git/list[1] front-page[2] threads[3] people[4] search[5] about
 

Re: [PATCH v2 1/2] Documentation/config: fix typos

From
KHKristoffer Haugsbakk <kristofferhaugsbakk@fastmail.com>
Date
Sep 23, 2024, 19:13 UTC
Message-ID
<7944f06f-33d9-4988-85ab-c1cf563615f3@app.fastmail.com>
In-Reply-To
<CAPig+cSErj4ZB9bHB8mZfzNkiaN_EpjT6b4b=cfsf_+KMqytiA@mail.gmail.com>
On Mon, Sep 23, 2024, at 21:05, Eric Sunshine wrote:
Show 42 quoted lines
> On Mon, Sep 23, 2024 at 2:44 PM Kristoffer Haugsbakk
> <kristofferhaugsbakk@fastmail.com> wrote:
>> On Mon, Sep 23, 2024, at 19:51, Eric Sunshine wrote:
>> > Depending upon how dramatically the patch series changes from one
>> > version to the next, the range-diff may end up being unreadable
>> > gobbledygook, in which case you may instead want to include an
>> > interdiff ("git format-patch --interdiff").
>>
>> What’s the benefit of interdiff in that case? Neither
>> git-format-patch(1) nor git-range-diff(1) seems to discuss what the
>> differences between these two are.
>
> An interdiff is just a plain diff. If you have branch (or tag) "v1"
> which is the original version of a patch series, and "v2" which is the
> reroll of the series, then interdiff is simply:
>
>     git diff v1 v2
>
> Thus, it shows the difference between the final state of the code at
> v1 and the state at v2. Interdiffs are easy to read because they are
> just diffs. However, because they are only showing differences in file
> content, they don't show changes to commit messages or new or removed
> or reordered patches in a series.
>
> A range-diff is a diff-of-diffs. It is very, very roughly similar to this:
>
>     git format-patch -o v1-patches <common-base>..v1
>     git format-patch -o v2-patches <common-base>..v2
>     some-diff-dir-command v1-patches v2-patches
>
> It shows the diff of the patches themselves, including changes to
> commit messages and changes to changes, as well as inserted and
> removed and reordered patches.
>
> Range-diffs tend to be a good deal more difficult to read (at least at
> first) but help show the evolution of the _patch series_ itself
> between versions, whereas interdiffs show only the evolution of the
> _code_ between versions. As a reviewer, if you're primarily interested
> in how the code evolved, then interdiffs are much more easily
> digested, but most reviewers are also interested in the holistic
> aspects of a patch series for which range-diffs are more helpful. I
> periodically include both range-diff and interdiffs in my rerolls.

Thanks for that. I love when a good range-diff falls out of a reroll—and I love the tool—but of course that can’t be expected out of every reroll.

-- 
Kristoffer Haugsbakk
Previous: Eric SunshineNext: Kristoffer Haugsbakk
Message 8 of 9 in “Documentation/config: fix typos”
  1. 1/2 Documentation/config: fix typosAndrew Kreimer, Sep 23, 2024
  2. 2/2 Documentation: fix typosAndrew Kreimer, Sep 23, 2024
  3. Eric SunshineSep 23, 2024
  4. Eric SunshineSep 23, 2024
  5. Andrew KreimerSep 23, 2024
  6. Kristoffer HaugsbakkSep 23, 2024
  7. Eric SunshineSep 23, 2024
  8. Kristoffer HaugsbakkSep 23, 2024
  9. Kristoffer HaugsbakkNov 25, 2024

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.