash-based attachment de-duplication.
//
// The same file is frequently attached to multiple messages (e.g. a logo attached
// to every newsletter, or a PDF sent in a thread). Previously each attachment was
// stored as an independent BLOB row — N copies of the same bytes for N messages.
//
// This migration introduces content-addressable attachment storage:
// - attachment_blobs: stores each unique file exactly once, keyed by SHA-256 hash.
// - attachments.content_hash: FK into attachment_blobs. When set, the actual bytes
// are read from attachment_blobs instead of attachments.data.
//
// Backward-compatibility: existing rows keep their data in attachments.data (not NULL).
// The read path falls back to attachments.data when content_hash IS NULL. Going forward,
// new attachment writes set content_hash and leave data NULL, saving one copy per dupe.
`
CREATE TABLE IF NOT EXISTS attachment_blobs (
content_hash TEXT PRIMARY KEY,
data BLOB NOT NULL
);
ALTER TABLE attachments ADD COLUMN content_hash TEXT REFERENCES attachment_blobs(content_hash);
CREATE INDEX IF NOT EXISTS idx_attachments_content_hash ON attachments(content_hash);
`,
// Version 21: Clean up attachment de-duplication — drop legacy inline data column.
//
// Pre-release cleanup (no real users yet): removes the dual-storage fallback from v20.
// All attachment data now lives exclusively in attachment_blobs. The attachments table
// no longer has a data column; content_hash is NOT NULL.
//
// Migration steps:
// 1. Move any existing inline data (attachments.data) into attachment_blobs.
// 2. Rebuild the attachments table without the data column, content_hash NOT NULL.
// 3. Delete orphaned attachment rows that had no data and no content_hash.
`
PRAGMA foreign_keys = OFF;
BEGIN;
-- Step 1: Migrate any legacy inline data into attachment_blobs.
-- Compute SHA-256 would require app code, so we handle this in the JS migration hook.
-- For pure-SQL, we delete rows that have no data at all (no inline data AND no content_hash).
DELETE FROM attachments WHERE data IS NULL AND content_hash IS NULL;
-- Step 2: Rebuild attachments without the data column, content_hash NOT NULL.
CREATE TABLE attachments_v21 (
id INTEGER PRIMARY KEY AUTOINCREMENT,
message_id INTEGER NOT NULL REFERENCES messages(id) ON DELETE CASCADE,
filename TEXT,
content_type TEXT,
size INTEGER,
content_id TEXT,
content_hash TEXT NOT NULL REFERENCES attachment_blobs(content_hash)
);
INSERT INTO attachments_v21 (id, message_id, filename, content_type, size, content_id, content_hash)
SELECT id, message_id, filename, content_type, size, content_id, content_hash
FROM attachments
WHERE content_hash IS NOT NULL;
DROP TABLE attachments;
ALTER TABLE attachments_v21 RENAME TO attachments;
CREATE INDEX IF NOT EXISTS idx_attachments_message ON attachments(message_id);
CREATE INDEX IF NOT EXISTS idx_attachments_content_hash ON attachments(content_hash);
COMMIT;
PRAGMA foreign_keys = ON;
`,
// Version 22: Enforce connector-identity schema invariant.
//
// Every identity must be created alongside an outbound connector — there is
// no "identity without a connector" state. This migration tightens the schema:
//
// - outbound_connector_id becomes NOT NULL (was nullable).
// - ON DELETE CASCADE added so deleting a connector removes its identities.
//
// Pre-release cleanup: no real users, so any identities with a NULL
// outbound_connector_id are deleted rather than migrated (they are orphans).
`
PRAGMA foreign_keys = OFF;
BEGIN;
DELETE FROM identities WHERE outbound_connector_id IS NULL;
CREATE TABLE identities_v22 (
id INTEGER PRIMARY KEY AUTOINCREMENT,
name TEXT NOT NULL,
email TEXT NOT NULL,
outbound_connector_id INTEGER NOT NULL REFERENCES outbound_connectors(id) ON DELETE CASCADE,
default_view TEXT NOT NULL DEFAULT 'inbox',
created_at TEXT NOT NULL DEFAULT (datetime('now')),
updated_at TEXT NOT NULL DEFAULT (datetime('now'))
);
INSERT INTO identities_v22 (id, name, email, outbound_connector_id, default_view, created_at, updated_at)
SELECT id, name, email, outbound_connector_id, default_view, created_at, updated_at
FROM identities;
DROP TABLE identities;
ALTER TABLE identities_v22 RENAME TO identities;
COMMIT;
PRAGMA foreign_keys = ON;
`,
// Version 23: Compress html_body, raw_headers, and attachment blob data.
