From: Wolfgang Bumiller <w.bumiller@proxmox.com>
To: Gabriel Goller <g.goller@proxmox.com>
Cc: pbs-devel@lists.proxmox.com
Subject: Re: [pbs-devel] [PATCH pathpatterns v4] match_list: updated `matches()`, to only retrieve file mode if necessary
Date: Tue, 22 Aug 2023 15:07:03 +0200 [thread overview]
Message-ID: <nvvexy44tkm6mk7vbvg5ib6kx7wzvgueqlriaad7hjhq652hdn@gl6wf4kyrz22> (raw)
In-Reply-To: <20230821130826.147473-1-g.goller@proxmox.com>
LGTM now, 1 minor nit below since the other patch needs another version
anyway...
On Mon, Aug 21, 2023 at 03:08:25PM +0200, Gabriel Goller wrote:
> Updated `matches()` function, which now takes the `GetFileMode` trait and returns a
> Result. `GetFileMode` is a function that should return the `file_mode`. `matches()`
> will go through the patterns and match by path only until it finds a pattern which
> does not have `MatchFlag::ANY_FILE_TYPE`, in case it will call the
> `get_file_mode`, which will return a `file_mode`. This will ensure that the
> `get()` (which in our case executes `stat()`) will only be run once(at most),
> every pattern will only be processed once and that the order of the patterns
> will be respected. `GetFileMode` is also implemented for `Option<u32>` (so that
> we can simply pass options of the `file_mode`) and `u32` (returning
> `std::convert::Infallible`, so that we can pass a u32 mode directly).
>
> Signed-off-by: Gabriel Goller <g.goller@proxmox.com>
> ---
>
> changes v3:
> - changed `matches()` and `matches_exact()` to retrieve the file_mode lazily.
> Allowed these functions to take in a trait `GetFileMode`, so that we can pass
> closures and options that contain the `file_mode`.
>
> changes v2:
> - updated the `matches_path()` function to take a closure and renamed
> it to `matches_mode_lazy()`. This allows us to only match once, so we
> don't need to match before and after `stat()` (without and with
> `file_mode`) anymore.
>
> changes v1:
> - added `matches_path()` function, which matches by path only and returns an
> error if the `file_mode` is required.
>
>
> src/lib.rs | 72 +++++-----
> src/match_list.rs | 335 ++++++++++++++++++++++++++++++++++++++--------
> 2 files changed, 315 insertions(+), 92 deletions(-)
>
> diff --git a/src/match_list.rs b/src/match_list.rs
> index c5b14e0..69b1fdb 100644
> --- a/src/match_list.rs
> +++ b/src/match_list.rs
(...)
> @@ -383,30 +438,48 @@ where
> <&'a T as IntoIterator>::IntoIter: DoubleEndedIterator,
> <&'a T as IntoIterator>::Item: MatchListEntry,
> {
Since we have a T here already...
> - fn matches_do(&self, path: &[u8], file_mode: Option<u32>) -> Option<MatchType> {
> + fn matches<U, G>(&self, path: U, get_file_mode: G) -> Result<Option<MatchType>, G::Error>
maybe use `P` for the `path` ;-)
> + where
> + U: AsRef<[u8]>,
> + G: GetFileMode,
> + {
> // This is an &self method on a `T where T: 'a`.
> let this: &'a Self = unsafe { std::mem::transmute(self) };
>
> + let mut get_file_mode = Some(get_file_mode);
> + let mut file_mode = None;
> +
> for m in this.into_iter().rev() {
> - if let Some(mt) = m.entry_matches(path, file_mode) {
> - return Some(mt);
> + if file_mode.is_none() && m.entry_needs_file_mode() {
> + file_mode = Some(get_file_mode.take().unwrap().get()?);
> + }
> + if let Some(mt) = m.entry_matches(path.as_ref(), file_mode) {
> + return Ok(Some(mt));
> }
> }
> -
> - None
> + Ok(None)
> }
>
> - fn matches_exact_do(&self, path: &[u8], file_mode: Option<u32>) -> Option<MatchType> {
> + fn matches_exact<U, G>(&self, path: U, get_file_mode: G) -> Result<Option<MatchType>, G::Error>
^ here as well
> + where
> + U: AsRef<[u8]>,
> + G: GetFileMode,
> + {
> // This is an &self method on a `T where T: 'a`.
> let this: &'a Self = unsafe { std::mem::transmute(self) };
>
> + let mut get_file_mode = Some(get_file_mode);
> + let mut file_mode = None;
> +
> for m in this.into_iter().rev() {
> - if let Some(mt) = m.entry_matches_exact(path, file_mode) {
> - return Some(mt);
> + if file_mode.is_none() && m.entry_needs_file_mode() {
> + file_mode = Some(get_file_mode.take().unwrap().get()?);
> + }
> + if let Some(mt) = m.entry_matches_exact(path.as_ref(), file_mode) {
> + return Ok(Some(mt));
> }
> }
> -
> - None
> + Ok(None)
> }
> }
>
prev parent reply other threads:[~2023-08-22 13:07 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-08-21 13:08 Gabriel Goller
2023-08-21 13:08 ` [pbs-devel] [PATCH proxmox-backup v6] fix #4380: check if file is excluded before running `stat()` Gabriel Goller
2023-08-22 13:04 ` Wolfgang Bumiller
2023-08-22 14:07 ` Gabriel Goller
2023-08-22 13:07 ` Wolfgang Bumiller [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=nvvexy44tkm6mk7vbvg5ib6kx7wzvgueqlriaad7hjhq652hdn@gl6wf4kyrz22 \
--to=w.bumiller@proxmox.com \
--cc=g.goller@proxmox.com \
--cc=pbs-devel@lists.proxmox.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox