If you think a bug might affect users in the 57 release, please set the correct tracking and status flags for Release Management.

fontconfig is partially honored

UNCONFIRMED
Unassigned

Status

()

Core
Layout: Text
UNCONFIRMED
3 years ago
a year ago

People

(Reporter: Tom Yan, Unassigned)

Tracking

33 Branch
x86_64
Linux
Points:
---

Firefox Tracking Flags

(Not tracked)

Details

Attachments

(4 attachments)

(Reporter)

Description

3 years ago
Created attachment 8509658 [details]
testcase.png

User Agent: Mozilla/5.0 (X11; Linux x86_64; rv:33.0) Gecko/20100101 Firefox/33.0
Build ID: 20141016231145

Steps to reproduce:

Wrote a html file with style of different font family


Actual results:

Firefox never handled cursive and fantasy, and some font family string which works with fc-match was not matched in Firefox


Expected results:

Firefox should pick fonts exactly as the result of fc-match
Component: Untriaged → Layout: Text
Product: Firefox → Core

Comment 1

3 years ago
Could you attach the testcase, please.
Flags: needinfo?(tom.ty89)
Keywords: testcase-wanted
(Reporter)

Comment 2

3 years ago
You mean the html file in my attached screenshot or?

Some more perhaps useful info:
I'm with Arch Linux. My fontconfig version is 2.11.1. The "URW Chancery L" font is from here:
http://svn.ghostscript.com/ghostscript/tags/urw-fonts-1.0.7pre44/
I also got ttf-dejavu 2.34 installed.

Comment 3

3 years ago
(In reply to Tom Yan from comment #2)
> You mean the html file in my attached screenshot or?

Yes.

Comment 4

2 years ago
Created attachment 8720916 [details]
screen-copy.PNG


User Agent 	Mozilla/5.0 (X11; Linux x86_64; rv:47.0) Gecko/20100101 Firefox/47.0
Version 	47.0a1
Build ID 	20160217062621
Update Channel 	nightly
OS 	Linux 3.19.0-49-generic x86-64

Test case was given by the reporter but it was in PNG format. Rewritten html file is attached now. In addition, Screenshot of my fc-match and browser outputs is also attached. It is not reproducible in my machine. Fantasy and cursive fonts seem to match with the fc-match fonts.

Updated

2 years ago
Keywords: testcase-wanted

Comment 5

2 years ago
Created attachment 8720919 [details]
html
(Reporter)

Comment 6

2 years ago
@Abe - QA You were not even doing a fair test. You should at least make sure that cursive and fantasy is not matched to the universal default ("Dejavu Sans" "Book"). Otherwise the test is simply pointless. For example:

[tom@localhost ~]$ git diff /etc/fonts/conf.avail/45-latin.conf /etc/fonts/conf.d/45-latin.conf 
diff --git a/etc/fonts/conf.avail/45-latin.conf b/etc/fonts/conf.d/45-latin.conf
index 47a1837..ade6386 100644
--- a/etc/fonts/conf.avail/45-latin.conf
+++ b/etc/fonts/conf.d/45-latin.conf
@@ -167,6 +167,10 @@
   Fantasy faces
  -->
        <alias>
+               <family>URW Gothic</family>
+               <default><family>fantasy</family></default>
+       </alias>
+       <alias>
                <family>Impact</family>
                <default><family>fantasy</family></default>
        </alias>
@@ -186,6 +190,10 @@
   Cursive faces
   -->
        <alias>
+               <family>Chancery URW</family>
+               <default><family>cursive</family></default>
+       </alias>
+       <alias>
                <family>ITC Zapf Chancery Std</family>
                <default><family>cursive</family></default>
        </alias>
[tom@localhost ~]$ git diff /etc/fonts/conf.avail/60-latin.conf /etc/fonts/conf.d/60-latin.conf 
diff --git a/etc/fonts/conf.avail/60-latin.conf b/etc/fonts/conf.d/60-latin.conf
index 35600ea..e531aa9 100644
--- a/etc/fonts/conf.avail/60-latin.conf
+++ b/etc/fonts/conf.d/60-latin.conf
@@ -52,6 +52,7 @@
        <alias>
                <family>fantasy</family>
                <prefer>
+                       <family>URW Gothic</family>
                        <family>Impact</family>
                        <family>Copperplate Gothic Std</family>
                        <family>Cooper Std</family>
@@ -64,6 +65,7 @@
        <alias>
                <family>cursive</family>
                <prefer>
+                       <family>Chancery URW</family>
                        <family>ITC Zapf Chancery Std</family>
                        <family>Zapfino</family>
                        <family>Comic Sans MS</family>

Anyway, I confirmed that at least as of 44.0.2, firefox matches the same font as fc-match when dealing with "cursive" and "fantasy".

Although it still doesn't match the "indirect" alias "Zapf Chancery" to "Chancery URW" like fc-match does, but at least it does match the "direct" alias "ITC Zapf Chancery", which Chromium doesn't match:

(/etc/fonts/conf.d/30-metric-aliases.conf)
...
	<alias binding="same">
	  <family>Zapf Chancery</family>
	  <default>
	  <family>ITC Zapf Chancery</family>
	  </default>
	</alias>
...
	<alias binding="same">
	  <family>Chancery URW</family>
	  <default>
	  <family>ITC Zapf Chancery</family>
	  </default>
	</alias>
...
	<alias binding="same">
	  <family>ITC Zapf Chancery</family>
	  <accept>
	  <family>TeX Gyre Chorus</family>
	  <family>Chancery URW</family>
	  <family>URW Chancery L</family>
	  </accept>
	</alias>
...
(Reporter)

Comment 7

2 years ago
Created attachment 8720970 [details]
testcase_44_0_2.png

Again I won't provide the html file. It's clearly shown on the screenshots I uploaded. Not to mention that it is just a super simple html file anyone can make in a minute, and whoever want to reproduce this issue should write the test file according to their available fonts and fontconfig conf files being used. The request was so silly/absurb that I didn't even want to bother caring about this report anymore.

Seriously, I doubt anyone would bother to fix it if he don't even want to write a proper test file himself, not to mention that it is such a simple one.
Flags: needinfo?(tom.ty89)

Comment 8

2 years ago
Tom, thanks for taking on this issue. I was verifying if the two fonts (cursive and fantasy)match with my fc-match. I also noticed that other fonts do not match with fc-match. I will re-review it further using your inputs.
Flags: needinfo?(tom.ty89)

Comment 9

2 years ago
Hi, Tom - thank you for filing the bug, but this tone is unwarranted. If you would like to provide the file, or a reduced testcase, that reproduces the issue, that would help us prioritize and resolve this bug. Otherwise, we'll have to do the best we can with the resources available.

Updated

a year ago
Flags: needinfo?(tom.ty89)
You need to log in before you can comment on or make changes to this bug.