Open GoogleCodeExporter opened 8 years ago
I'm not sure exactly what's going on but it looks like a font issue. But I
don't know why it would only break in the pop-up and not the rest of the
internet.
I tried to reproduce and was not able to, so the Kanji stored in the database
is correct it's just being displayed wrong in your case for some reason.
Original comment by melin...@gmail.com
on 8 Apr 2011 at 2:34
What font is rikaikun using for the kanji screens?
Original comment by mihail.p...@gmail.com
on 11 Apr 2011 at 3:46
I was looking at this issue because its similar to another I have that
sometimes when there is a line break, rikaikun translates the kanji in the next
line and not the one i'm pointing.
In this particular case (the kanji in the picture), i looked in the wiktionary
and seems it's the same kanji. Look at this page:
http://en.wiktionary.org/wiki/%E7%BD%AE
Original comment by estevez....@gmail.com
on 10 Jun 2011 at 12:46
Yes, i agree - however each form has it's own encoding. I would expect
rikaikun to preserve the encoding of the character and not replace it
with a variant.
Original comment by mihail.p...@gmail.com
on 10 Jun 2011 at 12:59
It's hard to check on this because I can't reproduce it but I agree that the
problem is there for some reason. What's odd is that even if Rikaikun was
choosing it's own encoding that should make the problem pretty wide spread.
I obviously haven't spent quality time on rikaikun in too long, but I'll keep
this in mind. Mihail, is this problem very common for you?
Original comment by melin...@gmail.com
on 10 Jun 2011 at 1:45
No. I think it only happened twice.
Original comment by mihail.p...@gmail.com
on 10 Jun 2011 at 2:06
I've looked for the kanji of the popup on the Unihan Database (http://www.unicode.org/cgi-bin/GetUnihanData.pl?codepoint=7F6E) and it seems that if you change the default (sans-serif) font in the chrome options page the kanji turns to the other.
If the page doesn't specify a font, Chrome uses Arial by default so the
character that the popup shows it's the same as in unihan database. Maybe you
should use the same font that the character has in the page.
Original comment by estevez....@gmail.com
on 4 Jul 2011 at 12:47
You're saying that the default chrome font works fine right?
It only breaks if Chrome is set to sans-serif?
I'm confused. which font are you saying I should use?
Original comment by melin...@gmail.com
on 4 Jul 2011 at 12:53
If you go to the options page of chrome (Options->Under the Bonnet->Web
Content->Customize Font->Sans Serif Font), this is the font you should change
to see the other kanji in the sample page I put in my other comment. Some fonts
I found that is "incorrect" is MS Gothic or MS Mincho.
The suggerence I made is that to match the kanji (even if it's "incorrect") is
to take the font that the page is using. An example would be, if the page is
using MS Mincho the popup should use it too, if the page is using Arial the
popup should use Arial, etc.
Original comment by estevez....@gmail.com
on 5 Jul 2011 at 7:09
If you go to this page: http://aulex.org/ja-es/kanji.php you should see the
"incorrect" kanji.Sorry, but the page is in spanish. The kanji in question
should be under "13 trazos" in the 5th line (if zoom is at 100%).
If you use "Inspect element" on the kanji, you should see the "correct" kanji
in the inspector. Take note that the font-family in the styles is "MS Mincho",
"MS Gothic", "Arial Unicode MS".
If you follow my suggestion in this case you should use this font-family in the
popup.
Sorry for the long comment and my bad english.
Original comment by estevez....@gmail.com
on 5 Jul 2011 at 9:57
In the pdf on this link: http://www.unicode.org/charts/PDF/U4E00.pdf you can
see the differences in the kanjis. As you can see many kanji have different
font versions but very few are so evident as in the attach image. To find the
kanji use the unicode (7F6E).
For easy use, you should download the pdf instead of opening it, unless you
have a fast connection. The file has a size of 28 MB.
Original comment by estevez....@gmail.com
on 5 Jul 2011 at 11:03
The code to obtain the font family should be something like this:
var ff =
document.defaultView.getComputedStyle(element,null).getPropertyValue("font-famil
y");
Where "element" is the element you want to know the style of. In this case it
would be the element that contains the kanji. Later, you should use this value
to apply it to the kanji in the popup, this way the two kanjis will match.
The getComputedStyle() is convenient in this case because it will give you the
style being applied to the element not the one it has in his properties (that
might even not exist).
Original comment by estevez....@gmail.com
on 6 Jul 2011 at 7:37
Sorry, I think I went a little too fast. I have been looking for more
information on the kanji in the sample and it seems that the kanji shown is
incorrect. If you look in the pdf mentioned earlier, the two first columns to
chinese characters, the third column corresponds to japanese, the fourth to
korean and the fifth and sixth to vietnamese.
That means you should use a font or fonts that shows the other kanji, because I
don't think you will see the other kanji in any japanese website. Of course,
you could use the suggestion I mentioned earlier to ensure that the popup will
always match the one in the page.
Original comment by estevez....@gmail.com
on 7 Jul 2011 at 3:48
This isn't a huge priority but I can agree that showing the same Kanji that is
on the page is a good solution. Thanks for looking into this.
Original comment by melin...@gmail.com
on 8 Jul 2011 at 2:18
Original issue reported on code.google.com by
mihail.p...@gmail.com
on 8 Apr 2011 at 9:48Attachments: