Re: Framework metadata

2013-12-20 Thread Alex Merry
On 20/12/13 19:39, Aurélien Gâteau wrote: > I created $framework.yaml file for all frameworks, and updated kf5dot to make > use of it (hackish for now). You can find updated dependency diagrams at: > > http://agateau.com/tmp/kf5 > > I created placeholders for the missing README.md and rena

Re: Framework metadata

2013-12-20 Thread Cornelius Schumacher
On Friday 20 December 2013 Aurélien Gâteau wrote: > > Are we all OK with this short-term plan? I'm very much in favor with anything which gets us moving forward. Bikeshedding about the format is not one of these things. It would be great, if the fields needed for Inqlude would be available in a

Re: Framework metadata

2013-12-20 Thread Aurélien Gâteau
Le vendredi 20 décembre 2013 14:15:33 Aurélien Gâteau a écrit : > Mandatory files for a framework: > > # README > > Formatted in Markdown. Contains a high level overview of the framework and > links to: > - mailing list > - IRC channel > - git repo > - bugtracker > - reviewboard > > # INSTALL >

Re: Framework metadata

2013-12-20 Thread Aurélien Gâteau
Le jeudi 19 décembre 2013 20:50:43 Nicolás Alvarez a écrit : > 2013/12/19 Aurélien Gâteau : > > On Thu, 19 Dec 2013 12:13:08 +0200, Giorgos Tsiapaliokas wrote: > >> I definitely like the idea but I don't like xml/rdf. Why don't we just > >> use json? > > > > Because DOAP is an existing format, so

Re: Framework metadata

2013-12-20 Thread Aaron J. Seigo
On Thursday, December 19, 2013 21:58:36 Martin Graesslin wrote: > A standard which is used, is better than a not-used proprietary format. No > matter what language it is. In the end it's about being machine readable and > any machine readable language can be converted into another. Don't like > XML

Re: Framework metadata

2013-12-19 Thread Nicolás Alvarez
2013/12/19 Aurélien Gâteau : > On Thu, 19 Dec 2013 12:13:08 +0200, Giorgos Tsiapaliokas wrote: >> I definitely like the idea but I don't like xml/rdf. Why don't we just >> use json? > > Because DOAP is an existing format, so it avoids reinventing the wheel. Everyone please keep in mind that DOAP i

Re: Framework metadata

2013-12-19 Thread Martin Graesslin
On Thursday 19 December 2013 14:35:29 Aurélien Gâteau wrote: > On Thu, 19 Dec 2013 12:13:08 +0200, Giorgos Tsiapaliokas wrote: > > I definitely like the idea but I don't like xml/rdf. Why don't we > > just > > use json? > > Because DOAP is an existing format, so it avoids reinventing the wheel. R

Re: Framework metadata

2013-12-19 Thread Cornelius Schumacher
On Thursday 19 December 2013 14:58:36 Aurélien Gâteau wrote: > > > * list of licenses > > Project/license Can this be a list? -- Cornelius Schumacher ___ Kde-frameworks-devel mailing list Kde-frameworks-devel@kde.org https://mail.kde.org/mailman/lis

Re: Framework metadata

2013-12-19 Thread Aurélien Gâteau
On Thu, 19 Dec 2013 17:18:37 +0100, Cornelius Schumacher wrote: On Thursday 19 December 2013 14:58:36 Aurélien Gâteau wrote: > * list of licenses Project/license Can this be a list? Just checked: you can have multiple "license" elements in the "Project" element, see: https://github.com/ed

Re: Framework metadata

2013-12-19 Thread Martin Sandsmark
On Wed, Dec 18, 2013 at 09:35:33PM +0100, Christophe Giboudeaux wrote: > ./tier1/kjs/src/kjs/context.h: *No copyright* UNKNOWN This is a single line: #include I don't think this is copyrightable? > ./tier1/kjs/src/kjs/dtoa.cpp: UNKNOWN /*

Re: Framework metadata

2013-12-19 Thread Aurélien Gâteau
On Wed, 18 Dec 2013 21:35:33 +0100, Christophe Giboudeaux wrote: On Wednesday 18 December 2013 19:50:58 Alex Merry wrote: On 18/12/13 17:54, Aurélien Gâteau wrote: > We need to have at least a COPYING file in there, with the full content of > the license. I believe this was already done befor

Re: Framework metadata

2013-12-19 Thread Aurélien Gâteau
On Thu, 19 Dec 2013 08:47:01 +0100, Cornelius Schumacher wrote: On Wednesday 18 December 2013 Aurélien Gâteau wrote: The information in the DOAP file can also be used to generate manifest files for Inqlude (http://inqlude.org/) For this to work we need at least the following data in the DOA

