RPGWatch - Moving to a new server completed

Originally Posted by Wisdom
If you mean the characters Chien, Fluent, Joxer and Dart?
I think those characters were broken before the move.

In SWTOR lingua, "broken" can also mean "vastly overpowered"

Originally Posted by purpleblob
Just a few things I'd like to mention.

Logging on takes me to RPGWatch main page, instead of forums now.

Also, some characters are now broken, especially when you use quote function.

Can you give an example of characters that are broken?


Odd that it has this changed behavior. We moved the forum software from 3.8.10 to 3.8.11, so it is strange if this new behavior is caused by that. Perhaps it is some odd thing because of PHP 7 or there is a dependency somewhere I'm not aware of.

Well don't know if it helps but when I quote a reply or an article I get a bunch of ?'s.

Recently, after a promotional trip paid by Square Enix, Brendan Caldwell of Rock Paper Shotgun wrote a very unflattering article about the Final Fantasy XIV Fan Festival which insulted both the game and it?s fans. As this is becoming a common occurrence I decided that it?s about time I start applying my own special brand of opinion and critical commentary to the actions of Games Journalists as it is becoming more and more readily apparent that they hate Gamers.

Just one example from my news thread.

Also noticed you cant click on the folder icon to clear new reply's either.

Originally Posted by Myrthos
Can you give an example of characters that are broken?


Odd that it has this changed behavior. We moved the forum software from 3.8.10 to 3.8.11, so it is strange if this new behavior is caused by that. Perhaps it is some odd thing because of PHP 7 or there is a dependency somewhere I'm not aware of.

An example is this post from Silver: https://www.rpgwatch.com/forums/show…0&postcount=11

In your post that he quoted, although it looks like you used an apostrophe, what you really used was a unicode (UTF-8) right single quote mark. In hex, E2808B (’). When Silver quoted it, it just came out as a "?". Probably quoting of all non-ASCII characters is hosed?

In your post I'm quoting now, you just used a regular ASCII apostrophe, so it'll come out fine.

Originally Posted by Stingray
An example is this post from Silver: https://www.rpgwatch.com/forums/show…0&postcount=11

In your post that he quoted, although it looks like you used an apostrophe, what you really used was a unicode (UTF-8) right single quote mark. In hex, E2808B (?). When Silver quoted it, it just came out as a "?". Probably quoting of all non-ASCII characters is hosed?

In your post I'm quoting now, you just used a regular ASCII apostrophe, so it'll come out fine.

Quoting myself, watch the special quote I put in parentheses become a "?"

Maybe something is set to use ASCII or ISO-8859-1 or whatnot, instead of UTF-8, somewhere in the PHP, MySQL, or vBulletin config? Just a guess, I've never admin'ed vBulletin.

Thanks for the feedback. You are probably right in that it now is using a different default, probably in MySQL.

Now I just have to figure out what the heck I changed that would cause this.

its clearly something to do with the encoding - but even the source is showing the "?" and not the actual character. Changing your view from Western to Unicode does nothing at the client level.

The new PHP does make some sense - after all it is a Parser

But his guy thinks thinks its a MySQL problem as it defaults to latin1 instead of Unicode

https://stackoverflow.com/questions/…ows-as-diamond

Ope - now I see other strange characters in @Alrik Fassbauer's signature

??? Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius ??? and a lot of courage ??? to move in the opposite direction.??? (E.F.Schumacher, Economist, Source)

https://stackoverflow.com/questions/…e2%82%ac-issue

Originally Posted by BalusC
This will occur when the conversion between characters and bytes has taken place using the wrong charset. Computers handles data as bytes, but to represent the data in a sensible manner to humans, it has to be converted to characters (strings). This conversion takes place based on a charset of which there are many different ones.

In the particular ??? example, this is a typical CP1252 representation of the Unicode Character 'RIGHT SINQLE QUOTATION MARK' (U+2019) ? which was been read using UTF-8. In UTF-8, that character exist of the bytes 0xE2, 0x80 and 0x99. If you check the CP1252 codepage layout, then you'll see that those bytes represent exactly the characters ?, ? and ?.

