Fix handling of Arabic script in SVGTextContentElement.getSubStringLength

Fix handling of Arabic script in SVGTextContentElement.getSubStringLength

Pass the direction of the line box instead of the layout object when
measuring the sub string.

BUG=470326
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=192670

[fs, 2015-03-26 14:42:19]

fs@opera.com changed reviewers:
+ ed@opera.com, pdr@chromium.org

[Erik Dahlström (inactive), 2015-03-26 16:36:22]

lgtm

Might be good with a tests that spans more than one textbox (with different
directions), just to show that it works.

[fs, 2015-03-26 17:17:01]

On 2015/03/26 16:36:22, Erik Dahlström wrote:
> lgtm
> 
> Might be good with a tests that spans more than one textbox (with different
> directions), just to show that it works.

That's pretty tangential to this fairly simple drive-by fixup... =) I can look
into it though (I suspect that will require more fixing though...)

[fs, 2015-03-27 09:58:47]

On 2015/03/26 17:17:01, fs wrote:
> On 2015/03/26 16:36:22, Erik Dahlström wrote:
> > lgtm
> > 
> > Might be good with a tests that spans more than one textbox (with different
> > directions), just to show that it works.
> 
> That's pretty tangential to this fairly simple drive-by fixup... =) I can look
> into it though (I suspect that will require more fixing though...)

Maybe unsurprisingly, but this method has other issues then... Ok to land this
in the mean time while looking into that?

[fs, 2015-03-27 10:02:29]

On 2015/03/27 09:58:47, fs wrote:
> On 2015/03/26 17:17:01, fs wrote:
> > On 2015/03/26 16:36:22, Erik Dahlström wrote:
> > > lgtm
> > > 
> > > Might be good with a tests that spans more than one textbox (with
different
> > > directions), just to show that it works.
> > 
> > That's pretty tangential to this fairly simple drive-by fixup... =) I can
look
> > into it though (I suspect that will require more fixing though...)
> 
> Maybe unsurprisingly, but this method has other issues then... Ok to land this
> in the mean time while looking into that?

"SVG text code - where the fun never ends"

/me goes off to start a theme park with rides such as: "<textPath> of horror",
"Delta adjustments" and "BiDi text query bungee".

[Erik Dahlström (inactive), 2015-03-27 10:14:06]

On 2015/03/27 10:02:29, fs wrote:
> On 2015/03/27 09:58:47, fs wrote:
> > On 2015/03/26 17:17:01, fs wrote:
> > > On 2015/03/26 16:36:22, Erik Dahlström wrote:
> > > > lgtm
> > > > 
> > > > Might be good with a tests that spans more than one textbox (with
> different
> > > > directions), just to show that it works.
> > > 
> > > That's pretty tangential to this fairly simple drive-by fixup... =) I can
> look
> > > into it though (I suspect that will require more fixing though...)
> > 
> > Maybe unsurprisingly, but this method has other issues then... Ok to land
this
> > in the mean time while looking into that?
> 
> "SVG text code - where the fun never ends"
> 
> /me goes off to start a theme park with rides such as: "<textPath> of horror",
> "Delta adjustments" and "BiDi text query bungee".

I think it's fine to land this now, and to address the remaining issues in
followup CLs.

[fs, 2015-03-27 10:14:39]

On 2015/03/27 10:14:06, Erik Dahlström wrote:
> On 2015/03/27 10:02:29, fs wrote:
> > On 2015/03/27 09:58:47, fs wrote:
> > > On 2015/03/26 17:17:01, fs wrote:
> > > > On 2015/03/26 16:36:22, Erik Dahlström wrote:
> > > > > lgtm
> > > > > 
> > > > > Might be good with a tests that spans more than one textbox (with
> > different
> > > > > directions), just to show that it works.
> > > > 
> > > > That's pretty tangential to this fairly simple drive-by fixup... =) I
can
> > look
> > > > into it though (I suspect that will require more fixing though...)
> > > 
> > > Maybe unsurprisingly, but this method has other issues then... Ok to land
> this
> > > in the mean time while looking into that?
> > 
> > "SVG text code - where the fun never ends"
> > 
> > /me goes off to start a theme park with rides such as: "<textPath> of
horror",
> > "Delta adjustments" and "BiDi text query bungee".
> 
> I think it's fine to land this now, and to address the remaining issues in
> followup CLs.

I think I'll make a new bug for it even...

[fs, 2015-03-27 10:14:43]

The CQ bit was checked by fs@opera.com

[commit-bot: I haz the power, 2015-03-27 10:15:05]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1032363002/1

[commit-bot: I haz the power, 2015-03-27 11:44:21]

Committed patchset #1 (id:1) as
https://src.chromium.org/viewvc/blink?view=rev&revision=192670