//
// Emails contain a lot of text that compresses very well. HTML bodies and raw
// headers are often 5-10x larger than the information they carry. Compressing
// these fields at the application level (zlib deflate) typically reduces the
// messages table size by 50-70%.
//
// What is compressed:
// - messages.html_body: HTML email bodies (not in FTS5 index)
// - messages.raw_headers: full raw MIME headers (not searched)
// - attachment_blobs.data: binary attachment content
//
// What is NOT compressed:
// - messages.text_body: indexed by FTS5 via content=messages triggers.
// Compressing it would break full-text search.
//
// SQLite is dynamically typed, so TEXT columns can hold BLOBs. The application
// layer detects the type on read: strings pass through (legacy uncompressed),
// Buffers are inflated (compressed). No schema DDL changes needed — the
// pre-migration JS hook batch-compresses existing data in place.
//
// This is a no-op SQL migration; all work happens in PRE_MIGRATION_HOOKS[23].
"SELECT 1",
// Version 24: Backfill text_body for HTML-only messages.
//
// Emails with only an HTML body (no plain-text MIME part) had text_body = NULL,
// making them invisible to FTS5 search. This migration extracts searchable
// plain text from html_body for those messages.
//
// Going forward, the write paths (imap-sync, email-storage, send) use
// htmlToText() as a fallback when parsed.text is absent.
//
// No-op SQL — the pre-migration hook does the work.
"SELECT 1",
// Version 25: Add html_text_body column and include it in FTS5.
//
// Adds a dedicated column for text extracted from HTML bodies, separate from
// text_body (the original plain-text MIME part). Both are indexed by FTS5 so
// search finds content regardless of which MIME part it lives in. This solves
// the case where plain-text is a stub ("click here to view online") but the
// HTML body contains the real content.
//
// The pre-migration hook adds the column and backfills it from html_body.
// The SQL migration recreates the FTS table and triggers to include the new
// column, then rebuilds the index.
`
DROP TRIGGER IF EXISTS messages_ai;
DROP TRIGGER IF EXISTS messages_ad;
DROP TRIGGER IF EXISTS messages_au;
DROP TABLE IF EXISTS messages_fts;
CREATE VIRTUAL TABLE messages_fts USING fts5(
subject,
from_address,
from_name,
to_addresses,
text_body,
html_text_body,
content=messages,
content_rowid=id,
tokenize='porter unicode61'
);
CREATE TRIGGER messages_ai AFTER INSERT ON messages BEGIN
INSERT INTO messages_fts(rowid, subject, from_address, from_name, to_addresses, text_body, html_text_body)
VALUES (new.id, new.subject, new.from_address, new.from_name, new.to_addresses, new.text_body, new.html_text_body);
END;
CREATE TRIGGER messages_ad AFTER DELETE ON messages BEGIN
INSERT INTO messages_fts(messages_fts, rowid, subject, from_address, from_name, to_addresses, text_body, html_text_body)
VALUES ('delete', old.id, old.subject, old.from_address, old.from_name, old.to_addresses, old.text_body, old.html_text_body);
END;
CREATE TRIGGER messages_au AFTER UPDATE ON messages BEGIN
INSERT INTO messages_fts(messages_fts, rowid, subject, from_address, from_name, to_addresses, text_body, html_text_body)
VALUES ('delete', old.id, old.subject, old.from_address, old.from_name, old.to_addresses, old.text_body, old.html_text_body);
INSERT INTO messages_fts(rowid, subject, from_address, from_name, to_addresses, text_body, html_text_body)
VALUES (new.id, new.subject, new.from_address, new.from_name, new.to_addresses, new.text_body, new.html_text_body);
END;
INSERT INTO messages_fts(messages_fts) VALUES ('rebuild');
`,
// Version 26: Labels use full IMAP folder path instead of leaf name.
