[TUHS] Introduction to {t,r,g}roff & co...

Grant Taylor gtaylor at tnetconsulting.net
Tue Sep 5 12:42:00 AEST 2017


On 09/04/2017 08:19 PM, Larry McVoy wrote:
> Hi Grant, 

Hi Larry,

> Somehow I missed your initial email, time to check my spam filters
> I guess.

Ah, spam filters, the never ending battle.  I've been doing that for ...
much longer than I care to admit.  (Double digit years.)

> All this rambling below boils down to one thing: if you need help
> with roff, I'm your guy.  Tell me what you want to do and I can
> probably come up with some example stuff that you can play with.

Thank you Larry, your offer is very much appreciated.

I don't have a project that I'm working on per say.  Rather I've always
respected *roff and the recent threads on the TUHS list have stirred a
long standing interest.

> I would suggest groff as a good start.  James did a great job.  There is
> the heirloom stuff, I've played with it, my take is that it is like
> Keith's nvi stuff, true to the origin but not useful because the world
> has moved on.  Groff is my goto roff tool.

I'm okay learning some history while learning new things.  What I don't
learn initially, I like to circle back and learn more.  -  Sort of like
why I subscribe and participate in TUHS.

> Anyhoo, I *love* troff and the preprocessors, I can draw pictures in my
> head and then draw them in pic (I've done a lot of pic, got James to put
> an extension in gnu pic so that you could iterate through the N things
> you just drew, I can show you an exampe).

I'd be interested in seeing an example, if it's handy.

I was going through "troff and its companion programs" (troff and its
companion programs) briefly at work and found it to be fairly easy to
follow to see some initial results.

> I _think_ I have the sources to the troff docs, I feel like I did a
> project at one point to modernize how they looked.  

I have a dead tree copy of "UNIX Text Processing" somewhere and have
thumbed through it multiple times.

I was pleasantly surprised to see m4 in there, something I occasionally
choose to use for new projects.

> So you've gotten some good suggestions, I'm a fan of the original
> docs though.  I still have the stack of docs that I bought at the 
> UW Madison computing center - n/troff doc, pic, eqn, tbl.  Then 
> there were various others, like grap, chem, etc.

I've already started lifting an eyebrow at things like the fact that
chem is an awk script.  -  I've done more in awk than some, but am
impressed, and want to learn more.  -  What it does, how it does it, and
how I might be able to apply that methodology to other things.

> I love all that stuff because it was designed at a time where you did
> your markup and you sent it to the lab where the printer was and you
> got it the next day or so.  There were no bitmapped displays, all this
> stuff was done on 80x24 CRTs.  So the markup language, the pic stuff,
> the eqn stuff, it all had to be something that you could see in your
> head and put down in text.

I'm cool with that.

One of the current questions is how, and why, people chose different
macro packages.

I do see why someone would use (or write / modify) macros to do some
basic things in *roff.  -  I suspect it's similar to what I've hard of
people do in assembly programming.  Namely write in the macro language
that is then expanded to the lower layer *roff.

My knee jerk reaction for expanding short text (macros) into longer text
with logic would be m4.  But I want to learn the *roff world before I
get off course.

> That fits really well with how I think, I love the roff ecosystem to
> this day (and I've done conference proceedings in roff and in LaTex,
> I much prefer roff and the funny thing is when I show LaTex people
> roff they go, wow, simple).

:-)



-- 
Grant. . . .
unix || die

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 3717 bytes
Desc: S/MIME Cryptographic Signature
URL: <http://minnie.tuhs.org/pipermail/tuhs/attachments/20170904/1305d6bf/attachment.bin>


More information about the TUHS mailing list