SynWeb - advanced multi PHP, (X)HTML, WML, CSS, JS syntax highlighter for SynEdit/UniSynEdit and other tools

News Download Project page | Browse Submit New | Admin

Project:

#ID	58
Title	SynWeb can not draw/parse unicode WideGlyph characters
Description	hi, on the description of this component, is says its Unicode enabled but it only works with halfwidth unicode characters. i'm currently developing an JAPANESE editor with UniSynEdit and without being able to draw/parse japanese wide glyphs ( like, KANJI ) the unicode support would be completely meaningless (for me and alot others from asia). UniSynEdit itself works perfectly with wide glyps. i am sure a lot of people need this too, if anybody knows a quick fix it would help (me and asians ) alot. thanks in advance, i really like SynWeb but it's sad it doesn't work with Japanse(and other languages with WideGlyphs). sorry for my sloppy english, Hiroshi Horie
Status	Closed
Type	Bug
Resolution	Resolved
Priority	High
Posted	2007-02-23 18:36
Last modified	2008-07-25 19:57
Closed	2007-04-02 16:02
Submited by	Hiroshi Horie
E-mail	(hidden)
Summary	Fixed parsing unicode wide characters in asian versions of Windows (reported by Hiroshi Horie)
Available in	v1.4 (SVN rev. 140)
Back

Total comments: 5 ,Add comment
#ID	52	Posted By	Krystian Bigaj	Posted	2007-02-24 19:10
Hi, Thanks for report. Could you give me an example file and screenshots with/without SynWeb? SynWeb internally always works in ANSI (even in Unicode mode). I know that this is not best (recommended?) way, but for highlighter like SynWeb it (should?) doesn't matter. In that way it works faster, and it was much easier to implement. It's not like that: Unicode input ---convert to ANSI---> [internal SynWeb work on ANSI] ---conver ANSI to Unicode---> Unicode output (in that way all Unicode chars will be gone (changed to "?") but it's like that: Unicode input ---convert to ANSI---> [internal SynWeb work on ANSI] ---get data from Unicode input---> Unicode output I made some small tests and it seems to work fine (copied text from your email domain server www http://www.s-center.net/): UniSynEdit: http://flatdev.republika.pl/unisynedit.png UniSynEdit with SynWeb: http://flatdev.republika.pl/unisynedit_wSynWeb.png
#ID	53	Posted By	Hiroshi Horie	Posted	2007-02-25 08:13
hi, thanks for your reply. humm...thats odd.. you're saying SynWeb should work right? it looks like you can display japanese correctly, i wonder why.. these are the screen shots of SynWeb on and off on : http://www.s-center.net/c_temp/synweb_with.png off : http://www.s-center.net/c_temp/synweb_without.png (i might delete these pics in the near future) as you can see, it draws garbage characters after the text. and when selecting, it selects text by half-width only, should be full-width on Japanese chars. i tested both SynWeb demo executable from SourceForge, and the one i compiled myself but they both have the same problems. just for your information, OS is Windows XP Pro Japanese SP2, MEM 2G ps. i checked out PHP Designer 2007 that uses UniSynEdit and SynWeb, but they have the same problem too. thanks in advance, Hiroshi Horie
#ID	54	Posted By	Krystian Bigaj	Posted	2007-02-25 10:57
I've downloaded exe from sf.net (to use same .exe) and made same test as you did (same text): http://flatdev.republika.pl/unisynedit2.png http://flatdev.republika.pl/unisynedit2_wSynWeb.png I'm using WinXP Pro SP2 EN. I really don't know what could be wrong if on same exe on one comp. work, but on another doesn't work properly. Also I tried select some text, and it selects as it should, on widechars it select whole wide char. Try latest PHP Designer 2007 5.2.1, Michael Pham in that version uses latest SynWeb (SVN) and latest UniSynEdit (CVS), as I remember in UniSynEdit in one of last commit was some changes related to painting tokens. You could also recompile synweb demo, and replace highlighter with SynHiglighterHtml and check results.
#ID	58	Posted By	Krystian Bigaj	Posted	2007-03-27 19:27
Confirmed. After changing regional settings in "Control Panel/Regional and Language Options - Standards and formats" to for example to: Chinese (Taiwan) problem appear. Change that setting to: English to temporary workaround.
#ID	59	Posted By	Krystian Bigaj	Posted	2007-03-28 02:38
Fixed in svn revision #135 It was Delphi bug (conversion WideString->String with specific regional settings). In Turbo Delphi 2006 Pro with latest hot fixes it worked fine without that fix. Thanks to Michael Pham to possibility of remote desktop debug :)
#ID	(New)	Posted By
Verification code			Retype string from image here - case sensitive string, only A-Z, a-z chars
BBCode: [b]...[/b], [i]...[/i], [u]...[/u], [color=...]...[/color], [bgcolor=...]...[/bgcolor], [code]...[/code], [email]...[/email], [email=...]...[/email], , [url]...[/url], [url=...]...[/url] Post new comment