You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@hc.apache.org by Oleg Kalnichevski <ol...@apache.org> on 2013/09/01 12:52:24 UTC
[VOTE] HttpComponents Client 4.2.6 release
Please vote on releasing these packages as HttpComponents Client 4.2.6.
The vote is open for the at least 72 hours, and only votes from
HttpComponents PMC members are binding. The vote passes if at least
three binding +1 votes are cast and there are more +1 than -1 votes.
Packages:
https://dist.apache.org/repos/dist/dev/httpcomponents/httpclient-4.2.6-RC1/
Release notes:
https://dist.apache.org/repos/dist/dev/httpcomponents/httpclient-4.2.6-RC1/RELEASE_NOTES-4.2.x.txt
Maven artefacts:
https://repository.apache.org/content/repositories/orgapachehttpcomponents-129/org/apache/httpcomponents/
SVN tag:
https://svn.apache.org/repos/asf/httpcomponents/httpclient/tags/4.2.6-RC1/
(Revision: 1519258)
--------------------------------------------------------------------------
Vote: HttpComponents Client 4.2.6 release
[ ] +1 Release the packages as HttpComponents Client 4.2.6.
[ ] -1 I am against releasing the packages (must include a reason).
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
For additional commands, e-mail: dev-help@hc.apache.org
Re: [VOTE] HttpComponents Client 4.2.6 release
Posted by Karl Wright <da...@gmail.com>.
Passed all MCF tests.
+1 from me.
Karl
On Sun, Sep 1, 2013 at 6:52 AM, Oleg Kalnichevski <ol...@apache.org> wrote:
> Please vote on releasing these packages as HttpComponents Client 4.2.6.
> The vote is open for the at least 72 hours, and only votes from
> HttpComponents PMC members are binding. The vote passes if at least
> three binding +1 votes are cast and there are more +1 than -1 votes.
>
> Packages:
> https://dist.apache.org/repos/dist/dev/httpcomponents/httpclient-4.2.6-RC1/
>
> Release notes:
>
> https://dist.apache.org/repos/dist/dev/httpcomponents/httpclient-4.2.6-RC1/RELEASE_NOTES-4.2.x.txt
>
> Maven artefacts:
>
> https://repository.apache.org/content/repositories/orgapachehttpcomponents-129/org/apache/httpcomponents/
>
> SVN tag:
> https://svn.apache.org/repos/asf/httpcomponents/httpclient/tags/4.2.6-RC1/
> (Revision: 1519258)
>
> --------------------------------------------------------------------------
> Vote: HttpComponents Client 4.2.6 release
> [ ] +1 Release the packages as HttpComponents Client 4.2.6.
> [ ] -1 I am against releasing the packages (must include a reason).
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
> For additional commands, e-mail: dev-help@hc.apache.org
>
>
Re: [VOTE] HttpComponents Client 4.2.6 release
Posted by "Asankha C. Perera" <as...@apache.org>.
> Vote: HttpComponents Client 4.2.6 release
> [X] +1 Release the packages as HttpComponents Client 4.2.6.
> [ ] -1 I am against releasing the packages (must include a reason)
regards
asankha
--
Asankha C. Perera
AdroitLogic, http://adroitlogic.org
http://esbmagic.blogspot.com
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
For additional commands, e-mail: dev-help@hc.apache.org
Re: [VOTE] HttpComponents Client 4.2.6 release
Posted by Karl Wright <da...@gmail.com>.
Agreed.
The vote is about whether to release the source bits. EOL-style
considerations between SVN and the source release really do not apply
here. It's a nice-to-have, not a make-or-break.
Karl
On Mon, Sep 2, 2013 at 1:28 PM, Oleg Kalnichevski <ol...@apache.org> wrote:
> On Mon, 2013-09-02 at 21:59 +0530, Asankha C. Perera wrote:
> > On 09/02/2013 06:06 PM, Oleg Kalnichevski wrote:
> > > On Mon, 2013-09-02 at 11:53 +0100, sebb wrote:
> > > ...
