Some things can be done using existing built-in functions, such as the FORMAT (or PRINTF) function, although some do not.
I have also made some SQLite extensions, such as one to produce PostScript tokens as output, as well as a PIPE function (uses an external program to process the data), 2D CSV, and others.
In some cases, things are better done by things to be made in SQLite itself.
For example, you cannot use ORDER BY in non-window aggregate functions (even though PostgreSQL allows it); this may be useful with some functions, such as GROUP_CONCAT. I would want them to fix this, but they did not do this.
I have written a patch for SQLite to work with non-Unicode text properly. One feature of SQLite is that it uses Unicode, so I fixed it so that it uses non-Unicode. I did not implement a patch of the other one, since that would require remaking the amalgamation, I think.
go-sqlite3-stdlib is influenced by sqlean but is written in Go (not C) so should be easier to contribute to, has 95% test coverage, and is easier to integrate with Go programs than C extensions.
I have also made some SQLite extensions, such as one to produce PostScript tokens as output, as well as a PIPE function (uses an external program to process the data), 2D CSV, and others.
In some cases, things are better done by things to be made in SQLite itself.
For example, you cannot use ORDER BY in non-window aggregate functions (even though PostgreSQL allows it); this may be useful with some functions, such as GROUP_CONCAT. I would want them to fix this, but they did not do this.
I have written a patch for SQLite to work with non-Unicode text properly. One feature of SQLite is that it uses Unicode, so I fixed it so that it uses non-Unicode. I did not implement a patch of the other one, since that would require remaking the amalgamation, I think.