improve human readable output by eworm-de · Pull Request #459 · RsyncProject/rsync

eworm-de · 2023-03-21T09:45:52Z

Calculate in a loop, use dynamic precision.

eworm-de · 2023-04-29T19:00:20Z

As there is a new pre-release... Can I have a quick feedback and state on this?

WayneD · 2023-05-01T00:48:36Z

The current release only has bugfixes in the main code (it's only the adjunct scripts & manpages that got enhancements).

I'm very cautious in making output-affecting changing, but will be giving this a closer look.

eworm-de · 2023-05-02T08:52:59Z

I can understand you are cautious about making output-affecting changing, guess this is due to breaking scripts that expect specific output?
My changes effect human readable output only, though. 😉 I hope this will not break anything.

tridge · 2024-04-06T20:16:51Z

@eworm-de could you add a description of the before and after of what this change does? Needs a bit more context than "improve" :-)

eworm-de · 2024-04-06T21:20:52Z

The commit messages have the details. Are you missing something there or just here in the pull request?

eworm-de · 2024-04-07T18:02:23Z

This is the output of git log master..:

commit 75ea671e822b347cd4a6656d9913eb9fa89ff600 (HEAD -> human)
Author: Christian Hesse <mail@eworm.de>
Date:   Thu Mar 23 10:33:33 2023 +0100

    human-readable: add an "i" to unit to indicate binary (1024) base

commit 4f215ad2cf7b0dff74db74202412793a9947cadb
Author: Christian Hesse <mail@eworm.de>
Date:   Wed Mar 22 11:07:15 2023 +0100

    human-readable: also handle num < mult with the same code
    
    ... just make sure no precision is added.

commit d4784655035dcfdf02efb30164b181f47546713b
Author: Christian Hesse <mail@eworm.de>
Date:   Tue Mar 21 10:17:09 2023 +0100

    human-readable: use dynamic precision length
    
    Let's lower precision for huge numbers.
    
    The output used to be:
    
      3.45M -> 46.73M -> 523.11M -> 1.24G -> ...
    
    With this change the code always gives the three most significant digits:
    
      3.45M -> 46.7M -> 523M -> 1.24G -> ...

commit 53e4e68e9d5d195679db74e178c4d9e915094236
Author: Christian Hesse <mail@eworm.de>
Date:   Tue Mar 21 10:13:47 2023 +0100

    human-readable: calculate numbers in a loop
    
    This drops a lot of `else if` blocks and extends units
    by "E", "Z" & "Y".

tridge · 2024-04-07T21:23:25Z

@eworm-de ok, thanks, I normally expect it in the PR message
@WayneD are we concerned about users having parses these messages? Also, should we have a unit test for this?

eworm-de · 2024-04-09T08:18:01Z

Just added another commit:

commit 7f26b1d2dd34ec39a566d0ba42a43bf22d6d5c46 (HEAD -> human, origin/human)
Author: Christian Hesse <mail@eworm.de>
Date:   Tue Apr 9 09:56:39 2024 +0200

    human-readable: also use it to format rate
    
    Let's also simplify the code for rate in progress, and benefit from
    same functionality.

eworm-de · 2024-04-12T05:44:52Z

are we concerned about users having parses these messages?

I still stand by my point: This is about human readable output. We should not be worried too much about compatibility with parsers.

eworm-de · 2024-05-15T14:14:02Z

Any news on this? Does it help if I split this across several pull requests?

steadytao

Thanks for sticking with this. I like the direction of making the formatter less hand-coded but I would request changes on the current patch.

Alongside my inline comments, the manpage progress examples are now stale and the current tests dont appear to cover the exact human-readable/progress output being changed here. Given this changes user-visible output, I think it needs focused formatter/progress coverage before merge alongside the bug fixes.

steadytao

Quick overall thing as well as the inlines. The manpage progress examples are still stale and the tests still only check very broad output properties. Since this PR intentionally changes user-visible formatting, I think it needs very focused tests for the formatter and at least one exact progress/final-stats shape before merge.

eworm-de · 2026-06-10T20:05:04Z

The manpage progress examples are still stale

Ups, I was not aware that the dynamic precision never made it into the man page. Updated... Is that sufficient?

and the tests still only check very broad output properties. Since this PR intentionally changes user-visible formatting, I think it needs very focused tests for the formatter and at least one exact progress/final-stats shape before merge.

Agreed. But the test suite is (mostly?) python, and I am not very familiar with that. Any help appreciated...

steadytao

LGTM. Awaiting @tridge review.

eworm-de · 2026-06-11T08:32:26Z

Oh, you pushed commit into my pull request. Gladly I noticed before I force-pushed my rebase. 😄️

Some of your changes are more cleanups for my commits. Is it ok for you if I clean up the history a bit?

Also some of the code is heavily influenced by you. Are you fine if I add Co-authored-by for you into my commits?

steadytao · 2026-06-11T08:59:28Z

Some of your changes are more cleanups for my commits. Is it ok for you if I clean up the history a bit?

Feel happy to. You can even make changes if you notice anything. It is your branch 😁

Also some of the code is heavily influenced by you. Are you fine if I add Co-authored-by for you into my commits?

I wouldn't mind either way but thank you for the offer, of course you can!

