Searching a file for an unrecognized character

[Eric Van Tassel 101233.342@xxxxxxxx]

Hope this isn't a tangent to the thread of 20-9 or 21-9, which I deleted
before writing this:

Whenever I want to do a search for an unrecognizable character in a file
that someone has sent me, I put the cursor on the character and run the
following program (from the kbd file):

{..} = function all, ≪...≫ = guillemets

(1)    {DF}{CR}{DF}≪SV999≫{ES}{BC}sea |≪GT999≫|

If I know that the unrecognized character is may e-mail program's
substitution for a known character (e.g. when I get Greek theta and know
that it's supposed to be e+acute), I run

(2)    {DF}{CR}{DF}≪SV999≫{ES}{SI}{BC}cva
|≪GT999≫||{CL}{GT}{CL}{CL}{GH}

- which forces typing into "insert" mode and sends the cursor left to input
the "solution" to the problem.

As it happens, the search in (1) lands only on the next appearance of the
unrecognized character _after_ the one where I had placed the cursor. In
(2), the last four func.calls {GT}{CL}{CL}{GH} put the cursor down into the
text, move it back to before the unrecognized character, and then returns
the cursor to the CM line, so (2) does offer first to convert the character
at the place where I'd positioned the cursor. I could also have had those
four func.calls at the beginning of (1), but I don't find it as useful that
way.

Neither of these ends with {XC} to execute, because I prefer to look at the
CM line first and see if it's OK.

Does this work only in Xy4? or can it be adapted for use in Xy3+?

Cheers
Eric Van Tassel