Thanks for the detail, Richard. Like you, I wish things were as we want, 
not as they are. That's why I'm wondering if Mathieu's problem could be 
solved a different way, namely with a user-defined variable.

For example, if XYZ were always bold-by-override, perhaps a variable 
something like <Default Para Font><NewBold>XYZ<Default Para Font> pasted 
from the clipboard - by Replace By Pasting - could work.

Again, I'm not able to test this right now, but maybe there's a chance?

If it is successful, and variables aren't desired for the final state of 
the document, they can be replaced by text, which would requite one more 
test to see that this works as required.

HTH
________________
Regards,

Peter Gold
KnowHow ProServices

Combs, Richard wrote:
> Peter Gold wrote:
>  
>   
>> Richard's point is excellent!
>>     
>
> Thank you.
>  
>   
>> I believe that one way to avoid this problem is to create a 
>> named character format for the purpose, whose properties are 
>> only those you want to apply. In this example, you'd define 
>> the "NewBold" format by clicking outside any text, or 
>> choosing Commands > Set Window To As Is in the character 
>> designer,  then setting only the Bold attribute.  I haven't 
>> tested this, but I think that pasting a named character 
>> format that's copied to the clipboard applies only the 
>> properties of the named format, not every property of the 
>> text. A quick test should clarify the behavior.
>>     
>
> Unfortunately, no. You can only copy to the clipboard an _instance_ of
> the format, not the format as an abstraction. 
>
> FM interprets "As Is" quite literally. As soon as you instantiate the
> char format by applying it to a text string, the attributes defined as
> "As Is" are set to the existing values for that string. So, your Bold +
> As Is char format becomes Bold + Palatino, 10 pt, Angle Regular,
> Variation Regular, Color Black, etc. If you have FrameScript, you can
> verify this by retrieving the property list of a string with a char
> format applied. 
>
> I wish it weren't so. I'd like all the undefined char properties to
> contain pointers to the corresponding pgf properties (or to be
> null/empty, and interpreted as pointers to the pgf properties). 
>
> Richard 
>   


Reply via email to