> > >
> > > Further httpclient/src/test/resources/suffixlist.txt has got EOL=LF in
> > > the zip file so it disagrees with the SVN tag checkout.
> > >
> > > Also, the *.properties files ought to be CRLF on Windows (they are
> > > native in SVN)
> > > Likewise *.xsl and *.css and SPNEGO.svg
> > > Otherwise checkouts of the tag don't agree with the Zip
> > >
> > > I am sorry but I have no idea what this is all about. I have no idea
> why
> > > they should agree in the first place.
> > I guess Sebastian's concern is that if you unzip the Zip on a Windows
> > box, and compare the result with a checkout of the tag on the same
> > Windows system, the line endings would be different between the two for
> > some files. Although that maybe a minor issue, if the differences are
> > related only to the line endings of a few files, I would not consider it
> > a blocker to the release. Personally I consider differences in line
> > endings as ignorable, and many IDEs or file compare tools also tend to
> > ignore them. I guess one who would look for the source releases would
> > prefer to get it directly via the SVN tag, and ones who download the
> > source archive of a release would generally want to refer to the code
> > for understanding or debugging etc - so the line endings would not be
> > that critical - my 2c
>
> Asankha
> I understand the technical bits. What I do not understand is why that
> would matter at all, why anyone in their sane mind would want to compare
> SVN tag with the content of ZIP archive and why on earth we need to
> waste our collective time on stuff like that?
>
> Oleg
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
> For additional commands, e-mail: dev-help@hc.apache.org
>
>
Re: [VOTE] HttpComponents Client 4.2.6 release
Posted by sebb <se...@gmail.com>.
On 3 September 2013 12:34, Oleg Kalnichevski <ol...@apache.org> wrote:
> On Tue, 2013-09-03 at 10:57 +0100, sebb wrote:
>> On 3 September 2013 10:48, Oleg Kalnichevski <ol...@apache.org> wrote:
>> > On Tue, 2013-09-03 at 02:03 -0400, Karl Wright wrote:
>> >> Hi Sebb,
>> >>
>> >> "It is essential that the files in a source release can be matched
>> >> against those in SVN."
>> >>
>> >> I think that's technically not necessary according to Apache rules; what's
>> >> necessary is that the bits in the tar or zip be based *solely* on the svn
>> >> tag and upstream dependencies. But you really vote not on the tag but on
>> >> the tar and zip.
>> >>
>> >> "But the main issue I have is that the PNG file in the ZIP archive is
>> >> corrupt."
>> >>
>> >> I think that should be fixed. It's also worth considering that perhaps
>> >> fixing up line endings for Windows users is maybe not worth the pain in the
>> >> behind at release time that it is. Most people use editors these days that
>> >> don't care.
>> >>
>> >> Karl
>> >>
>> >
>> > Karl
>> > Line delimiters of all human readable files as well as source code get
>> > adjusted based on the type of the distribution. The whole thing boils
>> > down to a just a few _resource_ files, which I personally think should
>> > never be meddled with given that line delimiters may be semantically
>> > significant.
>>
>> If a fixed eol is required by resource files, they should not have
>> eol-style:native in SVN.
>> They should have either LF or CRLF depending on which eol is required
>> by the format.
>>
>> But if the format requires different EOL for Windows and Un*x, then
>> native is correct and so is conversion to the appropriate ending.
>>
>> I am not saying that every text file needs to be converted.
>> But ones which are converted by SVN (i.e. native) should be converted
>> for the corresponding archive.
>>
>
> All which the assembly script has no way of knowing.
In theory the build script could parse the output of svn pl -v and use
that to control the eol massaging.
I agree that would be tedious; it's a lot easier to create the
archives, check the results and then fix any errors.
But the point is, we already know that all the .txt files and all the
.properties files are flagged as eol:native, so we should do the same
in the build script.