This drops a lot of `else if` blocks and extends units by "E" (Exa), which is 2^60 and thus the last to fit into int64. Co-authored-by: Zen Dodd <mail@steadytao.com>

Co-authored-by: Zen Dodd <mail@steadytao.com>

Let's lower precision for huge numbers. The output used to be: 3.45M -> 46.73M -> 523.11M -> 1.24G -> ... With this change the code always gives the three most significant digits: 3.45M -> 46.7M -> 523M -> 1.24G -> ... Co-authored-by: Zen Dodd <mail@steadytao.com>

... just make sure no precision is added. Co-authored-by: Zen Dodd <mail@steadytao.com>

Co-authored-by: Zen Dodd <mail@steadytao.com>

Let's also simplify the code for rate in progress, and benefit from same functionality. Co-authored-by: Zen Dodd <mail@steadytao.com>

... and also fix the formatting in man page. Co-authored-by: Zen Dodd <mail@steadytao.com>

eworm-de · 2026-06-11T09:29:13Z

Shuffled a bit... The result should be the same. Again, thanks a lot!

steadytao · 2026-06-12T04:09:35Z

Thank you for your work!

eworm-de force-pushed the human branch 2 times, most recently from 67d6240 to 42238c9 Compare March 23, 2023 09:51

eworm-de force-pushed the human branch from 42238c9 to 298f9fc Compare April 27, 2023 11:58

eworm-de force-pushed the human branch from 298f9fc to ac7f2dc Compare July 4, 2023 15:59

tridge force-pushed the human branch from ac7f2dc to 1ea4f0a Compare April 6, 2024 20:17

eworm-de force-pushed the human branch from 1ea4f0a to e3c6ef1 Compare April 6, 2024 20:43

eworm-de force-pushed the human branch 2 times, most recently from 933dc1f to 75ea671 Compare April 7, 2024 18:01

eworm-de force-pushed the human branch 3 times, most recently from 0ca5744 to 6abcd9b Compare April 9, 2024 07:06

eworm-de force-pushed the human branch 2 times, most recently from a20c3e6 to cb0b932 Compare April 11, 2024 06:31

eworm-de force-pushed the human branch from cb0b932 to c20bba0 Compare May 29, 2024 06:49

eworm-de force-pushed the human branch from c20bba0 to e21e5b3 Compare October 30, 2024 09:39

eworm-de force-pushed the human branch 2 times, most recently from 9bcdb1b to 050399c Compare November 11, 2024 15:43

eworm-de force-pushed the human branch 3 times, most recently from 46d2d98 to a5a0fa6 Compare November 20, 2024 08:37

eworm-de force-pushed the human branch 7 times, most recently from e05ae95 to 1b83598 Compare June 9, 2026 05:13

steadytao requested changes Jun 9, 2026

View reviewed changes

Comment thread progress.c Outdated

Comment thread lib/compat.c Outdated

Comment thread lib/compat.c Outdated

eworm-de force-pushed the human branch 2 times, most recently from 4b09949 to b833d8c Compare June 9, 2026 10:02

eworm-de requested a review from steadytao June 9, 2026 10:03

steadytao requested changes Jun 10, 2026

View reviewed changes

Comment thread lib/compat.c Outdated

Comment thread progress.c Outdated

eworm-de force-pushed the human branch 3 times, most recently from ad844de to e329ff1 Compare June 10, 2026 20:32

steadytao approved these changes Jun 11, 2026

View reviewed changes

eworm-de and others added 9 commits June 11, 2026 11:02

human-readable: calculate numbers in a loop

dd7f907

This drops a lot of `else if` blocks and extends units by "E" (Exa), which is 2^60 and thus the last to fit into int64. Co-authored-by: Zen Dodd <mail@steadytao.com>

human-readable: do not add trailing space in output

0e5fedc

Co-authored-by: Zen Dodd <mail@steadytao.com>

human-readable: also handle num < mult with the same code

8b9946d

... just make sure no precision is added. Co-authored-by: Zen Dodd <mail@steadytao.com>

human-readable: add an "i" to unit to indicate binary (1024) base

cfc82b2

Co-authored-by: Zen Dodd <mail@steadytao.com>

human-readable: also use it to format rate in progress

3170615

Let's also simplify the code for rate in progress, and benefit from same functionality. Co-authored-by: Zen Dodd <mail@steadytao.com>

human-readable: append suffix "B" for byte to ofs...

144f88b

... and also fix the formatting in man page. Co-authored-by: Zen Dodd <mail@steadytao.com>

human-readable: unbreak for do_big_dnum() with fract

d23b0ec

testsuite: cover human-readable output

00c8035

Conversation

eworm-de commented Mar 21, 2023

Uh oh!

eworm-de commented Apr 29, 2023

Uh oh!

WayneD commented May 1, 2023

Uh oh!

eworm-de commented May 2, 2023

Uh oh!

tridge commented Apr 6, 2024

Uh oh!

eworm-de commented Apr 6, 2024

Uh oh!

eworm-de commented Apr 7, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

tridge commented Apr 7, 2024

Uh oh!

eworm-de commented Apr 9, 2024

Uh oh!

eworm-de commented Apr 12, 2024

Uh oh!

eworm-de commented May 15, 2024

Uh oh!

steadytao left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

steadytao left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

eworm-de commented Jun 10, 2026

Uh oh!

steadytao left a comment

Choose a reason for hiding this comment

Uh oh!

eworm-de commented Jun 11, 2026

Uh oh!

steadytao commented Jun 11, 2026

Uh oh!

eworm-de commented Jun 11, 2026

Uh oh!

steadytao commented Jun 12, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

eworm-de commented Apr 7, 2024 •

edited

Loading