Closed lberezy closed 3 years ago
As an aside, I would recommend a full cargo fmt --all
across the project to ease further contributions.
@lberezy Thanks, this looks great!
Thanks for pointing out the issue with generating IDs. I haven't been following the development of sled (or tantivy) that closely recently, but it looks like this is now fixed by allowing TransactionalTree
to generate IDs (see https://github.com/spacejam/sled/pull/1146). I'll merge your PR, make this change myself, than release
I think I'd cargo fmtted everything, but I will be sure to check in my .rustfmt.toml
when I bump the version and release. Thanks!
@lberezy Oh, nevermind, I see you fixed it so the TransactionalTree
generates the IDs. Great!
This PR upgrades the versions of
tantivy
andsled
to their latest respective versions, addressing the issue where the current state of the crate is unable to build. This also removes the need to patch thesled
crate as those changes have been upstreamed.I've also upgrade and moved the
tempfile
dependency to a dev-dependency as it's only used in an example.There were a few minor breaking changes to both
tantivy
andsled
. I'm not 100% familiar with these crates so I have only done work to get the example to build and run, so not sure about any subtle bugs I may have introduced. My main concern is with how IDs are now generated from aTransactionalTree
, required to not hang inside of a transaction context.