Announcement

Collapse
No announcement yet.

Characters replaced with html encoding

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Wayne Luke
    replied
    You shouldn't have to rerun the utf8convert script.

    Leave a comment:


  • Grouper
    replied
    Originally posted by Wayne Luke View Post
    Please use tools.php to rebuild the serialized data in the database. These will be under the Datastore section of the menu.
    So run the datastore cache clear, then rerun the utf8convert.phar?

    Leave a comment:


  • Wayne Luke
    replied
    Please use tools.php to rebuild the serialized data in the database. These will be under the Datastore section of the menu.

    Leave a comment:


  • Grouper
    replied
    I did the migration but kept getting the same errors that were posted in the comments of empty array(), but it appears to be working.

    Leave a comment:


  • Wayne Luke
    replied
    No. vBulletin.com uses UTF8MB4 for the database and the UTF-8 HTML character set.

    We've been recommending UTF8 storage with vBulletin 5 for years now. We will explicitly set new installations to use the UTF8MB4 character set and the utf8mb4_general_ci collation. vBulletin 4 most likely used Latin1 because that was the default on new MySQL installations over a decade again. Back then, we just used the default database settings. Today's default installation on MySQL installations is utf8mb_900_ai_ci. Any utfmb4 collation will work though. However the collation determines how MySQL will sort searches.

    We do not change the character set and collation on upgrades because it can be an intensive operation and doing it improperly can cause irreparable data loss.

    Personally, I have been fighting for UTF-8 support for 15 years now and it is just recently been implemented in full for vBulletin 5.

    Leave a comment:


  • Grouper
    replied
    But doesn't vBulletin.com use latin1? I read in documentation vB recommends latin1 for most needed cases.

    Leave a comment:


  • Wayne Luke
    replied
    Originally posted by Grouper View Post
    some characters are replaced with HTML encoded versions, for example '<' is restored as '&lt;'. This happens when you restore a saved post or if you quote someone that has those characters in them.

    vb 5.5.6
    php 7.3
    db: mariadb 10.3.13
    testing....

    Leave a comment:


  • Wayne Luke
    replied
    This is by design and how vBulletin 3 and 4 worked to support UTF-8 and HTML characters. They are switched to HTML Entities so they can be stored properly in a Latin1 database. These characters should be avoided in titles.

    Converting your database to the UTF8MB4 character set will allow most of these characters to be stored properly. However since < and > are HTML tag delimiters they will continue to be converted to HTML Entities for security purposes. To convert your database to the UTF8MB4 character set, we provide a series of scripts in the /do_not_upload/dbtools directory. You can see how to use these here:
    https://forum.vbulletin.com/articles...database-tools

    Always make sure to create database backups.

    Leave a comment:


  • Grouper
    started a topic Characters replaced with html encoding

    Characters replaced with html encoding

    some characters are replaced with HTML encoded versions, for example '<' is restored as '&lt;'. This happens when you restore a saved post or if you quote someone that has those characters in them.

    vb 5.5.6
    php 7.3
    db: mariadb 10.3.13

Related Topics

Collapse

Working...
X