Otherwise the generated zip and tar.gz file contents will depend on
the EOL setting for the host where the SVN tag is checked out.
> Oleg
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
> For additional commands, e-mail: dev-help@hc.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
For additional commands, e-mail: dev-help@hc.apache.org
Re: [VOTE] HttpComponents Client 4.2.6 release
Posted by Oleg Kalnichevski <ol...@apache.org>.
On Tue, 2013-09-03 at 10:57 +0100, sebb wrote:
> On 3 September 2013 10:48, Oleg Kalnichevski <ol...@apache.org> wrote:
> > On Tue, 2013-09-03 at 02:03 -0400, Karl Wright wrote:
> >> Hi Sebb,
> >>
> >> "It is essential that the files in a source release can be matched
> >> against those in SVN."
> >>
> >> I think that's technically not necessary according to Apache rules; what's
> >> necessary is that the bits in the tar or zip be based *solely* on the svn
> >> tag and upstream dependencies. But you really vote not on the tag but on
> >> the tar and zip.
> >>
> >> "But the main issue I have is that the PNG file in the ZIP archive is
> >> corrupt."
> >>
> >> I think that should be fixed. It's also worth considering that perhaps
> >> fixing up line endings for Windows users is maybe not worth the pain in the
> >> behind at release time that it is. Most people use editors these days that
> >> don't care.
> >>
> >> Karl
> >>
> >
> > Karl
> > Line delimiters of all human readable files as well as source code get
> > adjusted based on the type of the distribution. The whole thing boils
> > down to a just a few _resource_ files, which I personally think should
> > never be meddled with given that line delimiters may be semantically
> > significant.
>
> If a fixed eol is required by resource files, they should not have
> eol-style:native in SVN.
> They should have either LF or CRLF depending on which eol is required
> by the format.
>
> But if the format requires different EOL for Windows and Un*x, then
> native is correct and so is conversion to the appropriate ending.
>
> I am not saying that every text file needs to be converted.
> But ones which are converted by SVN (i.e. native) should be converted
> for the corresponding archive.
>
All which the assembly script has no way of knowing.
Oleg
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
For additional commands, e-mail: dev-help@hc.apache.org
Re: [VOTE] HttpComponents Client 4.2.6 release
Posted by sebb <se...@gmail.com>.
On 3 September 2013 10:48, Oleg Kalnichevski <ol...@apache.org> wrote:
> On Tue, 2013-09-03 at 02:03 -0400, Karl Wright wrote:
>> Hi Sebb,
>>
>> "It is essential that the files in a source release can be matched
>> against those in SVN."
>>
>> I think that's technically not necessary according to Apache rules; what's
>> necessary is that the bits in the tar or zip be based *solely* on the svn
>> tag and upstream dependencies. But you really vote not on the tag but on
>> the tar and zip.
>>
>> "But the main issue I have is that the PNG file in the ZIP archive is
>> corrupt."
>>
>> I think that should be fixed. It's also worth considering that perhaps
>> fixing up line endings for Windows users is maybe not worth the pain in the
>> behind at release time that it is. Most people use editors these days that
>> don't care.
>>
>> Karl
>>
>
> Karl
> Line delimiters of all human readable files as well as source code get
> adjusted based on the type of the distribution. The whole thing boils
> down to a just a few _resource_ files, which I personally think should
> never be meddled with given that line delimiters may be semantically
> significant.
If a fixed eol is required by resource files, they should not have
eol-style:native in SVN.
They should have either LF or CRLF depending on which eol is required
by the format.
But if the format requires different EOL for Windows and Un*x, then
native is correct and so is conversion to the appropriate ending.
I am not saying that every text file needs to be converted.
But ones which are converted by SVN (i.e. native) should be converted
for the corresponding archive.
> Oleg
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
> For additional commands, e-mail: dev-help@hc.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
For additional commands, e-mail: dev-help@hc.apache.org
Re: [VOTE] HttpComponents Client 4.2.6 release
Posted by Oleg Kalnichevski <ol...@apache.org>.
