This solves a major problem that I built an npm package called "pgstrap"[1] for. It generates a "database structure" directory so that my database schema is available to LLMs (it also makes code review easier because you can see the changes to various tables). So I have a SQL file for each table in my database, neatly organized into directories for each schema. Rails has a similar idea with schema.rb
I'm not sure whether or not it's better to have your editor database-aware or to have your codebase have appropriate context committed. On one hand, less generated code/artifacts make for a cleaner codebase. On the other hand, not everyone uses VC Code or will know how to use this integration. Database browser GUIs have never really had a single winner. That said, VS Code does have enough dominance to potentially make themselves "the standard way to view a database in development"
I’m confused. Isn’t including the canonical state of the database schema in version control along with all the migrations that brought it to that point a completely standard part of every web framework?
It often is, but the schema might be written against a language-specific ORM.
That code might in turn have plugins or feature flags that mean you don't know the concrete sql schema until runtime.
Same for seed data and migrations.
So it depends on the use-case how useful this format is for tooling and discovery vs an actual connection to the database.
The _canonical_ state is what is in the production database(s). What's in version control is hopefully able to recreate it, with the obvious caveat of being highly unlikely to be able to repopulate the data.
For development, the actual data itself is not so important, but the features of the data are extremely important. Such as which fields have higher or lower cardinality, which fields are accessed often and which are barely touched.
Often times, the indexes will reflect this. Often times, not.
I'm more confused why the version control of the thing using the database is including the entire schema of the database in it's repository
The database schema of an app is tightly coupled enough to essentially be code, and migrations are also probably checked in. This lets you see how the db schema changes over time -- likely along with the queries using the schema.
This is only true when your app is the only thing that is using the database. In enterprise environments, databases are frequently shared between many different apps.
Sure, and in those cases you'd typically have a dedicated repository storing said schema and migrations.
It's important to manage your schema in code for various reasons from change control to standing up development databases, etc.
It makes it trivial to have pretty diffs of the net result of migrations.
That "works" for about as long as you have <10 employees and <3 customers or so. After that the railsapp doesn't get to be the sole owner of the db.
Providing access to other services is what APIs are for.
Jeff Bezos famously said[1] that anyone who does otherwise should be fired, and I agree.
SQL is also an API.
the reason he did that was for business reasons. He wanted to be able to expose any part of the stack as a public service to external customers, and vice versa, to let his internal service compete against the publicly available ones.
But this is only valid when you're trying to build AWS. Not everyone does that.
Relational databases have extensive permission systems for a reason.
It worked fine for GitLab when it had 2000+ employees and god knows how many customers. The same applies to many other large Rails shops.
Do you have multiple separate apps that can change a shared DB schema?
How do you keep that all in sync across your apps?
So usually one is the “main” but, personally speaking, I’d just say “don’t do it,” for the obvious issues I suspect you’re aware of ;-)
By “don’t do it,” I mean having multiple apps talk to one DB schema.
"don't do it" is the right answer. Others have pointed it out as well, many large SaaS companies I worked with, have had apps owning their databases. Anyone else needs anything - use APIs (and ETL if you need everything)
In situations like these, the database admins are the ones responsible for the schema; the apps are mere users.
Wow, this is precisely how ClickHouse stores table metadata! A set of .sql files in the directories, corresponding to databases.
That seems like a really pragmatic tool, thanks for sharing it!
I'm curious, do you output triggers, store procedures, and such? Many tools seem to stop after you've defined tables, columns, and indices, but I'd love some better tooling to make use of the rest of the DB's features.
Yep! It basically runs pg_dump and categorizes all of the output into different files so it should be comprehensive. I think there's `functions/function_name.sql`, `misc.sql`, `triggers.sql` etc.
You built it but you don’t know what it outputs?
I just use a MCP server (with copilot or cline) that has a read only login to my database.
Out of interest.. does the resultant data get used by the LLM or just generating SQL, executing and returning separately?
PM on the project here - The results from the query are generally not used by the LLM. In agent mode though, during query planning, the agent may retrieve sample of the data to improve precision of the queries. For example, getting distinct values from dimensional table to resolve filter condition from natural language statement.
Thanks. I worry about these kind of tools connecting to production databases.. Especially considering how easy it is to switch out LLM endpoints, where that data is going, how it is retained, the context etc becomes a bit of a privacy nightmare..
Absolutely valid concern. Our extension connects to LLMs through Github Copilot. Github Copilot is Microsoft product and offers variety of enterprise plans, which enables your IT to approve what can be used for what kind of data. This gives you a clear path towards compliance with your enterprise requirements.
Makes sense. Appreciate the responses. Honestly though, as a person outside the US, I'm removing my dependence on US company IT tools and infrastructure, GitHub, VSCode, AWS etc, enterprise or otherwise.. Congrats on the project though.
Which is strictly worse than just giving the LLM access to the source of truth for the database.
You're adding a round trip to the database and the LLM and inserting a tool call in the conversation before it even starts generating any code.
And the reference Postgres MCP implementation doesn't include Postgres types or materialized views, and is one of the most widely used packages: Zed.dev's MCP server for example, is seemingly just a port of it and has the same problem.
I don't see how a round trip of <500ms, which is equivalent to maybe 50 tokens, is worse than including many thousands more extra tokens in the prompt, just in case they might be useful. Not to mention the context fatigue.
If designed well - by suspending generation in memory and inserting a <function_result>, without restarting generation and fetching cache from disk - the round trip/tool call is better (costs the equivalent of 50 tokens for waiting + function_result tokens).
You're dealing with the full TTFT x2 + the tokens all the prompts of all your MCPs before you even get to that round trip to the DB.
And you don't have to wonder about "if designed well": the reference implementation that's getting 20k downloads a week and getting embedded in downstream editors is is not designed well and will make the round trip every time and still not give the LLM the full information of the table.
Most MCP implementations are crappy half-assed implementations in similar fashion because everyone was rushing to post how they added <insert DB/API/Data Source> to MCP.
And if you're worried about "context fatigue" (you mean LLMs getting distracted by relevant information...), you should 100% prefer a well known schema format to N MCP prompt definitions with tool usage instructions that weren't even necessarily tuned for the LLM in question.
LLMs are much more easily derailed by the addition of extra tools and having to reason about when to call them and the results of calling them, than they are a prompt caching friendly block of tokens with easy to follow meaning.
What source of truth? If you have access to the database then you have the actual truth right there.
MCP also gives the LLM access to your example data, which can add clarity beyond what your schema alone provides.
The schema in the db should be the source of truth and an MCP server like that is the most flexible, can work with any ORM set up
Oh woah this looks great.
Quite amazing they put the effort into this for Postgres instead of SQL Server. The demand must be a lot higher.
There is already a Microsoft SQL server extension for VS Code and this looks to effectively be a clone of it. After giving this a quick spin, it looks and feels the same as the SQL server extension, with the same menus, dialogs, etc. The SQL server extension I believe is what formed the basis of the now-deprecated Azure Data Studio (which was a VS code fork).
See here for the SQL server extension: https://marketplace.visualstudio.com/items?itemName=ms-mssql...
Since most databases expose similar schema views it shouldn't be too complicated. Feels like JetBrains has been doing this for a long time
I'm a developer advocate at Microsoft, and from my perspective, both teams have been putting in a bunch of effort improving their extensions. I participated in usability studies with both the teams behind the SQL Server extension and new PostgreSQL extension, and then once they were ready, I participated in bug bashes.
Both teams seem to very much want developers to enjoy their tools, so please do send them feedback on what you need out of the tools.
Follow Carlos Robles if you want SQL server extension news: https://www.linkedin.com/in/croblesm/
Follow Joshua Johnson for PostgreSQL server extension news: https://www.linkedin.com/in/johnsonjoshuae/
The problem customers have with these Shiny New Things that Microsoft keeps trying get us to switch to is that they drop features and entire product suites on the floor, even those that aren't officially deprecated and have no equivalent replacements. It's common to see SSRS, SSIS, SSAS in multidimensional mode, etc... simply forgotten about like they no longer exist.
Not to mention that SQL "SDK-style" projects only work properly in VS Code, so Visual Studio users are left out in the cold having to deal with an incomplete, half-baked solution.
I totally agree and I beg you to consider this feedback if you are reading @pamelafox.
The biggest problem with the usability of Microsoft products today is short-sighted thinking. New features, platforms, frameworks etc are launched and then forgotten about just a few years later with no effort to tie into the groundwork of what came before.
You might think this is only a problem for old customers who are already accustomed to the old technologies, but that's not true: it burdens new customers too. There's a few reasons for this that I can think of.
1) It's hard for new customers to know what technologies they should be reaching for in what situations when there's so many different choices.
2) It's hard to find the right documentation for the technology you've picked because you have to browse through a ton of out-of-date documentation that wrongly refers to the deprecated technologies and it's not clear what the current recommendations are.
