On 7/31/11 2:55 PM, Rob Beezer wrote:
Kudos to Jason and the students at Drake. This is very nice.
Thanks! And thanks again for testing!
I've tried to "hide" some of the diagnostic information, and style a few of the remaining elements. I've done this with my own crude CSS stylesheet and via the arguments to the singlecell creation function.
Disclaimer: I haven't tested hiding elements very extensively. I wouldn't be surprised if there was a bug or two in doing this.
Some questions and observations. 1. Starting from the current HTML page being served from sagemath.org, where would one add the reference to a "custom" CSS file? I have this mostly working right, but probably not correctly. In particular, setting many of the output parts to not display has totally broken interacts, so in that sense what I have done is definitely not correct.
You should be able to add a CSS link statement at the end of the head or the start of the body. We dynamically add CSS to the end of the head in the embedded_singlecell.js, so there might be problems with us overwriting your CSS rules if you have them in the head element. So maybe try putting them at the top of the body element?
Can you post your current working/nonworking version so I can test it too?
2. Using arguments to the singlecell creation function to turn-off the messages (three values set to false) seems to also break interacts.
Can you post your current file so I can test it too?
3. The default value of "outputDiv" listed in the documentation for the dictionary of options to the single cell creation caused an error for me, but a value of "#singlecell" does seem to work.
Thanks; I'll take a look. The documentation likely needs to be updated.
4. Maybe the singlecell_completion div can be not-displayed when "messages" is false?
At one point, Ira implemented tab-completion, but I don't think that's turned on now (I think we may have broken it with some extensive changes after he implemented it). I don't think singlecell_completion serves any purpose right now and can be removed.
5. I was able to put two apparently independent single-cells on the same page by running singlecell.makeSinglecell(alternate_config) where alternate_config specified a different id for a second div used to locate the cell. Maybe not news, but I did not see this documented anywhere. In particular I was just guessing about how to properly invoke the singlecell creation function.
Yes, we planned to support multiple single cells on a page, so I'm glad it's working.
6. Minor nit: CodeMirror does not handle the "F.<a> = GF(3^2)" syntax properly.
Thanks for the reminder. We had to custom-hack the codemirror in Sage to support that. I'll look at porting that hack over to codemirror2 and the singlecell.
7. Do interacts time-out after some period of inactivity? It seems so.
Yes. I think 60 seconds is the timeout we currently have set up. Computations are allowed 100 seconds of CPU time too, so that could also cause a computation/interact to time out.
Thanks, Jason -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
