refdb-users — List for RefDB users

Re: [Refdb-users] RisX import

[Bruce D'A. <bd...@fa...>]

On Sunday, August 24, 2003, at 12:27  PM, Markus Hoenicka wrote:

> And if Trang allows the conversion of DTD->Schema as
> well, then why should I bother now to write in Schema as long as my
> editor doesn't support it?

Perhaps cleaner, more future-proof, design?

Bruce

Re: [Refdb-users] RisX import

[Markus H. <mar...@mh...>]

Bruce D'Arcus writes:
 > This is a bit of a red herring though.  It is true that there is little 
 > support for RNG yet in XML editors, but this is not much of an issue 
 > when you consider that Trang allows for reliable conversion to DTDs and 
 > XML Schema, both of which are supported in editors.  And editor support 
 > will come in time.
 > 
 > The reason people are migrating development to RNG is precisely this 
 > power and flexibility.  DocBook in the future will be written in RNG, 
 > but this certainly does not mean they won't be releasing DTD versions 
 > of it.  Likewise with Open Office. Word supports import of arbitrary 
 > XML, but only if defined against a Schema; likewise almost surely to be 
 > the case with the next major version of OOo.
 > 

That's another red herring ;-). If you have to rely on the conversion
of your Schema to a DTD in order to use it for coding documents
anyway, you can't use any of the cool features that provide "power and
flexibility". And if Trang allows the conversion of DTD->Schema as
well, then why should I bother now to write in Schema as long as my
editor doesn't support it? I can convert it anytime as soon as it
does.

regards,
Markus

-- 
Markus Hoenicka
mar...@ca...
(Spam-protected email: replace the quadrupeds with "mhoenicka")
http://www.mhoenicka.de

[Refdb-users] ris -- is this ok?

[Markus H. <mar...@mh...>]

Bruce D'Arcus writes:
 > 
 > Am cleaning up the big ris db for import.  Is this OK?
 > 
 > PY  - 1975///September/October
 > 
 > Note slash in the months.
 > 

This is fine with RefDB. It checks only the first three slashes to
chunk the string, the remainder is imported as is.

regards,
Markus

-- 
Markus Hoenicka
mar...@ca...
(Spam-protected email: replace the quadrupeds with "mhoenicka")
http://www.mhoenicka.de

Re: [Refdb-users] Issues and Observations

[Bruce D'A. <bd...@fa...>]

	
On Thursday, August 21, 2003, at 07:12  AM, Justus H. Piater wrote:

> Or is there another/better way to create formatted, MS-Word-editable
> output from DocBook XML? (This is against my principles, but reality
> sometimes demands a dose of pragmatism...)

As a matter of fact, there is: HTML.

You can create HTML that uses CSS to force Word to apply it's internal 
styles to HTML code tagged with class attributes.  I've even figured 
out how to modify the DocBook stylesheets to produce code that Word 
will read as footnotes.  This is a far better solution IMHO than RTF.

If anyone is interested in this, perhaps we can package this up into 
something more than the hack it currently is?

Bruce

[Refdb-users] Issues and Observations

[<Jus...@UL...>]

Hi,

I've just started using RefDB, great stuff! Nevertheless, here are
some observations due to bugs, features, or my ignorance. Details of
my setup follow further down.

1. Multi-reference citations (as in <citation
   role=3D"REFDB">Kruger-AISB2003;Kruger-CNS2002-13</citation>); both
   journal styles, pdf, rtf, and html output:

   (a) All but the first two-reference citation miss the enclosing
       parentheses in the generated output (). This must be a bug.

   (b) For such multi-reference citations, only a single, common
       hyperlink is created that refers to the first reference. Is
       this a bug or a feature? If the latter, what's the rationale?

   (c) For single-reference citations, openjade complains that "xref
       to bibliomset unsupported", and produces empty brackets without
       links instead of hyperlinked text.  Multi-reference citations
       do work because they do not point to bibliomsets but to -
       whatever, bibliomixed, I think. This is weird because I've not
       found any evidence out there that anyone else in the universe
       has observed this. Note that here I use openjade (SGML) to
       process DocBook XML, which may be a rare thing to do.

2. Miscellaneous formatting problems, again involving both journal
   styles:

   (a) XSL-FO processing: I still have the "xsl:call-template :
       template component.separator not found" error reported by Marc
       Baaden on 2003-04-17 17:54.

   (b) XSL-FO processing: The heading "Reference List" is followed by
       the normal-sized text "Reference List", also reported by Marc
       Baaden in the same message.

   (c) All output formats: The rendering of the reference text of the
       AISB2003 reference (see my other posting for details) is weird,
       probably due to the missing page numbers. This may not
       literally be a bug, but I consider it a serious shortcoming
       because (at least in my field) it is extremely common to cite
       conference papers in proceedings without page numbers, and it
       is often very difficult to get these without consulting the
       physical volume.

