Opened 10 years ago

Closed 10 years ago

#1584 closed defect (fixed)

font configuration through gconf

Reported by: dsd Owned by: tomeu
Priority: Unspecified by Maintainer Milestone: Unspecified
Component: Sugar Version: Unspecified
Severity: Unspecified Keywords: r+
Cc: Distribution/OS: Unspecified
Bug Status: Unconfirmed

Description

Sugar currently has fonts configured in 2 places - gtkrc and style.py. Neither can be easily modified by implementors/deployers.

As a first step to implementing http://wiki.sugarlabs.org/go/Features/Font_configuration here are some patches which add gconf-based font configuration.

I learned some things while doing this so the feature proposal will benefit from some tweaks based on the outcome of this ticket (which I'll do at that point)

Attachments (2)

0001-Configure-fonts-through-GConf.patch (2.8 KB) - added by dsd 10 years ago.
sugar patch
0001-Apply-font-configuration-from-GConf.patch (1.9 KB) - added by dsd 10 years ago.
sugar-toolkit patch

Download all attachments as: .zip

Change History (9)

Changed 10 years ago by dsd

sugar patch

Changed 10 years ago by dsd

sugar-toolkit patch

comment:1 Changed 10 years ago by dsd

  • Keywords r? added

comment:2 Changed 10 years ago by sascha_silbe

You are replacing two different fonts (Sans Serif 10 in gtkrc, Bitstream Vera Sans 10 in style.py) with a single one (Sans Serif 10). That's probably fine (though I don't know where in the UI each of them was used), but there should be a conscious decision about it.

comment:3 Changed 10 years ago by dsd

Yes. I think it was only a bug that these faces were different. It wasn't like that in 0.82. Perhaps someone forgot to update style when they changed gtkrc.

The Vera Sans option in the style module is not used much across the desktop.

comment:4 Changed 10 years ago by tomeu

I'm a bit concerned about activity startup performance, can you check what difference it makes on the XO-1?

comment:5 Changed 10 years ago by dsd

Very little, gconf values are already read in that very same codepath.

I measure 1.5ms increase in startup time, but much of that 1.5ms will be the time needed to execute the system call to read the clock in my measurement code.

comment:6 Changed 10 years ago by tomeu

  • Keywords r+ added; r? removed

comment:7 Changed 10 years ago by dsd

  • Resolution set to fixed
  • Status changed from new to closed
Note: See TracTickets for help on using tickets.