// The pre-migration hook renames existing labels and re-links message_labels.
"SELECT 1",
// Version 27: Move attachment_blobs to separate hot/cold storage DB.
//
// Attachment blobs (the bulk of large mailbox storage) are moved to stork-blobs.db,
// which is ATTACHed as 'blobs'. The main stork.db shrinks dramatically — a 16 GB
// mailbox typically has 14+ GB in blobs, leaving a ~2 GB main DB that fits in RAM.
// Indices, message metadata, FTS, and labels stay in main and can be fully mmap'd.
//
// The attachments table is rebuilt without the cross-schema FK constraint (SQLite
// doesn't enforce FK references into attached databases, so app code ensures integrity).
// Data migration happens in PRE_MIGRATION_HOOKS[27].
`
PRAGMA foreign_keys = OFF;
BEGIN;
CREATE TABLE attachments_v27 (
id INTEGER PRIMARY KEY AUTOINCREMENT,
message_id INTEGER NOT NULL REFERENCES messages(id) ON DELETE CASCADE,
filename TEXT,
content_type TEXT,
size INTEGER,
content_id TEXT,
content_hash TEXT NOT NULL
);
INSERT INTO attachments_v27 (id, message_id, filename, content_type, size, content_id, content_hash)
SELECT id, message_id, filename, content_type, size, content_id, content_hash
FROM attachments;
DROP TABLE attachments;
ALTER TABLE attachments_v27 RENAME TO attachments;
CREATE INDEX IF NOT EXISTS idx_attachments_message ON attachments(message_id);
CREATE INDEX IF NOT EXISTS idx_attachments_content_hash ON attachments(content_hash);
DROP TABLE IF EXISTS attachment_blobs;
COMMIT;
PRAGMA foreign_keys = ON;
`,
// Version 28: Track user label overrides so sync doesn't re-add removed labels.
//
// When a user removes an IMAP-derived label (source='imap') from a message,
// we record it here. applyFolderLabelsToMessages() skips pairs present in
// this table, so the removal persists across syncs. Re-adding the label via
// the API clears the override.
`
CREATE TABLE IF NOT EXISTS label_overrides (
message_id INTEGER NOT NULL REFERENCES messages(id) ON DELETE CASCADE,
label_id INTEGER NOT NULL REFERENCES labels(id) ON DELETE CASCADE,
created_at TEXT NOT NULL DEFAULT (datetime('now')),
PRIMARY KEY (message_id, label_id)
);
`,
// Version 29: Add icon column to labels so users can customise label icons.
`ALTER TABLE labels ADD COLUMN icon TEXT;`,
// Version 30: Covering index on message_labels for O(1) paginated label views.
//
// The root cause of slow page loads on large mailboxes (5+ GB): every paginated
// label view (inbox, custom labels) required SQLite to materialise ALL matching
// message IDs from message_labels, join to messages, sort by date, then apply
// LIMIT/OFFSET. On a 5 GB mailbox with 2M+ message_labels rows this took 30-60s.
//
// Fix: add a `date` column to message_labels and a covering compound index
// (label_id, date DESC). SQLite can now satisfy ORDER BY date DESC with an index
// range scan in label_id order — no sort step, O(LIMIT) I/Os regardless of inbox
// size. The pre-migration JS hook backfills date for all existing rows in batches.
//
// Also adds cached_message_count / cached_unread_count to inbound_connectors.
// The cross-connector all-messages and unread-messages count endpoints previously
// did a full messages table scan with a LIKE on flags. These columns let those
// endpoints return in O(connectors) — a tiny table — instead of O(messages).
// Maintained by refreshLabelCounts() at the end of each sync cycle.
`
ALTER TABLE message_labels ADD COLUMN date TEXT;
CREATE INDEX idx_message_labels_label_date ON message_labels(label_id, date DESC);
ALTER TABLE inbound_connectors ADD COLUMN cached_message_count INTEGER DEFAULT NULL;
ALTER TABLE inbound_connectors ADD COLUMN cached_unread_count INTEGER DEFAULT NULL;
`,
];
|