Re: non-Latin-1 in iCab 3

On Sat, 22 Jul 2006, Sander Tekelenburg wrote:

> I put up some 'screensots' of how iCab 3 renders your
> pages: <http://santek.no-ip.org/~st/tests/iCab/characters/encoding/>.
> Does that look wrong to you?

Thank you very much!
I wonder whether *you* switched CSS off or whether iCab can't do.


ISO Baltic Test:

Okay - although letters with cedilla are from a different font than
the rest. Note especially "G g K k l".


Windows Hebrew Test:

The Hebrew text in <dd dir="rtl" class="rightalign"> should be
right aligned, but it is left aligned.

The parentheses are wrong.


ISO Arabic Test:

The Arabic text in <dd dir="rtl" class="rightalign"> should be
right aligned, but it is left aligned.

The digits should be displayed as "10 9 8 7 6 5 4 3 2 1 0".
The Arabic-Indic digits are missing, although the Persian digits
are displayed.


Unicode (UTF-8) Test:

Some Latin Extended letters are from another font. Note especially
"Gg" with cedilla, cir***flex, dot above.

Hebrew - errors as above.

Arabic - errors as above.

Devanagari:
The vowel signs for "ku kuu kr krr kl kai" are misplaced.

Gujarati:
The vowel signs for "ku kuu kr krr kai" are misplaced.

Gurmukhi:
The vowel signs for "ku kuu ke kai ko kau" are misplaced.
The letter "l. with nukta" is displayed as "ka" plus a square. ?!

Symbols:
The en dash looks more like a hyphen.


Conclusion:
iCab 3 does certainly better than iCab 2.x. It seems to work
well with the Latin, Greek, Cyrillic scripts. However, iCab is
still not ready for right-to-left scripts and for Indic scripts.