[Development] Implementing of a DNS feature under Windows [QTBUG-30166]

2013-09-26 Thread Mandeep Sandhu
Hi All,

I'm working on adding support for a custom DNS server in QDnsLookup.

The *nix implementation is complete, however I needed some pointers on
doing it under Windows.

The process of specifying a specific DNS server is sort of 'undocumented'.
I've looked at the current implementation of DNS lookup under windows (in
src/network/kernel/qdnslookup_win.cpp).

1. We're using the DnsQuery_W function. The docs of this function state
that the 4th param
(pExtra) is reserved for future use and must be set to NULL*.**
*
However, sample code on MSDN show this param being used for specifying a
custom DNS server*: *http://support.microsoft.com/kb/831226
My question is, can we use this feature (apparently the docs seem to
incorrect on this matter)?

2. For allocating memory for holding the IP addr of the nameserver, the
sample code (see above link) uses the 'LocalAlloc' function.

This function does NOT generate any exceptions and returns NULL (like
malloc) on failure. So in case of error, I'm not able to find a suitable
QDnsLookup::Error enum value for it as the current ones seem to be related
to DNS problems and not system related.

How should a failed mem alloc be handled in windows? (Should I use the
Heap* functions which are recommended for newer apps and generate
exceptions on failure?)

3. I'll be using the PIP4_ARRAY struct for specifying the nameserver (
http://msdn.microsoft.com/en-us/library/windows/desktop/ms682139(v=vs.85).aspx
).

I'm not sure if AddrArray field of this struct requires the IP octet in
host or network order. All the examples seem to be showing converting from
a string using the inet_addr() function. Since I already have the ip addr
as a quint32, I can set it directly. Though I'm a little unsure about the
byte order.

Any pointers are highly appreciated.

Thanks,
-mandeep
___
Development mailing list
Development@qt-project.org
http://lists.qt-project.org/mailman/listinfo/development


[Development] Qt5 releasing wiki

2013-09-26 Thread Heikkinen Jani
Hi all,

First version of Qt5 releasing wiki pages are available, see 
http://qt-project.org/wiki/Qt5Releasing. Target is to give basic information 
about releasing process, schedule and so on. Be free to add missing 
information and fix mistakes (at least definition of release phases should be 
written in own subpages, hoping someone can help with this...).

Comments are welcome!

Br,
Jani
___
Development mailing list
Development@qt-project.org
http://lists.qt-project.org/mailman/listinfo/development


Re: [Development] [Releasing] HEADS UP: Starting preparations for Qt 5.2.0 - merge 'dev' into 'stable'

2013-09-26 Thread Sergio Ahumada
 
 Hi,
 
 All merges are now done, so changes required for Qt 5.2.0 need to be
 pushed/re-targeted to the 'stable' branch from now on. No new features
 are allowed unless they are approved by Lars.
 
 Changes in the current 'dev' branch will be part of the Qt 5.3.0
 release. Bumping the Qt version in 'dev' starts with
 https://codereview.qt-project.org/66129 after 'dev' opens again in a
 couple of days.
 
 Cheers,
 

Hi,

Three days have passed so we are opening 'dev' again.

Please be aware that 'dev' is 5.3.0 now, 'stable' is 5.2.0 and release
is still '5.1.1'

Cheers,
-- 
Sergio Ahumada
Release Engineer - Digia, Qt
___
Development mailing list
Development@qt-project.org
http://lists.qt-project.org/mailman/listinfo/development


[Development] Novice question for first-time contributions

2013-09-26 Thread Mandeep Sandhu
Hi All,

Couple of days back I had fixed 2 bugs: one was a low priority one [
QTBUG-33439 https://bugreports.qt-project.org/browse/QTBUG-33439] and
another was not evaluated yet
[QTBUG-32911https://bugreports.qt-project.org/browse/QTBUG-32911],
though a similar bug
[QTBUG-31753https://bugreports.qt-project.org/browse/QTBUG-31753]
was evaluated as important (they were essentially due the same problem).

I had pushed my changes to the ref- refs/for/stable (as shown in the
Gerrit wiki page). Is that correct or do I need to send it to the 'dev'
branch?

Also, whats the correct way to solicit a review? Should I check with
reviewer about his/her availability before adding them on Gerrit?

Thanks,
-mandeep
___
Development mailing list
Development@qt-project.org
http://lists.qt-project.org/mailman/listinfo/development


Re: [Development] Novice question for first-time contributions

2013-09-26 Thread Paul Olav Tvete
On Thursday 26 September 2013 17:12:09 Mandeep Sandhu wrote:

 I had pushed my changes to the ref- refs/for/stable (as shown in the
 Gerrit wiki page). Is that correct or do I need to send it to the 'dev'
 branch?

Stable is correct for bug fixes that target the upcoming release.

