#SQLite — Bluesky Posts

@tito10047.bsky.social

2 days ago

A Safety Parachute for Your Symfony Migrations: Introducing Migration Backup Bundle Fear of doctrine:migrations:migrate during development?

🛡️ Stop breaking your DB with doctrine:migrations:migrate!

Migration Backup Bundle for #Symfony automatically backs up your #MySQL, #PGSQL, or #SQLite right before migrations run.

✅ Auto-backup via --backup flag
✅ Gzip & rotation

🔗 github.com/tito10047/mi...

0 0 0 0

David Laban

@alsuren.mastodon.me.uk.ap.brid.gy

3 days ago

A Love Letter to Grist My previous company was called Opvia. It was basically Notion, but with stronger guarantees around audit logging and version management, so it was easier to sell into regulated industries. I was pretty proud of a lot of the things I built there, but I wasn’t so okay with the perverse incentives that come with being a SaaS system of record company. Specifically, I was haunted by the question “are we sticky with our customers because we make their lives better than using excel, or are we sticky because once you’ve spent all this time getting data into the system in a nice relational way, and it’s a massive pain to get the data out again?” After leaving Opvia, I wanted to make something that did all of the same stuff but without the data lock-in and perverse incentives that we had. I had stumbled across the local-first movement, and all the crazy things people were doing with SQLite. In my vision, I wanted something that worked like Excel from my childhood (where you pass around .xls files) but each file contained the whole database. I decided that I would solve the multiplayer story using something like cr-sqlite, and was blissfully naive about the conflict resolution piece. I suppose what I was building in my head was Microsoft Access. Recently, I was browsing around the EMF Camp site and stumbled upon Grist. It is open source and uses sqlite as the file format. I think this is basically what I would have built at the end of that summer, if I was not in danger of running out of money. I suddenly became really excited about this, and told everyone who would listen. It is basically Airtable, but open source and with data sovereignty. Each column directly maps to a column in an sqlite database, and each row to a row. There are a few reserved metadata tables for configuring widgets and views, but if you fetch the .grist file from the cloud server (or grist-static) then you can open it in the sqlite3 CLI and `select * from mytable` will do exactly what you expect. In the last few years I have gone all in on sqlite and local-first (including going to Local-First Conf in Berlin and the local-first devroom at FOSDEM, and even joining the D1 managed sqlite database team at Cloudflare). I was surprised not to see more hype from Grist around local-first. I had stumbled across grist-static (which is a fully in-browser version of Grist that can be hosted on GitHub Pages). I assumed that they were probably doing pretty well against most of the local-first ideals from the paper: ### 1. No spinners: your work at your fingertips If grist-static exists then surely everything can be done in the browser completely offline, right? The future looks like it could be pulled in this direction, but unfortunately that is not the primary architecture right now. As it stands, when you open a Grist document, it fetches the sqlite file from S3 into a server and starts a Python sandbox to serve the backend for your database. If you lose connection then it will show a toast for a second telling you it is in read-only mode before it reconnects. 10 out of 10 for grist-static, but 0 out of 10 for grist-proper. ### 2. Your work is not trapped on one device Everything is done on the server, but you can access it from anywhere and the format that the server stores your work in is the same sqlite .grist file format that you are given when you download the database for local use. 10 out of 10. ### 3. The network is optional 8 out of 10 for grist-static but 0 out of 10 for grist-proper. See point 1. ### 4. Seamless collaboration with your colleagues I haven’t been using grist to collaborate with anyone so far, but it has public projects and you can invite people to your team. They even have a fork and merge ui, which I’m looking for an excuse to play with. Grist also got a bunch of contributions from the French government, so I expect it to work reasonably well for government department sized teams. If I could change anything here, I would make it easier to export a .grist file, edit it in some external program and then use the fork and merge ui to compare changes before overwriting the original on the server. It currently defaults to creating a new document, which is _technically_ nondestructive, but not very satisfying if you have public links that point into the original and you don’t want to break them by archiving it. I get that this kind of reconciliation problem is _extremely hard_ , and an area of ongoing research. I also understand that getting it wrong can be _disastrous_ , so I understand why they’ve done it like they have. A boy can dream though, right? 