controlling character entity transformations in Jade

[Robin Stephenson <robin.stephenson@xxxxxxxxxxxxxxx>]

I'm transforming SGML from one DTD to another with Jade.  I was
pleased to see Jade putting in literal equivalents for things like
é and so on, but it seems to be transforming some things to
Unicode, eg ` ' is transformed to `&8201;'.
  The thing (one of many things, in fact) that's confusing me is
I'm including the (unedited) ISOpub entities in the output DTD, so I
thought that ` ' would be mapped to `[thinsp]':

<!ENTITY thinsp SDATA "[thinsp]"--=thin space (1/6-em)-->

  How can I tell Jade not to do this?  After a bit of delving about in
AltaVista I came up with this command line, but it's still doing the
aforementioned mapping to a Unicode character entity:

SP_ENCODING=is8859-1 jade -d blacksci.dsl -t sgml sgml/Cei00401.sgm

-- 
 Robin Stephenson
 Auto Stop


 DSSSList info and archive:  http://www.mulberrytech.com/dsssl/dssslist