On Tue, 2013-09-03 at 02:03 -0400, Karl Wright wrote:
> Hi Sebb,
>
> "It is essential that the files in a source release can be matched
> against those in SVN."
>
> I think that's technically not necessary according to Apache rules; what's
> necessary is that the bits in the tar or zip be based *solely* on the svn
> tag and upstream dependencies. But you really vote not on the tag but on
> the tar and zip.
>
> "But the main issue I have is that the PNG file in the ZIP archive is
> corrupt."
>
> I think that should be fixed. It's also worth considering that perhaps
> fixing up line endings for Windows users is maybe not worth the pain in the
> behind at release time that it is. Most people use editors these days that
> don't care.
>
> Karl
>
Karl
Line delimiters of all human readable files as well as source code get
adjusted based on the type of the distribution. The whole thing boils
down to a just a few _resource_ files, which I personally think should
never be meddled with given that line delimiters may be semantically
significant.
Oleg
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
For additional commands, e-mail: dev-help@hc.apache.org
Re: [VOTE] HttpComponents Client 4.2.6 release
Posted by sebb <se...@gmail.com>.
On 3 September 2013 07:03, Karl Wright <da...@gmail.com> wrote:
> Hi Sebb,
>
> "It is essential that the files in a source release can be matched
> against those in SVN."
>
> I think that's technically not necessary according to Apache rules; what's
> necessary is that the bits in the tar or zip be based *solely* on the svn
> tag and upstream dependencies.
How can you check that the bits in the tar are based solely on the tag
if you cannot compare the tar contents against the tag?
> But you really vote not on the tag but on the tar and zip.
Yes, of course, but the files in the archive must have the correct license.
There are far too many files to check individually, so one relies on
the fact that files in the tag have been pre-approved.
Effectively SVN tag is a database of files that are believed OK at
that point in time.
> "But the main issue I have is that the PNG file in the ZIP archive is
> corrupt."
>
> I think that should be fixed.
Thanks.
> It's also worth considering that perhaps
> fixing up line endings for Windows users is maybe not worth the pain in the
> behind at release time that it is. Most people use editors these days that
> don't care.
We also fix up line endings for Un*x users if the RM happens to be
using Windows.
> Karl
>
>
>
> On Mon, Sep 2, 2013 at 9:38 PM, sebb <se...@gmail.com> wrote:
>
>> On 2 September 2013 18:28, Oleg Kalnichevski <ol...@apache.org> wrote:
>> > On Mon, 2013-09-02 at 21:59 +0530, Asankha C. Perera wrote:
>> >> On 09/02/2013 06:06 PM, Oleg Kalnichevski wrote:
>> >> > On Mon, 2013-09-02 at 11:53 +0100, sebb wrote:
>> >> > ...
>> >> >
>> >> > Further httpclient/src/test/resources/suffixlist.txt has got EOL=LF in
>> >> > the zip file so it disagrees with the SVN tag checkout.
>> >> >
>> >> > Also, the *.properties files ought to be CRLF on Windows (they are
>> >> > native in SVN)
>> >> > Likewise *.xsl and *.css and SPNEGO.svg
>> >> > Otherwise checkouts of the tag don't agree with the Zip
>> >> >
>> >> > I am sorry but I have no idea what this is all about. I have no idea
>> why
>> >> > they should agree in the first place.
>> >> I guess Sebastian's concern is that if you unzip the Zip on a Windows
>> >> box, and compare the result with a checkout of the tag on the same
>> >> Windows system, the line endings would be different between the two for
>> >> some files. Although that maybe a minor issue, if the differences are
>> >> related only to the line endings of a few files, I would not consider it
>> >> a blocker to the release. Personally I consider differences in line
>> >> endings as ignorable, and many IDEs or file compare tools also tend to
>> >> ignore them. I guess one who would look for the source releases would
>> >> prefer to get it directly via the SVN tag, and ones who download the
>> >> source archive of a release would generally want to refer to the code
>> >> for understanding or debugging etc - so the line endings would not be
>> >> that critical - my 2c
>> >
>> > Asankha
>> > I understand the technical bits. What I do not understand is why that
>> > would matter at all, why anyone in their sane mind would want to compare
>> > SVN tag with the content of ZIP archive and why on earth we need to
>> > waste our collective time on stuff like that?
