add meta ... Content-Type content="text/html; charset=utf-8 to site
Talk Bug Collectors
Join LibraryThing to post.
This topic is currently marked as "dormant"—the last message is more than 90 days old. You can revive it by posting a reply.
1gangleri
Hi! Sorry for this!
It does not make much sense to display random / unpredictable content.
I noticed that my posts at http://www.librarything.com/topic/99615 are displayed in an random fashion.
Message 1 does not display what was originaly posted. Message 8 looks fine now but how will it look tomorrow ...
I beleive that adding some meta tgas to the site
meta http-equiv="Content-Type" content="text/html; charset=utf-8"
could save a lot of time ond avoid many troubles.
Best regards Reinhardt
Here some test:
test:_à ,_á,_â,_ä,_Ã¥,_æ,_ç,_è,_é,-ë,_Ã,_ï,_ð,_ñ,_ó,_ö,_ø,_ú,_ü,_þ
Test:_%C3%A0,_%C3%A1,_%C3%A2,_%C3%A4,_%C3%A5,_%C3%A6,_%C3%A7,_%C3%A8,_%C3%A9,-%C3%AB,_%C3%AD,_%C3%AF,_%C3%B0,_%C3%B1,_%C3%B3,_%C3%B6,_%C3%B8,_%C3%BA,_%C3%BC,_%C3%BE
I used the first line as a title in Wikipedia. Then I copied and edited the resulting url using firefox.
It does not make much sense to display random / unpredictable content.
I noticed that my posts at http://www.librarything.com/topic/99615 are displayed in an random fashion.
Message 1 does not display what was originaly posted. Message 8 looks fine now but how will it look tomorrow ...
I beleive that adding some meta tgas to the site
meta http-equiv="Content-Type" content="text/html; charset=utf-8"
could save a lot of time ond avoid many troubles.
Best regards Reinhardt
Here some test:
test:_à ,_á,_â,_ä,_Ã¥,_æ,_ç,_è,_é,-ë,_Ã,_ï,_ð,_ñ,_ó,_ö,_ø,_ú,_ü,_þ
Test:_%C3%A0,_%C3%A1,_%C3%A2,_%C3%A4,_%C3%A5,_%C3%A6,_%C3%A7,_%C3%A8,_%C3%A9,-%C3%AB,_%C3%AD,_%C3%AF,_%C3%B0,_%C3%B1,_%C3%B3,_%C3%B6,_%C3%B8,_%C3%BA,_%C3%BC,_%C3%BE
I used the first line as a title in Wikipedia. Then I copied and edited the resulting url using firefox.
2gangleri
>1 gangleri: The bug might relate to the ediiting feature:
again:
test:_à,_á,_â,_ä,_å,_æ,_ç,_è,_é,-ë,_í,_ï,_ð,_ñ,_ó,_ö,_ø,_ú,_ü,_þ
Test:_%C3%A0,_%C3%A1,_%C3%A2,_%C3%A4,_%C3%A5,_%C3%A6,_%C3%A7,_%C3%A8,_%C3%A9,-%C3%AB,_%C3%AD,_%C3%AF,_%C3%B0,_%C3%B1,_%C3%B3,_%C3%B6,_%C3%B8,_%C3%BA,_%C3%BC,_%C3%BE
again:
test:_à,_á,_â,_ä,_å,_æ,_ç,_è,_é,-ë,_í,_ï,_ð,_ñ,_ó,_ö,_ø,_ú,_ü,_þ
Test:_%C3%A0,_%C3%A1,_%C3%A2,_%C3%A4,_%C3%A5,_%C3%A6,_%C3%A7,_%C3%A8,_%C3%A9,-%C3%AB,_%C3%AD,_%C3%AF,_%C3%B0,_%C3%B1,_%C3%B3,_%C3%B6,_%C3%B8,_%C3%BA,_%C3%BC,_%C3%BE
3gangleri
note: this might a a subsequent error or somthing different
The two Talk search differ only by the leading space at the first one:
+leading+space
leading+space
The first shows Unicode Character 'REPLACEMENT CHARACTER's:
gangleri in Bug Collectors : normalization: ignore �� leading / trailing spaces �� �� spaces / whitespace �� in �� Common Knowle (Jul 15, 3:04pm)
The second shows:
gangleri in Bug Collectors : normalization: ignore « leading / trailing spaces » « spaces / whitespace » in « Common Knowle (Jul 15, 3:04pm)
I do not know ho the highlighting might be involved becuase the fist search starts witha leading space.
The two Talk search differ only by the leading space at the first one:
+leading+space
leading+space
The first shows Unicode Character 'REPLACEMENT CHARACTER's:
gangleri in Bug Collectors : normalization: ignore �� leading / trailing spaces �� �� spaces / whitespace �� in �� Common Knowle (Jul 15, 3:04pm)
The second shows:
gangleri in Bug Collectors : normalization: ignore « leading / trailing spaces » « spaces / whitespace » in « Common Knowle (Jul 15, 3:04pm)
I do not know ho the highlighting might be involved becuase the fist search starts witha leading space.
4brightcopy
Your original two posts were about a bug that Tim fixed a few days ago (I can dig it up if you'd like).
For the new post, I tried your first example and got:
Warning: strpos(): Empty delimiter. in /var/www/html/search_talk.php on line 355
As the first line. That's definitely a bug. You should post this to Bug Collectors as a new post if you haven't already. Make sure you mention that error.
