I've made four attempts at cherry-picking, all of which resulted in error identical to the following:

seireitei:dmd ace$ git cherry-pick 19a756
error: Commit 19a756f99c610b2906cc87b0e10b63b42a49a4e2 is a merge but no -m option was given.
fatal: cherry-pick failed

The step followed for this particular attempt:

    cd dmd/
    git checkout master
    git pull
    git checkout release
    git cherry-pick 19a756

From what I understand, this is caused by the particular item being picked having multiple parents. The solution would be to identify which of the parents to use as the base:

    git cherry-pick -m 1 19a756

This would designate parent #1 as the base. My question is this: How do I know which parent is the base? Is it always parent #1?

Thanks,
Andrew
_______________________________________________
dmd-beta mailing list
dmd-beta@puremagic.com
http://lists.puremagic.com/mailman/listinfo/dmd-beta

If you use "git show" it will tell you the parents of the commit eg:

$ git show
commit 918f3de1c99697be7057c401a245bc57a788483f
Merge: 3b23866 3fce69c
Author: Daniel Murphy <yebblies@gmail.com>
Date:   Wed Jan 22 07:59:12 2014 -0800

    Merge pull request #3139 from 9rnsr/fix11968

    Issue 11968 - ICE(expression.c) Crash when deleting __FILE__

Using git show on the parents will tell you what each was:
$ git show 3b23866
commit 3b23866da32b98426bbb94011c19bf928102e01f
Merge: e1bb12f d031e96
Author: Daniel Murphy <yebblies@gmail.com>
Date:   Wed Jan 22 05:25:46 2014 -0800

    Merge pull request #3137 from 9rnsr/fix11967

    [REG2.065a] Issue 11967 - ICE(parse.c) Parser crash

Which makes it clear that commit (the first one, or -m1) was the master branch and the other was the pull request.

In general the first one will always be the old master commit and should be set as the mainline.

As always, make sure to check in gitk or github diff or something similar that what you wanted is what actually happened before pushing to upstream.

On Thu, Jan 23, 2014 at 4:00 AM, Andrew Edwards <edwards.ac@gmail.com>wrote:

>
> I've made four attempts at cherry-picking, all of which resulted in error identical to the following:
>
> seireitei:dmd ace$ git cherry-pick 19a756
> error: Commit 19a756f99c610b2906cc87b0e10b63b42a49a4e2 is a merge but no
> -m option was given.
> fatal: cherry-pick failed
>
> The step followed for this particular attempt:
>
>     cd dmd/
>     git checkout master
>     git pull
>     git checkout release
>     git cherry-pick 19a756
>
> From what I understand, this is caused by the particular item being picked having multiple parents. The solution would be to identify which of the parents to use as the base:
>
>     git cherry-pick -m 1 19a756
>
> This would designate parent #1 as the base. My question is this: How do I know which parent is the base? Is it always parent #1?
>
> Thanks,
> Andrew
> _______________________________________________
> dmd-beta mailing list
> dmd-beta@puremagic.com
> http://lists.puremagic.com/mailman/listinfo/dmd-beta
>

To the untrained eye, almost everything that comes across the airwaves resemble noise. Here is an update on all the items fixes I've merged picked:

[dmd] [REG2.065a] Issue 11931  <https://github.com/D-Programming-Language/dmd/pull/3107>
[dmd] [REG2.065a] Issue 11941  <https://github.com/D-Programming-Language/dmd/pull/3112>
[dmd] [REG2.065a] Issue 11963 - ICE(parse.c) Parser crash (#3135)  <https://github.com/D-Programming-Language/dmd/pull/3135>
[dmd] [REG2.064] Issue 11965 - Segfault on garbage (#3136)  <https://github.com/D-Programming-Language/dmd/pull/3136>
[dmd] [REG2.065a] Issue 11967 - ICE(parse.c) Parser crash (#3137)  <https://github.com/D-Programming-Language/dmd/pull/3137>

[installer] cleanup leftover from merge conflict (#40)  <https://github.com/D-Programming-Language/installer/pull/40>
[installer] fix expanding zip files created on Windows (#41)  <https://github.com/D-Programming-Language/installer/pull/41>
[installer] don't zip .git* and .DS_Store files (#42)  <https://github.com/D-Programming-Language/installer/pull/42>
[installer] add "dustmite" binary on deb/rpm packages (#43)  <https://github.com/D-Programming-Language/installer/pull/43>

[dlang.org] 2.065 changelog (#476)  <https://github.com/D-Programming-Language/dlang.org/pull/476>

The following needs to be reviewed and merged:

[Issue 11966] New: Regression 2.065.b1: inout(const(char))[] doesn't convert to inout(char)[]  <https://github.com/D-Programming-Language/dmd/pull/3138>


No doubt there has been a lot more merges/fixes since Saturday but it is not abundantly clear to me which are intended to address issues in this release so I have not picked them.