3) The new stuff is often built without consideration for the ways of thinking that the underlying platform was built with. Thus, you end up with weird idiosyncrasies as you move from one technology to another, which make it hard to learn and hard to use.
4) When you replace the old technologies you lose the benefit of community knowledge on platforms like Stack Overflow, you lose the ability to look at existing open-source projects for guidance, etc. You are basically going into uncharted territory where there are no clearly established patterns in the wild.
So, even new users coming on to your platform suffer from these deficits. That's not to say I don't appreciate all the work on these new powerful technologies like VS Code and .NET Platform and so on, but I think a more long-sighted vision for these products would go a long way. And it's not just a matter of looking forward, since you never know what's going to happen in the future with a product as organizational priorities change. It's also a matter of looking backwards at what came before, at what groundwork was laid by previous efforts, and how it can be best taken advantage of and re-used for future efforts. That is the biggest missing piece at Microsoft today in my opinion.
I speculate that it's because the MSSQL tools have been maintained as part of Azure Data Studio, and were in better shape.
ADS is being sunset, and I was surprised when trying to install the Postgres extension on VS Code to find that it had its last meaningful contribution 6 years ago [0]. It couldn't work on newer VS Code versions.
I use ADS with both Postgres and MSSQL, prior to this announcement, I kept using ADS because there was nothing to migrate to.
[0] https://github.com/microsoft/vscode-postgresql/commits/maste...
They built the postgres plugin in a way that nobody could usefully contribute to unless they worked at msft - like the rest of ADS the level of control they tried to maintain meant nobody wanted to work on it.
That is seemingly true of a bunch of tools. I am using an official Microsoft Python library - the repo is public on GitHub, but all of the CI or other backend integration is behind the Microsoft curtain, so it is impossible for the public to actually participate. The cherry on top is that the team that used to support the tool was impacted, so now nobody can maintain the thing.
I am curious: Which library?
They already have the SQL Server Management Studio[1], which seems to cover similar ground?
I'm assuming they might want to move SSMS to VSCode in time, so trying it out by covering new ground, PostgreSQL, makes sense to me.
[1]: https://learn.microsoft.com/en-us/ssms/sql-server-management...
But it's only for Windows
In my experience, MSSQL shops are far more likely to be using Windows already than PGSQL shops, so that's just yet another reason for why PGSQL was a good first choice for a VSCode plugin IMHO.
No, it has a different purpose; VS Code extension and the former ADS are targeted for development, while SSMS is for server and database administration. I am a heavy user of SSMS and can do everything I ever need there, I don't use the VS Code extension for MS SQL even if I have it installed and I use VS Code quite a lot. This is because I am also acting as a backup and supervisor for our DBA team, so I am involved in DBA work.
SSMS is also for Development, I've been using it for that for 20 years
Same here, much better than the VS tooling for SQL Server, including all the Transact-SQL support.
I still miss the tool it replaced, SQL Query Analyzer/Profiler. To this day it's my favourite SQL 'IDE'.
https://learn.microsoft.com/en-us/archive/msdn-magazine/2005...
Those features/tools are still part of SSMS today.
Yeah, it's specifically the old applications I miss.
They were really well designed, incredibly snappy and responsive. When SSMS was launched it was really slow and clunky on my computer.
I switched to Postgres around that time so I'm 20 years out of touch at this point.
I agree, I work on an application with a lot of business logic in SQL and SSMS ends up being my primary development IDE because of that.
I just said what Microsoft is targeting the tools for, not how they can be used.
Yes, it's good!
SSMS is what my nightmares are made of
SQL server is in "cash cow mode" at this point. Investing more into tooling is unlikely to increase revenues at this stage.
I can think of several improvements for Transact-SQL, starting with stored procedure packages.
I wish there was something similar for SQLite.
The copilot integrations look sweet, as does the schema view. The MSSQL extension doesn't have those, but the rest of it looks similar to the Postgres one.
Microsoft seems to be going all in on open source over the past 10-15 years.
From a consumer perspective, we're almost all benefiting.
From a business perspective, they get unpaid help and community brownie points.
> going all in on open source over the past 10-15 years.
Given that there are many Microsoft closed source extensions for VS Code, that cannot legally be used with the open source Version of VS Code, I would say they are not going all in. Knee deep maybe.
Because of the closed extensions situation, the open source part feels insincere. Sort of like a free plan up to 10 users and then pretty expensive situation. The purpose isn't the free plan, it's just an advertising measure to get people to where you really benefit eventually.
I recently called this (and other Microsoft behavior) out as being "fake" open source. The comment was highly controversial with quite a battle of up and down votes - so clearly not everyone agrees.
In my opinion, Microsoft wants the good vibes and PR that comes with open source, but they don't actually want to be open source. Its why many people still don't trust them in this arena.
Regarding fake open source, WSL2 comes to mind. It is entirely useless that it's open source except in one way, people can help Microsoft to replace Linux with Windows - for free.
At least they are making windows so bad that the incentive to go the other way around with Linux and wine looks better by the day. I personally made the transition last year. I have been playing with Linux since late 90s with dual boot, vms, wsl etc. But Linux never stuck as my main driver. I still don’t love it, but they managed to make me hate the windows experience so much that it feels natural to switch. I also have Mac, which I am using less and less for some other unidentified reason. Probably, that experience has degraded too but in more subtle ways.
Like Android for example?
Yes, Microsoft has an history, yet it isn't as if there is any big corporation doing full open source across all their products, the large majority only does the part that somehow brings good vibes, cuts down their own R&D costs, or is a kind of suble way to find out about possible new employees.
You're mixing up ideas.
Microsoft isn't "sincere" because it's just a business doing what businesses do, making money. They're not trying to be altruistic or principled. They're just doing business.
But I have personally benefited from this deal by having TypeScript and VS Code at my disposal.
I don't believe that all business is "just business". You have tobacco companies and Purdue Pharma on one end of the spectrum and, let's say, Mozilla Corporation and Valve (debatable but I think they're cool) on the other end. And, of course, large companies are kinda many different entities, really. Microsoft has a long history of dishonest behavior, some of it pretty sophisticated and with a long-term view. That makes it very hard to trust them, generally. Why is part of VS Code not FOSS anyway?
> Microsoft has a long history of dishonest behavior
Can you point out such recent behavior that isn't just echoing other peoples opinions from the anti-trust case from over two decades ago? It's been my experience that many people seem to "borrow" their opinion about Microsoft from things they've read rather than personal experience so we keep getting the same low-effort criticism ad nauseam.
VSCode's marketing was that it is an open source editor you could rely upon, complete with open source extensions for popular languages like Python. Then when once it became popular and vscodium was growing in popularity (a vscode fork), MS locked things down. Now the Python extensions are closed source, and MS has artificially prevented vscode forks from using those extensions. A bait and switch if I've ever seen one.
> VSCode's marketing
Point to this marketing. Here's the blog announcement on the 1.0 releases - what I can't find are any examples of Microsoft over-promising.
https://web.archive.org/web/20160422123116/https://code.visu...
> MS locked things down... and MS has artificially prevented vscode forks from using those extensions
Or like any growing and maturing project they established boundaries - one of which was that the plugin marketplace was proprietary, which is a perfectly reasonable position. Their existing and continued contributions to vscode are significant, so I think they can be allowed to keep some cards up their sleeves like the plugin marketplace or their Python extension. I'm just flabbergasted at this idea that somehow we're entitled to everything vscode-adjacent "just because", or that Microsoft is obligated to subsidize other billion-dollar business by giving them free features for their vscode forks.
> A bait and switch if I've ever seen one.
Where's the bait? Where's the switch? If the best you have is that they released a closed source plugin I'm going to bucket this as another borrowed opinion.
> Their existing and continued contributions to vscode are significant, so I think they can be allowed to keep some cards up their sleeves like the plugin marketplace or their Python extension.
It would have been fine if MS had started with their Python extension being proprietary, that would have been up front and transparent. Instead, they lured folks in (no small part due to open source), and once it became popular, they started turning the screws and making things proprietary and locking it down.
> I'm just flabbergasted at this idea that somehow we're entitled to everything vscode-adjacent "just because"
You're not arguing in good faith at this point. I don't think it is unreasonable to ask someone to make their intentions known up front do you? Instead MS waited until vscode became popular (partly because everything was open source) and then altering the deal Vader style closing off parts of vscode and extensions that were open. That doesn't feel particularly transparent.
> or that Microsoft is obligated to subsidize other billion-dollar business by giving them free features for their vscode forks.
I have no idea what you're talking about here. Vscodium is an entirely free and open source fork, no one makes any money from it afaik.
> Where's the bait? Where's the switch? If the best you have is that they released a closed source plugin I'm going to bucket this as another borrowed opinion.
