RE: [nmrg] web page for uml diagrams

["Sharon Chisholm" <[e-mail]>]

hi

I've managed to do PDF, but it is 36 pages since the diagrams are very
zoomed in. I similarly created a PDF from the MPLS dia on the website and it
was an insane number of pages and therefore not terribly useful.

I've attached the PDF for the dia Bert sent. Does anyone know how to make it
print (the method I use to create PDF) so you can fit more diagram to a
page?

Sharon

-----Original Message-----
From: Wijnen, Bert (Bert) [mailto:[e-mail]]
Sent: Saturday, July 26, 2003 6:41 PM
To: Frank Strauss; Marcus Brunner
Cc: [e-mail]; [e-mail]
Subject: RE: [nmrg] web page for uml diagrams


Mmm... the complete set is kind of undoable.
So I took just the LDP MIB.
I have a .dia file that presents a reasonable picture.
I do not have a proper tool to create a PDF file. 
Can one of you do so?

Attached is the .dia file (strange, when it generated the
file when I saved it, it seems to put virtually everything on one
long line. Oj well).

The associations in the MIB itself look awfully complex to me.
Other comments?

Thanks,
Bert 

> -----Original Message-----
> From: Frank Strauss [mailto:[e-mail]]
> Sent: woensdag 23 juli 2003 18:12
> To: Marcus Brunner
> Cc: [e-mail]; [e-mail]
> Subject: Re: [nmrg] web page for uml diagrams
> 
> 
> Hi!
> 
> Marcus> I tried to do it for the MPLS MIBs, but I miserably 
> failed even
> Marcus> parsing the MIBs in the latest drafts. But it might 
> be myself not
> Marcus> using the tools correctly.
>  
> >> Could you please point me (us) to the MIBs (or the I-Ds) you've fed
> >> to smidump? There are 36 documents in 
> internet-drafts/*mpls*mib*, and
> >> I don't want to check all of them. :-))
> 
> Marcus> draft-ietf-mpls-ftn-mib-07.txt
> Marcus> draft-ietf-mpls-ldp-mib-12.txt
> Marcus> draft-ietf-mpls-lsr-mib-11.txt
> Marcus> draft-ietf-mpls-tc-mib-08.txt
> Marcus> draft-ietf-mpls-te-mib-11.txt
> 
> Thanks. Here is a short description of what I did to check 
> the modules,
> to "repair" them so that they could be further processed, and 
> to generate
> a diagram. Maybe this helps others to do similar things...
> 
> 1. Get the documents.
> 
> 2. Extract the modules using smistrip. (The latest version from CVS of
>    smistrip now also accepts .gz files and is able to read documents
>    from a given directory):
> 
>         smistrip -i /ftp/docs/internet-drafts \
>                 draft-ietf-mpls-ftn-mib-07.txt \
>                 draft-ietf-mpls-ldp-mib-12.txt \
>                 draft-ietf-mpls-lsr-mib-11.txt \
>                 draft-ietf-mpls-tc-mib-08.txt \
>                 draft-ietf-mpls-te-mib-11.txt
> 
>    Note that documents that are formatted in an "unexpected" form, may
>    lead to warnings or even broken MIB modules, because smistrip has
>    to make some vague assumptions on how to cut off page footers and
>    headers.
> 
> 3. Check the extracted modules with smilint. If not otherwise set, the
>    local directory might have to be added to the paths in 
> which smilint
>    will search for modules.
> 
>         SMIPATH=.: smilint -i namelength-32 MPLS-*
> 
>    In this case I wanted to ignore all warnings about identifiers that
>    are longer than 32 chars. The errors that remain complain 
> about "XXX"
>    placeholders for a number of registrations.
> 
> 4. Fix errors, if required. In this case nothing has to be done to
>    produce the dia diagrams.
> 
> 5. Use smidump to generate whatever I want, e.g. metrics, 
> trees, diagrams:
> 
>         SMIPATH=.: smidump -u -f metrics MPLS-* | less
> 
>         SMIPATH=.: smidump -u -f tree MPLS-* | less
> 
>         SMIPATH=.: smidump -u -f cm MPLS-* > mpls-model.dia
> 
>    The -u option is used to get a "united" representation of all
>    input modules, instead of separated outputs per module.
> 
> 6. Use dia to drag items around to get a better layout and finally
>    print the diagram to a PDF file. I have not yet done this. It
>    seems to be quite some work in this case. :-) The non-beautified
>    diagram is at 
http://www.ibr.cs.tu-bs.de/users/strauss/mpls-model.dia.

 -frank
-- 
!! This message is brought to you via the `nmrg' mailing list.
!! Please do not reply to this message to unsubscribe. To unsubscribe or
adjust
!! your settings, send a mail message to <[e-mail]>
!! or look at https://www.ibr.cs.tu-bs.de/mailman/listinfo/nmrg.