   (d) All output formats: If no date is given, "nd" is inserted
       automatically. This makes it impossible to properly render
       certain UNPB references, e.g. submitted articles.

   (e) All output formats: I do not understand how THES references are
       supposed to work. The output contains superflouous dots. Is the
       degree-granting institution to be coded as the publisher? In
       that case, the output order would be
       unconventional. Eur.J.Pharmacol. adds a trailing "(nd)", even
       though the year does show up before.

   (f) In fact, Eur.J.Pharmacol. adds a trailing "(nd)" for (at least)
       CONF, UNPB, THES.

   (g) openjade-generated output: UTF-8-coded non-ASCII characters are
       not rendered correctly. This may not a RefDB problem however.

   (h) XSLT-generated HTML output: All citation keys and text appear
       in the generated HTML code. This leads to citations like
       (Eur.J.Pharmacol.):

       [IDClifton-CD1993-64-1099] (Clifton, Muir, Ashmead, and
       Clarkson, 1993) (Clifton, Muir, Ashmead, and Clarkson, 1993)
       Clifton, Muir, Ashmead, and Clarkson Clifton, Muir, Ashmead,
       and Clarkson (1993) Clifton R.K., D.W. Muir, D.H. Ashmead, and
       M.G. Clarkson, 1993, "Is Visually Guided Reaching in Early
       Infancy a Myth?", Child Dev. 64, 1099.

       J.Biol.Chem. has a similar problem.
       openjade-generated HTML does not have this problem.

   (i) XSLT-generated HTML output: Multi-reference citations generate
       empty reference entries at the end of the reference list in the
       HTML output, rendered as follows (for two two-reference
       citations, Eur.J.Pharmacol.):

       (Clifton, Muir, Ashmead, and Clarkson, 1993; McCarty et al. ,
       2001)Kr=FCger, Lappe, and W=F6rg=F6tter, 2003; Kr=FCger and W=F6rg=
=F6tter,
       2002

       J.Biol.Chem. has a similar problem.
       openjade-generated HTML does not have this problem.

3. Other remark:

   (a) In ris.el, ris-font-lock-keywords contains UNPG instead of UNPB
       which causes syntax highlighting problems. Serious bug!!!!!!
       (just kidding)


Here are the details of my setup:

- RefDB 0.9.3

- Document source: -//OASIS//DTD DocBook XML V4.2//EN

- To generate XML-FO:
  - Norman Walsh's DocBook XSL stylesheets, version 1.61.3.
  - xsltproc: Using libxml 20507, libxslt 10030 and libexslt 720
    xsltproc was compiled against libxml 20507, libxslt 10030 and libexslt =
720
    libxslt 10030 was compiled against libxml 20507
    libexslt 720 was compiled against libxml 20507

- to render pdf from XML-FO:
  - PassiveTeX, latest version (I believe)

- to render HTML:
  - either via openjade 1.3.2
  - or via xsltproc

- to render RTF:
  - via openjade 1.3.2

To make openjade work with XML, I use xml.dcl instead of the (SGML)
docbook.dcl. I (naturally) use the XML bibliography file, so I run
runbib twice; once with -t db31 to generate the DSSSL style file, and
then with -t db31x to generate the XML bibliography file. Weird, eh?
But it works, except for the quirks mentioned above (that may or may
not have to do with this procedure).

I would like to drop openjade, but I cannot because I never suceeded
at making JFOR work. It seems it cannot handle my (recent) DocBook
setup. If someone has sucessfully used JFOR with a DocBook XML setup
similar to mine, please contact me!

Or is there another/better way to create formatted, MS-Word-editable
output from DocBook XML? (This is against my principles, but reality
sometimes demands a dose of pragmatism...)

An off-topic, but related question: To write an XML file that is
compatible with both openjade and xsltproc, I have to specify a DTD
system identifier in the form of a file name. Correct? I'd rather
specify a URL, but openjade complains.


Congratulations to all those who have read all the way down to here! I
hope I have provided and will receive some useful feedback.

Cheers,
Justus

--=20
Justus H. Piater, Ph.D.         http://www.montefiore.ulg.ac.be/~piater
Institut Montefiore, B28        Phone: +32-4-366-2279
Universit=E9 de Li=E8ge, Belgium    Fax:   +32-4-366-2620

Re: [Refdb-users] How to create a scientific report with refdb/docbook ?

[Markus H. <mar...@mh...>]

Hi,

thanks for the examples. Frankly, I believe that RefDB can't do this
out of the box. What you have in mind seems to require quite heavy
stylesheet tweaking. E.g. there is currently no provision for
subdividing the bibliographic table.

regards,
Markus

Marc Baaden writes:
 > 
 > Hi,
 > 
 > I hope the following mail is not too long (but it's only text), and
 > it is just meant to give some examples of possible formatted paragraphs
 > that I would (ideally) like to generate from refdb entries.
 > 

-- 
Markus Hoenicka
mar...@ca...
(Spam-protected email: replace the quadrupeds with "mhoenicka")
http://www.mhoenicka.de

Re: [Refdb-users] RisX import

[Bruce D'A. <bd...@fa...>]

On Thursday, August 21, 2003, at 01:47  AM, Couwberghs Tom wrote:

> If you plan to develop a different DTD for each type of reference, what
> are you going to do with query's that return multiple types of
> references in risx?