>>
>> At present we create OS-specific archives - CRLF for zip (intended for
>> Windows) and LF for tar.gz (intended for Un*x).
>> Given that we do so, we ought to do it properly.
>>
>> It is essential that the files in a source release can be matched
>> against those in SVN.
>> That's how provenance is established, and how the reviewer knows that
>> the files are OK to release.
>> SVN is assumed to contain only files with correct licences etc.
>> It makes the reviewers job much easier if the line endings agree.
>>
>> It's also easier for Windows developers if text files are in CRLF
>> format. CRLF files are awkward on Un*x systems.
>>
>> But the main issue I have is that the PNG file in the ZIP archive is
>> corrupt.
>>
>>
>> > Oleg
>> >
>> >
>> >
>> > ---------------------------------------------------------------------
>> > To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
>> > For additional commands, e-mail: dev-help@hc.apache.org
>> >
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
>> For additional commands, e-mail: dev-help@hc.apache.org
>>
>>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
For additional commands, e-mail: dev-help@hc.apache.org
Re: [VOTE] HttpComponents Client 4.2.6 release
Posted by Karl Wright <da...@gmail.com>.
Hi Sebb,
"It is essential that the files in a source release can be matched
against those in SVN."
I think that's technically not necessary according to Apache rules; what's
necessary is that the bits in the tar or zip be based *solely* on the svn
tag and upstream dependencies. But you really vote not on the tag but on
the tar and zip.
"But the main issue I have is that the PNG file in the ZIP archive is
corrupt."
I think that should be fixed. It's also worth considering that perhaps
fixing up line endings for Windows users is maybe not worth the pain in the
behind at release time that it is. Most people use editors these days that
don't care.
Karl
On Mon, Sep 2, 2013 at 9:38 PM, sebb <se...@gmail.com> wrote:
> On 2 September 2013 18:28, Oleg Kalnichevski <ol...@apache.org> wrote:
> > On Mon, 2013-09-02 at 21:59 +0530, Asankha C. Perera wrote:
> >> On 09/02/2013 06:06 PM, Oleg Kalnichevski wrote:
> >> > On Mon, 2013-09-02 at 11:53 +0100, sebb wrote:
> >> > ...
> >> >
> >> > Further httpclient/src/test/resources/suffixlist.txt has got EOL=LF in
> >> > the zip file so it disagrees with the SVN tag checkout.
> >> >
> >> > Also, the *.properties files ought to be CRLF on Windows (they are
> >> > native in SVN)
> >> > Likewise *.xsl and *.css and SPNEGO.svg
> >> > Otherwise checkouts of the tag don't agree with the Zip
> >> >
> >> > I am sorry but I have no idea what this is all about. I have no idea
> why
> >> > they should agree in the first place.
> >> I guess Sebastian's concern is that if you unzip the Zip on a Windows
> >> box, and compare the result with a checkout of the tag on the same
> >> Windows system, the line endings would be different between the two for
> >> some files. Although that maybe a minor issue, if the differences are
> >> related only to the line endings of a few files, I would not consider it
> >> a blocker to the release. Personally I consider differences in line
> >> endings as ignorable, and many IDEs or file compare tools also tend to
> >> ignore them. I guess one who would look for the source releases would
> >> prefer to get it directly via the SVN tag, and ones who download the
> >> source archive of a release would generally want to refer to the code
> >> for understanding or debugging etc - so the line endings would not be
> >> that critical - my 2c
> >
> > Asankha
> > I understand the technical bits. What I do not understand is why that
> > would matter at all, why anyone in their sane mind would want to compare
> > SVN tag with the content of ZIP archive and why on earth we need to
> > waste our collective time on stuff like that?
