On 31 Dec 2003, Lars Magne Ingebrigtsen said: > Xavier Maillard writes: > >> It seems to me that 'gnus-header-face-alist' prevents other faces to >> be displayed as they are defined. >> >> For example if I set 'gnus-header-face-alist' to highlight 'From', >> 'To' and 'Subject' headers, then all the other headers are displayed >> with a default face. I don't like this behavior *at all* since I want >> all my header faces to be honored as defined. > > I don't quite follow you. Defined where by what? Ok. Let's say I have this as my g-h-f-a variable: (setq gnus-header-face-alist's value is (("From" nil gnus-header-from-face) ("Newsgroups:.*," nil gnus-header-newsgroups-face) ("" gnus-header-name-face gnus-header-content-face))) This would render my gnus-article with those Faces and it will show me the different part of the headers using those faces. Here only From and Newsgroups headers will be rendered with custom faces. The rest will use the faces defined in the "" part. Now let's imagine I also have something like this: (set-face-foreground 'gnus-header-subject-face "rosy brown") (set-face-bold-p 'gnus-header-subject-face t) With this I can expect Gnus to display the "Subject" header with rosy brown color and the text should be bold. Here is the problem, the subject part won't be displayed using the face gnus-header-subject-face but ones in the "" part of g-h-f-a. I suspect that it is not the expected behavior and so I was asking to have a slightly different one. That is to say respects the g-h-f-a variable and also the possibly defined other faces. In fact I was thinking to use g-h-f-a to add more stuff to be hilighted (for example adding new faces to display X- headers). But AFAIK it won't be possible with the current implementation. I hope you will understand now. The current workaround I am using now is to add face definitions with hooking (using gnus-article-prepare-hook). That way I can use either gnus-header-face-alist own face definitions and the default one with face-spec-set. Maybe this is the way to go normally but as I am not sure... zeDek -- Xavier Maillard 7 rue Jeanne Jugan, 51100 Reims, France phone: +33 3 26 77 02 21, mobile: +33 6 62 59 68 62 email: zedek@gnu-rox.org