I was wondering this myself.  I have a hunch it might be better to have 
a modular design that is built for extension.  I sent this to Markus 
already, but what I was thinking was top-level elements like titles, 
names, date, in (with type attributes that distinguish between 
monographs and serials), etc.

They'd then be plugged together something like (this using RNG because 
it's easier for me than DTD syntax):

Article = element article {
    	Name*, Titles, Date, In.Periodical
    	}

Chapter = element chapter {
    	Name*, Titles, Date, In.Monograph
    	}

Book = element book {
    	Name*, Title, Date, Publisher?
    	)

Note: the above isn't quite right because I think stuff like book and 
article should be defined in an attribute within an entry element, but 
I'm not sure how to do this :-)

Finally, you'd explicitly build in the extension, which would be 
defined in an external file with specifications defined however you 
want.

So, in other words, build this both for the immediate task -- to make 
entering risx data simpler -- but also for the longer term one: MODS.  
The translation logic is just built into the xslt.

Sample record, BTW, could be:

<entry type="article" id="Doe1999a">
    <name role="author">
       <givenName>John</givenName>
       <familyName>Doe</familyName>
    </name>
    <titles>
       <title>A Title</title>
       <subtitle>A Subtitle</subtitle>
    </titles>
    <date year="1999" month="6"/>
    <in type="periodical" genre="journal">
       <titles>
          <title>A journal</title>
       </titles>
       <part>
          <volume>3</volume>
          <issue>4</issue>
          <pages>
             <start>23</start>
             <end>54</end>
          </pages>
       </part>
    </in>
</entry>

Bruce

RE: [Refdb-users] RisX import

[Couwberghs T. <tco...@sc...>]

Markus,

If you plan to develop a different DTD for each type of reference, what
are you going to do with query's that return multiple types of
references in risx?

Just wondering, multiple DTD's could be a solution but might not work in
the example above.

Greetz Tom

-----Original Message-----
From: Markus Hoenicka [mailto:mar...@mh...]=20
Sent: Thursday, August 21, 2003 4:05 AM
To: Bruce D'Arcus
Cc: ref...@li...
Subject: Re: [Refdb-users] RisX import


Hi Bruce,

a DTD can certainly do this. I've mentioned this a while ago in private
mail that I planned to write "front-end" DTDs for the individual
reference types that would then be transformed to risx. I haven't gotten
round to do this yet.

regards,
Markus

Bruce D'Arcus writes:
 > Markus, is there no way for the DTD to control this stuff?
 >=20

--=20
Markus Hoenicka
mar...@ca...
(Spam-protected email: replace the quadrupeds with "mhoenicka")
http://www.mhoenicka.de



-------------------------------------------------------
This SF.net email is sponsored by Dice.com.
Did you know that Dice has over 25,000 tech jobs available today? From
careers in IT to Engineering to Tech Sales, Dice has tech jobs from the
best hiring companies. http://www.dice.com/index.epl?rel_code=3D104
_______________________________________________
Refdb-users mailing list
Ref...@li...
https://lists.sourceforge.net/lists/listinfo/refdb-users

Re: [Refdb-users] How to create a scientific report with refdb/docbook ?

[Marc B. <ba...@sm...>]

Hi Markus (and other interested minds),

I have started to work on this report and specifically the use of
refdb and "styled" citations. I am taking a step by step approach,
starting to do the things that are easily realisable first and then
go to the trickier ones.

That is why I have started with classical literature citations that
I want to make look like I need it.

The first problem, controlling the order of appearance, is solved.
Or let's say "hacked"

I use the *.id.xml files created by runbib and hand-edit them. This
is basically a list of references to be cited and dictates the order
they will appear in the bibliography. One can still cite them in the
text, the defined order seems to prevail and the right number is
cross-referenced. That is fine :)

To format them, I use a slightly changed version of runbib, which just
skips the step of creating the *.id.xml file, but starts from my hand-
edited one.

mar...@mh... said:
>> I see the problem. The order of the bibliographic entries in the
>> bibliography can be controlled by the style (appearance in the text
>> vs. chronological or reverse chronological). Adding entries that don't
>> appear in the text currently requires nasty hacks, e.g. by editing the
>> intermediate citationlistx XML file. We should think about a more
>> elegant mechanism to specify additional references. 


>> E.g. there is currently no provision for
>> subdividing the bibliographic table. 