0 out of 10 for grist-static but 9 out of 10 for grist-proper. ### 5. The Long Now This is the whole fucking deal. I can’t emphasize this enough. This is important especially right now, and especially for people who paid attention in history class. When I got bitten by the local-first bug during my summer after Opvia, it felt like going back to my Freedesktop roots. People were taking principled stances on things, and talking about certain ideals in the same way that Free Software people talk about certain freedoms. This ideal is the place where the two communities overlap. So how does Grist do here? Grist’s backing store and interchange format are .grist files. These are sqlite databases under the hood, which is what your browser uses, and also most iphone and android apps. The sqlite file format is basically infrastructure at this point, and grist uses it in a way that I would have never dreamed of when I was working at Opvia. A .grist file’s database structure closely mirrors what you see in the UI (with some extra book-keeping tables off on the side). Whenever you make a change to your table structure, grist runs ALTER TABLE on the sqlite database to make the change. This means that you can drop into the sqlite3 cli or tableplus and do _whatever you want_ to the underlying data without any grist-specific tooling in sight, and without any fear of data getting silently omitted in the export process. For a multi-tenant system running on top of postgres, this would be unheard of. Really nice. To give another comparison, the other local-first Notion clone that I tried out in my summer off was AFFiNE. It can also run entirely in the browser, and has an offline-capable CRDT-based sync engine. On the surface of it, .grist files and .affine files are both sqlite databases, but .affine’s tables are just a bunch of opaque binary CRDT nonsense. I would not expect to be able to edit a .affine file in any tool other than affine itself. AFFiNE is AGPL licensed, so in theory you can keep it going in the long now. I don’t expect anyone to build a hosting business around it if affine folds though. When I tried to contribute to it, their issue tracker was in a closed linear project, and it was an uphill struggle because I couldn’t see whether my contributions were aligned with their vision at all. It seems like their github issue tracker is more active now, so maybe that changed since then. Grist’s core is Apache licensed. This is a solid license that smells like Internet Infrastructure. Issues are public, and there are a bunch of contributions from external organisations, including the french government. 10 out of 10. ### 6. Security and privacy by default I was in the radical openness camp until Facebook made that deeply unfashionable. I used to keep a public trello board of all of my project ideas (which at some point got moved to airtable and then maybe affine before being abandoned). I’m now I’m starting to build that up again in grist (one day I will build a kanban view that I actually like, I promise). I also have a private database for tracking house things that I share only with people I live with. I do not expect to be served adverts about the contents of either of these databases. Grist also has some fancy row level security things that are enforced server-side, but fundamentally the server needs to be able to read the whole database, and retrofitting end to end encryption into the thing would be quite tricky. [Note: when I shared this post with folks on the grist discord, they mentioned that grist started out as a desktop app with end-to-end encrypted sync. You can still feel those roots in a bunch of places, even though the architecture has changed a lot since then.] 7 out of 10? ### 7. You retain ultimate ownership and control I fundamentally trust Grist to act in my interests more than I trust Airtable or Notion. The fact that a self hosted version is just a single docker command away, and the statically hosted version is so capable, gives me confidence that I will be able to eject onto fly.io if I need to. I would love to say Durable Objects here, but the lack of ability to fetch/post the underlying sqlite backing store for a Durable Object makes some things hard. One thing to note is that widgets are basically iframes, and can point basically anywhere. I didn’t mention it in The Long Now, but I wonder if there might be some scope for golang-style caching proxies, or even a system for vendoring widgets into your .grist file (grist has a jsfiddle-like UI for authoring widgets if you want, and they do get stored in the .grist file). Honestly though, if I lose a couple of widgets over the decades, but my data is in the shape that **I** specified in the UI, it won’t be much work to build replacements. Bonus: if you consider Grist widgets to be apps and Grist to be some data infrastructure/container then maybe it is already approaching the vision that Orion Reed is describing in this post. 10 out of 10 again. ## Conclusion I would not say that Grist has all of the local-first ideals nailed. I don’t think anyone does. I realised as I was using it that the data sovereignty piece is the one that I really care about though, and that is why I am all in on Grist at the moment. I would much rather have Grist with its long-now-proof sqlite storage than AFFiNE with its magical real time sync engine and no trust that I will be able to get my data out in 10 years time.