This can be caused by the website not having read in the original source properly (it should have used CP1252 for this), or is displaying an UTF-8 page with the wrong charset=CP1252 attribute in Content-Type response header (or the attribute is missing; on Windows machines the default charset of CP1252 would be used then).

But you probably already knew that.

Maybe some of our experts know what has happened here:

On my iPad I see a square at the end of posting #7.
When I quote that post to see what is in there I see the following (pay attention to the end):

Originally Posted by Myrthos
We are moved now.

There isn't that much difference. On the new server we have a newer version of PHP, but we could have installed that on the old server as well.
The main reasons are that one of the licenses that comes with the server expires by the end of the year and this server is newer and also slightly cheaper.

But a move like this always offers an opportunity to make some additional changes, which takes more time than planned.[ IMG ] [ url ] https://www.rpgwatch.com/forums/data…BJRU5ErkJggg==[ /url ] [ /IMG ]

I have pointed Myrthos to the peculiar square in his posting, but he sees nothing mysterious in his posting on his PC.

When previewing this posting of mine I only saw the rectangular again, in the quote, so I added some spaces to show you all what is at the end of his posting, brackets and a link, when I look closer at his posting.

Is it my new eye or do some of you see what I am seeing?

Other characters:

Originally Posted by Darth Tagnan
Yes, he was the only man who wasn’t tempted and that’s why I like him

Originally Posted by Alrik Fassbauer
The kind of "play with clichés" !

Ok, don't think I have anymore examples, thanks guys

Encoding are fun; what happened to the good old days when 8 bits was 8 bits.

Pages seem to load faster for me now… been experiencing intermittent weird 3 to 5 second "hesitation" for the last two years or so. But now pages load quickly!

The site does seem a bit quicker for me also, and I've not experienced any errors as of yet. I'd call it an upgrade, thanks for the work.

Originally Posted by Carnifex
The site does seem a bit quicker for me also, and I've not experienced any errors as of yet. I'd call it an upgrade, thanks for the work.

Agreed

Except sign on bringing to main page rather than from page i signed on from.

Originally Posted by Alrik Fassbauer
In SWTOR lingua, "broken" can also mean "vastly overpowered"

Now we're talking. Vastly overpowered, mwahaha.

P.S. It feels so…different here. Like we've…crossed over somehow. This is…most strange.

Also having issue with typing message, getting logged out, but when I submit and relog in, it wipes my message.
More testing required for me on desktop. Mobile isnt functioning well for me at the moment.

I've tried various collations, but none of them gave the desired result. So it isn't collation. Currently we are on UTF8, but I tried various others.

The issue as you have noticed, is that on posting a character like ’ is showing correctly. (In the database the character is translated to a HTML entity)
Editing a post works properly as well.

When quoting this post the character is changed into a question mark upon posting.

Editing that quoted post and changing the question mark to the original character works.

So, what has changed related to this functionality:
1) We moved from version 3.10 to 3.11 of vBulletin
2) We moved from PHP 5.x to 7.1.x
3) We changed from MySQL to MariaDB
4) We changed from MyISAM tables to InnoDB

I've tried option 4) and changed the table holding the posts back to MyISAM. That didn't help, so it doesn't look like it solves the issue.
After all the tests, I don't think option 3) is the actual issue. Inserts and updates work well after all.

So that leaves option 1) and 2), which I will be investigating further.

On a side note, I fixed the issue Silver reported and multi-quotes appear not to be working.

When you quote a post, you still see the special character in what you've quoted while editing your new post…it doesn't get changed to a "?" until you actually hit post, which seems kind of weird, why would the piece inside quote tags get treated so differently from the rest of the post?

Oh I see the difference, when you're posting it initially you're typing/pasting in the actual UTF-8 character then posting, it gets converted to an HTML entity as you mentioned (&#xxxx;).

But when you quote someone who posted the UTF-8 character, it's actually fed back to you (inside the post editor) as the HTML entity, not converted back to the original UTF-8 character. Then when you post it, vBulletin I guess refuses to accept that HTML entity and turns it into a "?" instead?