That can be done by using distinct *.id.xml files for every subdivision.
You just have to split your list in several files, define entities for
them and call them at the right spot. I'm fine with this.


mar...@mh... said:
>> thanks for the examples. Frankly, I believe that RefDB can't do this
>> out of the box. What you have in mind seems to require quite heavy
>> stylesheet tweaking.

Yes, there I'd need some/a bit of help.
I started to modify things to my gusto, but some questions remain open,
mainly also due to my inexperience with style questions.

One thing I would like, is have a line-break in a citation.
Eg first line authors
second line title
remaining lines reference

How can that be done ? Currently I use the citestylex.dtd, is this
appropriate (or to put it in another way: can I use it without modifying
it and do everything in the actual style xml file, or are there limitations
which will require to tweak the dtd also ?).

Another thing (not top priority, but still) would be to have the citation
number at the front, and the rest of the text starting at a given indentation,
so that the citation number is clearly visible. I mean:

1  blah balkdf kklkg ..d,g.,.,gf., 
   djkjf kjdkj kjk

instead of
1  blah balkdf kklkg ..d,g.,.,gf., 
djkjf kjdkj kjk

How could that be done ?


Thanks in advance.

  Marc Baaden


-- 
 Dr. Marc Baaden  - Institut de Biologie Physico-Chimique, Paris
 mailto:ba...@sm...      -      http://www.marc-baaden.de
 FAX: +49 697912 39550  -  Tel: +33 15841 5176 ou +33 609 843217

[Refdb-users] ris -- is this ok?

[Bruce D'A. <bd...@fa...>]

Am cleaning up the big ris db for import.  Is this OK?

PY  - 1975///September/October

Note slash in the months.

Bruce

[Refdb-users] Re: RisX import

[<Jus...@UL...>]

Markus Hoenicka wrote:

> I'd be glad if everyone could test the risx import/export with the
> weirdest citations they can find and tell me if the output has gross
> errors.

I have a case where risx export drops part of the output. Here is a
part of my hand-edited risx file:

<entry type=3D"CHAP" id=3D"4" citekey=3D"Kruger-AISB2003">
  <part>
    <title>Biologically Motivated Multi-Modal Processing of Visual Primitiv=
es</title>
    <author><lastname>Kr=C3=BCger</lastname><firstname>Norbert</firstname><=
/author>
    <author><lastname>Lappe</lastname><firstname>M</firstname></author>
    <author><lastname>W=C3=B6rg=C3=B6tter</lastname><firstname>Florentin</f=
irstname></author>
  </part>
    <publication>
      <title>AISB 2003 Convention: Cognition in Machines and Animals</title>
      <pubinfo>
	<pubdate type=3D"primary">
	  <date><year>2003</year></date>
	</pubdate>
      </pubinfo>
    </publication>
  <libinfo user=3D"piater">
    <reprint status=3D"NOTINFILE"/>
  </libinfo>
</entry>

<entry type=3D"CHAP" id=3D"10" citekey=3D"Piater-BMCV2000">
  <part>
    <title>Distinctive Features Should Be Learned</title>
    <author><lastname>Piater</lastname><firstname>Justus</firstname><middle=
name>H</middlename></author>
    <author><lastname>Grupen</lastname><firstname>Roderic</firstname><middl=
ename>A</middlename></author>
  </part>
  <publication>
    <title type=3D"gen">International Workshop on Biologically Motivated Co=
mputer Vision</title>
    <author><lastname>Lee</lastname><firstname>Seong-Whan</firstname></auth=
or>
    <author><lastname>B=C3=BClthoff</lastname><firstname>Heinrich</firstnam=
e></author>
    <author><lastname>Poggio</lastname><firstname>Tomaso</firstname></autho=
r>
    <pubinfo>
      <pubdate type=3D"primary">
        <date><year>2000</year><month>05</month></date>
      </pubdate>
      <volume>1811</volume>
      <startpage>52</startpage>
      <endpage>61</endpage>
      <publisher>Springer</publisher>
      <url>http://www-robotics.cs.umass.edu/cgi-bin/getfile/pub/papers/BMCV=
2000.ps.gz</url>
    </pubinfo>
  </publication>
  <set>
    <title type=3D"gen">LNCS</title>
  </set>
  <libinfo user=3D"piater">
    <notes>Seoul, Korea; 88 submissions, 25 oral + 30 poster</notes>
    <reprint status=3D"NOTINFILE"/>
  </libinfo>
</entry>

(If you see some weird characters in the names, that's because the
e-mail-transmission of UTF-8-encoded umlauts did not work right).

The complete, hand-edited file is valid XML. I imported it into an
empty database (sqlite backend).

If I export as ris, all is fine.

If I export as risx, the first entry misses the
<publication>...</publication> part, and therefore does not
validate. This is a bug. Or am I missing something? The second entry
does not have this problem.

I then corrected both entry types to CONF (since that's what they
are), and the results still have the same problem.