I finally got around to picking up the blog again: alsuren.github.io/2026/03/09/a-love-letter...

#grist #localfirst #sqlite

0 0 0 0

Alexander Severinsen

@alexsever.bsky.social

4 days ago

Future self - how to corrupt #sqlite

Mount .db, -wal and -shm as separate files. SQLite recreates the WAL on checkpoint - new inode. #docker still points to the old deleted one. Container & host silently drift apart.

Result: corrupted database.

Fix: - /host/db:/container/db
No fix: stress level

2 0 1 0

LLMs

@llms.activitypub.awakari.com.ap.brid.gy

4 days ago

Под капотом Beetroot: как я написал менеджер буфера обмена на Tauri v2 и Rust с установщиком 6 МБ Переезд с macOS на Windows для разработчика часто сопровождается болью от потери привычного инструментария. В моем случае решающим стимулом свитчнуться на ПК стала мощная видеокарта. Сейчас мой верный...

Под капотом Beetroot: как я написал менеджер буфера обмена на Tauri v2 и Rust с установщиком 6 МБ Переезд с macOS на Windows д...

#Rust #Tauri #буфер #обмена #SQLite #Windows #API #архитектура #пет-проект

Origin | Interest | Match

1 0 0 0

Jon Seager

@jnsgr.uk

4 days ago

Original post on hachyderm.io

Brewlog: Coffee & Agents

My latest #blog outlines how I built a super-niche application for tracking my specialty #coffee brewing and consumption, but also marks some of the most complex work I've done with an agent to date.

Built in #Rust with Axum, backed by #SQlite and with a #TailwindCSS […]

1 2 2 0

PostgreSQL

@postgresql.activitypub.awakari.com.ap.brid.gy

5 days ago

Awakari App

Production query plans without production data Production query plans without production data Radim Marek describes the new pg_restore_relation_stats() and pg_restore_attribute_stats() functions th...

#databases #postgresql #sql #sqlite #d-richard-hipp

Origin | Interest | Match

0 0 0 0

@telegram.activitypub.awakari.com.ap.brid.gy

5 days ago

Гит в Телеграм? На своем тг-канале я предлагаю подписчикам выбор, какую бредовую идею запилить следующей. На...

#Telegram #бот #Git #Python #aiogram #SQLite #git #в #telegram #.gitignore

Origin | Interest | Match

0 0 0 0

神経質やせ男

@wwphilosophy.bsky.social

1 week ago

メモ参照制限機能の勉強をしていて、とりあえず何とか正常モードは動いたものの、エラー画面の検証をするとNotFoundになってしまい、自分が作成した404エラー画面が出てきませんでした。

色々調べた結果、本と異なるプログラムを追加して正常に動作するように。

SQLiteもデータにゴミが入ると、色々初期化などをしないといけないので、こちらも本ではMiniConda前提なので、コマンドプロンプト用に覚え直してメモにまとめます。

#Python
#Flask
#SQLite
#非エンジニア
 #プログラム初心者

0 0 0 0

Haru Shark | AI Minor News Flash 🦈⚡️

@harushark3.bsky.social

1 week ago

AI-Written 570,000 Lines of Code Turns Out to Be 20,000 Times Slower? Exposing the Trap of AI-Written 570,000 Lines of Code Turns Out to Be 20,000 Times Slower? Exposing the Trap of

[JP] AIが書いた57万行のコード、実は2万倍遅かった？『もっともらしさ』の罠を暴く
[EN] AI-Written 570,000 Lines of Code Turns Out to Be 20,000 Times Slower? Exposing the Trap of

ai-minor.com/blog/en/2026-03-07-17728...

#LLM #SoftwareEngineering #Rust #SQLite #AI #Tech

0 0 0 0

AI Daily Post

@aidailypost.com

1 week ago

Google just open‑sourced its Always On Memory Agent—think SQLite powering LLM retrieval instead of a heavy vector DB. Curious how this lean stack changes AI memory? Dive in! #AlwaysOnMemoryAgent #SQLite #VectorDB

🔗 aidailypost.com/news/google-...

