refdb-users — List for RefDB users

Re: [Refdb-users] risx output problems

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

On Saturday, August 30, 2003, at 07:38  PM, Markus Hoenicka wrote:

>> Problem 1: characters like & are not converted, despite the fact that
>> the entity is defined.
>
> RefDB should export &, <, and > as entities. Can you provide examples
> where this fails?

I was seeing it in journal titles; all of them that included an & 
character.

>> Also, for some reason I am getting a unicode error from the parser 
>> with
>> this string:
>
> I'm not sure what to make of this. Is the unicode character valid? Is
> it changed by importing to and exporting from RefDB?

This is with data exported from Endnote w/Mac Roman encoding, and then 
cleaned up with the en2ris script.  So maybe something went wrong there?

>> I am also getting parser errors about an incorrect reference type of
>> HEAR.  Looking at the refdb docs, I realize there is no such reference
>> type.  However, if I look here, RefMan itself (at least v10) does
>> support this reference type.
>
> RefDB lacks a couple of features present in RefMan 10. I've been using
> my RefMan 8.01 docs so far as I've last used that version. Other
> issues are e.g. support for multiple UR fields and  support for L1-L4
> fields. I'll work on that.

For the short-term, perhaps you can just add HEAR as an option in the 
DTD?

>> Finally, going back to previous discussions, don't we need a "note" or
>> "annotation" DTD now?  It's quite awkward to have the "\n" characters
>> in the output, so we obviously need a paragraph element.  I'd opt for
>> these as well at minimum: quote, blockquote, emphasis, and perhaps a
>> generic span.  Also, I'd like attributes on all of these for citation
>> IDREFs, pages, etc.
>
> I currently lack the time for new features that are tricky to
> implement. I'd rather go through the quite impressive list of bug
> reports and get that right before starting other stuff. That is, I
> don't agree with "now" but I see the need for what you suggest. I also
> see problems to keep the output suitable for the target DTDs and for
> BibTeX unless I misunderstand your suggestion.

I can see this, and I understand "no time."

While I think there is some trickiness to this feature, it isn't too 
complicated, since the focus is simply on the notes; it doesn't extend 
to formatting.  Still, for BibTeX compatibility, you would need to map 
that stuff to equivalent constructs, so this isn't trivial.

I may put together the schema/dtd myself, as it's going to be needed, 
whether in risx or in mods.

Bruce

[Refdb-users] Issues and Observations

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

Hi,

some progress...

Justus H. Piater writes:
 > 1. Multi-reference citations (as in <citation
 >    role="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.
 > 

It is a bug, and it is fixed in CVS.

 >    (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.
 > 

I've looked at your examples and I'm as clueless as before. It looks
like a real limitation of openjade for reasons that I don't
understand.

 > 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.
 > 

This is fixed in CVS.

 >    (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.
 > 

I haven't found a reason for this yet. The RefDB driver files do not
specifically process the bibliography title, so it looks to me that
the DocBook stylesheets do this. Any help in debugging this is welcome.

 >    (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,b|(Bger, Lappe, and W,bv(Brg,bv(Btter, 2003; Kr,b|(Bger and W,bv(Brg,bv(Btter,
 >        2002
 > 
 >        J.Biol.Chem. has a similar problem.
 >        openjade-generated HTML does not have this problem.
 > 

After fixing some more attribute case mismatches, I don't see these
two problems in my testcases. The changes were committed to CVS.

 > 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)
 > 

This is also fixed in CVS.

regards,
Markus

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

[Refdb-users] risx output problems

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

 > Problem 1: characters like & are not converted, despite the fact that 
 > the entity is defined.

RefDB should export &, <, and > as entities. Can you provide examples
where this fails?

 > Problem 2 (and bigger): I'm losing some data too. This input:

Thanks for the examples. I'll try to fix that asap.

 > Also, for some reason I am getting a unicode error from the parser with 
 > this string:

I'm not sure what to make of this. Is the unicode character valid? Is
it changed by importing to and exporting from RefDB?

 > I am also getting parser errors about an incorrect reference type of 
 > HEAR.  Looking at the refdb docs, I realize there is no such reference 
 > type.  However, if I look here, RefMan itself (at least v10) does 
 > support this reference type.

