Closed Enselic closed 1 year ago
That was only a temporary issue because 1.46 had a terrible performance regression. I don't support old Rust versions otherwise, and crates-index won't work with <1.57.
However, I don't think MPL is a problem. AFAIK it's not viral, and you don't really need to care about it unless you modify the smartstring crate.
I don't need smol_str
's special handling of whitespace — spaces aren't even allowed in any of the strings I use it for, so smartstring's extra character of storage is a better trade-off.
Thanks for the response.
After learning more about MPL 2.0 and pinging some users of my project, this indeed seems fine. Good point about smartstring
having a technical advantage over smol_str
for the uses-cases of crates-index
, i.e. it is not only a matter of licensing.
Closing.
smartstring turned out to be problematic: #92
Hi and thank you for this very useful project :pray:
I noticed that in https://github.com/frewsxcv/rust-crates-index/commit/7e945b37dc94a442438bd1d8c2da9ee81ac3f0ad,
smol_str
was replaced bysmartstring
for MSRV reasons.What is the MSRV for this project? Is it higher than 1.45? If so, we should be able to go back to
smol_str
. I can runcargo test
onsmol_str
with 1.46, but not 1.45. So I think the effective MSRV ofsmol_str
0.1.23
is 1.46.The downside of
smartstring
is that it is licensed under MPL 2.0, which has a copyleft flavor to it, which some can find concerning.smol_str
isMIT OR APACHE-2.0
.Inte: The officiell MSRV policy of
smol_str
is "latest stable". Maybe that is a problem. Another option to this PR could be to add a feature wheresmol_str
ORsmartstring
can be chosen.Thank you in advance for any guidance or pointers.