Every time I use it I feels like I’m going back to the 90s. No variables, no functions; Oh but you can do a CTE or subquery…👍
UNION ALL, UNION ALL, UNION ALL… “There’s got to be a better way, surely…”
looks up better way
“Oh, what the fuck?!.. Nope, this will just be quicker…” UNION ALL, UNION ALL, UNION ALL…
Join in a table sharing column names… Everything breaks. You gotta put the new prefixes in front of all the headers you called in now. In every select, in every where, etc… Which is weird because that kinda works like a variable and it’s fine…
“When you see this little piece of text, it means all this, got it?”
“Okay. Yep. Easy.”
“So why can’t you do that with expressions?”
SQL SCREAMS MANICALLY
“Okay, okay, okay!.. Jesus…”
And then you try put a MAX in a where and it won’t let you because you gotta pull all the maxes out in their own query, make a table, join them in, and use them like a filter…
I hate it. It has speed, when you can finally run the script, but everything up to that is so…ugh.
It seems that you need to get better. There are plenty of valid complaints against SQL, but your problems seem to be all due to lack of familiarity.
No variables, no functions; Oh but you can do a CTE
Yeah, CTEs are more expressive than variables. And as somebody pointed, every database out there supports functions, you may want to look how they work.
UNION ALL, UNION ALL, UNION ALL… “There’s got to be a better way, surely…”
What do you mean by a “better way”? Union all is a perfectly valid operation.
And then you try put a MAX in a where and it won’t let you because you gotta pull all the maxes out in their own query, make a table, join them in, and use them like a filter…
Every major SQL implementation includes both of those things. Of course, it’s rarely needed or desirable if you know how to properly write SQL.
“So why can’t you do that with expressions?”
You can alias expressions.
And then you try put a MAX in a where and it won’t let you because you gotta pull all the maxes out in their own query, make a table, join them in, and use them like a filter…
Wtf are you talking about? For one, filtering by the output of an aggregate is what the HAVING clause is for. But even if that didn’t exist, you could just use a subquery instead. You don’t need to make table…
Tbh it just sounds like you don’t know SQL very well. Which is fine, but doesn’t make for a very compelling criticism. SQL does have warts (even though it’s great overall), but none of what you described are real problems.
While I agree that “SQL Enjoyer” seems like a weird category, I personally love SQL. I’ve been using it professionally for over 20 years, and I’ve yet to encounter a more elegant, efficient, and practical language for handling data in a relational database. Every attempt I’ve seen to replace it with something simpler has fallen far short.
Which database systems were you dealing with, that didn’t allow variables? My personal favorite is PostgreSQL, which does allow them on scripting languages, such as PLPGSQL.
See, I don’t have to worry about such details. I work in corporate software dev, which means that everything is an MSSQL database where most of the tables contain only an ID of a table-specific format and a JSON blob. Why use an ORM when you can badly reimplement NoSQL in a relational database instead?
It doesn’t arbitrarily double rows or something. For each row in the relation on the left of the join, it will produce 1 or more rows depending on how many rows in the relation on the right of the join match the join condition. The output relation of the join may have duplicate rows depending on the contents of each joined relation as well as what columns you are projecting from each.
If you want to remove duplicates, that’s what DISTINCT is for.
SQL enjoyer?
Every time I use it I feels like I’m going back to the 90s. No variables, no functions; Oh but you can do a CTE or subquery…👍
UNION ALL, UNION ALL, UNION ALL… “There’s got to be a better way, surely…”
looks up better way
“Oh, what the fuck?!.. Nope, this will just be quicker…” UNION ALL, UNION ALL, UNION ALL…
Join in a table sharing column names… Everything breaks. You gotta put the new prefixes in front of all the headers you called in now. In every select, in every where, etc… Which is weird because that kinda works like a variable and it’s fine…
“When you see this little piece of text, it means all this, got it?”
“Okay. Yep. Easy.”
“So why can’t you do that with expressions?”
SQL SCREAMS MANICALLY
“Okay, okay, okay!.. Jesus…”
And then you try put a MAX in a where and it won’t let you because you gotta pull all the maxes out in their own query, make a table, join them in, and use them like a filter…
I hate it. It has speed, when you can finally run the script, but everything up to that is so…ugh.
It seems that you need to get better. There are plenty of valid complaints against SQL, but your problems seem to be all due to lack of familiarity.
Yeah, CTEs are more expressive than variables. And as somebody pointed, every database out there supports functions, you may want to look how they work.
What do you mean by a “better way”? Union all is a perfectly valid operation.
Window functions exist.
Personally I feel like SQL syntax is upside down, and things are used before they are defined.
SELECT a.id -- what the fuck is a? , a.name , b.city -- and b?? from users a -- oh join city b on a.id = b.user_id -- oh here's bI’d expect it to instead be like
From users a join city b on a.id = b.user_id SELECT a.id, a.name, b.cityEvery major SQL implementation includes both of those things. Of course, it’s rarely needed or desirable if you know how to properly write SQL.
You can alias expressions.
Wtf are you talking about? For one, filtering by the output of an aggregate is what the
HAVINGclause is for. But even if that didn’t exist, you could just use a subquery instead. You don’t need to make table…Tbh it just sounds like you don’t know SQL very well. Which is fine, but doesn’t make for a very compelling criticism. SQL does have warts (even though it’s great overall), but none of what you described are real problems.
While I agree that “SQL Enjoyer” seems like a weird category, I personally love SQL. I’ve been using it professionally for over 20 years, and I’ve yet to encounter a more elegant, efficient, and practical language for handling data in a relational database. Every attempt I’ve seen to replace it with something simpler has fallen far short.
Which database systems were you dealing with, that didn’t allow variables? My personal favorite is PostgreSQL, which does allow them on scripting languages, such as PLPGSQL.
See, I don’t have to worry about such details. I work in corporate software dev, which means that everything is an MSSQL database where most of the tables contain only an ID of a table-specific format and a JSON blob. Why use an ORM when you can badly reimplement NoSQL in a relational database instead?
hey hey, there there. don’t worry. most of the major NoSQL DBs implement just as horrible of travesties
Yep.
PostgreSQL is where its at, everybody else just hasn’t figured that out yet.
LEFT JOIN
Includes empty entries, doubles others.
…
It sure is long due for an overhaul.
That’s the whole point of a left join? Anything else wouldn’t be a left join anymore.
Well I didn’t expect doubles. I’m sure not an expert.
It doesn’t arbitrarily double rows or something. For each row in the relation on the left of the join, it will produce 1 or more rows depending on how many rows in the relation on the right of the join match the join condition. The output relation of the join may have duplicate rows depending on the contents of each joined relation as well as what columns you are projecting from each.
If you want to remove duplicates, that’s what
DISTINCTis for.Thanks, I will kot forget that the next time I have to do SQL!
Still wild there are no simpler language that have grown in popilarity for databases though.