Re: Making it easier to change a font.

[pks timing com]

Marco Scholten writes:
 >
 > I don't think that something like that would come along unless you  
 > explicitly call modify_font on the table or one of its parent containers,  
 > or can you name another situation where this would happen?

I was scared by the (at the time seemingly drastic) change in semantics
from "modify" to "modify all children as well".  To mind came this (as it
turns it out not all equivalent) example:

Consider lack of show_all() and show_all_children(), and asuggestion to
make show() change semantics as the "show_all" counterparts do.  Applications
would be displaying widgets they didn't intend to (and didn't before the change)
all over the place in a heart beat.

going back to my original concern, I was worried that:

  PopulateTableFromExample(tbl);
  tbl.set_name("name_which_selects_rc_style")

and

  tbl.set_name("name_which_selects_rc_style")
  PopulateTableFromExample(tbl);

might yield different results.


I greped for modify_font in both gtk and gtkmm.  It appears to have been
written with only the library end user in mind, and doesn't appear to
be used at all internally.  Initially I didn't exclude the possiblity that
rc styles would employ the modify_*() functions in some form or another,
or that the grep would yield 200 matches ...

I doesn't sound like such a bad idea now that I've taken a closer look.
And considering that the change might well have me delete 300 fewer how
to change a button's font messages a year it sounds fantastic !!!