>
> At present we create OS-specific archives - CRLF for zip (intended for
> Windows) and LF for tar.gz (intended for Un*x).
> Given that we do so, we ought to do it properly.
>
> It is essential that the files in a source release can be matched
> against those in SVN.
> That's how provenance is established, and how the reviewer knows that
> the files are OK to release.
> SVN is assumed to contain only files with correct licences etc.
> It makes the reviewers job much easier if the line endings agree.
>
> It's also easier for Windows developers if text files are in CRLF
> format. CRLF files are awkward on Un*x systems.
>
> But the main issue I have is that the PNG file in the ZIP archive is
> corrupt.
>
>
> > Oleg
> >
> >
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
> > For additional commands, e-mail: dev-help@hc.apache.org
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
> For additional commands, e-mail: dev-help@hc.apache.org
>
>
Re: [VOTE] HttpComponents Client 4.2.6 release
Posted by sebb <se...@gmail.com>.
On 2 September 2013 18:28, Oleg Kalnichevski <ol...@apache.org> wrote:
> On Mon, 2013-09-02 at 21:59 +0530, Asankha C. Perera wrote:
>> On 09/02/2013 06:06 PM, Oleg Kalnichevski wrote:
>> > On Mon, 2013-09-02 at 11:53 +0100, sebb wrote:
>> > ...
>> >
>> > Further httpclient/src/test/resources/suffixlist.txt has got EOL=LF in
>> > the zip file so it disagrees with the SVN tag checkout.
>> >
>> > Also, the *.properties files ought to be CRLF on Windows (they are
>> > native in SVN)
>> > Likewise *.xsl and *.css and SPNEGO.svg
>> > Otherwise checkouts of the tag don't agree with the Zip
>> >
>> > I am sorry but I have no idea what this is all about. I have no idea why
>> > they should agree in the first place.
>> I guess Sebastian's concern is that if you unzip the Zip on a Windows
>> box, and compare the result with a checkout of the tag on the same
>> Windows system, the line endings would be different between the two for
>> some files. Although that maybe a minor issue, if the differences are
>> related only to the line endings of a few files, I would not consider it
>> a blocker to the release. Personally I consider differences in line
>> endings as ignorable, and many IDEs or file compare tools also tend to
>> ignore them. I guess one who would look for the source releases would
>> prefer to get it directly via the SVN tag, and ones who download the
>> source archive of a release would generally want to refer to the code
>> for understanding or debugging etc - so the line endings would not be
>> that critical - my 2c
>
> Asankha
> I understand the technical bits. What I do not understand is why that
> would matter at all, why anyone in their sane mind would want to compare
> SVN tag with the content of ZIP archive and why on earth we need to
> waste our collective time on stuff like that?
At present we create OS-specific archives - CRLF for zip (intended for
Windows) and LF for tar.gz (intended for Un*x).
Given that we do so, we ought to do it properly.
It is essential that the files in a source release can be matched
against those in SVN.
That's how provenance is established, and how the reviewer knows that
the files are OK to release.
SVN is assumed to contain only files with correct licences etc.
It makes the reviewers job much easier if the line endings agree.
It's also easier for Windows developers if text files are in CRLF
format. CRLF files are awkward on Un*x systems.
But the main issue I have is that the PNG file in the ZIP archive is corrupt.
> Oleg
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
> For additional commands, e-mail: dev-help@hc.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
For additional commands, e-mail: dev-help@hc.apache.org
Re: [VOTE] HttpComponents Client 4.2.6 release
Posted by Oleg Kalnichevski <ol...@apache.org>.
On Mon, 2013-09-02 at 21:59 +0530, Asankha C. Perera wrote:
> On 09/02/2013 06:06 PM, Oleg Kalnichevski wrote:
> > On Mon, 2013-09-02 at 11:53 +0100, sebb wrote:
> > ...
