jajcus at jajcus.net
Tue Dec 1 15:24:45 CET 2015
On 2015-12-01 15:05, Marcin Krol wrote:
>> Anyway - glen is right, rpm should reject that...
> Perhaps we should go debian-like route and accept only patches that
> apply cleanly? Fuzziness can cause code to be inserted/deleted/modified
> in wrong places which can cause compile time errors, segfaults, broken
> configs etc. I encountered such issues many times and most of our
> patches apply with fuzzines or offsets.
But this case was not about fuzzines or invalid patch. Patches are
allowed to contain trailing data – they might be part of an e-mail
message. That is why blues' patch was accepted. The line ignored was
technically outside of the patch.
More information about the pld-devel-en