threads / patch / 44620

patchclone,fetch: explain the shallow-clone option a little more clearly

Subject: [PATCH] clone,fetch: explain the shallow-clone option a little more clearly

## tl;dr

5 messages between Dec 4, 2016 and Dec 7, 2016. Diffs are folded; open one to read it.

replies: 4people: 3as markdown or json

Alex Henrie· Dec 4, 2016, 22:03 UTC · lore

"deepen by excluding" does not make sense because excluding a revision does not deepen a repository; it makes the repository more shallow.

Signed-off-by: Alex Henrie <alexhenrie24@gmail.com>
---
 builtin/clone.c | 2 +-
 builtin/fetch.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

Show changes to 2 files +2 −2

builtin/clone.c, builtin/fetch.c

diff --git a/builtin/clone.c b/builtin/clone.c
index 6c76a6e..e3cb808 100644
--- a/builtin/clone.c
+++ b/builtin/clone.c
@@ -99,7 +99,7 @@ static struct option builtin_clone_options[] = {
 	OPT_STRING(0, "shallow-since", &option_since, N_("time"),
 		    N_("create a shallow clone since a specific time")),
 	OPT_STRING_LIST(0, "shallow-exclude", &option_not, N_("revision"),
-			N_("deepen history of shallow clone by excluding rev")),
+			N_("deepen history of shallow clone, excluding rev")),
 	OPT_BOOL(0, "single-branch", &option_single_branch,
 		    N_("clone only one branch, HEAD or --branch")),
 	OPT_BOOL(0, "shallow-submodules", &option_shallow_submodules,
diff --git a/builtin/fetch.c b/builtin/fetch.c
index b6a5597..fc74c84 100644
--- a/builtin/fetch.c
+++ b/builtin/fetch.c
@@ -122,7 +122,7 @@ static struct option builtin_fetch_options[] = {
 	OPT_STRING(0, "shallow-since", &deepen_since, N_("time"),
 		   N_("deepen history of shallow repository based on time")),
 	OPT_STRING_LIST(0, "shallow-exclude", &deepen_not, N_("revision"),
-			N_("deepen history of shallow clone by excluding rev")),
+			N_("deepen history of shallow clone, excluding rev")),
 	OPT_INTEGER(0, "deepen", &deepen_relative,
 		    N_("deepen history of shallow clone")),
 	{ OPTION_SET_INT, 0, "unshallow", &unshallow, NULL,

-- 
2.10.2

Junio C Hamano· Dec 5, 2016, 18:28 UTC · re: Alex Henrie · lore

Re: [PATCH] clone,fetch: explain the shallow-clone option a little more clearly

Alex Henrie <alexhenrie24@gmail.com> writes:

> "deepen by excluding" does not make sense because excluding a revision
> does not deepen a repository; it makes the repository more shallow.

I think that an intuitive way the feature should work may be:

 - You started with "git fetch --depth=20" and then later say "I
   have only a very short segment of the recent history, but I want
   a history that dates back to v1.0" with "--shallow-exclude=v1.0".
   In this case, you would be deepening.

 - You instead started with "git fetch --depth=20000" that dated
   back to v0.5.  "--shallow-exclude=v1.0" you say today would mean
   "I have very old cruft I no longer look at.  I just want my
   history lead back to v1.0 and no earlier".  In such a case, you
   indeed would be making the repository shallower.

I however offhand do not think the feature can be used to make the repository shallower, and I agree your changes to the usage string probably describe what the option does more correctly.

I however suspect that the feature is simply buggy and it would eventually want to allow to shorten the history as well. At that point we may want to work on the verb 'deepen' there, too.