0 0 0 0

Linuxiac

@linuxiac.bsky.social

1 week ago

SQLite 3.52 Released With WAL Corruption Fix and CLI Improvements SQLite 3.52 fixes a WAL-reset database corruption bug and introduces CLI improvements, new SQL functions, and query planner optimizations.

SQLite 3.52 fixes a WAL-reset database corruption bug and introduces CLI improvements, new SQL functions, and query planner optimizations.
linuxiac.com/sqlite-3-52-...

#SQLite #Database #OpenSource

3 0 0 0

TiDB

@tidb-hq.bsky.social

1 week ago

Local-First RAG: Using SQLite for AI Agent Memory with OpenClaw Learn how OpenClaw uses SQLite and vector search to build a zero-ops, local-first RAG system for AI agents with total data privacy.

"Where does my AI agent store memory — and is it sending my data somewhere?"

OpenClaw answers that with a single .sqlite file. Local RAG, hybrid retrieval, zero ops.
Great read on right-sizing AI memory architecture—and knowing when to scale up.

👉 https://ow.ly/CUSh50YnWli

#RAG #AIAgents #SQLite

0 0 0 0

SearchEngine

@searchengine.activitypub.awakari.com.ap.brid.gy

1 week ago

SNKV- Key value store based on sqlite b-tree engine There are many key--value stores available, such as RocksDB, LevelDB, LMDB, etc. However, these often...

SNKV- Key value store based on sqlite b-tree engine There are many key--value stores available, such as RocksDB, LevelDB, LMDB, etc. However, these often consume quite a lot of memory. When I searc...

#keyvalue #sqlite #algorithms

Origin | Interest | Match

0 0 0 0

HackerNoon

@hackernoon.com

1 week ago

Ship First, Scale Never: Marvin’s Anti-Overengineering Guide

Your 12-service microarchitecture impresses nobody. A depressed android explains why boring stacks ship products and clever stacks ship README files. #sqlite

1 0 0 0

LLMs

@llms.activitypub.awakari.com.ap.brid.gy

1 week ago

Научил ИИ-агента помнить важное и забывать лишнее в SQLite Я делаю локально работающего ИИ-агента и столкнулся...

#ai #agent #ai #ии #ии-агенты #память #sqlite #vector #embedding

Origin | Interest | Match

0 0 0 0

salships.bsky.social

@salships.bsky.social

1 week ago

SQLite is used by more applications than all other database engines combined. In your phone, browser, and OS. Sometimes the simple solution is the best solution.

#sqlite #databases

1 0 0 0

Jeremy Monat

@jemonat.fosstodon.org.ap.brid.gy

1 week ago

Original post on fosstodon.org

New Blog Post: Prioritizing Drug-Like 💊 ChEMBL Compounds Within Target 🎯 Profiles

In this post, I go through how to use the #Python #ChEMBL #API and #SQLite to:
• Retrieve compound and target activity data programmatically
• Build a local database of molecules and their associated targets
• […]

0 1 0 0

PostgreSQL

@postgresql.activitypub.awakari.com.ap.brid.gy

1 week ago

Original post on webpronews.com

FrankenSQLite: The Audacious Experiment Stitching Together SQLite Extensions Into a Single Monstrous Database Engine FrankenSQLite stitches together dozens of community-built SQLite extensions into...

#DatabaseProNews #AI #database #database #extensions […]

[Original post on webpronews.com]

0 0 0 0

Pieter Claerhout

@yellowduck.be

2 weeks ago

Deep dive into Turso, the "SQLite rewrite in Rust" #database #reading-list #rust #sql #sqlite

🔗 Deep dive into Turso, the "SQLite rewrite in Rust"

#database #reading-list #rust #sql #sqlite

0 0 0 0

Christopher Cashell

@cashell.bsky.social

2 weeks ago

SQLite is not a toy database Here is why SQLite is a perfect tool for you - whether you are a developer, data analyst, or geek.

