Closed GoogleCodeExporter closed 9 years ago
And months, quarters could be in the bottom row of the timeline
Original comment by dbarashev
on 4 Dec 2008 at 11:40
I'd like the years split in months rather than weeks so the long projects could
be
easier to follow.
It would be really good to be able to choose weeks or months depending on the
length
of the project
Original comment by mfama...@gmail.com
on 20 May 2010 at 1:46
Since we need also to zoom out further for our project we made a *quick and
dirty*
patch for this.
It adds 3 extra zoom steps, with Month/Week timeline. The months get a
lowercase
letter for odd years and uppercase for even years. I added comments in the
patch to
indicate where you could change this behaviour.
This solves our problem for now, hopefully it will help other as well.
But we are also really looking forward for a proper fix for this issue.
Original comment by maarten....@gmail.com
on 25 May 2010 at 8:23
Attachments:
This issue was closed by revision c74bcda6f1.
Original comment by maarten....@gmail.com
on 14 Oct 2010 at 8:49
Original comment by dbarashev
on 15 Oct 2010 at 12:01
It seems to me that zooming in 2.10 and in Praha build 210 is still not
possible for more than 2 years (on a 2000x1000 pxl monitor). Has this 25yr
function been reverted?
Why have a limit below total project size for zoom out at all? Yes, labelling
may become ugly in some cases. But that would still be less frustrating than
being blocked with nothing or patchwork from exported files.
Original comment by Hoffm...@CPPM.In2p3.Fr
on 13 Jan 2011 at 11:27
It was published in Praha 306. And who told about 25yr?
Labeling and chart grid will *surely* become ugly in all cases, and your tasks
may skew, and most likely you won't be able to see anything but mess and
clutter anyway. For instance, weekend bars are already annoying with the
current implementation. Implementing it in a way which doesn't produce too much
clutter is by far non-trivial problem.
Original comment by dbarashev
on 13 Jan 2011 at 11:46
Yes, but I am preparing a report, and the first comment on the chart, before
any comment on the content, was "please put months on the scales".
You should not worry about weekends when looking at a year. They are relevant
on the very short term only. I would simply remove weekend bars, if the
displayed width of two days becomes smaller than one pixel. (Otherwise they
would be longer than two days on the display anyway, which is more confusing
than having none at all.)
25 yrs were stated in revision 403d15d512cccef1ef9d7ddaf536fa59b3003f68
Original comment by Hoffm...@CPPM.In2p3.Fr
on 14 Jan 2011 at 12:24
> You should not worry about weekends when looking at a year. They are relevant
on the very short term only.
Exactly. And that's where the tricky part of the implementation starts.
Original comment by dbarashev
on 14 Jan 2011 at 12:40
> 25 yrs were stated in revision 403d15d512cccef1ef9d7ddaf536fa59b3003f68
This revision has never made it into the default branch (ie added to the
application). It was/is an try out which did not fulfill all of our demands.
You *could* try to compile it yourself, and see if it is usable for exporting a
graph. But I thought the exporting/printing functionalities got broken (big
time) by this (or prior) issue_15 revisions.
Original comment by maarten....@gmail.com
on 14 Jan 2011 at 8:57
I tried the head version from CVS. If you would like a complete feedback
report, let me know. This ticket is probably not the right place for that.
Regarding zooming possibilities, it is still very limiting to have a maximum of
50 weeks on my 1280 pixel screen. After the maximum zoom, it still reacts on
the mouse wheel, but the effect is that it jumps to a higher resolution (two
steps possible)!??
Annoying or at least surprising is also the effect that a new zoom factor also
leads to the chart being reset to the begin of the project!
As for the zoom, there are some scales (zoom factors), where the chart is
exactly sticking to the mouse when clicked, as expected. In other scales
however it goes faster or slower than the mouse.
In some situation, there is a latency between mouse movements and display.
After the mouse control has been done, the chart still updates a couple of
times (repaint() calls?). Are you sure you are using the Swing event model
correctly. Earlier repaints should simply be skipped, if new repaint()s have
come in.
