Mercurial > hg
view rust/hg-core/Cargo.toml @ 46472:98e39f04d60e
upgrade: implement partial upgrade for upgrading persistent-nodemap
Upgrading repositories to use persistent nodemap should be fast and easy as it
requires only two things:
1) Updating the requirements
2) Writing a persistent-nodemap on disk
For both of the steps above, we don't need to edit existing revlogs.
This patch makes upgrade only do the above mentioned two steps if we are
only upgarding to use persistent-nodemap feature.
Since `nodemap.persist_nodemap()` assumes that there exists a nodemap file for
the given revlog if we are trying to call it, this patch adds `force` argument
to create a file if does not exist which is true in our upgrade case.
The test changes demonstrate that we no longer write nodemap files for manifest
after upgrade which I think is desirable.
Differential Revision: https://phab.mercurial-scm.org/D9936
author | Pulkit Goyal <7895pulkit@gmail.com> |
---|---|
date | Mon, 01 Feb 2021 00:02:00 +0530 |
parents | 2e2033081274 |
children | 2845892dd489 |
line wrap: on
line source
[package] name = "hg-core" version = "0.1.0" authors = ["Georges Racinet <gracinet@anybox.fr>"] description = "Mercurial pure Rust core library, with no assumption on Python bindings (FFI)" edition = "2018" [lib] name = "hg" [dependencies] bytes-cast = "0.1" byteorder = "1.3.4" derive_more = "0.99" im-rc = "15.0.*" lazy_static = "1.4.0" memchr = "2.3.3" rand = "0.7.3" rand_pcg = "0.2.1" rand_distr = "0.2.2" rayon = "1.3.0" regex = "1.3.9" twox-hash = "1.5.0" same-file = "1.0.6" crossbeam-channel = "0.4" micro-timer = "0.3.0" log = "0.4.8" memmap = "0.7.0" zstd = "0.5.3" rust-crypto = "0.2.36" format-bytes = "0.1.2" # We don't use the `miniz-oxide` backend to not change rhg benchmarks and until # we have a clearer view of which backend is the fastest. [dependencies.flate2] version = "1.0.16" features = ["zlib"] default-features = false [dev-dependencies] clap = "*" pretty_assertions = "0.6.1" tempfile = "3.1.0" [features] # Use a (still unoptimized) tree for the dirstate instead of the current flat # dirstate. This is not yet recommended for performance reasons. A future # version might make it the default, or make it a runtime option. dirstate-tree = []