git@vger.kernel.org mailing list mirror (one of many)
 help / color / mirror / code / Atom feed
From: "Philip Oakley" <philipoakley@iee.org>
To: "Junio C Hamano" <gitster@pobox.com>
Cc: "GitList" <git@vger.kernel.org>,
	"Pat Thoyts" <patthoyts@users.sourceforge.net>,
	"Eric Sunshine" <sunshine@sunshineco.com>,
	"Alexey Astakhov" <asstv7@gmail.com>
Subject: Re: [PATCH v3 4/4] git gui: allow for a long recentrepo list
Date: Fri, 18 Dec 2015 00:26:07 -0000	[thread overview]
Message-ID: <9731888BD4C348F5BFC82FA96D978034@PhilipOakley> (raw)
In-Reply-To: xmqqbn9obqne.fsf@gitster.mtv.corp.google.com

From: "Junio C Hamano" <gitster@pobox.com>
> Philip Oakley <philipoakley@iee.org> writes:
>
>> The gui.recentrepo list may be longer than the maxrecent setting.
>> Use the actual length determined earlier, now saved as $lenrecent.
>>
>> In an ideal world, the git gui would limit the number of entries
>> to the maxrecent setting, however the recentrepo config list may
>> have been extended outside of the gui, or the maxrecent setting changed
>> to a reduced value. Further, when testing the gui's recentrepo
>> logic it is useful to show these extra, but valid, entries.
>>
>> The list length will be trimmed to $maxrecent after the user selects
>> a repo to open.
>>
>> Signed-off-by: Philip Oakley <philipoakley@iee.org>
>> ---
>> V2:
>> word usage corrected.
>> Eric's comments $gmane/282432
>> V3:
>> Updated list length measure following Junio's comments $gmane/282669
>>
>> Replaces the previous V2 Patch 4/4.
>>
>> ---
>>  git-gui/lib/choose_repository.tcl | 5 +++--
>>  1 file changed, 3 insertions(+), 2 deletions(-)
>>
>> diff --git a/git-gui/lib/choose_repository.tcl 
>> b/git-gui/lib/choose_repository.tcl
>> index ad7a888..b4cc7dd 100644
>> --- a/git-gui/lib/choose_repository.tcl
>> +++ b/git-gui/lib/choose_repository.tcl
>> @@ -134,7 +134,8 @@ constructor pick {} {
>>  $opts conf -state disabled
>>
>>  set sorted_recent [_get_recentrepos]
>> - if {[llength $sorted_recent] > 0} {
>> + set lenrecent [llength $sorted_recent]
>> + if {$lenrecent > 0} {
>>  if {$m_repo ne {}} {
>>  $m_repo add separator
>>  $m_repo add command \
>> @@ -153,7 +154,7 @@ constructor pick {} {
>>  -background [get_bg_color $w_body.recentlabel] \
>>  -wrap none \
>>  -width 50 \
>> - -height $maxrecent
>> + -height $lenrecent
>>  $w_recentlist tag conf link \
>>  -foreground blue \
>>  -underline 1
>
> Natural questions that come to mind after seeing the above are like
> these:
>
> - Does maxrecent still get used after this change?

Yes, essentially as you describe....

>
> - Does the remainder of the code get upset if -height is set to too
>   small (lenrecent might turn out to be 1, for example)?

No it's OK, the code has at least checked it is positive..

>
> And grepping for maxrecent shows there is _append_recentrepos that
> does this:

It does three things:
- removes the current (selected) repo from the list of past repos (and 
removes any duplicates)
- places the current repo at the end
- trims the list to back to max repos.

Nominally, it can't do anything about non-selected valid repos which may be 
listed in the config twice because git config won't selectively remove just 
one of two identical key/values, however in Patch 3 I do remove both older & 
newer entries and re-read the config (some may consider that a small 
regression).

It's not easy working around the contrary assumptions of the gui.recentrepo 
list (once only) and the reality of the .gitconfig (duplicates possible) and 
'git config's --unset/--unset-all limitation.

Exactly how those extra entries get there is an open question, but it does 
happen (so say's google, as well as the user report, I've seen it myself ;-)

>
> proc _append_recentrepos {path} {
> set path [file normalize $path]
> set recent [get_config gui.recentrepo]
>
> ...
> if {[set maxrecent [get_config gui.maxrecentrepo]] eq {}} {
> set maxrecent 10
> }
>
> while {[llength $recent] > $maxrecent} {
> _unset_recentrepo [lindex $recent 0]
> set recent [lrange $recent 1 end]
> }
> }
>
> The last while loop looks to me like truncating the overlong recent
> list down to maxrecent.

Yes

>   Perhaps a similar logic needs to be at the
> end of _get_recentrepos?

The logic here is to show what is in the config to the user, *before* we 
trim the list, hence we wait until the user has selected a repo to open and 
do the purging in _append_recentrepos.

As an aside, there's also a long standing 'feature' that if one works both 
on and off network, then network repo paths get trimmed as soon as the user 
starts the gui while off-net (or the network goes down;-) because those 
recent repo paths become 'invalid' and are purged, but let's not go there.

--
Philip 

  reply	other threads:[~2015-12-18  0:26 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-12-16 23:58 [PATCH v2 0/4] Fix git-gui when recentrepo list has duplicates Philip Oakley
2015-12-16 23:58 ` [PATCH v2 1/4] git-gui: de-dup entries from gui.recentrepo Philip Oakley
2015-12-16 23:58 ` [PATCH v2 2/4] git gui: cope with duplicates in _get_recentrepo Philip Oakley
2015-12-16 23:58 ` [PATCH v2 3/4] git gui: de-dup selected repo from recentrepo history Philip Oakley
2015-12-16 23:58 ` [PATCH v2 4/4] git gui: allow for a long recentrepo list Philip Oakley
2015-12-17 19:32   ` Junio C Hamano
2015-12-17 20:09     ` Philip Oakley
2015-12-17 22:11       ` Philip Oakley
2015-12-17 22:40         ` [PATCH v3 " Philip Oakley
2015-12-17 23:29           ` Junio C Hamano
2015-12-18  0:26             ` Philip Oakley [this message]
  -- strict thread matches above, loose matches on Subject: below --
2017-01-22 19:52 [PATCH v3 0/4] " Philip Oakley
2017-01-22 19:53 ` [PATCH v3 4/4] " Philip Oakley

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=9731888BD4C348F5BFC82FA96D978034@PhilipOakley \
    --to=philipoakley@iee.org \
    --cc=asstv7@gmail.com \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=patthoyts@users.sourceforge.net \
    --cc=sunshine@sunshineco.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.
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).