> >
> > Further httpclient/src/test/resources/suffixlist.txt has got EOL=LF in
> > the zip file so it disagrees with the SVN tag checkout.
> >
> > Also, the *.properties files ought to be CRLF on Windows (they are
> > native in SVN)
> > Likewise *.xsl and *.css and SPNEGO.svg
> > Otherwise checkouts of the tag don't agree with the Zip
> >
> > I am sorry but I have no idea what this is all about. I have no idea why
> > they should agree in the first place.
> I guess Sebastian's concern is that if you unzip the Zip on a Windows
> box, and compare the result with a checkout of the tag on the same
> Windows system, the line endings would be different between the two for
> some files. Although that maybe a minor issue, if the differences are
> related only to the line endings of a few files, I would not consider it
> a blocker to the release. Personally I consider differences in line
> endings as ignorable, and many IDEs or file compare tools also tend to
> ignore them. I guess one who would look for the source releases would
> prefer to get it directly via the SVN tag, and ones who download the
> source archive of a release would generally want to refer to the code
> for understanding or debugging etc - so the line endings would not be
> that critical - my 2c
Asankha
I understand the technical bits. What I do not understand is why that
would matter at all, why anyone in their sane mind would want to compare
SVN tag with the content of ZIP archive and why on earth we need to
waste our collective time on stuff like that?
Oleg
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
For additional commands, e-mail: dev-help@hc.apache.org
Re: [VOTE] HttpComponents Client 4.2.6 release
Posted by "Asankha C. Perera" <as...@apache.org>.
On 09/02/2013 06:06 PM, Oleg Kalnichevski wrote:
> On Mon, 2013-09-02 at 11:53 +0100, sebb wrote:
> ...
>
> Further httpclient/src/test/resources/suffixlist.txt has got EOL=LF in
> the zip file so it disagrees with the SVN tag checkout.
>
> Also, the *.properties files ought to be CRLF on Windows (they are
> native in SVN)
> Likewise *.xsl and *.css and SPNEGO.svg
> Otherwise checkouts of the tag don't agree with the Zip
>
> I am sorry but I have no idea what this is all about. I have no idea why
> they should agree in the first place.
I guess Sebastian's concern is that if you unzip the Zip on a Windows
box, and compare the result with a checkout of the tag on the same
Windows system, the line endings would be different between the two for
some files. Although that maybe a minor issue, if the differences are
related only to the line endings of a few files, I would not consider it
a blocker to the release. Personally I consider differences in line
endings as ignorable, and many IDEs or file compare tools also tend to
ignore them. I guess one who would look for the source releases would
prefer to get it directly via the SVN tag, and ones who download the
source archive of a release would generally want to refer to the code
for understanding or debugging etc - so the line endings would not be
that critical - my 2c
regards
asankha
--
Asankha C. Perera
AdroitLogic, http://adroitlogic.org
http://esbmagic.blogspot.com
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
For additional commands, e-mail: dev-help@hc.apache.org
Re: [VOTE] HttpComponents Client 4.2.6 release
Posted by Oleg Kalnichevski <ol...@apache.org>.
On Mon, 2013-09-02 at 11:53 +0100, sebb wrote:
> Sorry, but -1
> Something has gone wrong with the EOL conversions.
>
Release veto because of a corrupt PNG file in source zip dist?
> The file
>
> httpclient/src/main/java/org/apache/http/impl/conn/tsccm/doc-files/tsccm-structure.png
>
> is different between SVN and the zip archive.
> The tar.gz copy looks fine.
> The SVN mime-type is image/png, which is fine.
>
> Looks like the PNG file was subject to EOL conversion, as it is bigger
> in the zip archive.
>
> Further httpclient/src/test/resources/suffixlist.txt has got EOL=LF in
> the zip file so it disagrees with the SVN tag checkout.
