feat: adopt node-pg-migrate for database migrations (#78)#88
feat: adopt node-pg-migrate for database migrations (#78)#88BillChirico merged 10 commits intomainfrom
Conversation
|
Caution Review failedThe pull request is closed. ℹ️ Recent review infoConfiguration used: Organization UI Review profile: ASSERTIVE Plan: Pro ⛔ Files ignored due to path filters (1)
📒 Files selected for processing (13)
📝 WalkthroughSummary by CodeRabbit
WalkthroughReplaces inline DB schema setup with node-pg-migrate migrations. Adds an initial migration, migration runner in Changes
Possibly related PRs
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
|
Note Docstrings generation - SUCCESS |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 8
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@migrations/001_initial-schema.cjs`:
- Around line 103-105: The index idx_ai_usage_user_created on ai_usage(user_id,
created_at) will include many NULL user_id rows; replace it with a partial index
that only indexes rows where user_id IS NOT NULL to avoid index bloat. Locate
the CREATE INDEX statement for idx_ai_usage_user_created and change it so the
migration drops or does not create the full index and instead creates a partial
index on (user_id, created_at) with a WHERE clause user_id IS NOT NULL (use the
same index name or a new one as appropriate) so queries filtering by user_id use
a compact index.
- Around line 62-74: The foreign key on column case_id in the CREATE TABLE for
mod_scheduled_actions currently reads "case_id INTEGER REFERENCES mod_cases(id)"
which defaults to RESTRICT; update that reference to include an explicit ON
DELETE policy (e.g., "case_id INTEGER REFERENCES mod_cases(id) ON DELETE
CASCADE" to remove scheduled actions when a case is deleted, or "case_id INTEGER
REFERENCES mod_cases(id) ON DELETE SET NULL" and make case_id nullable if you
prefer to keep actions but disassociate them). Modify the CREATE TABLE SQL in
the migrations/001_initial-schema.cjs diff (the mod_scheduled_actions
definition) to include the chosen ON DELETE clause and adjust NULL/NOT NULL on
case_id as required.
In `@package.json`:
- Around line 18-19: The package.json defines two identical npm scripts,
"migrate" and "migrate:up", which is redundant; update package.json to remove
the duplicate by either deleting one of the entries (e.g., remove "migrate:up")
or make "migrate" a canonical alias with a clear comment and keep only one
implementation so that only a single script (either "migrate" or "migrate:up")
runs node-pg-migrate up --migrations-dir migrations to avoid duplication.
- Line 31: The package.json declares "node-pg-migrate" at ^8.0.4 which requires
Node >=20.11.0 but the "engines" field still allows "node": ">=18.0.0"; update
the package.json "engines" entry to require Node >=20.11.0 (e.g. "node":
">=20.11.0") and ensure any other engine declarations (the other engines entries
referenced around lines 43-45) are updated consistently; also verify CI/test
matrix and any documentation mentioning Node versions are aligned with this new
minimum.
In `@TASK.md`:
- Around line 119-129: Update the acceptance criterion that currently reads
"Initial migration captures all 9 tables + all indexes" to "Initial migration
captures all 8 tables + all indexes" to match the spec and migration; ensure the
eight table names (config, conversations, mod_cases, mod_scheduled_actions,
memory_optouts, ai_usage, logs, bot_restarts) are reflected implicitly by the
corrected count so the acceptance checklist matches the actual migration file
and Section 3 of the spec.
- Around line 7-9: Update the stale metadata in TASK.md by replacing the PR
field value from "TBD" to "#88" and change the State field from "🔄 In Progress"
to a completed status (e.g., "✅ Done" or "Completed") so the PR and state
reflect the finished work.
- Around line 3-9: The file has markdownlint violations MD022/MD031 due to
missing blank lines around headings and fenced code blocks (e.g., the "##
Status" heading and other headings/blocks); update TASK.md by inserting a single
blank line before and after each heading and fenced code block so that every
heading (like "## Status") is separated from adjacent content and all fenced
code blocks have blank lines above and below, then re-run markdownlint to ensure
MD022 and MD031 are resolved.
In `@tests/db.test.js`:
- Around line 103-118: The test currently asserts migration runner options but
omits verifying the log callback; update the test that inspects
migrationMocks.runner.mock.calls[0][0] (runnerOpts) to also assert that
runnerOpts.log is a function, ensuring the log integration passed from src/db.js
remains present (use migrationMocks.runner and runnerOpts.log to locate the
check).