Original comment by Hoffm...@CPPM.In2p3.Fr
on 15 Jan 2011 at 10:10
> I tried the head version from CVS
Did you mean "Mercurial" ?
> it is still very limiting to have a maximum of 50 weeks on my 1280 pixel
screen
It is 5 years on my 1680px display. Are you sure that you checked out the right
version?
Original comment by dbarashev
on 15 Jan 2011 at 11:08
>> I tried the head version from CVS
> Did you mean "Mercurial" ?
I was surprised that you mix them up, indeed. No I meant CVS, as I followed the
"unstable" link on
http://www.ganttproject.biz/participate
which lead to
http://code.google.com/p/ganttproject/wiki/GPHeadForDevelopers
which points to CVS. Is that description out of date???
> It is 5 years on my 1680px display. Are you sure that you checked out the
right version?
Five years is a more reasonable time period (if you reason in USSR economic
plans ;-), but a decade would be nice. More than a decade is barely plannable
and may serve as a limit.
But why restrict to any number? It might still be one day someone likes to draw
a Gantt chart for evolution, geological development or a biological model can
be described like that. Then a zoom to a larger scale might come in handy.
Original comment by Hoffm...@CPPM.In2p3.Fr
on 16 Jan 2011 at 8:37
I thought that I said that *good* implementation of the large zoom scales is by
far not trivial. If you feel brave enough, go ahead, clone the repository,
implement and send us a changelist. Until that we have lots of other important
stuff to do.
Original comment by dbarashev
on 16 Jan 2011 at 12:38
I feel brave enough to implement a zoom, but not brave enough to waste my time
with outdated code or start from incomplete or conflicting implementations. I
would like to obtain code
a) which has a working implementation of zoom, which is working also for printing and
b) containing the latest (coherent) modifications.
If for b), it is correct to follow
http://code.google.com/p/ganttproject/source/checkout (selecting the "default"
tag, which seems to yield Praha 306), then unfortunately a) is not true. I hope
you understand that starting with debugging someone else's code (two files
without any source code comment I scanned up to now) is a much bigger hurdle
than bringing in my own ideas.
It looks to me like somebody is still working in the same place, and I should
wait until that work is finished. Can you please comment on that?
Also, if the CVS code is outdated, why does it allow to specify hours, as
opposed to days in Praha/Mercurial? Is the sub-day scheduling a move forward or
backward in development???
Original comment by Hoffm...@CPPM.In2p3.Fr
on 17 Jan 2011 at 6:26
Yes, somebody is still working. Yes, you should wait until the work is
finished. Yes, Praha is *alpha version* which I said a number of times. Yes,
alpha-quality software surely has bugs. Thanks for finding them, and please be
patient and wait until we fix them.
Anything else that I need to confirm?
Original comment by dbarashev
on 17 Jan 2011 at 6:31
Issue 189 has been merged into this issue.
Original comment by dbarashev
on 9 Feb 2011 at 8:20
With the new timeline up and running, this bug can be closed
Original comment by maarten....@gmail.com
on 4 Aug 2011 at 10:02
Original comment by dbarashev
on 4 Aug 2011 at 10:11
> Anything else that I need to confirm?
Yes, please confirm that following
http://code.google.com/p/ganttproject/source/checkout is the correct way to
feel free to test your latest developments!
Thanks.
Original comment by Hoffm...@CPPM.In2p3.Fr
on 10 Aug 2011 at 4:59
It is the right way to checkout, but with respect to testing, _FeelFreeToTest_
means that there is public build where the problem is fixed (the explanation is
unfortunately not visible when you view the issue but it appears if you try to
edit it and click "Status" field). It is the last available build, in our case
beta1:
http://ganttproject.blogspot.com/2011/07/ganttproject-praha-beta1-is-out.html
(although it was available in more earlier alpha builds too).
Original comment by dbarashev
on 10 Aug 2011 at 5:34
Original comment by dbarashev
on 22 Sep 2014 at 12:19
Original issue reported on code.google.com by
dbarashev
on 2 Dec 2008 at 12:38