Show 33 quoted lines

>
> Signed-off-by: Alex Henrie <alexhenrie24@gmail.com>
> ---
>  builtin/clone.c | 2 +-
>  builtin/fetch.c | 2 +-
>  2 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/builtin/clone.c b/builtin/clone.c
> index 6c76a6e..e3cb808 100644
> --- a/builtin/clone.c
> +++ b/builtin/clone.c
> @@ -99,7 +99,7 @@ static struct option builtin_clone_options[] = {
>  	OPT_STRING(0, "shallow-since", &option_since, N_("time"),
>  		    N_("create a shallow clone since a specific time")),
>  	OPT_STRING_LIST(0, "shallow-exclude", &option_not, N_("revision"),
> -			N_("deepen history of shallow clone by excluding rev")),
> +			N_("deepen history of shallow clone, excluding rev")),
>  	OPT_BOOL(0, "single-branch", &option_single_branch,
>  		    N_("clone only one branch, HEAD or --branch")),
>  	OPT_BOOL(0, "shallow-submodules", &option_shallow_submodules,
> diff --git a/builtin/fetch.c b/builtin/fetch.c
> index b6a5597..fc74c84 100644
> --- a/builtin/fetch.c
> +++ b/builtin/fetch.c
> @@ -122,7 +122,7 @@ static struct option builtin_fetch_options[] = {
>  	OPT_STRING(0, "shallow-since", &deepen_since, N_("time"),
>  		   N_("deepen history of shallow repository based on time")),
>  	OPT_STRING_LIST(0, "shallow-exclude", &deepen_not, N_("revision"),
> -			N_("deepen history of shallow clone by excluding rev")),
> +			N_("deepen history of shallow clone, excluding rev")),
>  	OPT_INTEGER(0, "deepen", &deepen_relative,
>  		    N_("deepen history of shallow clone")),
>  	{ OPTION_SET_INT, 0, "unshallow", &unshallow, NULL,

Duy Nguyen· Dec 6, 2016, 09:34 UTC · re: Junio C Hamano · lore

Re: [PATCH] clone,fetch: explain the shallow-clone option a little more clearly

On Tue, Dec 6, 2016 at 1:28 AM, Junio C Hamano <gitster@pobox.com> wrote:

> I however offhand do not think the feature can be used to make the
> repository shallower

I'm pretty sure it can, though it's a waste because you should be able to shorten your history without talking to a remote server. But that no-remote shortening is not implemented yet. And you probably want an option to check with remote anyway to make sure the part you cut out is available there, no history will be lost.

-- 
Duy

Junio C Hamano· Dec 6, 2016, 18:29 UTC · re: Duy Nguyen · lore

Re: [PATCH] clone,fetch: explain the shallow-clone option a little more clearly

Duy Nguyen <pclouds@gmail.com> writes:

Show 5 quoted lines

> On Tue, Dec 6, 2016 at 1:28 AM, Junio C Hamano <gitster@pobox.com> wrote:
>> I however offhand do not think the feature can be used to make the
>> repository shallower
>
> I'm pretty sure it can,...

I wrote my message after a short local testing, but it is very possible I botched it and reached a wrong conclusion above.

If we can use the command to make it shallower, then the phrase "deepen" would probably be what we need to be fixing in this patch:

>  	OPT_STRING_LIST(0, "shallow-exclude", &option_not, N_("revision"),
> -			N_("deepen history of shallow clone by excluding rev")),
> +			N_("deepen history of shallow clone, excluding rev")),

Perhaps a shorter version of:

    Adjust the depth of shallow clone so that commits that are
    decendants of the named rev are made available, while commits
    that are ancestors of the named rev are made beyond reach.

or something like that.  Here is my (somewhat botched) attempt:

    Adjust shallow clone's history to be cut at the rev

Duy Nguyen· Dec 7, 2016, 14:07 UTC · re: Junio C Hamano · lore

Re: [PATCH] clone,fetch: explain the shallow-clone option a little more clearly

On Wed, Dec 7, 2016 at 1:29 AM, Junio C Hamano <gitster@pobox.com> wrote:

Show 27 quoted lines

> Duy Nguyen <pclouds@gmail.com> writes:
>
>> On Tue, Dec 6, 2016 at 1:28 AM, Junio C Hamano <gitster@pobox.com> wrote:
>>> I however offhand do not think the feature can be used to make the
>>> repository shallower
>>
>> I'm pretty sure it can,...
>
> I wrote my message after a short local testing, but it is very
> possible I botched it and reached a wrong conclusion above.
>
> If we can use the command to make it shallower, then the phrase
> "deepen" would probably be what we need to be fixing in this patch:
>
>>       OPT_STRING_LIST(0, "shallow-exclude", &option_not, N_("revision"),
>> -                     N_("deepen history of shallow clone by excluding rev")),
>> +                     N_("deepen history of shallow clone, excluding rev")),
>
> Perhaps a shorter version of:
>
>     Adjust the depth of shallow clone so that commits that are
>     decendants of the named rev are made available, while commits
>     that are ancestors of the named rev are made beyond reach.
>
> or something like that.  Here is my (somewhat botched) attempt:
>
>     Adjust shallow clone's history to be cut at the rev

OK mine is "exclude the given tag from local history".

BTW the clone's string could be rephrased better because we know there's no local history to begin with, if we stick to any verbs that involves deepening or shortening.

-- 
Duy