ℹ️ Review info
Configuration used: Organization UI
Review profile: ASSERTIVE
Plan: Pro
⛔ Files ignored due to path filters (1)
pnpm-lock.yamlis excluded by!**/pnpm-lock.yaml
📒 Files selected for processing (14)
DockerfileTASK.mdmigrations/001_initial-schema.cjspackage.jsonsrc/config-listeners.jssrc/db.jssrc/index.jssrc/transports/postgres.jssrc/utils/restartTracker.jstests/config-listeners.test.jstests/db.test.jstests/index.test.jstests/transports/postgres.test.jstests/utils/restartTracker.test.js
💤 Files with no reviewable changes (4)
- src/config-listeners.js
- src/utils/restartTracker.js
- src/transports/postgres.js
- tests/index.test.js
📜 Review details
🧰 Additional context used
📓 Path-based instructions (3)
**/*.js
📄 CodeRabbit inference engine (AGENTS.md)
**/*.js: Use ESM modules only — useimport/export, neverrequire()
Usenode:protocol for Node.js builtins (e.g.import { readFileSync } from 'node:fs')
Always use semicolons
Use single quotes for strings
Use 2-space indentation
No TypeScript — use plain JavaScript with JSDoc comments for documentation
Files:
tests/config-listeners.test.jstests/db.test.jstests/transports/postgres.test.jssrc/db.jstests/utils/restartTracker.test.jssrc/index.js
tests/**/*.js
📄 CodeRabbit inference engine (AGENTS.md)
Test files must achieve at least 80% code coverage on statements, branches, functions, and lines
Files:
tests/config-listeners.test.jstests/db.test.jstests/transports/postgres.test.jstests/utils/restartTracker.test.js
src/**/*.js
📄 CodeRabbit inference engine (AGENTS.md)
src/**/*.js: Always use Winston for logging — import{ info, warn, error }from../logger.js
NEVER useconsole.log,console.warn,console.error, or anyconsole.*method in src/ files
Pass structured metadata to Winston logging calls (e.g.info('Message processed', { userId, channelId }))
Use custom error classes fromsrc/utils/errors.jsfor error handling
Always log errors with context before re-throwing
UsegetConfig(guildId?)fromsrc/modules/config.jsto read config
UsesetConfigValue(path, value, guildId?)fromsrc/modules/config.jsto update config at runtime
UsesplitMessage()utility for messages exceeding Discord's 2000-character limit
UsesafeSend()wrapper for outgoing Discord messages to sanitize mentions and enforce allowedMentions
Files:
src/db.jssrc/index.js
🧠 Learnings (4)
📚 Learning: 2026-02-25T02:39:33.506Z
Learnt from: CR
Repo: VolvoxLLC/volvox-bot PR: 0
File: AGENTS.md:0-0
Timestamp: 2026-02-25T02:39:33.506Z
Learning: Stateful objects requiring runtime config changes must use reactive `onConfigChange` wiring in `src/index.js` startup
Applied to files:
tests/config-listeners.test.js
📚 Learning: 2026-02-25T02:39:08.568Z
Learnt from: CR
Repo: VolvoxLLC/volvox-bot PR: 0
File: CLAUDE.md:0-0
Timestamp: 2026-02-25T02:39:08.568Z
Learning: After completing code changes, rebuild the Docker container using `docker compose build` but do not start it
Applied to files:
Dockerfile
📚 Learning: 2026-02-25T02:39:33.506Z
Learnt from: CR
Repo: VolvoxLLC/volvox-bot PR: 0
File: AGENTS.md:0-0
Timestamp: 2026-02-25T02:39:33.506Z
Learning: Applies to src/commands/*mod*.js : Moderation commands must follow the pattern: deferReply → validate inputs → sendDmNotification → execute Discord action → createCase → sendModLogEmbed → checkEscalation
Applied to files:
src/index.js
📚 Learning: 2026-02-25T02:39:33.506Z
Learnt from: CR
Repo: VolvoxLLC/volvox-bot PR: 0
File: AGENTS.md:0-0
Timestamp: 2026-02-25T02:39:33.506Z
Learning: Applies to src/commands/*.js : Use `parseDuration()` from `src/utils/duration.js` for duration-based commands (timeout, tempban, slowmode)
Applied to files:
src/index.js
🧬 Code graph analysis (3)
tests/config-listeners.test.js (1)
src/logger.js (1)
addPostgresTransport(252-267)
src/db.js (5)
src/index.js (2)
__filename(50-50)__dirname(51-51)src/logger.js (2)
__dirname(19-19)info(223-225)src/modules/config.js (2)
__dirname(14-14)path(430-430)src/commands/config.js (2)
path(108-108)path(283-283)src/modules/memory.js (1)