 Also, whats the correct way to solicit a review? Should I check with
 reviewer about his/her availability before adding them on Gerrit?

Just add people on gerrit. If you get it wrong it's no big deal, and it is 
easy to remove oneself from a change.

...and last but not least: Hi, and welcome to the Qt project :)

- Paul
___
Development mailing list
Development@qt-project.org
http://lists.qt-project.org/mailman/listinfo/development


Re: [Development] Novice question for first-time contributions

2013-09-26 Thread Thiago Macieira
On quinta-feira, 26 de setembro de 2013 17:12:09, Mandeep Sandhu wrote:
 Hi All,
 
 Couple of days back I had fixed 2 bugs: one was a low priority one [
 QTBUG-33439 https://bugreports.qt-project.org/browse/QTBUG-33439] and
 another was not evaluated yet
 [QTBUG-32911https://bugreports.qt-project.org/browse/QTBUG-32911],
 though a similar bug
 [QTBUG-31753https://bugreports.qt-project.org/browse/QTBUG-31753]
 was evaluated as important (they were essentially due the same problem).
 
 I had pushed my changes to the ref- refs/for/stable (as shown in the
 Gerrit wiki page). Is that correct or do I need to send it to the 'dev'
 branch?

Just follow the branch guidelines: stable is for bugfixes (though not all 
bugfixes), dev is for new features and for bugfixes that require a little more 
attention; release is for the release and release-critical bugfixes.

You always submit to the most-frozen branch that the patch applies to. So if 
you're fixing a bug and it applies to stable, that's where you send.

 Also, whats the correct way to solicit a review? Should I check with
 reviewer about his/her availability before adding them on Gerrit?

No, just add the reviewers, they'll review when they get the time. You can find 
them by doing git log in the files you changed. Worst case scenario, add the 
maintainer.

If you get no reviews in a while, add the maintainer (if that person isn't 
already there) and say you're requesting that the maintainer step in.

-- 
Thiago Macieira - thiago.macieira (AT) intel.com
  Software Architect - Intel Open Source Technology Center


signature.asc
Description: This is a digitally signed message part.
___
Development mailing list
Development@qt-project.org
http://lists.qt-project.org/mailman/listinfo/development


Re: [Development] Minified javascript libraries in Qt source code

2013-09-26 Thread Thiago Macieira
On quinta-feira, 26 de setembro de 2013 11:31:07, Lisandro Damián Nicanor 
Pérez Meyer wrote:
 This means that if Qt ships a minified javascript library, a Debian
 maintainer  should either:
 
 a) Repack the source tarball without the minified javascript libs and use
 the  system provided ones prior to start the building process.
 
 b) Repack the source tarball to include the unminified javascript libs and 
 either minify them upon build time or use them unminified (after all, they
 are  mostly used in examples).
 
 c) Upon proper configuration (like ./configure --use-system-js), use the 
 system lib to provide the (un)minified js.

Since we wouldn't be modifying the libraries, we'd just be passing them 
through, can we have option d:

 d) Ship the libraries unmodified from upstream and include a link to where 
they were obtained and where to obtain the non-minified sources.

Or is that not acceptable as per our own licensing? That is, since we're 
shipping our sources as LGPLv2.1, do we need to ship the 3rdparty JS libraries 
we use in their preferred form for modification?

PS: examples are licensed BSD.
-- 
Thiago Macieira - thiago.macieira (AT) intel.com
  Software Architect - Intel Open Source Technology Center


signature.asc
Description: This is a digitally signed message part.
___
Development mailing list
Development@qt-project.org
http://lists.qt-project.org/mailman/listinfo/development


Re: [Development] Minified javascript libraries in Qt source code

2013-09-26 Thread Lisandro Damián Nicanor Pérez Meyer
On Thursday 26 September 2013 09:42:01 Thiago Macieira wrote:
 On quinta-feira, 26 de setembro de 2013 11:31:07, Lisandro Damián Nicanor
 
 Pérez Meyer wrote:
  This means that if Qt ships a minified javascript library, a Debian
  maintainer  should either:
  
  a) Repack the source tarball without the minified javascript libs and use
  the  system provided ones prior to start the building process.
  
  b) Repack the source tarball to include the unminified javascript libs and
  either minify them upon build time or use them unminified (after all, they
  are  mostly used in examples).
  
  c) Upon proper configuration (like ./configure --use-system-js), use the
  system lib to provide the (un)minified js.
 
 Since we wouldn't be modifying the libraries, we'd just be passing them
 through, can we have option d:
 
  d) Ship the libraries unmodified from upstream and include a link to where
 they were obtained and where to obtain the non-minified sources.
 
 Or is that not acceptable as per our own licensing? That is, since we're
 shipping our sources as LGPLv2.1, do we need to ship the 3rdparty JS
 libraries we use in their preferred form for modification?

webkit-fancybrowser-jquery-qrc.html and webkit-fancybrowser-jquery-min-js.html 
contain a copy of the minified jquery. As the best of my knowledge, this are 
GFDL-NIV licensed (please correct me if I am wrong).
 
