On Sun, 20 Oct 2002, Jungshik Shin wrote:

> On Fri, 18 Oct 2002, Keith Packard wrote:

> > As per my comment above, I strongly prefer to make the contents of the
> > cache files independent of the configuration so that multiple
> > configurations can share the same cache files without difficulty.
>
>  Wouldn't it be possible to leave the content of cache file independent
> of the configuration  while still letting users override what
> fontconfig detects?   When a client requests for info.  on a font, what

<snip : a completely wrong guess >

> A better alternative would be  when charset(code range covered) or
> any other properties of fonts are explicitly specified in fonts.conf,
> fontconfig can ignore the values of those properties stored in font-cache
> (by just skipping them) during step 2 and *believe* what the configuration
> says about fonts.

  I'm sorry that I didn't read 'man fontconfig'(or rather not
refreshed my memory about how fontconfig works) before writing my
previous message.

  1. get a pattern from an application(fontconfig client)
  2. apply configuration-specified editing rules to the pattern.
  For each font:
    3. read in font properties from fc-cache or (directly from font if
     fc-cache is not present)
    4. measure the distance between the pattern and each font

  5. come up with the closest match or a list of matches
  6. post-processing with match(es)

  I'm wondering how difficult it would be to insert a step between
step 3 and step 4 above:
  (Or, step 3 and 3.5 can be done kinda in parallel.)

  3.5 apply configuration-specified changes to font properties
      obtained in step 3. (add/remove char(s) to/from charset,
      add/replace/remove lang(s), etc..)

  Would it require a major structural/design change in fontconfig? Is
this a crazy/stupid idea?  If not, could you tell me where to look and
begin if I want to experiment?

  TIA,

  Jungshik

_______________________________________________
Fonts mailing list
[EMAIL PROTECTED]
http://XFree86.Org/mailman/listinfo/fonts

Reply via email to