On Sun, 2003-02-02 at 22:12, Grant Jacobs wrote:
At 8:21 PM -0600 2/2/03, Richard D. Jackson wrote:
= Web Design in a Nutshell, in cause you didn't guess!
Oh you are talking about a book! No I did not know or guess. Which version is it. The local used book store may have a copy I can take a look at.
Mine is version the first edition (its now second Ed.) 4/99 print run.
Personally I'm going to leave texinfo out of mine, but that has to do with the fact that I want it to tie in with my little computer language project and I don't want that too dependent on other things as if it even come to completion it'll have to stand on its own two little feet ;-)
Then what are you going to use for hard copy production?
I probably should have written this better. There are two reason I guess.
I have my own data structure for playing around with text. Eventually, I'd like to try to use this for parsing, etc., but in the interim it seemed to me that doc notes would be simpler chance for me to try it out, etc. In that sense my aims are rather different.
The other is that I wanted to leave the dependencies downstream as much as possible and hence the exact mechanism of hard copy production downstream too. There has to be a mechanism obviously, but I don't want to tie it to one (eg. textinfo) right from the onset.
You might want to take a look at texinfo as it looks to have most everything you would need for both printed and hard copy. It is also very strait forward. It is also less combersome than XML and requires fewer tools to work with. My biggest dislike of texinfo has mostly been with the online viewers ( aka info ) versas with texinfo itself.
Its something of a contradiction in terms because obviously you have to tie it to _something_. Initially I want a basic key-word based internal text format that can be converted into whatever at a later stage. My initial thought is to use XML/HTML or the like. That said, I suppose it could just as easily be textinfo instead of XML... Precisely what it is could (and will) come later as its just a format after all. I want to focus on just getting a code-based layout going first. (In practice this will likely be a text rep. of the binary data structure, but that's another story; debugging, etc.) This project is a back burner as I have a hell of a lot of things to do to try make $$ -- I have a salary that comes out of the sky, so I can't afford to get _too_ sidetracked. That's why I thought it might be better for me to focus on the content of the docs and just making comments as far as the doc parser goes. [long-winded aren't I?!]
I think you're probably heading roughly the same way, albeit just using plain text, so perhaps this textinfo thing is misleading.
Out of curiosity: any thought to just allowing POD to work on GPC? Say by using //= as the lead-in (since // is a comment lead-in, and = is the POD lead-in. E.g. strip all ' '*// then hand off to POD). Spoils the fun though eh? Also I suppose dealing with the types/vars/etc might be clumsy. I suppose you could do the same with textinfo and //@. Sorry, I just had to throw this in ;-)
You could write a quick and dirty perl script to do that if it is what you want. But as you say dealing with the language ellements would be the tricky part.
Grant
Take a look at my reply to Frank in the next email... Richard