Post titles should include useful context.įor Rust questions, use the stickied Q&A thread.Īrts-and-crafts posts are permitted on weekends.Ĭriticism is encouraged, though it must be constructive, useful and actionable. For content that does not, use a text post to explain its relevance. Posts must reference Rust or relate to things using Rust.
#RUST SQLITE EXAMPLE CODE#
We observe the Rust Project Code of Conduct. Strive to treat others with respect, patience, kindness, and empathy. It looks highly promising though.Please read The Rust Community Code of Conduct The Rust Programming LanguageĪ place for all things related to the Rust programming language-an open-source systems language that emphasizes performance, reliability, and productivity. That said, the fact that sqlite-loadable is an unstable large set of unsafe code makes me not recommend it for general use.
#RUST SQLITE EXAMPLE PORTABLE#
It is designed after Rusqlite which means that pretty much all the work I did initally was portable within an hour. It took me a while to realise that Rusqlite was not able to handle loadable extensions which was a bit frustrating. In this naive implementation all values are TEXT but in a fancier version with a custom set of columns this would be the point to determine the right column affinity.Īnd that's it. #Īny extension requires an entry point with a shape as follows: # pub fn sqlite3_tomlvtab_init ( db : *mut sqlite3 ) -> Result. With that and the JSON functions and operators it is possible to query, normalise and manipulate the data freely. In other words, the extension should surface a virtual table constructor toml() with a parameter dirname to provide the root directory for locating TOML files.Īnd for each TOML file, create a row in the virtual table with its contents serialised as JSON.Ī future iteration could allow passing the list of expected columns instead but this approach keeps things simple with a fixed set of columns: filename and value. recipe.value, '$.ingredients ')) AS ingredient value, '$.name ') AS ingredient_nameįROM temp. The goal was to build a SQLite extension offering virtual table that would allow a user to query against a set of TOML files doing something like: CREATE VIRTUAL TABLE temp. #Ī virtual table for collections of TOML files So I built a SQLite extension that lets you query against a collection of TOML files using sqlite-loadable. Worth checking out sqlite-xsv and sqlite-regex as well as the article Introducing sqlite-loadable-rs: A framework for building SQLite Extensions in Rust. The author is using it to build a set of extensions already. Sqlite-loadable is a young (v0.0.5) attempt offering such feature using an interface very similar to the one offered by Rusqlite for building non-dynamic extensions. There is a long standing proposal (circa 2019) that suggests that eventually the project will allow for this feature but there is no expected delivery time.
This means that Rusqlite is not able to generate an artefact that can be dynamically loaded by, for example, the SQLite CLI or the Python SQLite module. However, as per v0.29, it only allows building extensions that are loadable by a Rust application using Rusqlite. In Rust, the main SQLite library is Rusqlite. The SQLite project offers a few examples of extensions, for example the FTS5 virtual table to enable full-text search.
Some examples are application-defined SQL functions, collating sequences, virtual tables and virtual file systems. SQLite has the ability to load extensions at runtime. This note captures my experience building a SQLite extension in Rust. A virtual table for collections of TOML files.
0 kommentar(er)