They released the Python stack as fully open source. Then released the proprietary one, deprecating the open source one. Then made double certain that vscodium or any of the other forks could not use it at all, even if the use manually downloaded the extension. How is that not a bait and switch?
> It would have been fine if MS had started with their Python extension being proprietary
Except that never happened. Pyright was released first and was and continues to be open source. Pylance was built on Pyright but has never been open source. No promises or commitments were made otherwise. Deprecating the open source Python Language Server in favor of Pylance is also a perfectly reasonable and valid decision - the community was more than welcome to continue maintaining it, but most people I know continue to rely on Pylance.
> Instead, they lured folks in
Saying this doesn't make it true.
> Instead, they lured folks in (no small part due to open source), and once it became popular, they started turning the screws and making things proprietary and locking it down.
Microsoft has not once backtracked on anything vscode-related that's been open sourced. Trying to villianize them for not making everything open source is an argument with no legs.
> I don't think it is unreasonable to ask someone to make their intentions known up front do you?
They have. Point me to a single actual example of Microsoft operating in bad faith, that isn't them deciding to keep some parts of the ecosystem proprietary while 99% remains FOSS.
> Vscodium is an entirely free and open source fork
Microsoft and the vscode team is not making long-term decisions with vscodium in mind. But they are probably worried about Windsurf and Cursor, the latter of which (a billion-dollar company) was caught violating MS's TOS around the plugin ecosystem.
Microsoft has spent over a decade investing in, curating, and improving the vscode first-party plugin ecosystem and being a rather good steward. I think they're perfectly reasonable in keeping it to themselves. Creators are free to upload their plugins to any alternative marketplace. I don't see any arguments being made that can diminish the open source contribution they've made with code - oss just because parts of the branded vscode are proprietary.
> They released the Python stack as fully open source.
Again, no they didn't. Pyright open source. Pylance always closed source. PLS deprecated. But you're entitled to what you borrowed.
If eight years is recent enough, Microsoft moved its German office to Munich and got the city of Munich to shut down its Linux migration in return. Not exactly dishonest, but a power move to destroy the competition that I happen to be rooting for.
Personal experience is irrelevant if the facts are not in doubt. One of these is that Microsoft was a pretty bad actor when nobody reigned them in, and I was around at the time.
As most of the big corps contributing to open source.
It's the contemporary version of the second E [0], if you will.
[0]: https://en.wikipedia.org/wiki/Embrace,_extend,_and_extinguis...
Also, given that they recently bought github, they have financial incentive to keep people there, who might upgrade to pro accounts or grow to need enterprise.
And most of us are invested in MSFT, so we benefit that way as well. (I don't hold any positions)
> Quite amazing they put the effort into this for Postgres instead of SQL Server.
Embrace, extend, extinguish.
MS SQL server is a legacy system. I don't think any business would create a new database using SQL server unless, for some technical reason, they don't have any other option.
Azure SQL Database for a long while has been the most cost-effective way of running SQL Server as a PaaS database, and still is if you choose the DTU-based modes, making it a very attractive option. Combined with the rich feature set and maturity and reliability of SQL Server, it is hardly legacy; in fact it's very capable and continues to get new updates like vector operations.
I've helped create apps that support millions to hundreds of millions of revenue on Azure SQL Databases that cost at most a few hundred dollars per month. And you can get started with a S0 database for $15/mo which is absolutely suitable for production use for simple apps.
Unfortunately, I think Microsoft realized how good of a value the DTU-based model was, and has started pushing everyone to the vCore model, which dramatically increases the barrier to entry for Azure SQL Database, making PostgreSQL a much more attractive option. If Microsoft ever kills off the DTU purchasing model of Azure SQL Database, I likely won't be recommending or choosing Azure SQL Database at all going forward. It'll 100% be PostgreSQL.
Yeah, I remember that option - basic tier of DTU DB with 250GB of storage - free for one year, then continue for $15/m.
When the client brought some 3rd party expert and he advised rewriting to MySQL, I quickly did the math and it was like $60/m, without a free year.
We continued with DTU MSSQL with Prisma ORM and never regreted.
There is a parallel world, called enterprise. The enterprise people in this world, like enterprise software. They were born to be enterprise oriented. This is fine. Not everyone is like you and it's ok to use a robust product like MSSQL.
SQL Server is technically very, very good.
But they charge you an arm and a leg for the pleasure, but it can be worthwhile for enterprise.
I spent the early part of my career in MSSQL and the current part in Pg.
MSSQL is extremely, extremely capable as a database engine.
But it also costs an arm and a leg. People who haven't used it don't know just how capable it is.
The Express Edition is a SQL database engine with some limitations on CPU count and memory and missing SQL agent. It is free. I installed the version 2017 on some servers 7 years ago and they still run some supplier portal on it, that team was too lazy to even upgrade to newer versions.
I have worked with multiple ex-SQL Server engine devs and holy shit are they good.
SQL Server is very good, while cheaper than Oracle. This is a good selling point for enterprises that care about cost (or are too cheap).
Why not Postgresql??? It is fre and the best.
Legal, compliance, high performance, and familiarity are all valid reasons. (I'm actually not a fan, but less opposed now that it can run on Linux)
SQL Server, like Oracle, is technically fantastic. They were both ahead of Postgres for a long time and many people would argue they still are.
The reason people don't use them more often is that they're not free or even inexpensive.
I mean they should be, they backed by huge corporation
plus they can take notes from Open source DB like postgress and improve their system better
The MS SQL Server engine is an improvement over the years of a branch that was always more technically advanced than Postgres. I don't think there is anything to steal from it.
You have it backwards. FOSS projects are still currently learning from SQL Server and Oracle. That may change in the next 5-10 years as Postgres has already become the lingua franca of the DB world.
SQL Server and Oracle are as actual as ever, regardless of the hate they get on FOSS circles.
>MS SQL server is a legacy system
that's every SQL server out there included Postgres. Is NoSQL considered as none-legacy?
>> all without ever leaving your favorite code editor
How do I install this on Neovim then? Is there a LSP? Or is this Microsoft proprietary? I wonder how much better without Copilot integration this is then the competition.
I'm using DBeaver CE currently. Does all I need (also for SQLite).
When for-profit companies write "Democratizes X for all" it means "for people who can pay and aren't embargoed". When they write "Simplify" it means "Get locked in to our ecosystem". When they write "Your favorite X" it means "The product we're selling to you", and so on.
Once you start reading press releases with this business-dictionary, it gets a lot easier to just close the tab and move on.
Even for DuckDB, I use DBeaver CE the last 9 years professionally. Also when I do Python Dev I reach for JupySQL in Jupyter when I need something quick and it works in VSCode through Jupyter plugin..
will definitely be taking a look at this. i started my career on mostly SQL Server and using SSMS fits my brain like a glove. i've been so dissatisfied with the typical options (pgadmin, dbeaver, datagrip, etc) for managing/querying postgres since i started using it probably like 10 years ago. postgres itself is great (don't get it twisted either, SQL Server is fantastic. just costs money) but i never understood why there wasn't more uproar in the community about its DBMS tooling ecosystem
I've found Datagrip to be far and away the most impressive universal database tool. I feel like I've tried them all, and they all have a quality of having been developed by database people, rather than IDE designers. The depth of capability, extensibility, pace of improvement--I'm a very happy customer.
I don't want to poop on open source, but pgadmin and dbeaver and not even close to playing in the same league.
I work in Oracle and Datagrip saved my sanity.
Are there any features in datagrip in particular that you like that aren't in dbeaver?
I used DBeaver (community edition) before Datagrip and I would have to give the nod to datagrip overall. I wouldn't have even tried it if I did not have the etbrains all product pack. DBeaver is great, don't get me wrong, but Datagrip just seems a lot more polished overall and the settings and the UI just seem more intuitive to me.
More streamlined UI. I find the graph viewer to display some things nicer, like coloring tables, bringing in related tables. Slightly better introspection and code formatting. Easier refactoring capabilities. Easier to get IdeaVim working then whatever Ecplise plugin is needed in DBeaver.
In my experience, DataGrip has been easier to get up and running out of the box and bringing the big IDE guns than DBeaver but DBeaver also does done this really well. For example I have never been able to setup DG with Access but DBeaver works pretty good out of the box with that garbage. Also it's free.
Both are solid
I'd love to try datagrip, what are some advantages over pgadmin or others?
Many people love Datagrip because it's got the polish and style of Jetbrains' other products. Others hate it and much prefer the more classic designs of DBeaver or the browser UI of PGAdmin.
I think it's worth downloading the 30 day trial and giving it a quick whirl. The people who dislike it seem to get hesitant quite quickly, so I doubt you'd need more than a day to decide if it's for you or not. There are all kinds of cool and fancy plugins you can set up, but I wouldn't bother with that if you decide the UX just isn't for you.
It supports quite literally every database I've thrown at it, which is pretty nice.
