From: Thomas Lamprecht <t.lamprecht@proxmox.com>
To: Proxmox Backup Server development discussion
<pbs-devel@lists.proxmox.com>,
Dominik Csapak <d.csapak@proxmox.com>
Subject: Re: [pbs-devel] [PATCH proxmox-backup 1/3] tools/systemd/time: implement some Traits for TimeSpan
Date: Wed, 17 Mar 2021 20:35:19 +0100 [thread overview]
Message-ID: <4d7e453a-ad33-07a9-a319-07276955fb5f@proxmox.com> (raw)
In-Reply-To: <20210316115623.9368-1-d.csapak@proxmox.com>
On 16.03.21 12:56, Dominik Csapak wrote:
> namely
> * From<Duration> (to convert easily from duration to timespan)
> * Display (for better formatting)
>
> Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
> ---
> if wanted, we can optimize the display trait a bit further, e.g.
> only showing the biggest two units instead
>
> src/tools/systemd/time.rs | 84 +++++++++++++++++++++++++++++++++++++++
> 1 file changed, 84 insertions(+)
>
> diff --git a/src/tools/systemd/time.rs b/src/tools/systemd/time.rs
> index 7cc42415..75fb0ea2 100644
> --- a/src/tools/systemd/time.rs
> +++ b/src/tools/systemd/time.rs
> @@ -141,6 +141,90 @@ impl From<TimeSpan> for f64 {
> }
> }
>
> +impl From<std::time::Duration> for TimeSpan {
> + fn from(duration: std::time::Duration) -> Self {
> + let mut duration = duration.as_nanos();
> + let nsec = (duration % 1000) as u64;
> + duration /= 1000;
> + let usec = (duration % 1000) as u64;
> + duration /= 1000;
> + let msec = (duration % 1000) as u64;
> + duration /= 1000;
> + let seconds = (duration % 60) as u64;
> + duration /= 60;
> + let minutes = (duration % 60) as u64;
> + duration /= 60;
> + let hours = (duration % 24) as u64;
> + duration /= 24;
> + let years = (duration as f64 / 365.25) as u64;
> + let ydays = (duration as f64 % 365.25) as u64;
> + let months = (ydays as f64 / 30.44) as u64;
> + let mdays = (ydays as f64 % 30.44) as u64;
> + let weeks = mdays / 7;
> + let days = mdays % 7;
> + Self {
> + nsec,
> + msec,
> + usec,
nit:
nano -> milli -> micro? While it works correct, that order above seems just wrong...
> + seconds,
> + minutes,
> + hours,
> + days,
> + weeks,
> + months,
> + years,
> + }
> + }
> +}
> +
> +impl std::fmt::Display for TimeSpan {
> + fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> Result<(), std::fmt::Error> {
> + let mut single = true;
> + if self.years > 0 {
> + write!(f, "{}y", self.years)?;
> + single = false;
> + }
> + let write_space = |single: bool, f: &mut std::fmt::Formatter| -> Result<bool, std::fmt::Error> {
> + if !single {
> + write!(f, " ")?;
> + }
> + Ok(false)
> + };
No sure if something like (untested):
let mut first = true;
let mut do_write = |v: u64, unit: &str| -> Result<(), std::fmt::Error> {
if !first {
write!(f, " ")?;
first = false;
}
write!(f, "{}{}", v, unit)
}
if self.months > 0 {
do_write(self.months, "m")?;
}
...
would be nicer IMO, especially as the "single" handling reads a bit weird.
Maybe the seconds field makes this not really feasible though...
> + if self.months > 0 {
> + single = write_space(single, f)?;
> + write!(f, "{}m", self.months)?;
> + }
> + if self.weeks > 0 {
> + single = write_space(single, f)?;
> + write!(f, "{}w", self.weeks)?;
> + }
> + if self.days > 0 {
> + single = write_space(single, f)?;
> + write!(f, "{}d", self.days)?;
> + }
> + if self.hours > 0 {
> + single = write_space(single, f)?;
> + write!(f, "{}h", self.hours)?;
> + }
> + if self.minutes > 0 {
> + single = write_space(single, f)?;
> + write!(f, "{}min", self.minutes)?;
> + }
> + let seconds = self.seconds as f64 + (self.msec as f64 / 1000.0);
> + if seconds >= 0.1 {
> + write_space(single, f)?;
> + if seconds >= 1.0 || !single {
> + write!(f, "{:.0}s", seconds)?;
> + } else {
> + write!(f, "{:.1}s", seconds)?;
> + }
> + } else if single {
> + write_space(single, f)?;
> + write!(f, "<0.1s")?;
> + }
> + Ok(())
> + }
> +}
>
> pub fn verify_time_span(i: &str) -> Result<(), Error> {
> parse_time_span(i)?;
>
next prev parent reply other threads:[~2021-03-17 19:35 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-16 11:56 Dominik Csapak
2021-03-16 11:56 ` [pbs-devel] [PATCH proxmox-backup 2/3] server/email_notifications: do not panic on template registration Dominik Csapak
2021-03-17 19:38 ` Thomas Lamprecht
2021-03-18 9:57 ` Dominik Csapak
2021-03-18 10:21 ` Thomas Lamprecht
2021-03-18 10:31 ` Dominik Csapak
2021-03-18 11:13 ` Thomas Lamprecht
2021-03-16 11:56 ` [pbs-devel] [PATCH proxmox-backup 3/3] api2/tape/backup: include a summary on notification e-mails Dominik Csapak
2021-03-17 19:35 ` Thomas Lamprecht [this message]
2021-03-18 9:49 ` [pbs-devel] [PATCH proxmox-backup 1/3] tools/systemd/time: implement some Traits for TimeSpan Dominik Csapak
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=4d7e453a-ad33-07a9-a319-07276955fb5f@proxmox.com \
--to=t.lamprecht@proxmox.com \
--cc=d.csapak@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