>
> Also, the *.properties files ought to be CRLF on Windows (they are
> native in SVN)
> Likewise *.xsl and *.css and SPNEGO.svg
> Otherwise checkouts of the tag don't agree with the Zip
>
I am sorry but I have no idea what this is all about. I have no idea why
they should agree in the first place.
Oleg
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
For additional commands, e-mail: dev-help@hc.apache.org
Re: [VOTE] HttpComponents Client 4.2.6 release
Posted by sebb <se...@gmail.com>.
Sorry, but -1
Something has gone wrong with the EOL conversions.
The file
httpclient/src/main/java/org/apache/http/impl/conn/tsccm/doc-files/tsccm-structure.png
is different between SVN and the zip archive.
The tar.gz copy looks fine.
The SVN mime-type is image/png, which is fine.
Looks like the PNG file was subject to EOL conversion, as it is bigger
in the zip archive.
Further httpclient/src/test/resources/suffixlist.txt has got EOL=LF in
the zip file so it disagrees with the SVN tag checkout.
Also, the *.properties files ought to be CRLF on Windows (they are
native in SVN)
Likewise *.xsl and *.css and SPNEGO.svg
Otherwise checkouts of the tag don't agree with the Zip
On 1 September 2013 11:52, Oleg Kalnichevski <ol...@apache.org> wrote:
> Please vote on releasing these packages as HttpComponents Client 4.2.6.
> The vote is open for the at least 72 hours, and only votes from
> HttpComponents PMC members are binding. The vote passes if at least
> three binding +1 votes are cast and there are more +1 than -1 votes.
>
> Packages:
> https://dist.apache.org/repos/dist/dev/httpcomponents/httpclient-4.2.6-RC1/
>
> Release notes:
> https://dist.apache.org/repos/dist/dev/httpcomponents/httpclient-4.2.6-RC1/RELEASE_NOTES-4.2.x.txt
>
> Maven artefacts:
> https://repository.apache.org/content/repositories/orgapachehttpcomponents-129/org/apache/httpcomponents/
>
> SVN tag:
> https://svn.apache.org/repos/asf/httpcomponents/httpclient/tags/4.2.6-RC1/
> (Revision: 1519258)
>
> --------------------------------------------------------------------------
> Vote: HttpComponents Client 4.2.6 release
> [ ] +1 Release the packages as HttpComponents Client 4.2.6.
> [ ] -1 I am against releasing the packages (must include a reason).
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
> For additional commands, e-mail: dev-help@hc.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
For additional commands, e-mail: dev-help@hc.apache.org
Re: [VOTE] HttpComponents Client 4.2.6 release
Posted by ant elder <an...@gmail.com>.
+1
...ant
On Sun, Sep 1, 2013 at 11:52 AM, Oleg Kalnichevski <ol...@apache.org> wrote:
> Please vote on releasing these packages as HttpComponents Client 4.2.6.
> The vote is open for the at least 72 hours, and only votes from
> HttpComponents PMC members are binding. The vote passes if at least
> three binding +1 votes are cast and there are more +1 than -1 votes.
>
> Packages:
> https://dist.apache.org/repos/dist/dev/httpcomponents/httpclient-4.2.6-RC1/
>
> Release notes:
>
> https://dist.apache.org/repos/dist/dev/httpcomponents/httpclient-4.2.6-RC1/RELEASE_NOTES-4.2.x.txt
>
> Maven artefacts:
>
> https://repository.apache.org/content/repositories/orgapachehttpcomponents-129/org/apache/httpcomponents/
>
> SVN tag:
> https://svn.apache.org/repos/asf/httpcomponents/httpclient/tags/4.2.6-RC1/
> (Revision: 1519258)
>
> --------------------------------------------------------------------------
> Vote: HttpComponents Client 4.2.6 release
> [ ] +1 Release the packages as HttpComponents Client 4.2.6.
> [ ] -1 I am against releasing the packages (must include a reason).
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
> For additional commands, e-mail: dev-help@hc.apache.org
>
>