From: Junio C Hamano <gitster@pobox.com>
To: Siddharth Kannan <kannan.siddharth12@gmail.com>
Cc: git@vger.kernel.org, Matthieu.Moy@imag.fr,
pranit.bauva@gmail.com, peff@peff.net, pclouds@gmail.com,
sandals@crustytoothpaste.ath.cx
Subject: Re: [PATCH 4/4 v4] sha1_name.c: teach get_sha1_1 "-" shorthand for "@{-1}"
Date: Thu, 16 Feb 2017 11:08:29 -0800 [thread overview]
Message-ID: <xmqq8tp6x8b6.fsf@gitster.mtv.corp.google.com> (raw)
In-Reply-To: <1487258054-32292-5-git-send-email-kannan.siddharth12@gmail.com> (Siddharth Kannan's message of "Thu, 16 Feb 2017 15:14:14 +0000")
Siddharth Kannan <kannan.siddharth12@gmail.com> writes:
> Instead of replacing the whole string, we would expand it accordingly using:
>
> if (*name == '-') {
> if (len == 1) {
> name = "@{-1}";
> len = 5;
> } else {
> struct strbuf changed_argument = STRBUF_INIT;
>
> strbuf_addstr(&changed_argument, "@{-1}");
> strbuf_addstr(&changed_argument, name + 1);
>
> strbuf_setlen(&changed_argument, strlen(name) + 4);
>
> name = strbuf_detach(&changed_argument, NULL);
> }
> }
>
> Junio's comments on a previous version of the patch which used this same
> approach but inside setup_revisions [1]
>
> [1]: <xmqqtw882n08.fsf@gitster.mtv.corp.google.com>
What I said is that when we know we got "-", there is no reason to
replace it with and textually parse "@{-1}".
> + if (*name == '-' && len == 1) {
> + name = "@{-1}";
> + len = 5;
> + }
> +
> ret = get_sha1_basic(name, len, sha1, lookup_flags);
If we look at get_sha1_basic(), it obviously is not prepared to
understand "-" as "@{-1}", and the primary obstacle is that the
underlying interpret_nth_prior_checkout() does two things. It
expects to take "@{-<num>}" as a string, and the first half parses
the <num> into "long nth". The latter half then finds the nth prior
checkout. We probably should factor out the latter half into a
separate function find_nth_prior_checkout() that takes "long nth" as
input, and call it from interpret_nth_prior_checkout(), as a
preparatory step. Once it is done, get_sha1_basic() can notice that
it was fed (len == 1 && str[0] == '-') and make a direct call to
find_nth_prior_checkout() without going through the "pass '@{-1}' as
text, have interpret_nth_prior_checkout() to parse it to recover 1",
which is a roundabout way to do what you want to do.
Having said all that, I do not think the remainder of the code is
prepared to take "-", not yet anyway [*1*], so turning "-" into
"@{-1}" this patch does before it calls get_sha1_basic(), while it
is not an ideal final state, is probably an acceptable milestone to
stop at.
It is a separate matter if this patch is sufficient to produce
correct results, though. I haven't studied the callers of this
change to make sure yet, and may find bugs in this approach later.
[Footnote]
*1* For example, the existing callsite in get_sha1_basic() that
calls interpret_nth_prior_checkout() does not replace "str" with
what was returned when the HEAD is not detached. The callpath
then depends on dwim_ref() to also understand "@{-1}" it got
from the caller. If we really want to keep what came from the
end user as-is so that error message can include it, we'd need
to teach dwim_ref() about the new "-" convention. The extent of
necessary change will become a lot larger. On the other hand,
if we allow error messages and reports to use a real refname
instead of parrotting exactly what the user gave us, I think we
may be able to arrange to replace str/len in get_sha1_basic()
when we call interpret/find_nth_prior_checkout() and get a ref,
without having to teach the new "-" convention all over the
place.
next prev parent reply other threads:[~2017-02-16 19:08 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-02-16 15:14 [PATCH 0/4 v4] WIP: allow "-" as a shorthand for "previous branch" Siddharth Kannan
2017-02-16 15:14 ` [PATCH 1/4 v4] revision.c: do not update argv with unknown option Siddharth Kannan
2017-02-16 16:48 ` Matthieu Moy
2017-02-16 18:11 ` Junio C Hamano
2017-02-16 18:22 ` Matthieu Moy
2017-02-16 19:39 ` Siddharth Kannan
2017-02-16 15:14 ` [PATCH 2/4 v4] revision.c: swap if/else blocks Siddharth Kannan
2017-02-16 15:14 ` [PATCH 3/4 v4] revision.c: args starting with "-" might be a revision Siddharth Kannan
2017-02-16 15:14 ` [PATCH 4/4 v4] sha1_name.c: teach get_sha1_1 "-" shorthand for "@{-1}" Siddharth Kannan
2017-02-16 19:08 ` Junio C Hamano [this message]
2017-02-20 14:21 ` Siddharth Kannan
2017-02-20 20:30 ` Junio C Hamano
2017-02-22 6:27 ` Siddharth Kannan
2017-02-16 16:41 ` [PATCH 0/4 v4] WIP: allow "-" as a shorthand for "previous branch" Matthieu Moy
2017-02-16 18:49 ` Junio C Hamano
2017-02-16 19:43 ` Siddharth Kannan
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
List information: http://vger.kernel.org/majordomo-info.html
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=xmqq8tp6x8b6.fsf@gitster.mtv.corp.google.com \
--to=gitster@pobox.com \
--cc=Matthieu.Moy@imag.fr \
--cc=git@vger.kernel.org \
--cc=kannan.siddharth12@gmail.com \
--cc=pclouds@gmail.com \
--cc=peff@peff.net \
--cc=pranit.bauva@gmail.com \
--cc=sandals@crustytoothpaste.ath.cx \
/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.
Code repositories for project(s) associated with this public inbox
https://80x24.org/mirrors/git.git
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).