The scheme use by Kenji in his submissions prove very helpful:

     [repo]  [milestone] Issue #### - Description (#pull)

I could only ask that others adapt the the same convention as it makes cherry-picking much simpler.

Regards,
Andrew

To Andrew Edwards:

Unfortunately, this merged list lacks following regression fixes.

SHA-1: 7ecbbce5fdbe36db303ac21b26aee7a226657488
* Merge pull request #3105 from 9rnsr/fix11930
[REG2.065a] Issue 11930 - Alias this not considered in is(T unused: U)
matching

SHA-1: 8409d8c25734f9569a687b18f8a4818ee5762775
* Merge pull request #3104 from 9rnsr/fix11896
[REG2.065a] Issue 11896 - isVirtualMethod related GitHub HEAD regression
(works with 2.064)

SHA-1: 3b37a573bf46a3463e8de41008ee4c9d81e75974
* Merge pull request #3097 from 9rnsr/fix11924
[REG2.065a] Issue 11924 - inout Variadic Template Parameters

SHA-1: 11f05b5dad2f5ef22a73dd851db1ec108508e022
* Merge pull request #3094 from WalterBright/fix11922
[REG2.065a] Issue 11922 - ICE on nonexistent identifier in templated auto
method

---------
If possible, also merging these _trivial_ ICE fixes to the release branch would be better:

SHA-1: ffb51147936181aa4f141afaccd83acba7dc229c
* Merge pull request #3125 from 9rnsr/fix11944
Issue 11944 - ICE(expression.c) Assertion `f' failed.

SHA-1: 918f3de1c99697be7057c401a245bc57a788483f
* Merge pull request #3139 from 9rnsr/fix11968
Issue 11968 - ICE(expression.c) Crash when deleting __FILE__

Kenji Hara



2014/1/23 Andrew Edwards <edwards.ac@gmail.com>

>  To the untrained eye, almost everything that comes across the airwaves
> resemble noise. Here is an update on all the items fixes I've merged picked:
>
> [dmd] [REG2.065a] Issue 11931 <https://github.com/D-Programming-Language/dmd/pull/3107>[dmd] [REG2.065a] Issue 11941 <https://github.com/D-Programming-Language/dmd/pull/3112>[dmd] [REG2.065a] Issue 11963 - ICE(parse.c) Parser crash (#3135) <https://github.com/D-Programming-Language/dmd/pull/3135>[dmd] [REG2.064] Issue 11965 - Segfault on garbage (#3136) <https://github.com/D-Programming-Language/dmd/pull/3136>[dmd] [REG2.065a] Issue 11967 - ICE(parse.c) Parser crash (#3137) <https://github.com/D-Programming-Language/dmd/pull/3137>
> [installer] cleanup leftover from merge conflict (#40) <https://github.com/D-Programming-Language/installer/pull/40>[installer] fix expanding zip files created on Windows (#41) <https://github.com/D-Programming-Language/installer/pull/41>[installer] don't zip .git* and .DS_Store files (#42) <https://github.com/D-Programming-Language/installer/pull/42>[installer] add "dustmite" binary on deb/rpm packages (#43) <https://github.com/D-Programming-Language/installer/pull/43>
> [dlang.org] 2.065 changelog (#476) <https://github.com/D-Programming-Language/dlang.org/pull/476>
>
> The following needs to be reviewed and merged:
>
> [Issue 11966] New: Regression 2.065.b1: inout(const(char))[] doesn't convert to inout(char)[] <https://github.com/D-Programming-Language/dmd/pull/3138>
>
>
> No doubt there has been a lot more merges/fixes since Saturday but it is not abundantly clear to me which are intended to address issues in this release so I have not picked them.
>
> The scheme use by Kenji in his submissions prove very helpful:
>
>     [repo]  [milestone] Issue #### - Description (#pull)
>
> I could only ask that others adapt the the same convention as it makes cherry-picking much simpler.
>
> Regards,
> Andrew
>
> _______________________________________________
> dmd-beta mailing list
> dmd-beta@puremagic.com
> http://lists.puremagic.com/mailman/listinfo/dmd-beta
>

On 1/22/14, 8:47 PM, Kenji Hara wrote:
> Unfortunately, this merged list lacks following regression fixes.
Done
> ---------
> If possible, also merging these _trivial_ ICE fixes to the release branch would be better:
And done.
_______________________________________________
dmd-beta mailing list
dmd-beta@puremagic.com
http://lists.puremagic.com/mailman/listinfo/dmd-beta

2014/1/23 Andrew Edwards <edwards.ac@gmail.com>

> On 1/22/14, 8:47 PM, Kenji Hara wrote:
>
>> Unfortunately, this merged list lacks following regression fixes.
>>
> Done
>
>  ---------
>> If possible, also merging these _trivial_ ICE fixes to the release branch would be better:
>>
> And done.


Thank you!

Kenji Hara