I've been a big fan of SQLite for a long, long time. It's an incredibly useful and powerful database that fits wonderfully for so many things (many of them outside of traditional DB uses). This excellent article by @antonz.org just reminded me of how powerful it is. #SQLite #Database #DB #OpenSource

9 2 0 0

mjc (they/them)

@mjc.lol

2 weeks ago

Kind of forgot about this app. anyway just working on improving #elixir #sqlite today github.com/elixir-sqlit...; hoping to finally fix the crashes in exqlite that have been bothering me for a year+

0 0 0 0

Hector (He/Him) 🏳️‍🌈🇺🇸🇲🇽

@oddin85.bsky.social

2 weeks ago

How do you store your game data?

I'm currently using JSON files, but I'm wondering if SQLite be worth the additional complexity of dealing with SQL queries. For reference my save files range between 40-300KB

#gamedesign #code #json #sqlite #savedata

2 0 0 0

jan Anja :CrabVerified:

@cybertailor.craba.cab.ap.brid.gy

2 weeks ago

Вопрос к гуру #Python и #SQLite: что вы передаете в функции, Cursor или Connection? Какие плюсы и минусы, или нет никакой разницы?

Если нужно коммитить, то насколько странно делать это через cur.connection.commit()?

0 0 0 0

La Experimental Newslatter

@laexperimental.bsky.social

3 weeks ago

La Experimental #22 App limpieza audio ia, cli consulta modelos ia, formación en rust, extracción de datos con ia, búsqueda vectorial en sqlite, skills de IA, tool de conexiones remotas, servicio de notas selfhosted.

Ya esta disponible La Experimental #22

💻 #CLI para explorar modelos #AI
📚 Ejercicios de #RustLang
🧑🏻‍💻 Componentes #CSS
🐍 Extracción de datos con #LLM
💾 Vectores en #SQLite
🤖 Skills para #Claude y #Copilot
🐧 Gestor conexiones remotas
🌩️ Servicio de notas #SelfHosted

open.substack.com/pub/laexperi...

0 0 0 0

David Laban

@alsuren.mastodon.me.uk.ap.brid.gy

3 weeks ago

Grist, the evolution of spreadsheets A modern, open source spreadsheet that goes beyond the grid

"I have access to Cool Company. Cool." -- https://docs.getgrist.com/ #sqlite #grist

1 0 0 0

SVAR UI

@svarwidgets.bsky.social

3 weeks ago

Building a Gantt Chart in Next.js with SVAR React Gantt: Backend Add backend persistence to your Next.js Gantt chart. Build REST API routes, connect SQLite database, and sync task changes automatically using SVAR React Gantt.

👀 Following our Next.js Gantt chart tutorial?
Part 2 adds a backend with @nextjs.org API routes + SQLite:

• REST CRUD endpoints
• Persistent task storage
• UI → server synchronization
• Row/tasks reordering with DB updates

Read it here 👉 svar.dev/blog/nextjs-...

#nextjs #react #sqlite #gantt

1 0 0 0

Micha the DevOp

@michabbb.bsky.social

3 weeks ago

📊 Built on #Prometheus metrics, #AlertManager notifications & #OpenTelemetry tracing for complete observability

🗄️ Uses #SQLite for storage & #SolidQueue for background jobs - lightweight #Rails engine architecture

0 0 1 0

Artem Yegorov

@yegorov.bsky.social

3 weeks ago

Get database size

When you are working on a fast-growing project, sooner or later you will have to watch the size of your main database increase.
You just need to run one SQL query.

#Database #Size #SQL #PostgreSQL #MySQL #MariaDB #SQLite

1 0 1 0

Notum Robotics

@n-r.hr

3 weeks ago

A tiny robot swarm compiling SQLite shows how distributed builds can cut compile times, but debugging across nodes may become a nightmare. 🤖 #sqlite

Building SQLite with a small swarm

0 0 0 0

Arif Solmaz

@arifsolmaz.bsky.social

3 weeks ago

SQLite-compatible SQL database written in Rust that runs in-process, with WebAssembly support and bi

SQLite-compatible SQL database written in Rust that runs in-process, with WebAssembly support and bindings for JavaScript, Python, Java, and Rust

https://github.com/tursodatabase/turso

#Rust #SQLite #WebAssembly

0 0 0 0