|
Greg Utas wrote: I don't know SQL
You're a fortunate soul. I've worn many hats in my development lifetime. "DBA" was one of my least favorite.
Real programmers use butterflies
|
|
|
|
|
I think you just outed where in the US you're originally from with that title. Ssshhh, don't leak our secret 
|
|
|
|
|
If you're gauging based on the title, I'm probably not where you think I'm from but it means what you think it does. I also use various britishism's like "whilst"
I steal vernacular from wherever. I'm a thief like that.
Real programmers use butterflies
|
|
|
|
|
It's a shame you're targeting MS SQL, it is weak in many areas compared to other SQL offerings. Postgres supports regex natively, as well as having a bunch more capability with string manipulation functions.
Is there an option for you to use .Net assemblies in MS SQL, or perhaps the R or Python environments? It seems MS is ignoring extending the base SQL capability, but offering these extensions.
Regardless, kudos for the tenacity required to do what your doing.
How are you finding the performance of the regex proc's in SQL?
Also, remember SQL is designed way back, before syntax highlighting/intellisense and many other modern expectations that would influence its design were around.
It's time to create SQLNext, and I've seen the odd concept put forward, but it would be really hard to shift the masses now.
Perhaps it'd best be done with a translation from SQLNext to the various SQL variants.
But ideally would have to exist closer to the database than the client application and eventually become a native database feature. Unfortunately, would be easiest to place in the client application - I'd hate to get into any DB Driver level stuff. Remember ODBC - yuck.
|
|
|
|
|
I intend to target others. I had to start somewhere, and I already had MSSQL, it's fairly popular, I'm familiar with it, and its many limitations, which actually made it a good place to start.
After all, I stuck as close to the SQL spec as a could, so I can port it to other databases, and I will be adding some sort of "flavor" switch to the app to specify which SQL to target. It's impossible to target SQL92 for this (i think) or I would.
Real programmers use butterflies
|
|
|
|
|
I didn't even respond to the rest of your questions! Whoops, let me try again.
I haven't tested the performance specifically. I wasn't impressed with the execution plan for the table version though. The compiled ones should perform "okay" once MSSQL gets done putting them into native code, but it won't be near as fast as outside of a DB.
I wouldn't use .NET's engine because it's not available everywhere. With Reggie you write the spec once, and it runs identically across any supported target. That's important for business schemas or even just application behavior consistency, if you're doing like, triple tier validation for example. Also I don't think you can use the CLR in the database without turning it on explicitly, and I could see a lot of people (if not most) leaving it off.
The takeaway here is Reggie will probably run where a .NET solution wouldn't, and it won't perform as well in the DB, unless you take the C# code reggie generates and run *that* in the database, which i suppose is doable as well.
So for a DB, this is best used for validation during updates that aren't intended to be super fast, full metal writes, though you can probably use this modestly in such a scenario for normalizing snatches of JSON and such.
I think I may solve my SQL problems with having structured chunks of sql that can render to multiple targets (like "advance the input position and move along a transition, with capture")
I've never heard of SQLNext.
Real programmers use butterflies
|
|
|
|
|
NeverJustHere wrote: It's a shame you're targeting MS SQL, it is weak in many areas compared to other SQL offerings. Postgres supports regex natively, as well as having a bunch more capability with string manipulation functions.
MS SQL supports that too.
NeverJustHere wrote: Is there an option for you to use .Net assemblies in MS SQL, or perhaps the R or Python environments? It seems MS is ignoring extending the base SQL capability, but offering these extensions.
Yes, you can do that (CLR), but local security restrictions may prevent you from doing so (like us - we can't do CLR because the security nazis said we couldn't).
NeverJustHere wrote: Also, remember SQL is designed way back, before syntax highlighting/intellisense and many other modern expectations that would influence its design were around.
Syntax highlighting is implemented by the editor, not the language. Syntax highlighting as been available in SSMS since at least 2007.
ANSI SQL has been around for years, but as corporate implementations are wont to have, extensions and variations to the language creep in (hence the "almost compatible" nature of many common functions - t's kinda like browsers). SQL variants are also updated at different times, so they tend to leap-frog each other in capabilities.
There's nothing really wrong with any of the SQL variants. You may have your preferences, but in the end, you work in the one that is responsible for your paycheck.
I think there are more important things to be concerned about - like why can't we use c# as a scripting language instead of javascript in ASP.NET web pages?
".45 ACP - because shooting twice is just silly" - JSOP, 2010
-----
You can never have too much ammo - unless you're swimming, or on fire. - JSOP, 2010
-----
When you pry the gun from my cold dead hands, be careful - the barrel will be very hot. - JSOP, 2013
|
|
|
|
|
Quote: Syntax highlighting is implemented by the editor, not the language. Syntax highlighting as been available in SSMS since at least 2007.
Yep, syntax highlighting is fine. It's intellisense where it gets messed up. Select can't list relevant columns because the referenced tables haven't been entered yet. It's why LINQ is the other way around. FROM .... SELECT ...
Quote: like why can't we use c# as a scripting language instead of javascript in ASP.NET web pages?
