spellerpages / coldfusion integration broken for non en_US languages

[tomalak]

spellerpages / coldfusion integration broken for non en_US languages

Server based spell checking with aspell fails at extended characters (German umlauts, for example). As a result, HTML source code and words get mixed up in the spellerpages dialog box because <cfexecute> seems to have a hard time with UTF-8 encoded characters on std_out.

There is a lack of consistency in treating all the data as UTF-8 during it's way on the server. I modified the source of spellchecker.cfm to make it work (see attachment).

These are the major changes I did to the page logic:

• changed from hard-coded temp directory to GetTempDirectory()
• aspell output is piped into a file rather than read by <cfexecute>
• output file is read with <cffile charset="utf-8">
• content-type of page changed from "text/html" to "text/html; charset=utf-8"

I did some general code clean-up as I went, all those changes are equivalent to the original meaning. The <cfloop list="#food#"...> part is rewritten to be easier to read and to fix a "index out of range" bug in the <cfset wrdList = mid(list,(LastIndexOf(...> part.

[tomalak]

modified spellchecker.cfm

[wwalc]

Fixed with [662]