Re: Framework metadata

2013-12-19 Thread Aurélien Gâteau
On Thu, 19 Dec 2013 12:13:08 +0200, Giorgos Tsiapaliokas wrote: I definitely like the idea but I don't like xml/rdf. Why don't we just use json?  Because DOAP is an existing format, so it avoids reinventing the wheel. If we decide we want to reinvent the wheel, then let's use YAML rather than

Re: Framework metadata

2013-12-19 Thread Ben Cooksley
On Thu, Dec 19, 2013 at 10:53 PM, John Layt wrote: > On 19 December 2013 08:47, Cornelius Schumacher wrote: >> On Wednesday 18 December 2013 Aurélien Gâteau wrote: >>> >>> The information in the DOAP file can also be used to generate manifest >>> files for Inqlude (http://inqlude.org/) >> >> For

Re: Framework metadata

2013-12-19 Thread Giorgos Tsiapaliokas
I definitely like the idea but I don't like xml/rdf. Why don't we just use json? Qt is already using json and the logical-module-structure

Re: Framework metadata

2013-12-19 Thread John Layt
On 19 December 2013 08:47, Cornelius Schumacher wrote: > On Wednesday 18 December 2013 Aurélien Gâteau wrote: >> >> The information in the DOAP file can also be used to generate manifest >> files for Inqlude (http://inqlude.org/) > > For this to work we need at least the following data in the DOAP

Re: Framework metadata

2013-12-19 Thread John Layt
On 19 December 2013 08:47, Cornelius Schumacher wrote: > On Wednesday 18 December 2013 Aurélien Gâteau wrote: >> >> The information in the DOAP file can also be used to generate manifest >> files for Inqlude (http://inqlude.org/) > > For this to work we need at least the following data in the DOAP

Re: Framework metadata

2013-12-19 Thread Ben Cooksley
On Thu, Dec 19, 2013 at 9:52 PM, Martin Graesslin wrote: > On Thursday 19 December 2013 21:22:36 you wrote: >> On Thu, Dec 19, 2013 at 9:15 PM, Martin Graesslin > wrote: >> > On Wednesday 18 December 2013 21:35:33 Christophe Giboudeaux wrote: >> >> On Wednesday 18 December 2013 19:50:58 Alex Merr

Re: Framework metadata

2013-12-19 Thread Martin Graesslin
On Thursday 19 December 2013 21:22:36 you wrote: > On Thu, Dec 19, 2013 at 9:15 PM, Martin Graesslin wrote: > > On Wednesday 18 December 2013 21:35:33 Christophe Giboudeaux wrote: > >> On Wednesday 18 December 2013 19:50:58 Alex Merry wrote: > >> > On 18/12/13 17:54, Aurélien Gâteau wrote: > >> >

Re: Framework metadata

2013-12-19 Thread Ben Cooksley
On Thu, Dec 19, 2013 at 9:15 PM, Martin Graesslin wrote: > On Wednesday 18 December 2013 21:35:33 Christophe Giboudeaux wrote: >> On Wednesday 18 December 2013 19:50:58 Alex Merry wrote: >> > On 18/12/13 17:54, Aurélien Gâteau wrote: >> > > We need to have at least a COPYING file in there, with th

Re: Framework metadata

2013-12-19 Thread Martin Graesslin
On Wednesday 18 December 2013 21:35:33 Christophe Giboudeaux wrote: > On Wednesday 18 December 2013 19:50:58 Alex Merry wrote: > > On 18/12/13 17:54, Aurélien Gâteau wrote: > > > We need to have at least a COPYING file in there, with the full content > > > of > > > the license. > > > > I believe t

Re: Framework metadata

2013-12-18 Thread Cornelius Schumacher
On Wednesday 18 December 2013 Aurélien Gâteau wrote: > > The information in the DOAP file can also be used to generate manifest > files for Inqlude (http://inqlude.org/) For this to work we need at least the following data in the DOAP file: * machine-readable name as identifier (all lower-case,

Re: Framework metadata

2013-12-18 Thread Christophe Giboudeaux
On Wednesday 18 December 2013 19:50:58 Alex Merry wrote: > On 18/12/13 17:54, Aurélien Gâteau wrote: > > We need to have at least a COPYING file in there, with the full content of > > the license. > > I believe this was already done before the split. > I added COPYING files in each frameworks which

Re: Framework metadata

2013-12-18 Thread Alex Merry
On 18/12/13 17:54, Aurélien Gâteau wrote: > We need to have at least a COPYING file in there, with the full content of > the > license. I believe this was already done before the split. > Then there is the README file. Historically we haven't been very good at > keeping them up to date, have a