RE: Sharing scripts

[MathCloud]

Hi,

In order for plot to work you have to allow pop-ups.

The reason it does not work with firefox seem to be that forefox does not
allow redirect from java script, which is used a lot on the site. Also, I
have not been able to make plot work for firefox so far. The other browsers
have worked fine, after allowing pop-ups.

Give it another try and look out for warnings from the browser that pop-ups
are blocked.

There is a problem when you get an error, you dont get any output from your
next command. But if you just repeat the command it should work. I will fix
this as soon as possible.

Best regards,

Anders Movert
MathCloud.se


dastew wrote:
> 
> 
> I see Michael Creel's email that he also had trouble after a plot command.
> so my test of  chrome and epiphany  were after the crash.
> 
> Doug Stewart
> 
> Date: Mon, 15 Mar 2010 08:30:34 -0400
> Subject: Re: Sharing scripts
> From: address@hidden
> To: address@hidden
> CC: address@hidden
> 
> I also tried mathcloud. 
> You say the firefox does not work- so I tried
> 
> firefox
>  epiphany
> chrome
> 
> and all seemed to have problems. 
> 
> What browser should I use from Ubuntu????
> 
>>From Firefox I did get logged in and was able to download a script (q2.m)
> 
> I then ran it and saw the correct numbers showed up on the screen, but the
> plot did not show.
> 
> 
>>> Welcome address@hidden You are now logged in.
>>>
> 
>  >> a=6; b=7;c=a*b
> c =  42
> 
>  >> q2
> l =  60
> phi =  2.0944
> th1 =  0.52360
> th2 =  1.0472
> 
>  >> plot(xx,w);print -dpng plot.png
>  >> plot.png;print -dpng plot.png
> There was an error evaluating your command : plot (y)
> 
>        plot (x, y, ...)
>        plot (x, y, fmt, ...)       plot (x, y, property, value, ...)
>  >> q2
>>> Bye! You have now logged out.
> 
> 
> But when I tried to see the print file (plot.png) it thought I was trying
> to plot again!!
> 
> 
> When I ran q2 the second time it did not display anything.
> 
> 
> Doug Stewart 
> 
> 
> 
> 
> On Mon, Mar 15, 2010 at 7:26 AM, Jaroslav Hajek <address@hidden> wrote:
> 
> On Mon, Mar 15, 2010 at 11:37 AM, MathCloud <address@hidden>
> wrote:
> 
> 
>>
> 
>> Hi Jaroslav,
> 
>>
> 
>> Now I have restarted MathCloud.se again, thanks for the input.
> 
>>
> 
>> I have filtered out feval and eval, which is quite sad.
> 
> 
> 
> Yes. Still it's not enough, unfortunately. I was just able to do:
> 
> 
> 
> s = str2func (["sys", "tem"]);
> 
> [st, out] = s("sysinfo");
> 
> 
> 
> to get complete information about your system. I could have also used
> 
> this to run malicious code (of course limited by account privileges).
> 
> 
> 
> 
> 
> At least the following functions pose a potential security risk as well:
> 
> 
> 
> dir
> 
> tar, untar, zip, unzip, gzip, gunzip
> 
> path, addpath, genpath, rmpath, pathdef, savepath
> 
> dlmread dlmwrite
> 
> save (?)
> 
> gnuplot_binary, info_program, makeinfo_program
> 
> 
> 
> maybe more...
> 
> 
> 
>> They are an
> 
>> important part of Octave so I will try to solve this in a more
>> sofisticated
> 
>> way.
> 
>>
> 
> 
> 
> Maybe you are aware that you can override built-in functions?
> 
> You can place system.m, popen.m etc somewhere and then add this to
> 
> path() at Octave's startup (.octaverc).
> 
> In this manner, you need not forbid eval and feval because they can't
> 
> influence the name lookup.
> 
> The problematic functions are "builtin" and "dispatch" which you
> 
> should either forbid or override to filter out forbidden functions
> 
> (because these two can alter the function lookup order).
> 
> 
> 
> 
> 
>> I will watch the progress with the restricted version of Octave with
>> great
> 
>> interest.
> 
>>
> 
> 
> 
> Yeah, I'm afraid you can't get much of a safety by simply filtering
> 
> the input, there's just too much options for workarounds. I believe a
> 
> solution from inside Octave will be much more safe and robust.
> 
> 
> 
> 
> 
> --
> 
> RNDr. Jaroslav Hajek, PhD
> 
> computing expert & GNU Octave developer
> 
> Aeronautical Research and Test Institute (VZLU)
> 
> Prague, Czech Republic
> 
> url: www.highegg.matfyz.cz
> 
> _______________________________________________
> 
> Help-octave mailing list
> 
> address@hidden
> 
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
> 
> 
>                                         
> _______________________________________________
> Help-octave mailing list
> address@hidden
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
> 
> 

-- 
View this message in context: 
http://old.nabble.com/Sharing-scripts-tp27847186p27904527.html
Sent from the Octave - General mailing list archive at Nabble.com.