The way it support projects is also rather useful to me as a developer. Most SQL tools seem to be focused on being an interface first and maybe having a few SQL files open second, but Datagrip's basis as an IDE makes it very easy to maintain collections of scripts (version migrations etc).
That said, I'm not sure if the 99 bucks (a year if you want updates, though you get a perpetual license for the current version) is worth it. I use it because it's part of Jetbrains' all products pack, which I paid to use other IDEs, essentially giving me Datagrip for free. If you're not already a Jetbrains customer you could definitely give it a go, but the value per dollar it provides is very different.
With Datagrip, you get all the niceties of a JetBrains IDE: massive customization, numerous plugins (e.g. IdeaVim, GitHub Copilot), lots of documentation. But you also get support for countless Db engines - I haven't seen anything yet which wasn't supported. The only one that was half-baked was Redis support, but it's not exactly a Db, either. Most importantly, it's that the UI doesn't feel clunky, unlike with PgAdmin. Everything feels streamlined and 1 or 2 clicks away, at most.
I doubt any new hotness could tear me away from Datagrip. I've used loads of database admin UIs over the years, and Datagrip is by far the most impressive.
This is going to be subjective, but the interface is better, especially if you're used to other jetbrain products. I haven't used pgAdmin for a while, but I remember the autocomplete being clunky to use and, quite frankly, quite bad.
My company had some leftover Datagrip licences, and it felt like moving from notepad to an IDE. I haven't looked back since.
Hmmm, interesting. I used Datagrip for some years, now DBeaver (as I don't have a JB subscription anymore). Datagrip was and probably still is very powerful with top intellisense. Now I'm using DBeaver and it's very solid. Ok, I'm not spending my whole day in it, but when I need it, it does the job well.
In my experience most SQL developers don't care too much about tooling. In most cases someone designs the database and tables, developers don't care about databases and care mostly about tables and views, rarely about indexes. The ones that care, and need tooling, are usually called development DBAs and they are very rare. Rare enough I was never able to hire one and keep them (we don't pay enough for how rare they are).
I've been in that camp for much of my career, and yes, tooling matters.
As you suggest however, tooling for that workload is pretty rare. I want something that focuses on enhancing the database development experience: that understands that there are development workflows for database code which are well controlled and rigorous. So many database tools are focused on being system administrative aids first, or giving you features for directly interacting and <ack> altering the running state of the database and its server from the tool.
The best tool I found for what I do was targeted at Oracle: Allround Automation's PL/SQL Developer (https://www.allroundautomations.com/products/pl-sql-develope...). It's a development oriented tool that, at least last I used it, was focused on serious development work rather than administrative work. Now, I haven't used it in almost 20 years when last I did Oracle development... but I haven't found anything for PostgreSQL that has that thoughtfully implemented database developer centric feature set.
Today I muddle through with DataGrip. DataGrip has just enough of what I need that it's marginally better to work with than just a simple text editor... and also narrowly avoids some misfeatures as not to negate it's utility.
All true.
I loved workgroup style app development. R:Base, Access, dBase, FoxPro. Then I switched to UI work for a stretch.
Circling back to back-end work, naive me embraced Hibernate (2004?), assuming it'd be familiar and good.
I was wrong.
Now that I have a lot of free time, I'm finally recreating the workgroup style experience, for the general dev population. Sort of.
Using Hibernate, our workflow became: rough in some ORM hack, capture the generated SQL, use SQL Query Analyzer and Toad to make it work (and performant), coerce Hibernate to regenerate the SQL we want. Totally backwards, right?
Eventually we gave up and just used HQL.
At that point, why even bother with ORM?
So I created a "SQL first" workflow. Treat your SQL (DML) as source code, use those explicit queries to generate the prepared statements (and typesafe DAOs, DTOs, etc). In other words, auto-generate all the things you'd do yourself, if only you had more time.
I used my tool for years. Am currently making it usable for other devs. eg Spent last week making my grammar for MySQL "good enough" for initial release. Already have PostgreSQL and SQLite, plus my original turrible "poor mans SQL" grammar (comparable SQL-92). Which I'll cull once I have "good enough" T-SQL and PL/SQL grammars.
Any way. Thanks for reading.
I only meant to confirm your experience with development DBAs. The only such person I was able to retain was near retirement and was tired of the hustle.
This is not available in Cursor. Extension ID "ms-ossdata.vscode-pgsql" returns no results. I wonder if Microsoft is blocking VSCode forks from having this extension?
Cursor was based on quite an old version of VS Code laat time I checked, so it's possible Microsoft just didn't target compatibility with a release from six months ago.
This is also what's said here by one of the devs: https://github.com/microsoft/vscode-pgsql/issues/15
Looks like manually loading the VSIX to bypass the compatibility check should work but who knows how stable that'll be.
Biggest thing that JetBrains has over VSCode for me was their very clean built in database tooling
You can also run the database tool separate (DataGrip). That's what I do.
I didn't know about DataGrip. Looks cool.
Every year or so I try to go back to VSCode but I can’t get past how good the git and database integrations are in JetBrains.
I think Code has fine git integration, but they way they implement it is just very Microsoft. The UI is very much like the one they use in their professional Visual Studio UI, and whether that's a good thing or not is definitely a very personal choice.
I much prefer JB's git integration, but I wouldn't discount it just because the UI is so completely different.
Congrats to the team on launching this! I was actually the first to demo it, as part of our sponsored session at Microsoft last week.
Here's the talk where I used it: https://www.youtube.com/watch?v=k6Vm2hakkV4
I also did a theater session at our MSFT booth, but the recording isn't up yet. You can follow the steps in this repo to check out all the features that I demo'd, however: https://github.com/Azure-Samples/postgresql-extension-playgr...
Let the team know about any issues here: https://github.com/microsoft/vscode-pgsql/issues
Hey, just a heads up that your website 404s if visted at "pamelafox.org", which is how it appears in your bio.
Thanks, I need to move that website to GitHub Pages soon, the current host isn't handling the naked domain well. Changed to www.pamelafox.org for now.
Postico has always been my defacto way to interact with Postgres. Curious if there are any Postico users who have tried this yet.
Very Mac-oriented, and I think IntelliJ built-in DB editor has way more functions / features.
The heavy Mac UX compliance is the reason why I enjoy using Postico for 99% of straightforward Postgres tasks
10+ year user of Postico. I will give this a try. I hope Copilot can start recognizing the schema when I use node-pg.
I wonder what is the most "valuable" IDE right now for MS. A few years ago VsCode was marketed essentially as "Visual studio for beginners", where you were supposed to move to Visual Studio after you became a real dev, but since then VSCode has been growing and growing and stands now as the most used "IDE", where Visual Studio is mostly seen as "legacy" (oversimplification, great IDE for CPP and .NET but still...).
Easily VSCode, if we're talking about developer reach. I'm not big into Microsoft stuff, but almost every 'serious' .Net developer I personally know is using Rider, so I can only assume that Visual Studio is retreating to the same space occupied by Eclipse and Netbeans, i.e. still used, but mostly only in places where change is hard.
I'm an emacs user and even I keep a copy of VSCode installed just because I occasionally have to interact with SQL Server and it's really the best way to do that on non-windows systems now that they're winding down ADS.
I work at a mostly .NET firm and almost all the developers on that side of things are on Visual Studio. Rider has less penetration than PyCharm has among the Python devs.
I only use Rider because it's cross-platform. It's not inherently better (or worse) than Visual Studio with Resharper installed.
it's fasterb and has a functional vim interface
Most .Net devs I know use VS, I used Rider because it’s so much less awful.
Monetarily, Visual Studio.
There are tons of enterprise development workflows, and plugins, that probably will never be ported into VSCode, from their .NET and COM implementations.
Now in terms of mindshare, and gateway drug into Microsoft ecosystem, definitely VSCode.
It is also the best Web IDE, for the return of timesharing development, sorry cloud.
That alone means everyone that is on Github and Azure, gets to use it as the modern version from X Windows and RDP/Citrix sessions.
Not bad, for Eclipse v2 (Enrich Gamma is one of the main architects), pity the whole Electron shell though.
VS Code is a downgrade from open source to freeware. At least the C++ plugin is freeware. And they block access to the extension store from any fork (self-compiling VS code is also considered a fork). So if you are an OSS purist, VS is bad. Other than that it's effing great.
It is also the best Web IDE, for the return of timesharing development, sorry cloud.
Also the best webdev IDE.
How is VSCode a "gateway drug" into the MS ecosystem? It's good PR, for sure, but it has little to no conceptual/GUI overlap with, say, Windows.
FWIW I use it via Linux .deb and integrate with a private GitLab.
Worrying about Windows is fighting the last war.
