Adam Majer wrote:
Adam Majer wrote:
kirstin penelope rhys wrote:
On Sunday 21 January 2007 13:11, Adam Majer wrote:
kirstin penelope rhys wrote:
Package: rails
Version: 1.1.6-3
Severity: normal
The 12_options patch which was added in version 1.1.6-2 and was supposed
to be removed for 1.1.6-3 (bug #406658) is still present, and the select
helper is still broken (no html name or id attributes are generated).
Hi,
I cannot reproduce this in 1.1.6-3. The select tag, like <%=
select('user','id',@users) %> seem to be working correctly.
Oh! That's apparently what happened. I ran a diff against your binary, and it
totally matches the one I made from the 1.1.6-3 source by manually removing
12_options.
Sorry! Nevermind! I didn't think to test that. I supposed this should be filed
against the source package then?
No, no. The only thing is, you'll have to remove the old source unpack
directory because it is the same one as the new one. Or at least do a
`fakeroot debian/rules clean` *before* you unpack the new version.
So,
1. remove the source tree
2. unpack the new -3 version with `dpkg-source -x ....`
3. build it with dpkg-buildpackage
Hmm, doesn't really make sense if I think about it, but do it anyway.
Sometimes weird things happen when you update sources in an already
build source tree.
The 12_options patch should not be applied to binary although it still
exists. If it does get applied, then it could be a problem with the
dpatch. What version of dpatch are you using?
Cheers,
- Adam
PS. do a `cat rails-1.1.6/debian/patches/00list` to see the list of
patches that *should* be applied.
> Hi again,
>
> So, did this work? If you rebuild from scratch (ie. remove the directory
> before unpacking new version), was the patch removed and thus the bug
fixed?
>
Ok, I built the package, from a fresh source package extractions, twice
in totally separate directories. Once from the pristine, once
with 12_options removed.
The resulting packages are, in fact, different, but only in their
documentation? Which, leaving aside whatever I must have done
originally, doesn't make much sense to me. Either the patch should have
been applied or not? How could the rdoc differ, while the source is
identical? The patch touches code, not just comments.
Best,
kirstin
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
