Re: [Groff] Preprocessor

[Tomohiro KUBOTA]

Hi,

At Mon, 06 Nov 2000 16:30:56 +0100 (CET),
Werner LEMBERG <address@hidden> wrote:

> Very nice!  I'll put it into the CVS (please send me the code
> privately again -- I believe that some tabs have been incorrectly
> converted to spaces, causing wrong indentation).  Which name of the
> preprocessor do you suggest?

The name space for command names is a limited and precious resource
and we have to be careful to name it.  Though I once suggested
'gpreconv' in the previous discussion, I don't know about traditional
or tacit principles on naming a command, if there are.

I think 'gpreconv' may be too general, though this preprocessor is
used only for Groff system.  However, other executables of Groff
system have long history and general names.

BTW, how will you join my preprocessor to the Groff source tree?
My interest is that 'I18N' directive should be determined by the
'configure' script.  And more, it needs some mechanism to have
a translation table between MIME encoding names and OS-specific
encoding names.  I have no idea how to implement that.  (Thus the
current version have an empty table for this translation.)


> Please note that there is more than one EBCDIC encoding.  I suggest
> that you stay with `cp1047' as the default.

Ok.


> Can you add a small man page?

I see. I will write.


> Thanks a lot for your contribution!  Unfortunately, I still don't have
> enough time to start with Unicode support in groff -- I first have to
> finalize FreeType 2.0...

Oh, you are so busy.
TrueType is also an important topic for Japanese unices users.

Then I will write a postprosessor.  It may share the source code (and
binary) with the preprocessor, i.e., check argv[0] and change the
behavior.  (I found that we have to determine the names:-)

---
Tomohiro KUBOTA <address@hidden>
http://surfchem0.riken.go.jp/~kubota/