Azure, Github, CoPilot, .NET (why do you think it is cross-platform), Java (yes, MS is back in Java land, they were the ones with initial ARM support), Go (they have their own FIPS compliant distro), Python (although with layoffs maybe not anymore), Rust, npm, Powershell, Powerapps, 365 AddIns, Teams plugins, clang/cmake (part of Visual Studio installer), Azure Linux, Sphere OS,....
Azure and GitHub are large Microsoft revenue sources. Both have first-class VSC integration.
Because it opens the gate up to the rest of Microsoft tools.
Once you're gonna play with azure, GitHub, vsc, you're bit by bit invested in the ecosystem and opening the wallet for that other feature or integration.
Erich Gamma
Thanks for the correction.
Visual Studio is still widely used in the games industry, being pretty much a requirement for targeting some platforms.
It is becoming common for some to use Rider primarily, but VS is still used as part of the build system.
it has to be VS Code by a long shot. They don't charge for it, but it serves as an enormous draw to keep people in the MS ecosphere and keeps MS in the developer game.
That's ironic since I develop most of my Microsoft related tech with JetBrains products and only use vs code for frontend/node - non Microsoft stuff.
> where you were supposed to move to Visual Studio after you became a real dev
It was never marketed like that, for the simple reason that popular VSCode languages like Python/HTML/Javascript were never well supported by regular Visual Studio, so there is no way to move to "proper" Visual Studio if you do Python/web development.
VS could do web okay 5 years ago, haven't run it since.
It still does when Web == .NET Web, and way better than VSCode at it.
> A few years ago VsCode was marketed essentially as "Visual studio for beginners", where you were supposed to move to Visual Studio after you became a real dev,
When was that?
I wont likely find the video but I remember watching the PM for both VSCode and VS (at the time was the same one, not sure now) recommending people to move to Visual Studio "eventually". I clearly remember it because it didn't make any sense, even if the names were similar there were/are nothing alike UI-wise and supported-language-wise. I said few years ago but it was prob around 8 years ago, vsc was still pretty young.
Yeah I’ve been using it since it was released and can’t ever remember it being marketed as such.
Yeah I only really saw vscode as a "let's capture atom and sublime users, bevause they will not use Visual Studio anyway" approach.
Is there a similar feature available for SQLite? Will there need to be a totally new extension for every DB, or is there a shared portion?
Microsoft Access meets PostgreSQL, 3 decades later?
Ah man, Access. Got a website to over 1m users/month with a dead simple Access DB and ASPv3. Back when a million users was a million users.
Is there something related to Access in this that I missed in this or you just being salty, because I have to work with that garbage and if there's something better than DBeaver I can use for that I'm all ears
Looks nice enough. I only wish there was an easy way to spread VS Code across multiple monitors. I commonly will work in code on one monitor, and the database tool (DataGrip currently) on another.
VS Code has had multi monitor support for a while now https://code.visualstudio.com/docs/configure/custom-layout#_...
You can already do that. I run VS Code on multiple monitors.
Any tab called be pulled into its own window and moved to a different screen.
Even the terminal panel can be popped out into a tab or panel or window. (The UI is not obvious but once you see it you can’t unsee it)
It’s pretty cool to have the code and terminal side by side in the editor window. (Of course this was always possible with emacs)
I noticed yesterday that you can pull out a tab into its own window. It's not "natural" cross monitor, but you can place the windows on both. Not sure if that helps
Have you thought about just solving the problem with hardware? I.e. one giant 38" wide screen?
Tried it, queries work but i can't even show the columns and other meta data. it says no connnection even though i have a connection to do queries.
Is there a similar extension for MySQL in VS Code?
Database Client is fairly similar. I don't think it does the schema diagrams or AI stuff, but it can do notebooks. Has a shareware pricing model, free for up to three active connections. https://database-client.com/
Remember the old adage, embrace, extend, extinguish.
They will extinguish postgres?
Looks amazing- and the point they're making in the article is correct. Switching back and forth from VS to PG Admin creates friction that this seems to solve in a much nicer way
I wonder how this compares to pgModeler (https://pgmodeler.io/) which I've been using the most in the recent years, would love is someone who had tried both could share some observations.
The AI integration is interesting to me. I have been trying to get Claude to help me with postgres from within my product and have found its ability and understanding of postgres/sql to be significantly worse than with common programming languages.
Maybe it's because my tolerance for imperfection is much lower for databases than web apps, but it is so bad that I can't trust it for anything database-related beyond text transformations and generating boilerplate queries. It will incorrectly tell me things like postgres doesn't support TABLE OF <TYPE>, or make syntax errors for ON CONFLICT, and immediately agree with any "what about" or "are you sure" I throw at it.
Curious if anybody else has run into this. Obviously LLMs are not always great in specialized domains like this but the poor performance with something as popular as postgres is pretty uncharacteristic IMMO.
I think it's more that LLMs are a given percentage (let's say 30%) inaccurate in general, and the results of this inaccuracy are just more immediately obvious with SQL
I used Azure Data Studio with SQL Server for a while—decent, but still clunky. I believe this plugin is the evolution of that, so it’s probably solid too—but can't beat out what I'm currently using since it targets only Postgres.
It’s going to be tough to beat SQLTools for Sublime Text (https://code.mteixeira.dev/SublimeText-SQLTools/). It’s simple, fast, does the basics right, and crucially, works for many of the usual suspects.
Most IDEs try to do everything and fail, yet only target one db vendor. They’re _all_ bloated and buggy, with endless trees and menus, bells and whistles. Results grids are slow, copy/paste is janky, formatting is never quite right—or outright misleading.
SQLTools stays out of the way and just works.
...also, I don’t want AI-powered nonsense. I’ll reach for that when I need it. Get off my lawn.
So it's just Azure Data Studio where it lacks SSH tunneling...
I got my hopes up for nothing
I've never used any of these kind of graphical SQL editors. I'm reasonably ok with SQL but I've always just done text queries.
Does anyone here feel that the graphical editors actually save them time or make their lives easier?
Just today I've been dealing with an issue trying to update an XML column because the XML I'm trying to insert was escaping the string in the update query.
But with DataGrip I could just pull the row I was interested and paste in the string from the UI.
Surely there's another workaround that wouldn't involve a UI but this made it pretty easy.
The schema visualization is a feature of the extension, not the primary mode of navigation. Lots of people love visualization, but not everyone does - totally cool to not use it. I think it's a fun way to browse a database that's new to me, to see roughly how things are connected.
Using a UI to explore data is like looking at the picture instead of someone describing the picture with words.
I know, what I was asking is if this is something professionals actually use.
Like, you can make websites graphically with DreamWeaver (or something), but I think most people say that you're better off using HTML and CSS and whatnot if you know how to use that.
Is this GUI stuff something people who are good at SQL actually use or is this mostly for new people?
I used to write a ton of SQL and I could write a query for anything you asked for, no matter how incompatible the schema was to your request.
My first step was to look at the data using the UI. Then when I tested my queries, I also just ran them in the UI.
It’s just faster.
I never used the UI for “simple” queries since there wasn’t anything to test.
Yet the picture at the top of the article is incomprehensible to me. Why is the "session_speakers" table positioned there? It relates speakers and sessions and then has a geometric layout that completely fails to convey that.
You put too many bezier curves on a page on my eyes go cross. They don't even bother to make the lines different colors. Just a bunch of white curves against a fixed grid layout. Wut?
It's taking well structured data and then destructuring it into an inaccurate sloppy picture. I don't get it.
Isn't it a problem with many-to-many relationships in relational algebra? You have to have this artificial table to relate things in many-to-many way. How would you display it in a better way?
The tables should be better ordered. "speakers" is obviously the primary table here and should be top left. The "session_speakers" should be between it and the table it maps "sessions." Then "reviews" and "events" on the bottom. There's an obvious "top end" and "bottom end" of this structure that this graph completely gets wrong.
You could argue that "events" are primary in which case just switch the order of that and "speakers."
I'd color the table headers differently and different from each other and have the arrows leaving that box match the color of the box it's leaving. Also drop "public:". It's expected and default so it does not need to be displayed here.
Drop the key icon and just make the key fields in bold or with a different style.
Don't make me hunt around to match information and don't be afraid to use differential styling.
That's great feedback! Thank you! I wonder if we can auto-layout in the way you describe. Should be doable. We will look into it.
We build privileged access management tools. Majority of our users do not like to use CLI. They know workbenches, pgadmin etc. at a wizard level.
In my opinion, i see that graphical editors are more product for major section of devs and admins.
I haven't used it for relational databases, but I can imagine it's useful to visualize complex relationships between tables.
You don't use it but still it's nice feature to have, isn't it?
I think he went a long way to simply ask what do these tools do that a cli won't.
Oh sure, I'm not criticizing its existence.
Doesn't look like there's much beyond what's currently possible in DataGrip yet (which is far beyond any other SQL client I've used), but nice to see a competitor in the space - especially one that will push JetBrains on the AI assistance side of things.
