• We have a handful of blogs using the Subscribe2HTML plugin (v 10.9). On one blog (only), emailed post content will occasionally get mangled with odd character substitutions that appear to be an encoding issue.

    Random characters get replaced with = signs; others with “?” in a black diamond. Here are some excerpts:
    helped make last week=E2��s dance a huge success
    5:00 �=80� 8:00: Girls’ Basketball

    This does not appear in the WordPress post — only in the email, but not with every email!

    It can happen when text is entered directly, and when copied & pasted from Word or an email.

    Please help!

    https://wordpress.org/plugins/subscribe2/

Viewing 7 replies - 1 through 7 (of 7 total)
  • @albcowebmgr

    This is probably due to differences in encoding in the email and database. Ideally both should be using the same and ideally UTF. You can check the email by viewing the source of the email for content type and use PHPMyAdmin to check the database tables and rows.

    Thread Starter AlbCoWebMgr

    (@albcowebmgr)

    Thanks for the rapid response, mattyrob!

    Both the email sent by Subscribe2HTML and the database are using utf-8. The database is UTF8_general_ci, specifically.

    The odd thing is that the Posts as displayed by WordPress do not have the issue. Only the email version.

    All the post emails are authored by one person, using Internet Explorer. Some are messed up, others are fine.

    Other suggestions, please!

    @albcowebmgr

    Have you checked the content containing rows also? See here for more info:

    http://codex.wordpress.org/Converting_Database_Character_Sets

    Thread Starter AlbCoWebMgr

    (@albcowebmgr)

    The blogs were all created with UTF8 as the default collation.

    @albcowebmgr

    Have you checked the collation within each table for each row? For example the posts->post_content row?

    Does the email show like this in all email clients?

    On the affected email what is the ‘Content-Transfer-Encoding’ set as?

    Thread Starter AlbCoWebMgr

    (@albcowebmgr)

    posts > post_content collation is also utf8_general_ci. Content of the email itself is UTF. Client doesn’t make any difference. Same in Apple Mail, MSOutlook, smartphone, & webaccess.

    In any case, again, the posts themselves in WordPRess don’t display this problem — only some of the HTML email versions.

    The emails are not stored in the DB as far as I can tell. Where are they processed? If in a temp folder on the server, could something in the SERVER’s file structure be corrupting the content?

    I’m thinking it’s not encoding per se, but some corruption in the plug-in scripts. I just deleted and reinstalled the plug-in for the second time. I don’t expect it will help, but we’re pretty desperate.

    If another email gets corrupted, my next step will be to create a new blog site for the School and start over with fresh data.

    I’d welcome any further suggestions.

    Thanks.

    @albcowebmgr

    I’ve had a check through the code again and I can’t see anywhere in Subscribe2 where the email content has and encoding changes made.

    Are the emails coming from your own server or via ReadyGraph?

Viewing 7 replies - 1 through 7 (of 7 total)
  • The topic ‘Subscribe2HTML – Intermittent encoding problems’ is closed to new replies.