On Fri, Sep 16, 2016 at 10:32 AM, Israel <israeldahl@xxxxxxxxx
<mailto:israeldahl@xxxxxxxxx>> wrote:
On 09/15/2016 05:58 PM, Ali/amjjawad wrote:
Dear all,
Hi,
My original email was:
I'm honoured and glad to announce the beginning of the next phase
of ToriOS development cycle.
ToriOS 1.0 has been officially released:
http://torios.top/torios-1-0-has-been-released/
<http://torios.top/torios-1-0-has-been-released/>
And now, we need to move forward and start a new page. It's time
for more achievements and challenges.
I'll make this as short as possible and ask for your kind vote.
* *ToriOS 1.xx* series/releases/versions will bring no major
changes but bug fixes + updated packages.
* *ToriOS 2.0* will be bring major changes or new features NOT
only updates, fixes, etc.
* *ToriOS 2.xx* series/releases/versions will bring updates +
bug fixes.
That's a very general description and we haven't yet started to
go so deep.
I just want to draw the outlines of the future. I just want to
let you know what I have in mind for 1.xx series and 2.xx series.
If you have any Q, please ask :)
My vote is: +1 for the above.
If you're unsure, please ask!
Thank you so much ...
--
Remember: "All of us are smarter than any one of us."
Best Regards,
Ali/amjjawad <https://wiki.ubuntu.com/amjjawad>
What you have replied below:
Hi Ali,
The only major change I can see is a switch in the
jwm-settings-manager backend. I have been working on switching it
from tinyxml2 to pugixml.
The major reason is that with tinyxml2 if a node does not exist it
will return NULL, which in turn (without a freakish amount of
testing which made the code incredibly hard to read/maintain and
made it very complicated) would make the program randomly crash or
not open.
So this is a major change, but the UX will look the same. There
will be a few changes to some things though. I have been fixing
issues in the original code, as well as beginning to implement a
few of the other features, but these changes will be minor. And
once I get to the point of release I will send it out for major
testing.
Currently I have these parts (mostly) working... I say mostly
because I have not done any major testing yet.
autostart (I would like to handle the XDG autostart directories,
as well as handle making desktop files to do this, which I already
have a program for this, also using FLTK and I could easily bring
the code in and adapt it to fit)
icons (only the directories are listed currently, I have most of
the code to support Icon themes, and will offer the option to do
either which is a big improvement IMO)
keyboard ( I *would like* to add layout/language support but this
will be put off for a while)
desktop (better support for some of the file managers
nautilus,thunar and of course pcmanfm.... however rox will not be
supported in this version as it is no longer developed and
non-standard in the way it does things)
mouse (this now shows a window without synaptic touchpad things...
i.e. JWM ONLY, if there is no touchpad... which is an improvement)
fonts (this actually works much better now, as I handle choosing
the 'bold' or 'italic' fonts, in the current version they do not work)
themes (everything works that I tested)
window (window grouping does not work yet)
shutdown (works just as it should, though I'd like to hide
suspend/hibernate if the computer does not support it)
What is left:
panel
menu
supporting the <Inlcude> and <Dynamic> tags, as well as adding in
the new features JWM has, though this last point will wait until
the program is stable and well tested.
This new version is already MUCH more stable, and will not crash
if something is missing. Everything now uses a common 'backend'
source/include and this can make a transition to a different XML
library much easier in the future (if needed). It also means that
most of the UI is decoupled from the backend code, so using a
different toolkit (like gtk) in the future could be implemented
much more easily. Though I will need to wrap things in
#ifdef
#endif
blocks to remove FLTK code, etc...
But that is not something for anytime in the future, just a nice
feature that came with making the code highly modular and much
easier to maintain.
This has been long overdue, and has made for many headaches using
the settings manager. The new version is faster to load, doesn't
crash (ever yet) and is going to be much easier to find/fix bugs
in the future, as well as add new features easily. The code to
support new JWM features is probably already in place, as most of
the different things use the same code base for everything.
Once this is in place I will move the scripts for configuration
into a separate package, which will support PPC users (they can
simply install torios from a PPA that way without the need to
compile things, or download deb packages).
That covers most of my update. I have been busy with real life so
I have been quite recently, but I have also been busy with c++ too.
If anyone wants to look at the code, or try a really early alpha
of jwm-settings-manager 2.0 let me know, but it is NOT ready for
production yet, and may mess things up (though I haven't had that
happen yet).
--
Regards
Unfortunately, is not what I was expecting.
We must learn from our previous mistakes:
* We must work harder on our internal communications.
* We must have a plan to follow.
The main reason why ToriOS 1.0 took 3 years to release is because we
haven't followed the above 2 main points, IMHO.
Moving forward, can we please stick to the topic in each and every email?
Can we save the long details for somewhere else and some other time?
I'll be on Slack, ping me whenever you have time ;)
Thank you!
--
Remember: "All of us are smarter than any one of us."
Best Regards,
Ali/amjjawad <https://wiki.ubuntu.com/amjjawad>