There seems to be a bug — I can't re-run the query from the query history panel; it returns no results.
New query could still get results.
Overall, it should have been a separate app because you can't really see all the results in a small panel.
Please file that bug in the issue tracker: https://github.com/microsoft/vscode-pgsql/issues
The team is very receptive to feedback, in my experience.
I use DataGrip, but only for very simple tasks and only against PG and occasionally SQLite and it has always felt like way overkill for me (and it's a heavy app). This will be a more convenient option!
Beekeeper for small stuff, that's my rule. However, I feel this may start to change that for me.
Data grip for big serious stuff.
DataGrip would be perfect if it had a community edition. As someone who connects to a database only two or three times a week, I’m not willing to pay for it.
Yeah I already have the all-products pack for Goland, Clion, and Resharper, I wouldn't have purchased DataGrip separately
Yeah same. As great as Datagrip is, I would have stuck with DBeaver if I didn't have the All Products Pack.
try galaxy instead ;) getgalaxy.io
Nice. The few db managers I’ve tried in VS Code are so awkward, creating files for queries and opening multiple panes that barely fit in the crowded IDE space
It makes me wish for something like phpmyadmin or adminer
Isn't that exactly what pgadmin is?
This is amazing.
I imagine later LLM have access to my table metadata and then perhaps down the road, LLM can suggest me better queries based on execution plans that it saw.
Have you read the article?
the copilote being aware of postgres schema is a thing I manually have to deal with in cursor rules. I keep all the SQL DLL files that created any table in context but then also its best to have cursor rules to tell it to use one orm if possible
I wonder how well this IDE works if you use AWS or GCP instead of Azure.
PM on the project here - All clouds and on-prem/local/docker work great! We test those, so if you have issues, please report in the repo we'll fix them.
I assume you run cloud-sql-proxy as usual (for proper TLS and passwordless IAM authentication), and then it's just another postgresql port for you. Any tool like psql can connect to it.
I run a personal Postgres server so I'm wondering how well it does there
> Password-less authentication with Entra Id
I don't understand why the first four points under this heading are basically the same.
Very nice and long awaited. Thanks Ms. There are basic features available now (visualisations and query). Looking ahead for more, such table/data editing etc.
I like that I can run this remotely with the vscode ssh extension to my remote systems. But compared to dbeaver, I miss some of the features of dbeaver. Like showing an estimate of large each table is, and seeing how many rows in the 'properties tab'. And being able to see the ERD graphics for individual tables (to show foreign keys, etc).
Also like with dbeaver that you can take the results and export them in many different formats. like directly to excel for some of my co-workers who live and die in excel.
I wonder if this will work with Redshift , which is based on Postgres?
Just tried it, wish there was better refactoring tools available!
If love for there to me a "rename variable" feature.
Graphical schema image generation was in IntelliJ for like 15 years at least? Since the times with all the craze about UML "disrupting" software engineering.
UML was the late 90s! Rational Rose[0] was the glorious future of class design. Draw the diagram and let the juniors fill in the methods.
[0] affectionately Crashional Rose
I was a junior back then, and was afraid to speak up that Rational Rose is just a waste of time. Don't you dare say that about LLMs!
No worries, nowadays we have Enterprise Architect
Graphical schema image generation was already a thing in 1990's Windows IDEs for databases.
People have to stop pretending there were no IDEs before IntelliJ .
it was in Access 95 and FoxPro (and likely before that too)
slowly development tools are catching up with the 90s
Automatic Schema Visualization always miss the biggest point.
When you generate visualization, you should only show key table and key fields within the table and hide all the helper, secondary fields and tables.
In an LLM world this should be easy for LLMs to pick out key tables and fields and only display those
Some fields are important for HR, others for Security, others for BI. Gotta have different visualizations for everyone.
LLM here would be like another person or another team, that decides what's important for me, what I should see and what I should not. Might work, but I'd rather see the full picture and decide myself.
Why would you need an LLM to do this? The whole point is that SQL describes it's own relationships. You literally have everything you need out of the box.
LLM All THe ThiNgS!
and just like that, 90% of the reason i pay for a jetbrains license just...disappeared
The difference? JetBrains won't pull the rug on you, or abandon the product at the next turn in the road.
This is so cool. A big reason I used prisma was for prisma studio. Having this kind of support in vscode is nice to see
(prisma team member)
We're going to be rolling out Prisma Studio as a VS Code extension soon!
Ooooh can’t wait for that, and for further investment in Prisma Studio generally — already a really nice tool, I just want some slightly more power-user-y stuff!
Does something like this exist for mysql?
DBeaver?
For vscode? I thought dbeaver was its own app.
Your understanding is correct, dbeaver is a stand-alone application.
Shame it's not a generic thing that can work with MySQL
This looks great. Would all the functionality work if the databases are hosted on other cloud providers?
I am PM from the project - short answer is yes - you can connect to any Postgres endpoint. There are some Azure specific features, like EntraID supported by the extension that don't exist in other cloud providers.
Thank you, this is great
Nice that they’ve got it working so well with copilot, the only thing keeping me from buying a premium sub is that they don’t bundle GitHub copilot with office copilot.
It seems like all their copilots are seperate subs, which seems like a missed opportunity honestly.
Too bad I switched to PGlite for local dev (not really because I love PGlite). I unfortunately don't see any PGlite VS Code extensions.
Looks promising, but I'll probably stick to `psql`
I wonder about people’s development workflows. If you are using a tool like this, how much time are you spending in the command-line (where all such tools can be interfaced)? Are most tools used wrapped in some layer like this?
I spend most of my time in the command line:
- neovim for file editing,
- zsh (+zoxide) for navigation / file management,
- plain git to manage my repos,
- plain text note taking and accounting, etc.
Try lazygit. It is truly amazing.
There’s a lot of developers that are scared of the command-line. Truth is you don’t really need these IDEs if you truly know SQL and your database, writing queries isn’t difficult. Keeping a file with common queries isn’t hard either. But most developers just keep a very shallow pool of knowledge and lean into ORMs etc.
I'm using CLIs like A LOT, but still would be happy to get _good_ autocomplete for SQL.
`psql` is pretty bad at it and in `\e` you will just end up in an editor, which will probably don't know about your schema.
I've tried many tools, but seems like I like DataGrip (or databases in PyCharm Professional) the most, so I use EAP from time to time, when I'm going to write a lot of SQL.
Hm, is there a psql extension to augment the CLI and provide better autocomplete, maybe even interface with LLM? And then it just stores whatever metadata (like queries you want to save) in its own tables...
I’m really surprised that some rabid rustacean or something hasn’t written an entirely new and aesthetic CLI replacement for psql with all the modern comforts. Autocomplete menus, graphic icons, colors, etc…
Exactly, unleash the crabs.
I am very comfortable in the command line and still work with databases in IDEA. It gives you:
— autocompletion for everything — table/function names, types; very helpful on projects with hundreds to thousands of tables
— navigation ("jump to referenced table", "find foreign keys to this column", etc)
— data export in two dozen formats (configurable)
— exactly the same UI for working with 30 database engines (or however many it supports, I'm too lazy to count). Especially helpful with databases that have atrocious CLI clients, like Oracle.
— a nice tree-structured view of your database; or you can generate a (possibly vector) diagram for the rare case when that helps
— high quality autoformatter that works for every SQL dialect it supports, and in the same way
— minor things like the ability to extract a subquery with a couple of key presses, or rename a table alias
Probably something else I'm forgetting.
Saving a couple of keystrokes when writing SQL has little to do with it.
Try https://www.pgcli.com/install if you haven't already, it's a nice improvement over pure psql.
Would you be interested if this extension supported all of the psql commands directly in the vscode editor?
I will be using this. 100%. very handy
This is great!
I don't really understand the UI though, there's no way to provide a host
am I the only one to have that problem?
Figure 6 in the article shows it.
thanks a lot
I really wonder why they would call it "server name" instead of database host though
True it's a little confusing, I think it's probably because they use that generic term a lot around code for other places that may have a context specific name for the "server" is called.
does this work for non-Azure?
PM on the project here - Yes! It works on any Postgres, on-prem or cloud
Hi maxlux, is it possible to configure a connection over SSH?
Not directly in the extension connection object, but it's in on the roadmap. In the meantime, you can create ssh tunnel using vscode remotes and then just connect through it.
This might already exist, but seems like AI would be a natural fit and efficient at recommending indexes based off queries. I’ve been building a Python SQLite3 app and I gave ChatGPT the entire schema along with all queries and it did a fantastic job at recommending queries and explaining them. Compound indexes, unique indexes, compound primary keys. Taken a step further, if a process ran live inspecting queries in real time and then sent notifications of missing indexes would be super useful.
I think you are describing Xata Agent :) https://github.com/xataio/agent It watches metrics/logs, queries pg_stat_stetements for slow queries, the sends slack notification with suggestions, including indexes, but can be more varied than that.