Justus

--=20
Justus H. Piater, Ph.D.         http://www.montefiore.ulg.ac.be/~piater
Institut Montefiore, B28        Phone: +32-4-366-2279
Universit=E9 de Li=E8ge, Belgium    Fax:   +32-4-366-2620

RE: [Refdb-users] RisX import

[Markus H. <mar...@mh...>]

Couwberghs Tom writes:
 > PS: for the time being: Does anybody have an overview of which elements
 > are available with the different kinds of references?
 > 

I agree that this piece of documentation is badly missing at the
moment. I intend to add this (and the additional "front-end" DTDs) to
0.9.4.

regards,
Markus

-- 
Markus Hoenicka
mar...@ca...
(Spam-protected email: replace the quadrupeds with "mhoenicka")
http://www.mhoenicka.de

Re: [Refdb-users] RisX import

[Bruce D'A. <bd...@fa...>]

On Thursday, August 21, 2003, at 04:00  PM, Markus Hoenicka wrote:

>> Also, consider that two high-profile xml efforts -- DocBook and
>> OpenOffice -- are moving to Relax NG as their canonical coding 
>> schemas.
>
> This is yet another chicken-and-egg problem. Relax NG will not be
> widely accepted unless you've got tools that support it.

This is a bit of a red herring though.  It is true that there is little 
support for RNG yet in XML editors, but this is not much of an issue 
when you consider that Trang allows for reliable conversion to DTDs and 
XML Schema, both of which are supported in editors.  And editor support 
will come in time.

The reason people are migrating development to RNG is precisely this 
power and flexibility.  DocBook in the future will be written in RNG, 
but this certainly does not mean they won't be releasing DTD versions 
of it.  Likewise with Open Office. Word supports import of arbitrary 
XML, but only if defined against a Schema; likewise almost surely to be 
the case with the next major version of OOo.

The way I see it, RNG is to DTDs what XML is to HTML: a reliable 
long-term investment that still allows you to work with the world as it 
currently exists.

Bruce

Re: [Refdb-users] RisX import

[Markus H. <mar...@mh...>]

Bruce D'Arcus writes:
 > 2)  Having recently been playing with Relax NG, I really think this is 
 > where new schema coding is at.  The Compact syntax is easy to read, 
 > elegant, and well, compact, and Relax NG itself has all the power of 
 > XML Schemas.  Moreover, it is trivial to convert RNG and RNC files to 
 > DTD and XML Schema (though the first route loses some fidelity of 
 > course).
 > 
 > Also, consider that two high-profile xml efforts -- DocBook and 
 > OpenOffice -- are moving to Relax NG as their canonical coding schemas.
 > 

This is yet another chicken-and-egg problem. Relax NG will not be
widely accepted unless you've got tools that support it. Tools will be
written only if something will be widely used. I'm still waiting for
PSGML or something like it to support Schema, let alone Relax NG. I'm
kinda picky about the tools I use - they ought to be Free, and they
must be usable in a networked environment. Either they must work in a
terminal (like Emacs+PSGML), or they must at least be X Window
applications that allow me to display the interface on any X server on
any platform. I don't want to rely on any platform-dependent stuff
that requires you to sit in front of a particular box. And I don't
want to switch to a different application just because that box runs a
different OS.

I'll be happy to think about Relax NG once this tiny problem is
solved.

regards,
Markus

-- 
Markus Hoenicka
mar...@ca...
(Spam-protected email: replace the quadrupeds with "mhoenicka")
http://www.mhoenicka.de

Re: [Refdb-users] Redhat testing/installation woes

[Markus H. <mar...@mh...>]

Hi Jason,

you're probably missing the development files. I've never used RedHat,
but many distributions use different packages for the runtime
libraries and for the development libraries+headers. A binary
installation of the mysql client requires only the runtime library,
that's why it works fine. In order to build a driver, you need the
additional development package.

regards,
Markus

Jason Schlumbohm writes:
 > Am I missing something? Do I need to run configure with additional  
 > options, and if so, what were they?

-- 
Markus Hoenicka
mar...@ca...
(Spam-protected email: replace the quadrupeds with "mhoenicka")
http://www.mhoenicka.de

Re: [Refdb-users] How to create a scientific report with refdb/docbook ?

