212137 - ::first-letter doesn't work properly when ::before/::after generated content is present

Status: VERIFIED FIXED

(Core :: CSS Parsing and Computation, defect, P4)

Description

[Anne (:annevk)]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.5a) Gecko/20030707
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.5a) Gecko/20030707

I made to testcases about this problem: 
:before / ::before
http://www.annevankesteren.nl/css/testcase/pseudo_before.html
:after / ::after
http://www.annevankesteren.nl/css/testcase/pseudo_after.html

The second table on each page shows the problem.

Unrelated: You can't select generated text, Mozilla applies the css on invalid
attributes.

Reproducible: Always

Steps to Reproduce:

Comment 1

[Josh Birnbaum]

selecting generated content is bug 12460

Comment 2

[Mats Palmgren (inactive)]

From http://www.w3.org/TR/CSS21/selector.html#first-letter :
"The :first-letter pseudo-element matches parts of block-level elements only."

If you change EM to P in the examples they work better (except "css3-test-21"
and "css3-test-22" for which you have a typo in the style rules).

There is one thing that doesn't seem to work though, the text-decoration
property isn't applied to the :first-letter.

Comment 3

[Anne (:annevk)]

Typos and incorrect implementation from the standard have been fixed in both
examples. Changed the summary als o a bit so it fits better to the problem.

//
I had also another problem with generated content in mozilla. Opera shows
something, but does not apply everything.

That was also about an inline element (code), but I set it to a block-level
element with css. Should I put this into it's own bug? Example:
http://www.annevankesteren.nl/css/testcase/generated_content.html

Comment 4

[Mats Palmgren (inactive)]

> Should I put this into it's own bug?

Yes, please open a new bug on that.

I don't really see what the problem is in this bug though (apart from text-
decoration and selection not working).  I see a green "G" everywhere in the
lower part of both your examples.  Would you mind explaining "actual behaviour"
and "expected behaviour" for one of those examples?

Comment 5

[Anne (:annevk)]

Well actually is was an extended testcase for the CSS3 testcase:
http://www.w3.org/Style/CSS/Test/CSS3/Selectors/current/html/full/flat/css3-modsel-39.html
http://www.w3.org/Style/CSS/Test/CSS3/Selectors/current/html/full/flat/css3-modsel-39a.html
http://www.w3.org/Style/CSS/Test/CSS3/Selectors/current/html/full/flat/css3-modsel-39c.html
http://www.w3.org/Style/CSS/Test/CSS3/Selectors/current/html/full/flat/css3-modsel-39b.html

In those (39a and 39c) tests it seems that Mozilla doesn't support it. The
text-decoration wasn't really the expected behavior, but now it's a bug :).

The expected behaviour stands above the tables. I you're pointing at something
else please notice me, since this is my very first Mozilla bug report.

Thank you.

Comment 6

[Mats Palmgren (inactive)]

OK, thanks.  Looking at those tests though (39a and 39c), it seems they have
a bug in them.  There is a comma in between the rules that shouldn't be there.
I'm going to attach cleaned up versions of those two tests.

Comment 7

[Mats Palmgren (inactive)]

Attachment: Testcase css3-modsel-39a.html

Comment 8

[Mats Palmgren (inactive)]

Attachment: Testcase css3-modsel-39c.html

Comment 9

[Mats Palmgren (inactive)]

AFAICT, those tests are still bogus since the selectors have equal specificity
and then the last (red) should be the background-color to use, no?

Comment 10

[Anne (:annevk)]

The W3C says this:

"When the ::first-letter and ::first-line pseudo-elements are combined with
::before and ::after, they apply to the first letter or line of the element
including the inserted text."

Source:
http://www.w3.org/TR/css3-selectors/#gen-content

So I think it should be totally green and not red.

Comment 11

[Mats Palmgren (inactive)]

I've spawned of bug 212149 for the text-decoration problem.

Comment 12

[Hixie (not reading bugmail)]

(tests have been fixed, expect the new tests uploaded soon)

Comment 13

[Mats Palmgren (inactive)]

This bug was fixed by the patch in bug 212149, see bug 212149 comment 4.

Comment 14

[Mats Palmgren (inactive)]

Verified fixed, 2003-07-11-22 trunk Linux