RefDB lacks a couple of features present in RefMan 10. I've been using
my RefMan 8.01 docs so far as I've last used that version. Other
issues are e.g. support for multiple UR fields and  support for L1-L4
fields. I'll work on that.

 > Finally, going back to previous discussions, don't we need a "note" or 
 > "annotation" DTD now?  It's quite awkward to have the "\n" characters 
 > in the output, so we obviously need a paragraph element.  I'd opt for 
 > these as well at minimum: quote, blockquote, emphasis, and perhaps a 
 > generic span.  Also, I'd like attributes on all of these for citation 
 > IDREFs, pages, etc.

I currently lack the time for new features that are tricky to
implement. I'd rather go through the quite impressive list of bug
reports and get that right before starting other stuff. That is, I
don't agree with "now" but I see the need for what you suggest. I also
see problems to keep the output suitable for the target DTDs and for
BibTeX unless I misunderstand your suggestion.

regards,
Markus

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

[Refdb-users] xhtml support

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

Hi all,

I've checked in a couple of changes that add xhtml support for the
getref command (use getref -t xhtml) and for DocBook XML
bibliographies (use make xhtml with the refdbnd-generated
Makefiles). If you find some time to test the CVS version, please
report any problems you bump into.

I'll create a new prerelease once I fixed the most important bugs that
were reported lately.

regards,
Markus

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

Re: [Refdb-users] import problems

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

Hi,

Bruce D'Arcus writes:
 > More on this.  Below is what I have in my terminal.  How do I interpret 
 > this, and what do I do with the other 1000+ records in this file?  Do I 
 > need to break this up into smaller chunks?  If so, how large?
 > 

I know you apparently solved this problem. Just for the record: RefDB
has no built-in limit for how many references you can import at a
time. If anyone finds a limit, it is a bug.

 > Also, is there no equivalent to the getau command in refdbc for titles?

No there isn't. Using getref :TI:=whatever should do the trick as
well.

regards,
Markus

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

[Refdb-users] risx output problems

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

I just exported my DB as risx, and ran into problems.

Problem 1: characters like & are not converted, despite the fact that=20
the entity is defined.

Problem 2 (and bigger): I'm losing some data too. This input:

TY  - NEWS
ID  - Comm1968a
T2  - Commonweal
T1  - Jailing Hayden?
N1  - cited in Dee (1990) @104:
=93the government is about to crack down on the New Left by singling Tom=20=

Hayden out for arrest.  The FBI has been scouring the country for=20
evidence to indict Hayden under the infamous =91Rap Brown=92 =
law=97crossing=20
state lines to incite to riot.=94 @393
PY  - 1968/12/20/
ER  -

Yields this output in risx, and an error from the xml parser (note=20
dropped publication element):

<entry type=3D"NEWS" id=3D"9" citekey=3D"Comm1968a">
   <part>
     <title>Jailing Hayden?</title>
   </part>
   <libinfo user=3D"darcusb">
     <notes>cited in Dee (1990) @104:\n=93the government is about to =
crack=20
down on the New Left by singling Tom Hayden out for arrest.  The FBI=20
has been scouring the country for evidence to indict Hayden under the=20
infamous =91Rap Brown=92 law=97crossing state lines to incite to riot.=94=20=

@393</notes>
     <reprint status=3D"NOTINFILE"/>
   </libinfo>
</entry>

I have at 10 records that are screwed up enough to yield a parser=20
error.  Here's another:

TY  - CASE
ID  - USvDellinger1968
PY  - 1968
T1  - United States v. David T. Dellinger et al
T2  - U.S. v. Dellinger
T3  - appeal
KW  - anti-riot
ER  -

...gets:

<entry type=3D"CASE" id=3D"15" citekey=3D"USvDellinger1968">
   <set>
     <title type=3D"gen">appeal</title>
   </set>
   <libinfo user=3D"darcusb">
     <reprint status=3D"NOTINFILE"/>
   </libinfo>
   <contents>
     <keyword>anti-riot</keyword>
   </contents>
</entry>

Also, for some reason I am getting a unicode error from the parser with=20=

this string:

<publisher>Routledge=1D</publisher>

I am also getting parser errors about an incorrect reference type of=20
HEAR.  Looking at the refdb docs, I realize there is no such reference=20=

type.  However, if I look here, RefMan itself (at least v10) does=20
support this reference type.

http://www.refman.com/support/risformat_reftypes.asp

I *really* need this, ASAP in the DTD (in case I need to transform the=20=

data), and soon after in refdb itself.