[Markus H. <mar...@mh...>]

Hi Marc,

I see the problem. The order of the bibliographic entries in the
bibliography can be controlled by the style (appearance in the text
vs. chronological or reverse chronological). Adding entries that don't
appear in the text currently requires nasty hacks, e.g. by editing the
intermediate citationlistx XML file. We should think about a more
elegant mechanism to specify additional references.

regards,
Markus

Marc Baaden writes:
 > But here my need is different:
 > Basically I want to create a bibliography in a certain order,
 > with *some* entries cited somewhere in the text, but other
 > entries *not cited* anywhere, but present in the bibliography.
 > I also want the numbering to be independent from the order
 > of eventual appearance in the text, but defined by myself (eg chronological or 
 > whatever criterion might apply).
 > 
 > So what I did not find in the documentation is how to create a bibliography
 > using refdb without having cited an entry in the text.
 > 
 > (I hope I make myself clear, please feel free to ask where my
 >  explanations leave questions)
 > 

-- 
Markus Hoenicka
mar...@ca...
(Spam-protected email: replace the quadrupeds with "mhoenicka")
http://www.mhoenicka.de

Re: [Refdb-users] RisX import

[Markus H. <mar...@mh...>]

Hi Bruce,

a DTD can certainly do this. I've mentioned this a while ago in
private mail that I planned to write "front-end" DTDs for the
individual reference types that would then be transformed to risx. I
haven't gotten round to do this yet.

regards,
Markus

Bruce D'Arcus writes:
 > Markus, is there no way for the DTD to control this stuff?
 > 

-- 
Markus Hoenicka
mar...@ca...
(Spam-protected email: replace the quadrupeds with "mhoenicka")
http://www.mhoenicka.de

Re: [Refdb-users] How to create a scientific report with refdb/docbook ?

[Marc B. <ba...@sm...>]

>>> "Markus Hoenicka" said:
 >> I'd suggest to have a close look at the tutorial [..]
 >> Refer to chapter 3, section "writing documents". [..]

Thanks Markus. I was well aware of the tutorial and in general I have
no problems with the examples you give.

But here my need is different:
Basically I want to create a bibliography in a certain order,
with *some* entries cited somewhere in the text, but other
entries *not cited* anywhere, but present in the bibliography.
I also want the numbering to be independent from the order
of eventual appearance in the text, but defined by myself (eg chronological or 
whatever criterion might apply).

So what I did not find in the documentation is how to create a bibliography
using refdb without having cited an entry in the text.

(I hope I make myself clear, please feel free to ask where my
 explanations leave questions)

Thanks in advance,

Marc

-- 
 Dr. Marc Baaden  - Institut de Biologie Physico-Chimique, Paris
 mailto:ba...@sm...      -      http://www.marc-baaden.de
 FAX: +49 697912 39550  -  Tel: +33 15841 5176 ou +33 609 843217

Re: [Refdb-users] How to create a scientific report with refdb/docbook ?

[Marc B. <ba...@sm...>]

Hi,

I hope the following mail is not too long (but it's only text), and
it is just meant to give some examples of possible formatted paragraphs
that I would (ideally) like to generate from refdb entries.


bd...@fa... said:
>> Also, Marc, can you be more specific  about exactly what you need in
>> your examples?  I'm particularly  wondering what you mean by your
>> first item.

ok, I'll give you some examples of what I'd like to be in the final
document:


* Research articles (nothing new here)

15.  M. Baaden, C. Meier and M. S. P. Sansom; "A molecular dynamics
     investigation of mono- and dimeric states of the outer membrane
     enzyme Ompla.", J. Mol. Biol., 331, 2003, 177-189


* Scientific Presentations

either indicating directly the type of presentation or subdivided
into chapters like: oral presentations, poster presentations and others
like seminars, lectures, teaching of the approximate form:


EXAMPLE 1:

19.  M. Baaden, L. Troxler and G. Wipff, "Interfacial aspects in
     liquid-liquid ion extraction."

     o    Lecture, 26th International conference on solution chemistry
          (26ICSC), Fukuoka, Japan, 27 July 1999.

31.  M. Baaden, M. S. P. Sansom, "Barrels & corks in a computer,
     modelling bacterial iron transport systems."

     o    Poster presentation at the Kollman Memorial Symposium, UCSF,
          San Francisco, 21-22 February 2002.

     o    Poster presentation at the 46th Annual Meeting of the
          Biophysical Society, San Francisco, 23-27 February 2002.
          (Biophys. J. Part 2, 82, 2002, 2610)

     o    Poster presentation at the 6=E8mes Journ=E9es Francophones des
          Jeunes Physico-Chimistes, Marseille, 3-5 July 2002.

32.  M. Baaden, "Modelling the ligand-gated protein channel and ferric
     enterobactin receptor FepA."

     o    Seminar, Laboratory of Molecular Biophysics, Oxford, 5
          February 2002

35.  M. Baaden, "La dynamique mol=E9culaire in silico."

     o    Course and tutorial, Ecole Th=E9matique du CNRS, Outils pour
          =E9tudier la structure et la dynamique des peptides et des
          prot=E9ines, Arcachon, 18-24 Mai 2003.

EXAMPLE 2:

PUBLICATIONS

Journaux avec Comit=E9 de Lecture

J. D. Campbell, P. C. Biggin, M. Baaden, M. S. P. Sansom (2003)
Extending the structure of an ABC transporter to atomic resolution:
modelling and simulation studies of Msba. Biochemistry, 42, 3666-3673
[..]


