[LUDOC-226] charset declared in in html manual does not match content

• Status: Open
• Project: Lustre Documentation
• Type: Bug
• Priority: Minor
• Reporter: Christopher Morrone
• Assignee: Lustre Manual Triage
• Resolution: Unresolved
• Severity: 3

Description

LUDOC-7 sort of covered this problem, but I'm not sure the conclusions in that ticket were correct.

The character that is showing up most frequently in the html version of the document is 0xc2a0. This is the UTF "No-break space" character. I am guessing that it is a special space character that tells the renderer not to break the line at that space.

Note that this is clearly a multi-byte UTF character. The html document, however, clearly declares the document as "charset=ISO-8859-1". This is wrong. If you manually change that to "charset=UTF-8" the problem goes away.

The xhtml version of the document correctly declares the file as encoding="UTF-8", which is why there are no web browser rendering problems with the xhtml.

So the problem is entirely in the document, and has nothing to do with jenkins or the web server.

I haven't looked into the solution, but at least now the problem is clearly understood.