Finally, going back to previous discussions, don't we need a "note" or=20=

"annotation" DTD now?  It's quite awkward to have the "\n" characters=20
in the output, so we obviously need a paragraph element.  I'd opt for=20
these as well at minimum: quote, blockquote, emphasis, and perhaps a=20
generic span.  Also, I'd like attributes on all of these for citation=20
IDREFs, pages, etc.

Bruce

[Refdb-users] risx output problems

[Bruce D'A. <da...@mu...>]

I just exported my DB as risx, and ran into problems.

Problem 1: characters like & are not converted, despite the fact that=20
the entity is defined.

Problem 2 (and bigger): I'm losing some data too. This input:

TY  - NEWS
ID  - Comm1968a
T2  - Commonweal
T1  - Jailing Hayden?
N1  - cited in Dee (1990) @104:
=93the government is about to crack down on the New Left by singling Tom=20=

Hayden out for arrest.  The FBI has been scouring the country for=20
evidence to indict Hayden under the infamous =91Rap Brown=92 =
law=97crossing=20
state lines to incite to riot.=94 @393
PY  - 1968/12/20/
ER  -

Yields this output in risx, and an error from the xml parser (note=20
dropped publication element):

<entry type=3D"NEWS" id=3D"9" citekey=3D"Comm1968a">
   <part>
     <title>Jailing Hayden?</title>
   </part>
   <libinfo user=3D"darcusb">
     <notes>cited in Dee (1990) @104:\n=93the government is about to =
crack=20
down on the New Left by singling Tom Hayden out for arrest.  The FBI=20
has been scouring the country for evidence to indict Hayden under the=20
infamous =91Rap Brown=92 law=97crossing state lines to incite to riot.=94=20=

@393</notes>
     <reprint status=3D"NOTINFILE"/>
   </libinfo>
</entry>

I have at 10 records that are screwed up enough to yield a parser=20
error.  Here's another:

TY  - CASE
ID  - USvDellinger1968
PY  - 1968
T1  - United States v. David T. Dellinger et al
T2  - U.S. v. Dellinger
T3  - appeal
KW  - anti-riot
ER  -

...gets:

<entry type=3D"CASE" id=3D"15" citekey=3D"USvDellinger1968">
   <set>
     <title type=3D"gen">appeal</title>
   </set>
   <libinfo user=3D"darcusb">
     <reprint status=3D"NOTINFILE"/>
   </libinfo>
   <contents>
     <keyword>anti-riot</keyword>
   </contents>
</entry>

Also, for some reason I am getting a unicode error from the parser with=20=

this string:

<publisher>Routledge=1D</publisher>

I am also getting parser errors about an incorrect reference type of=20
HEAR.  Looking at the refdb docs, I realize there is no such reference=20=

type.  However, if I look here, RefMan itself (at least v10) does=20
support this reference type.

http://www.refman.com/support/risformat_reftypes.asp

I *really* need this, ASAP in the DTD (in case I need to transform the=20=

data), and soon after in refdb itself.

Finally, going back to previous discussions, don't we need a "note" or=20=

"annotation" DTD now?  It's quite awkward to have the "\n" characters=20
in the output, so we obviously need a paragraph element.  I'd opt for=20
these as well at minimum: quote, blockquote, emphasis, and perhaps a=20
generic span.  Also, I'd like attributes on all of these for citation=20
IDREFs, pages, etc.

Bruce=

Re: [Refdb-users] import problems

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

OK, I'm not sure what happened, but I deleted the DB, recreated it, and 
reimported the data.  This time everything went fine. Sorry for the 
distraction...

Bruce

Re: [Refdb-users] import problems

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

On Saturday, August 30, 2003, at 09:09  AM, Bruce D'Arcus wrote:

>
> Markus,
>
> What do I need to know about the RIS file I import?  Refdbc is choking 
> on it.
>
> I am wondering if the problem might be with my notes fields, which 
> often start with newlines, and have them sprinkled throughout.  At 
> least I hope this is the problem...