msg(72-72)
tests/utils/restartTracker.test.js (1)
src/utils/restartTracker.js (1)
recordRestart(25-41)
🪛 markdownlint-cli2 (0.21.0)
TASK.md
[warning] 3-3: Headings should be surrounded by blank lines
Expected: 1; Actual: 0; Below
(MD022, blanks-around-headings)
[warning] 17-17: Headings should be surrounded by blank lines
Expected: 1; Actual: 0; Below
(MD022, blanks-around-headings)
[warning] 42-42: Headings should be surrounded by blank lines
Expected: 1; Actual: 0; Below
(MD022, blanks-around-headings)
[warning] 47-47: Headings should be surrounded by blank lines
Expected: 1; Actual: 0; Below
(MD022, blanks-around-headings)
[warning] 52-52: Headings should be surrounded by blank lines
Expected: 1; Actual: 0; Below
(MD022, blanks-around-headings)
[warning] 53-53: Fenced code blocks should be surrounded by blank lines
(MD031, blanks-around-fences)
[warning] 57-57: Headings should be surrounded by blank lines
Expected: 1; Actual: 0; Below
(MD022, blanks-around-headings)
[warning] 60-60: Headings should be surrounded by blank lines
Expected: 1; Actual: 0; Below
(MD022, blanks-around-headings)
[warning] 77-77: Headings should be surrounded by blank lines
Expected: 1; Actual: 0; Below
(MD022, blanks-around-headings)
[warning] 84-84: Headings should be surrounded by blank lines
Expected: 1; Actual: 0; Below
(MD022, blanks-around-headings)
[warning] 89-89: Headings should be surrounded by blank lines
Expected: 1; Actual: 0; Below
(MD022, blanks-around-headings)
[warning] 94-94: Headings should be surrounded by blank lines
Expected: 1; Actual: 0; Below
(MD022, blanks-around-headings)
[warning] 95-95: Fenced code blocks should be surrounded by blank lines
(MD031, blanks-around-fences)
[warning] 104-104: Headings should be surrounded by blank lines
Expected: 1; Actual: 0; Below
(MD022, blanks-around-headings)
[warning] 110-110: Headings should be surrounded by blank lines
Expected: 1; Actual: 0; Below
(MD022, blanks-around-headings)
[warning] 114-114: Headings should be surrounded by blank lines
Expected: 1; Actual: 0; Below
(MD022, blanks-around-headings)
[warning] 119-119: Headings should be surrounded by blank lines
Expected: 1; Actual: 0; Below
(MD022, blanks-around-headings)
[warning] 131-131: Headings should be surrounded by blank lines
Expected: 1; Actual: 0; Below
(MD022, blanks-around-headings)
[warning] 137-137: Headings should be surrounded by blank lines
Expected: 1; Actual: 0; Below
(MD022, blanks-around-headings)
🔇 Additional comments (12)
tests/utils/restartTracker.test.js (1)
49-96: LGTM — test assertions correctly reflect the removedensureTableindirection.The assertion at Line 59 now validates the first query is the
INSERT INTO bot_restartsdirectly (index[0]), and Line 80 does the same for the defaults test. This accurately mirrors the simplified production code where no precedingCREATE TABLEcall exists.Dockerfile (1)
23-27: LGTM — migrations directory is correctly copied into the production image.This aligns with the programmatic migration runner in
src/db.jswhich reads migration files at startup.tests/transports/postgres.test.js (1)
1-3: LGTM —initLogsTableimport and test suite correctly removed.The test file now accurately mirrors the production
src/transports/postgres.jssurface area post-refactor.src/index.js (1)
41-41: LGTM —initLogsTableimport cleanly replaced with justpruneOldLogs.The startup sequence in
startup()(Lines 354–370) correctly wires up the transport viaaddPostgresTransportand prunes old logs, with schema creation now handled by the migration runner ininitDb().tests/config-listeners.test.js (1)
268-291: LGTM — error path correctly updated to simulateaddPostgresTransportthrowing.
addPostgresTransportis synchronous (persrc/logger.jslines 251–266), so a synchronous.mockImplementationOnce(() => { throw ... })is the correct mock strategy here.migrations/001_initial-schema.cjs (1)
11-12:⚠️ Potential issue | 🔴 CriticalMigrations must use ESM syntax for node-pg-migrate v8 compatibility.