* Attendance at conferences

PARTICIPATION AUX CONGRES

Congr=E8s Internationaux

M. Baaden (2003) "Molecular dynamics simulations of outer membrane
phospholilpase A". Euresco Conference on Bionanotechnology -
Euroconference on Biomolecular Devices, Grenade, Espagne (poster).


In the case where you present a poster or so at a conference you might
have it appear twice, once as "dissemination" in one form, and then a
second time as participation at the conference in the form above. Where
participation at a conference would also include non-disseminating meetin=
gs
eg also workshops and courses.


* Other

Eventually also other information - all that is related to a name,
a subject, a place and a date in general (like awards and prices, or
like mobility in different labs eg) would be nice to have in a database,
too. Although this is much further than real references and changes
less frequently.

Examples for awards or teaching eg:

PRIX

M. Baaden (2003) Prix Jeune-chercheur du Groupe de Graphisme et
Mod=E9lisation Mol=E9culaire (GGMM)


ENSEIGNEMENT

M. Baaden (2003) Cours et atelier. La dynamique mol=E9culaire in
silico. Ecole Th=E9matique du CNRS "Outils pour =E9tudier la structure et=

la dynamique des peptides et des prot=E9ines". Arcachon.


  Marc Baaden


-- =

 Dr. Marc Baaden  - Institut de Biologie Physico-Chimique, Paris
 mailto:ba...@sm...      -      http://www.marc-baaden.de
 FAX: +49 697912 39550  -  Tel: +33 15841 5176 ou +33 609 843217

Re: [Refdb-users] How to create a scientific report with refdb/docbook ?

[Bruce D'A. <bd...@fa...>]

On Tuesday, August 19, 2003, at 04:28  PM, Markus Hoenicka wrote:

>> In particular I wonder about the
>> - participation in conferences
>> - oral presentations
>> - poster presentations
>>
>> Can (/should) these be put as a bibliographic entry into refdb ? Or 
>> is there
>> a better way ?
>
> The idea of using a bibliography tool is to keep the references in a
> database to make them reusable for any number of documents. We'd have
> to look at your requirements in terms of formatting though.

Actually, the other issue is storage.  How do you code these records 
with the ris data model Markus?  Also, Marc, can you be more specific 
about exactly what you need in your examples?  I'm particularly 
wondering what you mean by your first item.

There is the CONF (conference proceeding) reference type, but this 
tends to be for papers presented at a conference, and subsequently 
published in the proceedings.

Bruce

Re: [Refdb-users] RisX import

[Bruce D'A. <bd...@fa...>]

I've attached a Relax NG version of risx just in case anyone is 
interested.  It was auto-converted with Trang, so is no more strict 
than the DTD of course.

But what I was wondering is if you have this as the top level 
definition...

entry =
       element entry {
             attlist.entry,
             (part?, publication, set?, libinfo*, contents?)+
       }

...is it possible to specify the spec for each entry type?

So, for book, it'd be:

(publication, set?, libinfo*, contents?)

For a (journal/magazine/newspaper) article or book chapter, it'd be...

(part, publication, set?, libinfo*, contents?)

...etc.

I don't know if you can do this based on the type attribute with DTDs, 
though....

Bruce

Re: [Refdb-users] RisX import

[Bruce D'A. <bd...@fa...>]

On Wednesday, August 20, 2003, at 02:59  AM, Couwberghs Tom wrote:

> The problem is that I'm just a developper: You can't expect me to know
> all these exceptions/rules.

Yes, you're right.  Also, sometime in the mid-term, we need to think 
about a simple input-oriented Schema (see below) that can be validate 
records that be transformed to either risx or mods.

> I agree with Bruce here that 'this stuff' should be controlled, but I
> think DTD schemes are too limited to do the trick and XML-schemes would
> be more appropriate.

Two things:

1)  Actually, if you create separate definitions for each entry type, 
shouldn't this be doable with a DTD?  Or maybe not (am not a 
developer!).

2)  Having recently been playing with Relax NG, I really think this is 
where new schema coding is at.  The Compact syntax is easy to read, 
elegant, and well, compact, and Relax NG itself has all the power of 
XML Schemas.  Moreover, it is trivial to convert RNG and RNC files to 
DTD and XML Schema (though the first route loses some fidelity of 
course).

Also, consider that two high-profile xml efforts -- DocBook and 
OpenOffice -- are moving to Relax NG as their canonical coding schemas.

Bruce

RE: [Refdb-users] RisX import

[Couwberghs T. <tco...@sc...>]

I agree that a book shouldn't have a part section and I'm sure that
there are other examples of references where certain elements are not
relevant.

The problem is that I'm just a developper: You can't expect me to know
all these exceptions/rules.
I agree with Bruce here that 'this stuff' should be controlled, but I
think DTD schemes are too limited to do the trick and XML-schemes would
be more appropriate.

Greetz Tom

PS: for the time being: Does anybody have an overview of which elements
are available with the different kinds of references?

