Earlier quoted context omitted.
You can google SQL grammar. But here is the 2025: https://standards.iso.org/iso-iec/9075/-2/ed-6/en/
Thank you! My Google foo did not find this.
Google search sucks really bad now for very specific searches.
Show HN: SQL-tString a t-string SQL builder in Python
41–43 of 43 posts
Re: Show HN: SQL-tString a t-string SQL builder in Python
#42The columns feature seems somewhat dangerous at first glance. What if I wanted different sets of allowed columns for different parts of the query? Would I have to provide the superset of columns supported in both subqueries / tables?
The `sql_context` feature is meant to protect against SQL injection. I think for your usecase you would need to supply all columns, but you could used nested t-strings included based on whatever logic you need.
Re: Show HN: SQL-tString a t-string SQL builder in Python
#43Earlier quoted context omitted.
I might have misunderstood your point, but scoping isn’t related to what I’m trying to say. What I’m saying is that, regardless of how it works, I don’t think string templating for SQL is a good idea because it looks almost exactly like string concatenation. It makes more difficult to distinguish beteeen the right approach and the wrong approach (or learn about it)
No it was me who misunderstood you. And I kind of agree. I've never been a fan of tagged template literals. It gives you no autocompletion, no type checking, no syntax highlighting, nothing. And it requires a runtime string parser. I get why people like it, and maybe it's fine if you don't need those things and don't mind the cost of runtime parsing, but I need them and I do mind.
It really does not need runtime string parsing unless you want to do some kinds of manipulation. I built Notion’s sql`…` literal and “all” it does is concat the string parts arrays and concat the argument arrays so you can pass { query: string, args: PostgresArg[] } to the database. The database is the only one who needs to parse the query string.
As for syntax highlighting, that’s available in VS Code, we auto install the appropriate extension.