The migration file
migrations/001_initial-schema.cjsuses CommonJS syntax (exports.up/down) with a.cjsextension. However, node-pg-migrate v8 (ESM-only) expects migrations as.jsfiles with ESM syntax (export const up/down). The.cjsapproach is undocumented and unsupported; v8's CLI and documentation do not mention.cjsmigration files at all, onlyjs,ts, orsql.Rename the file to
migrations/001_initial-schema.jsand convert to ESM:/** `@param` {import('node-pg-migrate').MigrationBuilder} pgm */ export const up = (pgm) => { // migration code }; export const down = (pgm) => { // rollback code };⛔ Skipped due to learnings
Learnt from: CR Repo: VolvoxLLC/volvox-bot PR: 0 File: AGENTS.md:0-0 Timestamp: 2026-02-25T02:39:33.506Z Learning: Applies to **/*.js : Use ESM modules only — use `import`/`export`, never `require()`src/db.js (3)
6-10: LGTM — imports follow ESM andnode:protocol conventions.All new imports use ESM syntax and the
node:prefix for built-in modules. Winston logger is correctly imported for use withinsrc/.
106-125: Migration failure cleanup path is correct and well-structured.The
try/catcharound both the connection test and migration run properly cleans up the pool on any failure (Line 122), preventinggetPool()from returning a stale instance. Thepool = nullreset on Line 123 ensures subsequentinitDb()calls can retry cleanly.
59-73: 🧹 Nitpick | 🔵 TrivialUse the structured
loggerobject pattern instead oflogcallback to differentiate log levels.The
databaseUrlandlogoptions are correct per thenode-pg-migrateprogrammatic API. However, usinglog: (msg) => info(msg)forwards all migration output (including verbose per-statement logs) atinfolevel, which could be noisy in production. Instead, use theloggerobject pattern to separate verbose logs atdebuglevel from important summary messages:log: undefined, logger: { debug: (msg) => debug(msg), info: (msg) => info(msg), warn: (msg) => warn(msg), error: (msg) => error(msg), }This allows per-migration logs at
debuglevel and keeps the summary message on Line 72 atinfolevel.⛔ Skipped due to learnings
Learnt from: CR Repo: VolvoxLLC/volvox-bot PR: 0 File: AGENTS.md:0-0 Timestamp: 2026-02-25T02:39:33.506Z Learning: Applies to src/**/*.js : Always use Winston for logging — import `{ info, warn, error }` from `../logger.js`tests/db.test.js (3)
13-20: Migration mock setup is clean and correctly hoisted.The
vi.hoistedpattern ensures the mock is available before module loading, and the mock correctly mirrors the named export structure ofnode-pg-migrate.
163-168: Migration failure cleanup test is well-structured and mirrors the connection failure test.This correctly verifies that when
runnerrejects, the pool is ended and the error propagates. Good addition that covers the new failure path introduced by the migration step.
120-127: Idempotency test correctly validates single migration run.Asserting
migrationMocks.runneris called exactly once across twoinitDb()calls confirms the pool-caching logic prevents redundant migration attempts.
There was a problem hiding this comment.
Actionable comments posted: 2
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
src/db.js (1)
124-129:⚠️ Potential issue | 🟡 MinorLog context before rethrowing DB init failures.
This catch cleans up and rethrows but does not emit contextual error telemetry for whether the failure happened during connection test vs migration run. Please log before rethrowing.
Suggested patch
} catch (err) { + logError('Database initialization failed', { + stage: 'connection-test-or-migrations', + error: err.message, + }); // Clean up the pool so getPool() doesn't return an unusable instance await pool.end().catch(() => {}); pool = null; throw err; }As per coding guidelines: "Always log errors with context before re-throwing."