You can - it's just called Blazor rather than ASP. Just starting exploring it myself.
|
|
|
|
|
NeverJustHere wrote: It's intellisense where it gets messed up. Select can't list relevant columns because the referenced tables haven't been entered yet. It's why LINQ is the other way around. FROM .... SELECT ...
That's not SSMS, it's the nature of the language. They can't do SQL radically different from the ANSI standard because nobody would use the product. Yeah, it's annoying, but you learn to deal with it, like with any other tool.
I personally think that SQL does stuff backwards from the way your typical developer thinks about coding. It's unfortunate that a lot of DBAs are not really qualified to write SQL code - and I've actually been told by a DBA (when asked about a SQL Server-specific technique) that he "wasn't a developer and couldn't help me". As a result, developers are forced into the SQL dev role as a result.
".45 ACP - because shooting twice is just silly" - JSOP, 2010
-----
You can never have too much ammo - unless you're swimming, or on fire. - JSOP, 2010
-----
When you pry the gun from my cold dead hands, be careful - the barrel will be very hot. - JSOP, 2013
|
|
|
|
|
honey the codewitch wrote: It's impossible to remember it without using it every day
I USE SQL every day a week (Ok 5 out of 7)... Still there are things I have to look-up...
"The only place where Success comes before Work is in the dictionary." Vidal Sassoon, 1928 - 2012
|
|
|
|
|
honey the codewitch wrote: The syntax is inexplicable
honey the codewitch wrote: It's impossible to remember it without using it every day,
As a long time web developer I feel that's true for most all languages. As a web developer I do html, css, javascript, jquery, xml, json, c#, and t-sql. I often have to google for syntax. But SQL is the one I probably look up least often.
|
|
|
|
|
The web is even worse.
And it's not true of most languages. Duck typing is for people that like spending all day debugging a typo. Hard types are for closers. Compilers keep people honest.
Real programmers use butterflies
|
|
|
|
|
I'm thinking that the people that designed COBOL had a hand in SQL?
The less you need, the more you have.
Even a blind squirrel gets a nut...occasionally.
JaxCoder.com
|
|
|
|
|
Asking SQL to be like C# ... One wonders why one doesn't simply do C# and LINQ to SQL.
I'm sure SQL was a lot purer 40 years ago; before the community got hold of it. The pureness is still there, if you choose to ignore the other stuff.
Quote: 1999 SQL:1999 SQL3 Added regular expression matching, recursive queries (e.g. transitive closure), triggers, support for procedural and control-of-flow statements, nonscalar types (arrays), and some object-oriented features (e.g. structured types), support for embedding SQL in Java (SQL/OLB) and vice versa (SQL/JRT)
It was only in wine that he laid down no limit for himself, but he did not allow himself to be confused by it.
― Confucian Analects: Rules of Confucius about his food
modified 1-Nov-21 13:20pm.
|
|
|
|
|
The point of the exercise (in my case) was to get SQL to do non-trivial field validation on strings - to verify at the final tier as well as the middleware.
Real programmers use butterflies
|
|
|
|
|
IMHO...
SQL is for optimized for data storage and retrieval, not field validation (beyond making sure a string isn't too long for the column you're storing it in). The middleware is always going to be better at transforming/validating the data. By the time you get to saving the data, it should already be valid because at some point,you have to just "go with it" and assume the data you're working with is okay.
".45 ACP - because shooting twice is just silly" - JSOP, 2010
-----
You can never have too much ammo - unless you're swimming, or on fire. - JSOP, 2010
-----
When you pry the gun from my cold dead hands, be careful - the barrel will be very hot. - JSOP, 2013
|
|
|
|
|
Which is fine until your middleware is compromised or buggy.
Plus I like wherever possible to keep the rules enforcing the integrity of the data in the database itself, that way the database is a complete standalone unit in terms of operating to a spec.
I know what you're saying. I hear you. I just also think that when you can get away with it, you *should* be validating in the DB tier as well.
Real programmers use butterflies
|
|
|
|
|
VS2022 has preview version. I have the latest preview.
It also has RC which, it turns out, are different product to preview. Now I have preview 7. And RC3 was just released at the same time. And what is the difference between both I wonder?
It feels like it was just meant to confused me... :/
|
|
|
|
|
Now you know why I wait.
Real programmers use butterflies
|
|
|
|
|
Such wisdom and patience! 
|
|
|
|
|
First service pack or second?
I used to enjoy riding the bleeding edge, but it's been decades since I enjoyed being a beta (or alpha) tester!
|
|
|
|
|
Usually the 1st. I'm not ultra paranoid. I used to avoid anything with version "4" in it coming from microsoft as to this day .NET 4.x is maybe the only version 4 of anything they've produced that was solid, but medication helps.
Real programmers use butterflies
|
|
|
|
|
"Preview" means "Riddled with bugs we're working on"
"RCx" means "Riddled with bugs we're not going to fix."
"I have no idea what I did, but I'm taking full credit for it." - ThisOldTony
"Common sense is so rare these days, it should be classified as a super power" - Random T-shirt
AntiTwitter: @DalekDave is now a follower!
|
|
|
|
|
|
The RC version is meant for production use and will continue to get the production releases going forward. These are fully supported by Microsoft.
The Preview version is not for production use and will get continue to get future preview releases of the product going forward. Preview releases may have experimental features that may not make it to the release version.
|
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