More on this.  Below is what I have in my terminal.  How do I interpret 
this, and what do I do with the other 1000+ records in this file?  Do I 
need to break this up into smaller chunks?  If so, how large?

Also, is there no equivalent to the getau command in refdbc for titles?

Bruce

refdbc: addref -k out1.ris
try to add set as reference 1
Citation key: SpecialOps2001a
Adding set 0 successful
try to add set as reference 2
Citation key: Anonymous0
Adding set 0 successful
try to add set as reference 3
Citation key: Anonymous0a
Adding set 0 successful
try to add set as reference 4
Citation key: Anonymous1967
Adding set 0 successful
try to add set as reference 5
Citation key: Senate1967b
Adding set 0 successful
try to add set as reference 6
Citation key: Senate1967c
Adding set 0 successful
try to add set as reference 7
Citation key: Senate1967a
Adding set 0 successful
try to add set as reference 8
Citation key: USNews1968a
Adding set 0 successful
try to add set as reference 9
Citation key: Comm1968a
Adding set 0 successful
try to add set as reference 10
Citation key: ARL1968
Adding set 0 successful
try to add set as reference 11
Citation key: MinTr1968a
Adding set 0 successful
server error: 201 dataset(s) added, 0 skipped, 0 failed

[Refdb-users] import problems

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

Markus,

What do I need to know about the RIS file I import?  Refdbc is choking 
on it.

I am wondering if the problem might be with my notes fields, which 
often start with newlines, and have them sprinkled throughout.  At 
least I hope this is the problem...

Bruce

[Refdb-users] Participation at European Patents Demo?

[Jakob B. <j....@bw...>]

Hi everybody,

since many of the RefDB users seem to be European based, I wonder what 
you think about participating at the Anti-Patent Protests (Marcus, 
Bruce, all?). I know that the actual decision has been postponed from 
Sep. 1st, so maybe this would be up for quite some time.

I have compiled a sample frontpage, to have a look at
http://www.uni-weimar.de/~beetz/refdb_patents/

Best regards
Jakob
-- 
_____________________________________________________________________
Jakob Beetz - PhD student           Ontwerp Systemen | Design Systems
Eindhoven University of Technology            Faculty of Architecture 

mail: j....@bw...  phone: +31 (0)40 247 4934    loc: VRT 09H08

Re: [Refdb-users] Re: refdb

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

Hi Phillip,

thanks for firing up gdb.

Philipp v. Hilgers writes:
 > I debugged  "refdbd":
 > Program received signal SIGSEGV, Segmentation fault.
 > 0x400c483e in fflush () from /lib/i686/libc.so.6
 > (gdb)
 > (gdb) backtrace
 > #0  0x400c483e in fflush () from /lib/i686/libc.so.6
 > #1  0x0804b27b in main (argc=6, argv=0xbffff3f4) at refdbd.c:747
 > #2  0x4007e4c2 in __libc_start_main () from /lib/i686/libc.so.6
 > 

I'm afraid this bug has popped up now and then, but I couldn't track
it down as it never hit me. I've forgot to reset fp_log_file
appropriately, so under some particular circumstances refdbd tries to
write to or to flush a file that isn't open. Please use the attached
one-line patch (run "patch -p0 < refdbd.c.diff" in refdb/src and then
make && make install) to fix this problem.

 > Now "refdba" returns: "could not connect to database server"

This is an entirely different (and harmless) problem. This is usually
caused by inappropriate access rights or incorrect username/password
settings. Did you try this with sqlite? It is quite helpful to start
refdbd like this: refdbd -s -e 0 -l 7. This will print out the
connection parameters that refdbd uses to connect to the database
server. Please check also the database server log (if you use MySQL)
as this has further hints why the server won't let you in. Feel free
to send the refdbd and the mysql log output to get further
assistance.

regards,
Markus

Re: [Refdb-users] Re: refdb

[Philipp v. H. <p.v...@rz...>]

Markus Hoenicka wrote:

>Hi,
>
>doesn't look like a MySQL version problem then if it happens with
>SQLite as well. I'm afraid we'll have to debug this on your box
>because I haven't seen this problem on all platforms that I have
>access to (FreeBSD, Debian, Cygwin, OSX). Would you mind using gdb to
>trace back where the app crashes? Let me know if you need step-by-step
>instructions how to do that.
>
>regards,
>Markus
>

