[ d20statblock.org ] a grammar

[Luke]

I think codifying things to the point of having them in some kind of BNF/EBNF/DTD is a great idea.

I think there are some important areas not yet covered by the style guide, and it's mostly because of the informality of the approach.

For example:
- How are templates supposed to be displayed (eg. Half-Dragon)?
- How are special powers supposed to be described in magical weapons or armor (eg "+2 longsword (defending)")?

If things can be better codified, and that becomes standard, it also helps "corral" d20 publishers creating new material. This makes more likely that new material will fit a well defined "database format".

Coming from another angle, For compl

Regards,

 

[CRGreathouse]

I think codifying things to the point of having them in some kind of BNF/EBNF/DTD is a great idea.

Yeah, so do I. I was thinking of making a DTD, but never actualy did anything. Jmettraux's been doing everything - thanks a ton, jmettraux!

- How are templates supposed to be displayed (eg. Half-Dragon)?
- How are special powers supposed to be described in magical weapons or armor (eg "+2 longsword (defending)")?

I'll certainly work on this, but not until the basics are covered. the nice thing about ENBF is that it's easy to work with - anyone can tweak them once they're set up.

If things can be better codified, and that becomes standard, it also helps "corral" d20 publishers creating new material. This makes more likely that new material will fit a well defined "database format".

That's my hope!

 

[MJEggertson]

I'm very intersted in all this, and it's actually on my list of things I want to do for my project. Although, it's a more distant goal. My current understanding of xml (etc...) is pretty minimal, so I'm trying to implement other things first.

I've been looking for a formal standard to emerge, and my lack of kowledge of these markups (and limited time to learn) has kept me at lurker status in all these discussions.

/me watches with confused interest.

But I do think an agreed standard would go a long way to making everyone's software much more accepted. Most people, it seems, like every piece of software for a few reasons, and don't use any one product exclusively. Transparent data-porting I think would be a great goal for everyone to work on.

 

[jmettraux]

...Bison...

I wasn't having any luck downloading this. Is it just me, or is the link dead?

Edit: The download link, not the link you provided.

Charles,

Go to http://www.cygwin.com/ and install cygwin, you'll have all the unix tools on your windows desktop. If you need assistance for the installation, just send me a private email.

But, well... I don't think you will need bison right away, we want to turn your standard vision into a "E"? "BNF" document at first, any parser will come later. So your referee skills are required now


- The grammar -

Latest version :
http://cvs.sourceforge.net/cgi-bin/...?rev=1.5&content-type=text/vnd.viewcvs-markup
Diff :
http://cvs.sourceforge.net/cgi-bin/...port_samples/standard.ebnf.diff?r1=1.3&r2=1.5

 

[CRGreathouse]

But, well... I don't think you will need bison right away, we want to turn your standard vision into a "E"? "BNF" document at first, any parser will come later. So your referee skills are required now

I think that testing it with a parser generator could help us get there, that's all.

 

[jmettraux]

I think that testing it with a parser generator could help us get there, that's all.

I've made a few changes (especially to the size and type section);

OK ! The sooner we get to a format used by a parser generator, the less work we will have.

I will commit your changes to the grammar monday morning. Well done !

What would you think of :

languages: "Languages spoken: " language (", " language)*
language: String

 

[CRGreathouse]

What would you think of : [language section omitted]

It's an oft-requested feature. Actually, now that I think about it, I don't know why I didn't just put it in from the start - normally I don't weant to add more options than needed, but this one would be easy enough for a person or program to ignore since it's broken out.

 

[jmettraux]

New version of the grammar :
http://cvs.sourceforge.net/cgi-bin/...rev=HEAD&content-type=text/vnd.viewcvs-markup
diff :
http://cvs.sourceforge.net/cgi-bin/...port_samples/standard.ebnf.diff?r1=1.5&r2=1.7

Thanks Charles ! Your refinements are smart.


Two propositions :

1) I'm sure you already thought about it, but shouldn't this thread appear somehow in the d20 publishers forum ?

2) We could run a sourceforge project just for d20statblock.org's documents, so we could more easily collaborate. (Though the grammar itself will reach a 'stable' state sooner or later)

 

[CRGreathouse]

1) I'm sure you already thought about it, but shouldn't this thread appear somehow in the d20 publishers forum ?

2) We could run a sourceforge project just for d20statblock.org's documents, so we could more easily collaborate. (Though the grammar itself will reach a 'stable' state sooner or later)

1. I'd prefer to wait until it's a little more mature. Right now we're just working out the major bugs.
2. I've never worked with sourceforge. What are the advantages? What things would be moved there? In general, I have no objection - but I'd like to know more.

 

[jmettraux]

1. I'd prefer to wait until it's a little more mature. Right now we're just working out the major bugs.
2. I've never worked with sourceforge. What are the advantages? What things would be moved there? In general, I have no objection - but I'd like to know more.

1)

2)

You would be able to commit your changes directly.

We should move there the grammar itself and any other document related. A lot of examples should be stored there too.

The source code for a validation parser (a parser that just says 'yes this stat block is standard') could be a nice addon.
(We could run it online somewhere).

You could also store there your web site like I do for dgsh :
http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/dgsh/dgsh/doc/