Github mirror push rejected

Andrew Jeffery <[email protected]> Sat, 26 Mar 2016 00:18:28 +1030
Newsgroups	org.ozlabs.lists.ccan
Message-ID	<[email protected]>
Hey Rusty,

I've just pushed some patches to the strgrp module to fix some
anomalies in the generated clusters, but I encountered the error below
in the process:

    $ git push ozlabs HEAD:master
    ...
    remote: + rm -rf /home/ccan/tmp/ccan-24657
    remote: + cd /home/ccan/ccan
    remote: + git push
    remote: To     [email protected]    :rustyrussell/ccan.git
    remote:  ! [rejected]        master -> master (non-fast-forward)
    remote: error: failed to push some refs to     '[email protected]    :rustyrussell/ccan.git'
    remote: hint: Updates were rejected because the tip of your current branch is behind
    remote: hint: its remote counterpart. Integrate the remote changes (e.g.
    remote: hint: 'git pull ...') before pushing again.
    remote: hint: See the 'Note about fast-forwards' in 'git push --help' for details.
    To     [email protected]    :ccan
       d23fb57..e8f7a97  HEAD -> master

Seems the ozlabs.org repo had d23fb57c8e276090325162966b094ebf71e73e68
at the tip of master (before my push) whilst your github repo
has b75c31a84fd24e975f8cf409f35e5b3d08061f47. The merge-base was the
commit immediately prior in both
repos, bb88463014092a5550ea4c51a037e12e49446722. From the diff below it
looks like an earlier revision made it into github before a reroll was
pushed to ozlabs.org?

Cheers,

Andrew

PS: The diff:

    $ git diff d23fb57c8e276090325162966b094ebf71e73e68 b75c31a84fd24e975f8cf409f35e5b3d08061f47
    diff --git a/ccan/crypto/shachain/design.txt b/ccan/crypto/shachain/design.txt
    index b463976..8dc9743 100644
    --- a/ccan/crypto/shachain/design.txt
    +++ b/ccan/crypto/shachain/design.txt
    @@ -63,12 +63,9 @@ generate_from_seed(index):
     The Receiver's Tree
     -------------------
     
    -To derive the value for a index N, you need to have the root of a tree
    -which contains it.  That is the same as needing an index I which is N
    -rounded down in binary: eg. if N is 0b001100001, you need 0b001100000,
    -0b001000000 or 0b000000000.
    -
    -Pseudocode:
    +You can derive a lesser index if it only involves setting bits
    +(ie. you can generate index 1 from index 0, but not the other way
    +around).  Pseudocode:
     
     # Can we derive the value for to_index from from_index?
     can_derive(from_index, to_index):
    @@ -91,7 +88,7 @@ derive(from_index, to_index, from_value):
         return value
     
     If you are receiving values (in reverse order), you need to remember
    -up to 64 of them to derive all previous values.  The simplest method
    +up to 63 of them to derive all previous values.  The simplest method
     is to keep an array, indexed by the number of trailing zeroes in the
     received index:
     
    @@ -111,7 +108,7 @@ you can derive the value you want, eg:
     
     # Find an old value
     regenerate_value(index):
    -    for i in known:
    +    for i in remembered:
             if can_derive(i.index, index):
                 return derive(i.index, i.value, index)
         fail

PPS: Seems I'm not without a failure as well - the commits I pushed are
missing the S-O-B line since I wrote them before I updated my git
aliases to add the flag in :/

_______________________________________________
ccan mailing list
[email protected]
https://lists.ozlabs.org/listinfo/ccan
signature.asc (application/pgp-signature, 819 B)