-----Original Message-----
From: Bruce D'Arcus [mailto:bd...@fa...]=20
Sent: Wednesday, August 20, 2003 4:16 AM
To: ref...@li...
Subject: Re: [Refdb-users] RisX import



On Tuesday, August 19, 2003, at 02:35  AM, Couwberghs Tom wrote:

> Could you explain the reason why? Imho, It would make more sense if=20
> the risx structure was kept as it was.

Well, one obvious issue is that your input record is incorrectly coded.=20
  A book shouldn't have a part element.  That doesn't explain everything

of course (you certainly shouldn't lose the title), and you'll have to=20
wait for Markus to give a fuller example.

Markus, is there no way for the DTD to control this stuff?

Bruce



-------------------------------------------------------
This SF.net email is sponsored by Dice.com.
Did you know that Dice has over 25,000 tech jobs available today? From
careers in IT to Engineering to Tech Sales, Dice has tech jobs from the
best hiring companies. http://www.dice.com/index.epl?rel_code=3D104
_______________________________________________
Refdb-users mailing list
Ref...@li...
https://lists.sourceforge.net/lists/listinfo/refdb-users

Re: [Refdb-users] RisX import

[Bruce D'A. <bd...@fa...>]

On Tuesday, August 19, 2003, at 02:35  AM, Couwberghs Tom wrote:

> Could you explain the reason why? Imho, It would make more sense if the
> risx structure was kept as it was.

Well, one obvious issue is that your input record is incorrectly coded. 
  A book shouldn't have a part element.  That doesn't explain everything 
of course (you certainly shouldn't lose the title), and you'll have to 
wait for Markus to give a fuller example.

Markus, is there no way for the DTD to control this stuff?

Bruce

Re: [Refdb-users] Redhat testing/installation woes

[Jason S. <st...@co...>]

I finally had time to try and get this working again tonight. I re-did  
the libdbi framework, and now am trying to reinstall the drivers  
themselves. When I type:
./configure --with-mysql
Things seem to go well, but then the configuration turns sour:
checking for libdbi framework... yes: headers in /usr/local/include
checking for MySQL support... no
configure: error: Invalid MySQL directory - include files not found.

This is just what happened before. From this it looks like libdbi was  
installed correctly, but that MySQL is not? This is a standard MySQL  
installation on RedHat. I know MySQL works, as I'm able to login, make  
users, &c., and use the MySQL installation of Movable Type, so I'm  
fairly certain that things re well with MySQL.

Am I missing something? Do I need to run configure with additional  
options, and if so, what were they?

Thanks, Jason


On Monday, Aug 18, 2003, at 11:01 US/Pacific, Markus Hoenicka wrote:

> Hi,
>
> Alan Anderson writes:
>> First, you did also download the dbi driver files for MySQL, right?   
>> libdbi requires the drivers to
>> actually communicate with the database.
>>
>
> I'm afraid this can't be pointed out often enough. When RefDB 0.9.2
> was released, the current libdbi version contained the MySQL and
> PostgreSQL drivers. This is reflected in the RefDB manual as
> well. However, the current version of libdbi provides only the
> framework, and we moved all drivers to the libdbi-drivers project. The
> next version of the RefDB manual will of course be updated to reflect
> this. It will be an intermittent problem I reckon. Moreover, refdbd as
> of 0.9.3 will list the available libdbi drivers at startup. The
> problem and the solution will then be way more obvious.
>
> And yes, while we're at it: <mantra>You'll have to specify the
> driver(s) on the libdbi-drivers configure command line, otherwise
> nothing will be built</mantra> ;-)
>
>> Second, I discovered that RedHat does not seem to set LD_LIBRARY_PATH  
>> environment variable.  I had
>> to set this explicitly in my startup script to make RefDB work since  
>> the libdbi files are not in
>> /usr/lib.  This probably isn't you're problem if you don't have the  
>> actual shared library files.
>>
>
> Actually LD_LIBRARY_PATH (or an appropriate entry in ld.so.conf) is
> required to find libdbi.so. This in turn either uses a compiled-in
> path or one provided by the application using the library to locate
> the drivers. That is, the directory holding the drivers need not be
> listed in LD_LIBRARY_PATH.
>
> regards,
> Markus
>
> --  
> Markus Hoenicka
> mar...@ca...
> (Spam-protected email: replace the quadrupeds with "mhoenicka")
> http://www.mhoenicka.de
>
>
>
> -------------------------------------------------------
> This SF.Net email sponsored by: Free pre-built ASP.NET sites including
> Data Reports, E-commerce, Portals, and Forums are available now.
> Download today and enter to win an XBOX or Visual Studio .NET.
> http://aspnet.click-url.com/go/psa00100003ave/ 
> direct;at.aspnet_072303_01/01
> _______________________________________________
> Refdb-users mailing list
> Ref...@li...
> https://lists.sourceforge.net/lists/listinfo/refdb-users
>