I'm sure nobody cares, but I just independently stumbled upon this an hour ago and wondered why more people haven't used it.
It has a proprietary license[0]. That makes it a non-starter. Too bad: it looks nifty!
> The software is licensed, not sold. Microsoft reserves all other rights. Unless applicable law gives you more rights despite this limitation, you will not (and have no right to):[…] d) use the software for commercial, non-profit, or revenue-generating activities
Oops. Better not install this on your work laptop!
[0] https://marketplace.visualstudio.com/items/ms-ossdata.vscode...
Dev on the project here - I'll mention that this language is due to corporate boilerplate public preview licensing. It is absolutely available and encouraged for use in commercial etc. activities. The licensing language needs to be fixed.
That's good to know! Right now it's pretty much illegal for anyone to use it in the situations most people would want to use it for. Any idea what the eventual license will be?
We're working the details out right now, but the change will make it clear that the extension has a free license without restrictions. Stay tuned!
Right on. I'll keep an eye out for it!
You're a dev on the project and consider yourself authorized to speak on behalf of the project, on a legal matter, on Hacker News.
Are you authorized to do the same on a blog hosted on microsoft.com? A lot of people would treat that as authoritative even if bigger enterprises will wait for the shrink-wrap to be updated.
Solid point. That would be a "promissory estoppel" defense if MS changed their mind and decided to run amok with this.
Good: "Your honor, here's a copy of their official blog where they said the license terms are a temporary glitch but that we're fully allowed and encouraged to use the product."
Not good: "Your honor, an anonymous new account on Hacker News said it's totally fine to use this even though the license forbids it."
I'd cheerfully take my chances with the former. The latter? Not so much.
(As mentioned elsewhere, I don't for a second think MS is going to track me down and sue me for using this against the terms of the license. I'd feel a whole lot better if someone officially put that in writing, though.)
Not an MS fan, but you guys rock for making VS Code, really changed software development for me. Especially combined with good language servers (Redhat for Java, intelliphense for PHP, Clangd for C/C++, and some python stuff). Yay! Sure, Eclipse could do it, but things like search, which I do a 1000 times a day, really benefits from good UI design choices. It simply speeds up the work massively.
The blog has been updated, see the section "Feedback and Support".
[flagged]
Please don't harangue new users as soon as they show up to HN. Whatever legitimate point you have is drowned out by the atmosphere of hostility you create.
Not only that but it damages the reputation of this community.
You (<-- I don't mean you personally, but all of us) should be welcoming to new users and assume good faith, as the site guidelines ask (https://news.ycombinator.com/newsguidelines.html). There's no reason why you can't make your substantive points while doing so. (Edit: you needn't look far for a good example: https://news.ycombinator.com/item?id=44074135.)
It might also be good to remember that everyone makes mistakes, that project launches don't always go perfectly, and that HN is for discussing the interesting aspects of a submission.
I didn't mean it to be haranguing, but OP was giving what amounted to legal advice that was actually wrong and dangerous. I do think that needs to be called out and clarified—not as a personal attack on OP but just as a matter of safety for readers.
I'll acknowledge that kstrauser may have done a better job of sounding friendly about it, but I don't think my question was out of line nor even particularly aggressive given the circumstances.
I'm sorry to press the point, but I don't think you're correctly assessing the impact of comments like your GP post, especially on a legit new user like lossyrob.
You may not have meant to be haranguing, but intent doesn't communicate itself—at least not in the tiny textblobs which are all we have here. It has to be encoded in the actual message.
When I look at your comment from that point of view, I notice that it leads with a hostile personal trope ("You do recognize...?"), followed by a putdown of everything this team is probably hoping for ("no one should use this"), followed by a personal attack ("you shouldn't be encouraging people to do so"), followed by a pedantic hammer-blow ("legally the license is the license") that takes the spotlight away from anything new or exciting about their work. That is followed by a sentence that basically shames them for what was obviously just an oversight. How is a newcomer (or anyone, for that matter) supposed to feel when they encounter that?
You're a good HN member and I'm sure you didn't intend to condemn or humiliate anyone. The problem is that people routinely underestimate the provocation in their own comments and overestimate the provocation in others'. If the error is 10x in each direction, that's a huge skew [1]. That's why it's hard to track the impact that one's posts have—especially the righteously indignant sort of post [2].
I suppose one of the moderators' jobs is to step in and try to articulate that explicitly, in the hope of persuading enough users to generate a bit of a system correction.
[1] https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...
[2] which, by the way, I can feel a bit of in my own comment just now, and I'm probably underestimating it too.
OP was giving what amounted to legal advice
That is a huge overstatement that you can't really use to justify the haranguing. They just popped in to address the potential issue and let people know they're trying to sort it out. Nobody is giving legal advice and nobody is going to end up in legal trouble because Microsoft messed up their license boilerplate for a bit.
[flagged]
"Please don't post insinuations about astroturfing, shilling, bots, brigading, foreign agents and the like. It degrades discussion and is usually mistaken. If you're worried about abuse, email hn@ycombinator.com and we'll look at the data."
https://news.ycombinator.com/newsguidelines.html
https://hn.algolia.com/?sort=byDate&dateRange=all&type=comme...
Huh, I'm not sure I said anything like the above.
It's in the same general space, no?
If it helps at all, there's also this: "Assume good faith" - https://news.ycombinator.com/newsguidelines.html
Not to belabor the point too much, but assuming good faith is different from trusting anonymous advice from people claiming to be official advocates that could do real damage to your company if legal wanted to bone you.
Yes they are a new user, but they appear to have made the account just to make that comment.
but they appear to have made the account just to make that comment
People making accounts so they can talk about their work is one of the best reasons for people to make an account and a big part of what makes HN threads interesting.
people claiming to be official advocates that could do real damage to your company if legal wanted to bone you.
That seems like the opposite of assuming good faith.
> People making accounts so they can talk about their work is one of the best reasons for people to make an account and a big part of what makes HN threads interesting.
Uhuh but coming in to give bad legal advice as your first comment for the benefit of the largest most litigious corp on the planet, is that what makes HN threads interesting?
> That seems like the opposite of assuming good faith.
That wasn't a statement towards the individual. That was a statement of what could befall someone who took their advice.
It's not legal advice. People can just be wrong and you can tell them you think that without coming off as a jerk.
That wasn't a statement towards the individual. That was a statement of what could befall someone who took their advice.
No, I don't think that's true - it's just a catastrophizing rationalization for reflexive dickishness. We all suffer from reflexive dickishness so of course it happens and it's not that big of a deal but trying to pass it off as some sort of virtue is a mistake.
Can't use it for commercial, non-profit, or revenue-generating activities? Uh, this actually seems insane to me? What is it for then?
> Can't use it for commercial, non-profit, or revenue-generating activities? Uh, this actually seems insane to me? What is it for then?
Its a public preview, for which Microsoft probably does not wish to accept non-disclaimable liabilities for defects when used in those circumstances.
It is for previewing. By people who are interested in what is in the pipeline for a more general release.
> non-disclaimable liabilities
So you are saying that MIT/ISC/GPL/Apache2 and all the other OSS licenses do open you up to liabilites?
I am saying being a merchant in the field of software and supply software opens you up to liabilities, and saying “Not my responsibility” does not, in most jurisdictions, actially completely shield you from all of them, correct.
This may also, to a lesser extent, be true of people who are not merchants in the field of the product supplied.
It’s not the license creating the liability, in either case.
So you are saying that the parts of for example MIT or similar licenses that in clear terms say:
"THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE."
do not constitute something that frees you from those liabilities?
Cause if so I think basically most of open-source would just shut down tomorrow.
I'm with you on this one. Consider all the GPL'ed software that IBM distributes via Red Hat.
This doesn't make sense to me. They could just say the software is provided as-is and Microsoft holds no liability. Which they do say elsewhere. This license goes much farther to say Microsoft can sue you if you use it.
> This doesn't make sense to me. They could just say the software is provided as-is and Microsoft holds no liability.
You cannot effectively disclaim certain liability for uses of a product you supply, even with an as-is presentation (exactly what liability depends on jurisdiction and often other context). Merely claiming to have no liability does not make it so (what it will usually do is disclaim all yhe liability you can disclaim, except for particular liabilities that may require separate explicit specific waivers to be effective.)
OTOH, if the product you provide is a software license that doesn't cover specific uses, using the software for the excluded uses may not be seen as a use of the product provided at all, and may not trigger the non-disclaimable liabilities, and even it doesn't avoid those liabilities, in the event someone sues over them, it also enabled the product supplier to countersue for infringement damages and mitigate the liabilities.
My initial reaction is "entrapment".
Edit: That's uncharitable of me. I strongly doubt that's the plan. But it genuinely was the first thing that came to mind.
