When I load DrGeo pr file (http://download.gna.org/istoa/drgeo/) and save it again under Etoys, I have different result when I use Windows or Linux image.
1. The format seems to be different (check both .pr file at http://download.gna.org/istoa/drgeo/) 2. The one produced from the linux etoys is smaller and broken: the pr. cannot be reloaded, get a syntax error 3. the one produced with the etoys windows is just fine
Now when I take the windows image and use it in linux I can get proper .pr file I did not succeed to use the linux image with my stock VM: missing primitive
It may look like an image problem with the linux version.
I have not tested under the XO with latest build what is happening (if loading/saving/reloading is ok)
Hilaire Fernandes wrote:
When I load DrGeo pr file (http://download.gna.org/istoa/drgeo/) and save it again under Etoys, I have different result when I use Windows or Linux image.
- The format seems to be different (check both .pr file at
http://download.gna.org/istoa/drgeo/) 2. The one produced from the linux etoys is smaller and broken: the pr. cannot be reloaded, get a syntax error 3. the one produced with the etoys windows is just fine
Now when I take the windows image and use it in linux I can get proper .pr file I did not succeed to use the linux image with my stock VM: missing primitive
It may look like an image problem with the linux version.
I have not tested under the XO with latest build what is happening (if loading/saving/reloading is ok)
I loaded the project on the XO made a triangle and saved it. Loading again gives a syntax error, but if I proceed the syntax error it loads fine. (Proceed by right click on the red text in the syntax error window.)
I'm not sure what is going on.
Karl
2008/9/21 Karl Ramberg karlramberg@gmail.com:
When I load DrGeo pr file (http://download.gna.org/istoa/drgeo/) and save it again under Etoys, I have different result when I use Windows or Linux image.
Me either...
Hilaire
At Sun, 21 Sep 2008 16:39:51 +0200, Karl Ramberg wrote:
Hilaire Fernandes wrote:
When I load DrGeo pr file (http://download.gna.org/istoa/drgeo/) and save it again under Etoys, I have different result when I use Windows or Linux image.
- The format seems to be different (check both .pr file at
http://download.gna.org/istoa/drgeo/) 2. The one produced from the linux etoys is smaller and broken: the pr. cannot be reloaded, get a syntax error
Yes, and that seems to be a legitimate syntax error. What are the versions of these images? I tried it on the latest Etoys on my Fedora 9 and it seems to be working as it should be, if I start from DrGeoII.pr and not -buggy.pr one. Yes, I saved the project file with some content on Fedora 9 with fully updated etoys-dev.image, and restart the image and load the project and it worked. (and arcTan2: method is there.)
The linux installation from the site may have a problem. And what is the language setting of your Linux installation?
The image files won't be bit-identical, even on the same platform. But it shouldn't cause any problem.
Now when I take the windows image and use it in linux I can get proper .pr file I did not succeed to use the linux image with my stock VM: missing primitive
It may look like an image problem with the linux version.
It should not require any special primitive.
-- Yoshiki
2008/9/23 Yoshiki Ohshima yoshiki@vpri.org:
At Sun, 21 Sep 2008 16:39:51 +0200, Karl Ramberg wrote:
Hilaire Fernandes wrote:
When I load DrGeo pr file (http://download.gna.org/istoa/drgeo/) and save it again under Etoys, I have different result when I use Windows or Linux image.
- The format seems to be different (check both .pr file at
http://download.gna.org/istoa/drgeo/) 2. The one produced from the linux etoys is smaller and broken: the pr. cannot be reloaded, get a syntax error
Yes, and that seems to be a legitimate syntax error. What are the versions of these images? I tried it on the latest Etoys on my Fedora 9 and it seems to be working as it should be, if I start from DrGeoII.pr and not -buggy.pr one. Yes, I saved the project file with some content on Fedora 9 with fully updated etoys-dev.image, and restart the image and load the project and it worked. (and arcTan2: method is there.)
The linux installation from the site may have a problem. And what is the language setting of your Linux installation?
It is French; but the .pr does not come with language catalog (another problem to fix) Now, I think it would be just easier if I could provide a mechanism to let the user install DrGeo permanently in the Image. This will solve the other issue like slow to load drgeo project files. Do you know any preferred way to do that? I know the SAR way to do it as proposed by José, so I may look at it now and drop the .pr distribution.
Hilaire
Am 22.09.2008 um 23:49 schrieb Hilaire Fernandes:
Now, I think it would be just easier if I could provide a mechanism to let the user install DrGeo permanently in the Image. This will solve the other issue like slow to load drgeo project files. Do you know any preferred way to do that? I know the SAR way to do it as proposed by José, so I may look at it now and drop the .pr distribution.
