[WWdevel] Fwd: [WWcvs] pg: problemPreserveAnswers.pl

[Davide P. C. <dp...@un...>]

Forgot to copy the list on this response:

EV3P is unrelated to the $ in answer blanks, so no, EV3P doesn't do  
anything about that.  (The flaw is in the various answer rule macros  
that strip those characters out rather than convert them to HTML  
entities.)

The the test case is not in the problem file itself, but in what you  
enter into it.  If you enter a $ in an answer, submit it, and it  
shows up as part of the answer already showing in the answer rule,  
then the problem is solved.  If the dollar sign is missing, then the  
problem still exists.

Davide

On Jul 8, 2008, at 12:04 PM, Michael Gage wrote:

> Thanks Davide,
>
> On another issue -- problemPreserveAnswers.pl  keeps $ from being  
> erased in student sticky answers.
> Does using EV3P  fix this problem directly in PGbasicmacros.pl?
>
> I don't have a test case to work on to see if this issue is fixed  
> in rel-2-4-patches.  If it's not already fixed then
> one will still have to use problemPreserveAnswers.pl for this release.
>
> Take care,
>
> Mike
> On Jul 8, 2008, at 11:41 AM, Davide Cervone wrote:
>
>> I think this was a file that John Jones wrote originally, so perhaps
>> he knows where they are used.  I am not aware of any that use it, and
>> don't see any by grepping the libraries.  The error was probably
>> introduced by me last summer when I was trying to encapsulate the
>> initialization code and make sure it runs from the _init() routine
>> rather than in the .pl file itself.
>>
>> As for whether there is a better way, currently this is the right way
>> to do it.  One of the things that came out of the AIM conference is
>> that there needs to be a better way to add features to a context
>> (rather than just load a completely new context).  I have some ideas
>> for that, but haven't completed the work on it.  After that, it would
>> be reasonable to make this a "context patch" rather than a separate
>> context, but for now, this is the right way.
>>
>> I think the name change to contextCombinatorics would be fine.
>>
>> Davide