 PS: examples are licensed BSD.

So shipping then minified should not be a problem for Qt, at least in the 
example's case. And also in the best of my knowledge.

-- 

Lisandro Damián Nicanor Pérez Meyer
http://perezmeyer.com.ar/
http://perezmeyer.blogspot.com/


signature.asc
Description: This is a digitally signed message part.
___
Development mailing list
Development@qt-project.org
http://lists.qt-project.org/mailman/listinfo/development


Re: [Development] Minified javascript libraries in Qt source code

2013-09-26 Thread Lisandro Damián Nicanor Pérez Meyer
On Thursday 26 September 2013 15:07:28 Lisandro Damián Nicanor Pérez Meyer 
wrote:
 On Thursday 26 September 2013 09:42:01 Thiago Macieira wrote:
  On quinta-feira, 26 de setembro de 2013 11:31:07, Lisandro Damián Nicanor
  
  Pérez Meyer wrote:
   This means that if Qt ships a minified javascript library, a Debian
   maintainer  should either:
   
   a) Repack the source tarball without the minified javascript libs and
   use
   the  system provided ones prior to start the building process.
   
   b) Repack the source tarball to include the unminified javascript libs
   and
   either minify them upon build time or use them unminified (after all,
   they
   are  mostly used in examples).
   
   c) Upon proper configuration (like ./configure --use-system-js), use the
   system lib to provide the (un)minified js.
  
  Since we wouldn't be modifying the libraries, we'd just be passing them
  
  through, can we have option d:
   d) Ship the libraries unmodified from upstream and include a link to
   where
  
  they were obtained and where to obtain the non-minified sources.
  
  Or is that not acceptable as per our own licensing? That is, since we're
  shipping our sources as LGPLv2.1, do we need to ship the 3rdparty JS
  libraries we use in their preferred form for modification?
 
 webkit-fancybrowser-jquery-qrc.html and
 webkit-fancybrowser-jquery-min-js.html contain a copy of the minified
 jquery. As the best of my knowledge, this are GFDL-NIV licensed (please
 correct me if I am wrong).
 

qt4's tests/benchmarks/script/sunspider/tests/string-unpack-code.js also seems 
to contain minified js, but I don't know for sure the license of that code.


-- 

Lisandro Damián Nicanor Pérez Meyer
http://perezmeyer.com.ar/
http://perezmeyer.blogspot.com/


signature.asc
Description: This is a digitally signed message part.
___
Development mailing list
Development@qt-project.org
http://lists.qt-project.org/mailman/listinfo/development


Re: [Development] Novice question for first-time contributions

2013-09-26 Thread Mandeep Sandhu
On Thu, Sep 26, 2013 at 10:06 PM, Thiago Macieira thiago.macie...@intel.com
 wrote:

 On quinta-feira, 26 de setembro de 2013 17:12:09, Mandeep Sandhu wrote:
  Hi All,
 
  Couple of days back I had fixed 2 bugs: one was a low priority one [
  QTBUG-33439 https://bugreports.qt-project.org/browse/QTBUG-33439] and
  another was not evaluated yet
  [QTBUG-32911https://bugreports.qt-project.org/browse/QTBUG-32911],
  though a similar bug
  [QTBUG-31753https://bugreports.qt-project.org/browse/QTBUG-31753]
  was evaluated as important (they were essentially due the same problem).
 
  I had pushed my changes to the ref- refs/for/stable (as shown in the
  Gerrit wiki page). Is that correct or do I need to send it to the 'dev'
  branch?

 Just follow the branch guidelines: stable is for bugfixes (though not all
 bugfixes), dev is for new features and for bugfixes that require a little
 more
 attention; release is for the release and release-critical bugfixes.

 You always submit to the most-frozen branch that the patch applies to. So
 if
 you're fixing a bug and it applies to stable, that's where you send.


Thanks for this info! I'm going to keep this handy. Maybe we can add this
to the wiki too (Gerrit Flow/Commit policy pages).



  Also, whats the correct way to solicit a review? Should I check with
  reviewer about his/her availability before adding them on Gerrit?

 No, just add the reviewers, they'll review when they get the time. You can
 find
 them by doing git log in the files you changed. Worst case scenario, add
 the
 maintainer.

 If you get no reviews in a while, add the maintainer (if that person isn't
 already there) and say you're requesting that the maintainer step in.


Yeah, that's what I did, added the maintainer. But now I'm thinking that
they might already have a long queue and might not get time to review
non-critical stuff.
How do I find the next-best person for review? Check a file's git log and
see who's committed to it most often?

-mandeep
___
Development mailing list
Development@qt-project.org
http://lists.qt-project.org/mailman/listinfo/development