Fontconfig - Jun 2008 - Search for fonts by full name

Hi,

I''d like to be able to use full font names in match pattern. The first 
attached patch implements this. It is to be applied on top of Isaiah 
Beerbower''s patch (the one with postscript names).

There is a problem with it that also happens when matching fonts by postscript 
names. Not every font have a full name or a postscript name. Such fonts get a 
score of 0 (the best possible), while fonts that have a non-matching name get 
a score of 100. This results in fonts being selected based on other 
parameters (because both fonts with matching name and without a name get 0 
for full name test).

To fix this, the second patch gives the worst possible score to those match 
values that were not updated. Alternatively, the third patch makes sure that 
every font has a full name, making it up from it''s family and style if 
needed. Either one of them is enough to make this feature work.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 01-score-missing-elements-as-1e99
Type: text/x-diff
Size: 972 bytes
Desc: not available
Url :
http://lists.freedesktop.org/archives/fontconfig/attachments/20080608/fb5cfb6c/attachment.diff
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 02-search-by-fullname
Type: text/x-diff
Size: 4352 bytes
Desc: not available
Url :
http://lists.freedesktop.org/archives/fontconfig/attachments/20080608/fb5cfb6c/attachment-0001.diff
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 03-always-set-fullname
Type: text/x-diff
Size: 911 bytes
Desc: not available
Url :
http://lists.freedesktop.org/archives/fontconfig/attachments/20080608/fb5cfb6c/attachment-0002.diff

On Sunday 08 June 2008 22:46:29 Evgeniy Stepanov wrote:
> Hi,
>
> I''d like to be able to use full font names in match pattern. The
first
> attached patch implements this. It is to be applied on top of Isaiah
> Beerbower''s patch (the one with postscript names).
>
> There is a problem with it that also happens when matching fonts by
> postscript names. Not every font have a full name or a postscript name.
> Such fonts get a score of 0 (the best possible), while fonts that have a
> non-matching name get a score of 100. This results in fonts being selected
> based on other parameters (because both fonts with matching name and
> without a name get 0 for full name test).
>
> To fix this, the second patch gives the worst possible score to those match
> values that were not updated. Alternatively, the third patch makes sure
> that every font has a full name, making it up from it''s family and
style if
> needed. Either one of them is enough to make this feature work.
Any response? Should I send these patches to bug tracker instead?