I don't get quite the results you do, but I think that error may be screwing some things up.
For the new post, I tried your first example and got:
Warning: strpos(): Empty delimiter. in /var/www/html/search_talk.php on line 355
As the first line. That's definitely a bug. You should post this to Bug Collectors as a new post if you haven't already. Make sure you mention that error.
I don't get quite the results you do, but I think that error may be screwing some things up.
5gangleri
>4 brightcopy: Thanks for your time! http://www.librarything.com/topic/100109
I will make some tests to confirm if the original bug is fixed / "works for me".
I will make some tests to confirm if the original bug is fixed / "works for me".
6gangleri
>1 gangleri:, >2 gangleri:, >4 brightcopy: :
The characters from the test:_à,_á,_â,_ä,_å,_æ,_ç,_è,_é,-ë,_í,_ï,_ð,_ñ,_ó,_ö,_ø,_ú,_ü,_þ are reserved now when the message is edited again. Thanks Tim for fixing! Thanks brightcopy for the time. Would be nice to have the Bug Tracking number; however there is no hurry.
This editing is fixed / "works for me".
----
However I beleive that the title request meta ... Content-Type content="text/html; charset=utf-8" to site would help avoiding a lot of subsequent bugs:
http://www.librarything.com/topic/95972 might relate to such "browser misunderstandigs".
Many sites are missing this: at http://www.nndb.com/people/811/000091538/ I can see
Mother: Marija Javer�ek
Son: Mi�o Broz (by Hertha)
The characters from the test:_à,_á,_â,_ä,_å,_æ,_ç,_è,_é,-ë,_í,_ï,_ð,_ñ,_ó,_ö,_ø,_ú,_ü,_þ are reserved now when the message is edited again. Thanks Tim for fixing! Thanks brightcopy for the time. Would be nice to have the Bug Tracking number; however there is no hurry.
This editing is fixed / "works for me".
----
However I beleive that the title request meta ... Content-Type content="text/html; charset=utf-8" to site would help avoiding a lot of subsequent bugs:
http://www.librarything.com/topic/95972 might relate to such "browser misunderstandigs".
Many sites are missing this: at http://www.nndb.com/people/811/000091538/ I can see
Mother: Marija Javer�ek
Son: Mi�o Broz (by Hertha)
8gangleri
If I look at the the source code of this page I can see only
<meta name="description" content="LibraryThing catalogs your books online, easily, quickly and for free.">
<meta name="keywords" content="librarything, library, thing, catalog your books, catalogue your books, book cataloging, library, free book catalog, catalogue">
at a / any page from en.Wikipedia I can see:
<html xmlns="http://www.w3.org/1999/xhtml" lang="en" dir="ltr">
....
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />
<meta http-equiv="Content-Style-Type" content="text/css" />
<meta name="generator" content="MediaWiki 1.16wmf4" />
<meta name="robots" content="noindex,follow" />
This is a clear indication. Not everything will / may aply here ...
Oops there are five meta lines at Wikipedia
<meta name="description" content="LibraryThing catalogs your books online, easily, quickly and for free.">
<meta name="keywords" content="librarything, library, thing, catalog your books, catalogue your books, book cataloging, library, free book catalog, catalogue">
at a / any page from en.Wikipedia I can see:
<html xmlns="http://www.w3.org/1999/xhtml" lang="en" dir="ltr">
....
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />
<meta http-equiv="Content-Style-Type" content="text/css" />
<meta name="generator" content="MediaWiki 1.16wmf4" />
<meta name="robots" content="noindex,follow" />
This is a clear indication. Not everything will / may aply here ...
Oops there are five meta lines at Wikipedia
9brightcopy
Sorry, I'm talking about the characters that weren't properly escaped on editing. I'm not talking about the charset issue.
10gangleri
>7 brightcopy:, >9 brightcopy: I received a phone call. I noticed later. Thanks
11PaulFoley
http://validator.w3.org finds lots of errors in LT pages...
12gangleri
content="text/html; charset=utf-8" would assume to fix also a lot of utf-8 issues:
http://www.librarything.com/topic/82757 old bug
http://www.librarything.com/topic/100160 (Bug Tracking)
http://www.librarything.com/topic/82757 old bug
http://www.librarything.com/topic/100160 (Bug Tracking)
13PaulFoley
Putting the charset on the content header, while a good idea in itself, wouldn't help at all here. You're already seeing UTF-8 -- that's the problem: the character on the page is not UTF-8. It would need to send "charset=iso-8859-1" if you wanted that to display correctly (but then other characters on the page that are UTF-8 would be screwed). LT seems to have bad data in its language names.
16gangleri
http://www.librarything.com/topic/102441 refers here
17gangleri
originally posted: nov 22, 2010, 7:21am
I just found http://www.librarything.com/profile/Perlovka . The talk is illegible ...
2012-03-27: P.S. re: "CharSetUTF-8"
Today one and a half years later using another computer and "Mozilla/5.0 (Windows NT 5.1; rv:11.0) Gecko/20100101 Firefox/11.0" I can see the text in Russian properly.
I just found http://www.librarything.com/profile/Perlovka . The talk is illegible ...
2012-03-27: P.S. re: "CharSetUTF-8"
Today one and a half years later using another computer and "Mozilla/5.0 (Windows NT 5.1; rv:11.0) Gecko/20100101 Firefox/11.0" I can see the text in Russian properly.