I'm sure these restrictions will lift once it's out of preview. They have a huge Postgres hosting business in Azure that couldn't benefit from this if it's restricted to non-commercial.
I assume that restriction is due to the public preview status. But yeah MS really ought to at least allow businesses to evaluate it for potential subsequent use after preview.
> PRE-RELEASE SOFTWARE. The software is a pre-release version. It may not operate correctly. It may be different from the commercially released version.
Probably a copypasta mistake?
Why would they spend money into this extension if 99% of developers can't use it?
Because most people will use it without reading the license, giving them the upper hand when they might needed it? Might sound a but harsh, but we're talking about Microsoft.
Is there a single example of MS doing something like this in the last 10 years?
VS Code forks like Cursor that use their extensions?
That's not precisely a developer using the extension...
Maybe, but darned if I'm going to be the one to take the legal risk of installing it.
I'm willing to put good money that MS is not going to sue anyone for using this
I am extremely doubtful that they would. And yet, using it under that license is taking the legal risk that they'll act kindly even though they have the right, under the license, to be jerks about it.
Is MS going to be a jerk about it? Almost certainly not. Could they if they wanted to? Sure seems like it.
I worry that with the threat of Cursor this will be more common now. Microsoft's business interest will prevent them from funding development work that directly benefit the popular forks (for years they probably assumed no fork could ever gain traction...).
I suspect you're right. And thus came the end of the era of free VSCode.
Microsoft owns shares of Cursor through OpenAI
does anyone actually think like this?
Anyone who works in risk management and/or threat modeling for a living does.
thats such a small percentage of people
There aren't that many doctors, either, as a percentage of the population, but if they tell you not to eat paint, you might consider their opinion.
I go further than that and say screw Microsoft's partially open source stuff. Because VS Code isn't fully FOSS, which is a bit weird, isn't it.
https://github.com/Microsoft/vscode-pgsql
no source, not a promising privacy policy
I imagine this is part of Microsoft's plan to the vscode forks: open sourcing copilot to rally developers round it, along with pushing premium closed sourced extensions, which will almost definitely be incompatible with the forks.
> [...] all without ever leaving your favorite code editor.
That's cool. How do I get this into JetBrains IDEs?
This is Microsoft catching up to what JetBrains has had all along (in their paid products). There's DataGrip but I was also using all these features (minus the AI stuff) in IntelliJ Ultimate and PyCharm 5+ years ago.
The main difference is that JetBrains supports a bunch of databases, not just one.
Microsoft has had official Microsoft SQL Server, Azure Storage, CosmosDB and more extensions for VS Code for a long while now. The surprise is building an official branded Postgres extension rather than leaving that to all the (many) third party and/or open source extensions. I suppose it is a nice side-effect of "AI all the things" efforts that "GitHub Copilot support" was enough reason to grab resources for a postgres query editor and other tools.
(I thought it was particularly nice when you could install most to the VS Code DB extensions standalone as "Azure Data Studio"; you can build separate VS Code profiles for your coding and DBA hats of course, but it's not quite the same feel as launching a separate, dedicated application. Though "Azure Data Studio" was often overlooked because it worked just fine on non-Azure hosted databases.)
I feel like JetBrains' competing product is DataGrip, which even though I'm an Emacs user, I've happily paid for forever. I believe that if you just use the combined suite (IntelliJ?) then you get SQL completion based on your loaded database everywhere in the IDE. Something fun I accidentally learned this way is that the completion is so accurate I didn't even notice that I made a mistake in a database table I recently added. I used camelCase column names, which Postgres doesn't accept without quotes. Datagrip always added the quotes where I needed them, so I didn't notice, but it annoyed people on my team that used different database tools and they renamed them ;)
JetBrains has always done their completion / language integration differently than VSCode + LSP, but it seems to work well. My only complaint is that GoLand organizes imports differently than gopls / goimports, but my knowledge there is a few years out of date. I've worked on teams with a lot of GoLand users and nothing has really bothered me on this front recently, so they probably fixed it years ago.
> I believe that if you just use the combined suite (IntelliJ?) then you get SQL completion based on your loaded database everywhere in the IDE.
Yes, Intellj will recognize that the string in my code is an SQL statement and use auto complete and validation when it is connected to a database.
I've looked at the announcement and it seems that the functionality of Jetbrains is similar.
Out of curiosity: how do you set up your CI checks (linter, formatting) if the opensource tools behave differently from JetBrains'? Do you use Qodana[0]?
One benefit of using VSCode/Sublime Text/vim/emacs language integrations is that I roughly know what command to run to get the same results in the terminal as I get in the editor. With JetBrains does-it--all IDEs, I have no clue.
[0] https://www.jetbrains.com/help/qodana/getting-started.html
As far as I went with linting this sort of thing was to check that "gofmt -s" yields the code that's checked in. https://github.com/pachyderm/pachyderm/blob/master/src/inter...
It does not care if the imports are organized differently. I like to do:
import (
# standard libraries
"fmt"
"io"
# our stuff
"github.com/pachyderm/pachyderm/src/internal/whatever"
# packages
"example.com/foo/bar"
"github.com/whatever/whatever/foo"
)
You click on the button labeled database in the right-hand side panel.
You download Datagrip
Let me guess: proprietary, like Pylance, and unavailable in VSCodium?
Looks like it
That's the point of the VS Code strategy, yes.
I still think Jetbrains has the gold standard in IDE - Database interaction
Interesting in knowing why you think that
Because of a rich feature set and amazing integration with DB providers.
Good starting point: https://www.jetbrains.com/pages/intellij-idea-databases/
It even lints your SQL queries written in other languages. Truly gold standard.
And autocompletes, syntax highlight it. I couldn't imagine being without this.
Datagrip, as an extension, lets you work with SQL, highlighting, autocompletion, and more, inside non-SQL files, such as your programming language files. I think they call this 'language injection'.
I've been using DataGrip for a few weeks and admit it is a nice upgrade to DBVisualizer that I've been using for 10 years. The intellisense and features like being able to select the query in the current window are big time savers for me. I'm still on a trial and not certain I'll purchase it just because things are moving so fast in this field. I feel like not having it in my VSCode Agent loop is a huge negative at this point
Isn't PGAdmin good enough? Not hating, but I'm not a database guy either so just curious why create a VSCode extension for GUI stuff.
The longer I work in engineering the fewer windows I want open on my screen
If this is “PGAdmin, but 50% worse, but it just lives as another tab in your editor instead of another application” that is genuinely a huge win for me
Probably that’s because my day-to-day isn’t actually database administration, it’s application development where I need to be able to work with a database from time to time — so the preservation of context, however marginal, is really nice.
PgAdmin I think is more for working with data, I use Intellij DB editor mostly for writing complex SQL (with autocomplete, coloring, auto-formatting).
It also recognizes your regular programming code strings as SQL and can also autocomplete/highlight typos, but I don't use that, because we're using JOOQ that generates code stubs from live DB.
I love PGAdmin. For me it works very well. And then there are other tools when I need to work with schema etc. I always feel PGAdmin is an open source product that the team is most actively working on, always keep releasing fixes and updates. I don't want it in my IDE yet.
Few years ago PGAdmin was almost universally hated, especially in HN discussions. I think it was mostly around 3 -> 4 "upgrade". I see the sentiment has changed, although for me PGAdmin's UI is still very clunky and I avoid it if I can.
it's nice to have copilot in the db admin tool. I use pgadmin every day, and it's great, but lately i have found myself wishing to have AI support in writing certain queries.
Can you setup entra authentication with PgAdmin? I'm more of a MS Sql person so I don't know, but if not the security improvement from this would be a huge improvement
Has there been a final solution to the VSCode phoning-home problem?
Last time I tried to use it in my environment, it triggered too many 'external network requests' for the liking of our IT guy. We relegated it to the "Do Not Use" pile as a result.
Has this been resolved? Nobody in my team wants to use an IDE that sends data back to its masters ..
Ages behind Jetbrains and more of electron dumpster fire. No thanks
Glad to see them releasing it under an MIT license.
It's not. There is no actual code in the repo.
Super cool for them to finally add in VSCode. My team is trying to build even more here by building a dedicated SQL editor with a context aware AI copilot, and with sharing and collaboration so we don’t need to send queries in slack anymore :)
Check it out and get on the waitlist getgalaxy.io/explore/product-tour
happy to chat live with anyone if interested, support@getgalaxy.io
The GIF on your landing page is very jarring, with the constant zooming in and out. Way too fast to read what you're zooming into.
Your whole comment history seems to be spamming for your app.
Not a good way to generate potential user interest or generally good feelings towards it.
Just saying.
I agree, I immediately felt that their product was cheapened for me. The only other "newish" developer tool I used were Beekeeper and Bruno and that too only because there were a lot of users naturally shilling for them (like I am now) on HN.