weird standalone save behavior

[Richard Lague]

On Apr 2, 2006, at 11:57 PM, J. Landman Gay wrote:

> Richard Gaskin wrote:
>> Richard Lague wrote:
>>> On Apr 1, 2006, at 12:00 PM, Richard Gaskin wrote:
>>>> Richard Lague wrote:
>>>>> In an effort to figure out what is going on I have been using  
>>>>> the   following script for saving:
>>>>> ON mouseUp
>>>>>      save this stack
>>>>>      IF the result is not empty THEN
>>>>>          put the result into resultVar
>>>>>          answer resultVar
>>>>>      ELSE
>>>>>          answer "save confirmed"
>>>>>      END IF
>>>>> END mouseUp
>>>>> When I can not save data I always get the same result: “Can’t   
>>>>> open  stack file”.
>>>>
>>>> Are you sure that in the context it's being called "this stack"   
>>>> refers to the stack you want and not the standalone?
>>>
>>> The script is in a button on the sub-stack. I have been testing  
>>> this  with a very simple standalone -- a one-card main stack/ 
>>> application  that has one sub-stack. The sub-stack is also only  
>>> one card that has  a "save" button and a text field.
>> On Windows and UNIX systems, the executable file cannot modify  
>> itself. The Rev engine enforces this on Mac OS for consistency.
>> If I read your description correctly the substack is part of the  
>> executable stackfile.
>> If you break that substack out into its own stackfile you will be  
>> able to save it.
>
> I think something else is going on. He wrote to support recently  
> and I didn't have a clue what the problem might be. I did take a  
> look at his sample application, and the substack is saved out as a  
> separate file. When I tested it, it saved fine for me. I understand  
> that Sarah also tested it for him, and it saves fine for her too.  
> On his own machines, it saves okay sometimes and not others, and  
> seems to behave differently depending on whether he has recently  
> copied the files or not, and to which drive he has copied them.
>
> The idea of it being a permissions problem was mine, because I  
> believe the "can't open stack file" error is due to the inability  
> of the engine to create the temporary backup file during the save  
> operation. This only happens to him sometimes, and only under  
> certain conditions. It seems to be something specific to his setup,  
> as I haven't ever heard of anything like this except in cgi-bin  
> folders (which generally have permissions set to forbid file  
> creation.)
>
> Richard (Lague): did you try running Repair Permissions using the  
> Disk Utility app? It can't hurt.
>
> Also, when the save fails, is your app located somewhere in your  
> own user folder? You aren't trying to save a file to another user  
> folder, right?

I run the Repair Permissions from Disk Utility at least once a day.  
So, if it is a permissions problem it at least seems to be one that  
is beyond the ability of Disk Utility to fix.

The save fails in the folder where I keep all my Revolution stacks.  
It also fails if I launch Revolution by double clicking on Revolution  
in the top level Application folder and then try to create a  
standalone in a sub-folder within the Applications folder. Usually,  
my standalones will save if I either create them on the DeskTop or  
drag them to the DeskTop. I have a few old stacks that will make  
standalones that will save data in my regular Revolution folder, but  
when dragged to the DeskTop stop saving data.

Thanks,

Rich