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@src/db.js` around lines 124 - 129, The catch block that cleans up the pool and rethrows (around the pool.end() call and variable pool) must log the error with context before rethrowing; capture the thrown err and call the existing telemetry/logger (or console.error) with a message that includes the error and which phase failed (e.g., "connection test" vs "migration run") — implement this by adding a local phase variable (set to "connection test" before the connection check and "migration" before running migrations) and logging `phase` plus `err` right before `await pool.end().catch(() => {}); pool = null; throw err;` so the log shows whether the failure occurred during the connection test or migration step.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@src/db.js`:
- Around line 64-67: runMigrations currently passes a raw connection string to
node-pg-migrate's runner(), which ignores the SSL options used by the app pool;
change runMigrations to pass a pg Client config object (e.g., {
connectionString: databaseUrl, ssl }) so runner() receives the same SSL config
returned by getSslConfig(connectionString), and update the call site to pass
getSslConfig(connectionString) along with connectionString; also enhance the
catch in initialize (or the function that calls runMigrations) to log a
contextual error message (including err.message) before cleaning up the pool and
rethrowing to preserve diagnostics (reference symbols: runMigrations, runner,
getSslConfig, connectionString, pool).
In `@src/utils/restartTracker.js`:
- Around line 21-24: Update the JSDoc for the async functions recordRestart and
getRestarts to include the missing pool parameter and reflect async return
types: add a `@param` {Pool} pool (or appropriate pool type) description to both
docblocks and change the `@returns` to Promise<number|null> for recordRestart and
Promise<Array|whatever the actual type> or
Promise<RestartRow[]>/Promise<Array<Object>> for getRestarts (use the existing
restart row shape), ensuring the descriptions match their runtime behavior and
keep plain-JS JSDoc style.
---
Outside diff comments:
In `@src/db.js`:
- Around line 124-129: The catch block that cleans up the pool and rethrows
(around the pool.end() call and variable pool) must log the error with context
before rethrowing; capture the thrown err and call the existing telemetry/logger
(or console.error) with a message that includes the error and which phase failed
(e.g., "connection test" vs "migration run") — implement this by adding a local
phase variable (set to "connection test" before the connection check and
"migration" before running migrations) and logging `phase` plus `err` right
before `await pool.end().catch(() => {}); pool = null; throw err;` so the log
shows whether the failure occurred during the connection test or migration step.
ℹ️ Review info
Configuration used: Organization UI
Review profile: ASSERTIVE
Plan: Pro
📒 Files selected for processing (4)
src/config-listeners.jssrc/db.jssrc/index.jssrc/utils/restartTracker.js
📜 Review details
🧰 Additional context used
📓 Path-based instructions (2)
**/*.js
📄 CodeRabbit inference engine (AGENTS.md)
**/*.js: Use ESM modules only — useimport/export, neverrequire()
Usenode:protocol for Node.js builtins (e.g.import { readFileSync } from 'node:fs')
Always use semicolons
Use single quotes for strings
Use 2-space indentation
No TypeScript — use plain JavaScript with JSDoc comments for documentation
Files:
src/config-listeners.jssrc/utils/restartTracker.jssrc/index.jssrc/db.js
src/**/*.js
📄 CodeRabbit inference engine (AGENTS.md)
src/**/*.js: Always use Winston for logging — import{ info, warn, error }from../logger.js
NEVER useconsole.log,console.warn,console.error, or anyconsole.*method in src/ files
Pass structured metadata to Winston logging calls (e.g.info('Message processed', { userId, channelId }))
Use custom error classes fromsrc/utils/errors.jsfor error handling
Always log errors with context before re-throwing
UsegetConfig(guildId?)fromsrc/modules/config.jsto read config
UsesetConfigValue(path, value, guildId?)fromsrc/modules/config.jsto update config at runtime
UsesplitMessage()utility for messages exceeding Discord's 2000-character limit
UsesafeSend()wrapper for outgoing Discord messages to sanitize mentions and enforce allowedMentions
