460441 - Dynamically added "…"(intl.ellipsis) breaks surrogate pair

Status: RESOLVED DUPLICATE of bug 404856

(Core :: XUL, defect)

Description

[Masahiro YAMADA]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.1; ja; rv:1.9.0.3) Gecko/2008092417 Firefox/3.0.3
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; ja; rv:1.9.0.3) Gecko/2008092417 Firefox/3.0.3

Patches for bug 403484 and related bugs does not handle surrogate pair correctly.
intl.ellipsis can be inserted after first byte of surrogate pair.

Reproducible: Always

Steps to Reproduce:
1.
2.
3.



It would be better to implement Core APIs that does not break surrogate pair.

Comment 1

[Masahiro YAMADA]

In C++ code, we can use NS_IS_LOW_SURROGATE() and NS_IS_HIGH_SURROGATE().
nsIndexedToHTML.cpp uses them to keep surrogate pair correctly.

Comment 2

[Masahiro YAMADA]

For example:
http://www.google.co.jp/search?q=%F0%9D%90%B4%F0%9D%90%B4%F0%9D%90%B4%F0%9D%90%B4%F0%9D%90%B4%F0%9D%90%B4
(On Windows, It needs Cambria Math font, cantained in "MS Office" or "Power Point Viewer")

Comment 3

[Masahiro YAMADA]

(In reply to comment #2)
> For example:
> http://www.google.co.jp/search?q=%F0%9D%90%B4%F0%9D%90%B4%F0%9D%90%B4%F0%9D%90%B4%F0%9D%90%B4%F0%9D%90%B4
> (On Windows, It needs Cambria Math font, cantained in "MS Office" or "Power
> Point Viewer")

It can be reproduced on title text of tab.

Comment 4

[Makoto Kato [:m_kato]]

Attachment: fix v1

Comment 5

[Makoto Kato [:m_kato]]

Attachment: fix v1.1

Comment 6

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 540006 [details] [diff] [review]
fix v1.1

>-    // XXX and clusters!

This comment needs to stay.

>         case CropLeft:

In this case, ucs4 and isSurrogate should still be declared inside the for loop.

Can we try to factor out that identical-looking code instead of copy/pasting it?