I am trying to write a xychange transformation that matches on an em-dash that is preceded or followed by a space and outputs an em-dash without the space. Transforming Data doc (Table 3-2) shows several ways to represent an em-dash but I can't get my match to recognize any of them. Seems simple enough but apparently too tough for me... Can anyone help?
And should I add the following to this story?
Do not use Xychange to transform XML files.
Why? Because Xychange does not understand unicode and things like utf8 encoding.
Go the extra mile and use a perl transform.
Yes it is a bit harder at first but it will be a lot better and simpler.
Why? Because Perl is fully unicode aware and does understand utf8 encoding.
Like this, Bart?
$str =~ s/\xe2\x80\x94/ ...
Jay, you got it!
So if you want to match the string:
EMDASH<focus_end focus_idref="*" focus_number="*"/><focus_start itemref="*" focus_id="*" focus_number="*"/>
You simply do:
$str =~ s/\xe2\x80\x94\x3c\x66\x6f\x63\x75\x73\x5f\x65\x6e\x64\x20\x66\x6f\x63\x75\x73\x5f\x69\x64\x72\x65\x66\x3D\x22\x2A\x22\x20\x66\x6\xF6\x37\x57\x35F\x6E\x75\x6D\x62\x65\x72\x3D\x22\x2A\x22\x2F\x3E\x3C\x66\x6F\x63\x75\x73\x5F\x73\x74\x61\x72\x74\x20\x69\x74\x65\x6D\x72\x65\x66\x3D\x22\x2A\x22\x20\x66\x6F\x63\x75\x73\x5F\x69\x64\x3D\x22\x2A\x22\x20\x66\x6F\x63\x75\x73\x5F\x6E\x75\x6D\x62\x65\x72\x3D\x22\x2A\x22\x2F\x3E/....
It can not get much simpler than that!
Jay, you got it!
So if you want to match the string:
EMDASH<focus_end focus_idref="*" focus_number="*"/><focus_start itemref="*" focus_id="*" focus_number="*"/>
You simply do:
$str =~ s/\xe2\x80\x94\x3c\x66\x6f\x63\x75\x73\x5f\x65\x6e\x64\x20\x66\x6f\x63\x75\x73\x5f\x69\x64\x72\x65\x66\x3D\x22\x2A\x22\x20\x66\x6\xF6\x37\x57\x35F\x6E\x75\x6D\x62\x65\x72\x3D\x22\x2A\x22\x2F\x3E\x3C\x66\x6F\x63\x75\x73\x5F\x73\x74\x61\x72\x74\x20\x69\x74\x65\x6D\x72\x65\x66\x3D\x22\x2A\x22\x20\x66\x6F\x63\x75\x73\x5F\x69\x64\x3D\x22\x2A\x22\x20\x66\x6F\x63\x75\x73\x5F\x6E\x75\x6D\x62\x65\x72\x3D\x22\x2A\x22\x2F\x3E/....
It can not get much simpler than that!
But remember, this ONLY works with UTF-8! With Tridion Docs, we have to contend with UTF-16 a lot. But then there's JPCIN!
And just to be totally prepared, enter the strings in both utf-16LE and in utf-16BE format!
Better save then sorry...
Translate