Files:
src/config-listeners.jssrc/utils/restartTracker.jssrc/index.jssrc/db.js
🧠 Learnings (3)
📚 Learning: 2026-02-25T02:39:33.506Z
Learnt from: CR
Repo: VolvoxLLC/volvox-bot PR: 0
File: AGENTS.md:0-0
Timestamp: 2026-02-25T02:39:33.506Z
Learning: Applies to src/**/*.js : Always use Winston for logging — import `{ info, warn, error }` from `../logger.js`
Applied to files:
src/config-listeners.js
📚 Learning: 2026-02-25T02:39:33.506Z
Learnt from: CR
Repo: VolvoxLLC/volvox-bot PR: 0
File: AGENTS.md:0-0
Timestamp: 2026-02-25T02:39:33.506Z
Learning: Applies to src/commands/*mod*.js : Moderation commands must follow the pattern: deferReply → validate inputs → sendDmNotification → execute Discord action → createCase → sendModLogEmbed → checkEscalation
Applied to files:
src/index.js
📚 Learning: 2026-02-25T02:39:33.506Z
Learnt from: CR
Repo: VolvoxLLC/volvox-bot PR: 0
File: AGENTS.md:0-0
Timestamp: 2026-02-25T02:39:33.506Z
Learning: Applies to src/commands/*.js : Use `parseDuration()` from `src/utils/duration.js` for duration-based commands (timeout, tempban, slowmode)
Applied to files:
src/index.js
🧬 Code graph analysis (1)
src/db.js (3)
src/index.js (2)
__filename(50-50)__dirname(51-51)src/logger.js (2)
__dirname(19-19)info(223-225)src/modules/config.js (1)
__dirname(14-14)
🔇 Additional comments (2)
src/config-listeners.js (1)
33-38: JSDoc improvement is clear and accurate.This doc block correctly captures the enable/recreate/disable behavior and the
changePathlogging context.src/index.js (1)
41-41: Startup flow cleanup looks correct for migration-first schema management.Dropping runtime table-init coupling and documenting prune-on-startup behavior is consistent with the new migration-driven DB initialization.
Also applies to: 316-319
Captures all 8 tables and indexes from existing inline DDL: config, conversations, mod_cases, mod_scheduled_actions, memory_optouts, ai_usage, logs, bot_restarts. Uses IF NOT EXISTS throughout for idempotency on existing DBs. Down migration drops in reverse FK order.
- src/db.js: replaced all CREATE TABLE/INDEX/ALTER TABLE with programmatic node-pg-migrate runner call - src/transports/postgres.js: removed initLogsTable() - src/utils/restartTracker.js: removed ensureTable() and self-healing 42P01 catch block - src/index.js: removed initLogsTable() call - src/config-listeners.js: removed initLogsTable() call
- package.json: migrate, migrate:up, migrate:down, migrate:create - Dockerfile: copy migrations/ directory into image
- db.test.js: mock node-pg-migrate runner, test migration runs on init, test migration failure cleanup - transports/postgres.test.js: remove initLogsTable tests (function removed) - utils/restartTracker.test.js: remove ensureTable and self-healing tests - index.test.js: remove initLogsTable mock and assertions - config-listeners.test.js: remove initLogsTable references
…dex for nullable user_id
…1.0 for node-pg-migrate v8
BillChirico
force-pushed
the
feat/database-migrations
branch
from
58ff5dc to
e62e11b
Compare
1 participant
Summary
Replaces all inline DDL scattered across
db.js,postgres.js, andrestartTracker.jswith proper versioned migrations vianode-pg-migrate.Closes #78
Changes
node-pg-migrate+ initial migration (001_initial-schema.cjs) capturing all 8 tables and all indexessrc/db.js: Removed ~200 lines of inline DDL frominitDb(). Now: connect → test → run pending migrations → return poolsrc/transports/postgres.js: RemovedinitLogsTable()— table created by migrationsrc/utils/restartTracker.js: RemovedensureTable()and self-healing DDL — table created by migrationsrc/index.js+src/config-listeners.js: RemovedinitLogsTableimport/callsmigrate,migrate:up,migrate:down,migrate:createto package.jsonKey Decisions
.cjsmigration files — node-pg-migrate usesexports.up/downwhich conflicts with ESM. CJS files work perfectly viarequire()initDb(). CLI scripts available for manual useCREATE TABLE IF NOT EXISTSthroughout, safe on existing production databasesFiles Changed
migrations/001_initial-schema.cjssrc/db.jssrc/transports/postgres.jsinitLogsTable()src/utils/restartTracker.jsensureTable()+ self-healingsrc/index.jsinitLogsTableimport/callsrc/config-listeners.jsinitLogsTableimport/callpackage.jsonDockerfileTesting
Breaking Changes
None — existing databases migrate seamlessly. The initial migration is idempotent.