Well, the actual problem I see is that it's sub-optimal to always save the DrGeo code with any pr file. This was just the simplest way to do it.
The code should always be loaded from the DrGeo bundle, whereas a saved project should be just the DrGeo figure.
Loadable extensions like DrGeo could be much better supported by Etoys, but there has been no time to work on that from the core developers yet.
- Bert -
An easy solution would be to include the Drgeo directly in the image. This will solve all the problem and it will be just fine for me. Will it be ok for you?
Hilaire
2008/9/23 Bert Freudenberg bert@freudenbergs.de:
Am 22.09.2008 um 23:49 schrieb Hilaire Fernandes:
Now, I think it would be just easier if I could provide a mechanism to let the user install DrGeo permanently in the Image. This will solve the other issue like slow to load drgeo project files. Do you know any preferred way to do that? I know the SAR way to do it as proposed by José, so I may look at it now and drop the .pr distribution.
Well, the actual problem I see is that it's sub-optimal to always save the DrGeo code with any pr file. This was just the simplest way to do it.
The code should always be loaded from the DrGeo bundle, whereas a saved project should be just the DrGeo figure.
Loadable extensions like DrGeo could be much better supported by Etoys, but there has been no time to work on that from the core developers yet.
- Bert -
Am 23.09.2008 um 07:22 schrieb Hilaire Fernandes:
An easy solution would be to include the Drgeo directly in the image. This will solve all the problem and it will be just fine for me. Will it be ok for you?
Not at this point in the OLPC release cycle. The next release will be managed by Michael, he will surely comment.
- Bert -
At Tue, 23 Sep 2008 07:42:36 -0700, Bert Freudenberg wrote:
Am 23.09.2008 um 07:22 schrieb Hilaire Fernandes:
An easy solution would be to include the Drgeo directly in the image. This will solve all the problem and it will be just fine for me. Will it be ok for you?
Not at this point in the OLPC release cycle. The next release will be managed by Michael, he will surely comment.
Yes, and I'll be on the side of pursuading him to agree on that.
-- Yoshiki
2008/9/23 Bert Freudenberg bert@freudenbergs.de:
Am 23.09.2008 um 07:22 schrieb Hilaire Fernandes:
An easy solution would be to include the Drgeo directly in the image. This will solve all the problem and it will be just fine for me. Will it be ok for you?
Not at this point in the OLPC release cycle. The next release will be managed by Michael, he will surely comment.
Thanks for this clear answer. I appreciate it. As I am concerned, I really have to pause DrGeoII development until next summer, I have other duty to work on. In the meantime many think may have changed in the overall Squeak community, and at this time I may decide to port to Pharo and its new UI, so it will make it definitively incompatible with Etoys.
Bye,
Hilaire
2008/9/23 Bert Freudenberg bert@freudenbergs.de:
Am 23.09.2008 um 07:22 schrieb Hilaire Fernandes:
An easy solution would be to include the Drgeo directly in the image. This will solve all the problem and it will be just fine for me. Will it be ok for you?
Not at this point in the OLPC release cycle. The next release will be managed by Michael, he will surely comment.
Thanks for this clear answer. I appreciate it. As I am concerned, I really have to pause DrGeoII development until next summer, I have other duty to work on. In the meantime many think may have changed in the overall Squeak community, and at this time I may decide to port to Pharo and its new UI, so it will make it definitively incompatible with Etoys.
Bye,
Hilaire
Hi all,
Just a small bug I encountered while exploring the FishAndPlancton-project in the Gallery of Example Projects:
Tiles like 'start script: script' don't display the script to be started. See for example Plancton>>initialSetup (Screenshot attached).
Markus ----------------------------------------------- Markus Schlager m.slg@gmx.de
I can confirm this bug.
I did a quick review of the oter projects, here is a list of issues:
SalmonSniff project uses old random tiles. The BetterMovieUI and BallDropAnalyzes use UI buttons that are too small on XO. BallDropAnalyzes have no reset for the little blue ball. SimpleSpring have a script with a tile variable with bad layout. RandomRacing uses old random tiles. Most number tiles in particle projects have a a few pix off layout. Desktop project is unfinished
Karl
On 9/24/08, Markus Schlager m.slg@gmx.de wrote:
Hi all,
Just a small bug I encountered while exploring the FishAndPlancton-project in the Gallery of Example Projects:
Tiles like 'start script: script' don't display the script to be started. See for example Plancton>>initialSetup (Screenshot attached).
Markus
Markus Schlager m.slg@gmx.de
squeakland@lists.squeakfoundation.org
-
Bert Freudenberg -
Hilaire Fernandes -
karl ramberg -
Karl Ramberg
-
Markus Schlager -
Yoshiki Ohshima