Hi Markus,
I debugged  "refdbd":
Program received signal SIGSEGV, Segmentation fault.
0x400c483e in fflush () from /lib/i686/libc.so.6
(gdb)
(gdb) backtrace
#0  0x400c483e in fflush () from /lib/i686/libc.so.6
#1  0x0804b27b in main (argc=6, argv=0xbffff3f4) at refdbd.c:747
#2  0x4007e4c2 in __libc_start_main () from /lib/i686/libc.so.6

The debugger leads in "refdbd.c"  to the  function  for writing the log 
file:
 fflush(fp_log_file);
In order to prevent  pushing log file data to stderr, I started   
"/usr/local/bin/refdbd -s"in this mode.
Now "refdba" returns: "could not connect to database server"
This are messages in the  log file :
 6:pid=24242:Sat Aug 30 07:38:09 2003:Available libdbi database drivers:
6:pid=24242:Sat Aug 30 07:38:09 2003:mysql
6:pid=24242:Sat Aug 30 07:38:09 2003:sqlite
6:pid=24242:Sat Aug 30 07:38:09 2003:application server started
6:pid=24242:Sat Aug 30 07:38:09 2003:server waiting n_max_fd=7
6:pid=24242:Sat Aug 30 07:38:15 2003:adding client on fd 11
6:pid=24242:Sat Aug 30 07:38:15 2003:server waiting n_max_fd=11
6:pid=24247:Sat Aug 30 07:38:15 2003:serving client on fd 11 with 
protocol version 1
6:pid=24247:Sat Aug 30 07:38:15 2003:dbi is up
4:pid=24247:Sat Aug 30 07:38:15 2003:failed to connect to database server
6:pid=24247:Sat Aug 30 07:38:15 2003:child finished client on fd 11
6:pid=24242:Sat Aug 30 07:38:15 2003:parent removing client on fd 11
6:pid=24242:Sat Aug 30 07:38:15 2003:server waiting n_max_fd=7
6:pid=24242:Sat Aug 30 07:38:15 2003:child exited with code 0
6:pid=24242:Sat Aug 30 07:38:15 2003:server waiting n_max_fd=7
6:pid=24242:Sat Aug 30 07:38:16 2003:adding client on fd 11
6:pid=24242:Sat Aug 30 07:38:16 2003:server waiting n_max_fd=11
6:pid=24248:Sat Aug 30 07:38:16 2003:serving client on fd 11 with 
protocol version 1
6:pid=24248:Sat Aug 30 07:38:16 2003:dbi is up
4:pid=24248:Sat Aug 30 07:38:16 2003:failed to connect to database server
6:pid=24248:Sat Aug 30 07:38:16 2003:child finished client on fd 11
6:pid=24242:Sat Aug 30 07:38:16 2003:parent removing client on fd 11
6:pid=24242:Sat Aug 30 07:38:16 2003:server waiting n_max_fd=7
6:pid=24242:Sat Aug 30 07:38:16 2003:child exited with code 0
6:pid=24242:Sat Aug 30 07:38:16 2003:server waiting n_max_fd=7
6:pid=24242:Sat Aug 30 07:38:18 2003:adding client on fd 11
6:pid=24242:Sat Aug 30 07:38:18 2003:server waiting n_max_fd=11
6:pid=24249:Sat Aug 30 07:38:18 2003:serving client on fd 11 with 
protocol version 1
6:pid=24249:Sat Aug 30 07:38:18 2003:dbi is up
4:pid=24249:Sat Aug 30 07:38:18 2003:failed to connect to database server
6:pid=24249:Sat Aug 30 07:38:18 2003:child finished client on fd 11

Do you have some idea, what is going on?

Regards,

Philipp

[Refdb-users] Re: refdb

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

Hi,

doesn't look like a MySQL version problem then if it happens with
SQLite as well. I'm afraid we'll have to debug this on your box
because I haven't seen this problem on all platforms that I have
access to (FreeBSD, Debian, Cygwin, OSX). Would you mind using gdb to
trace back where the app crashes? Let me know if you need step-by-step
instructions how to do that.

regards,
Markus

Philipp v. Hilgers writes:
 > I have mysql  Ver 11.18 Distrib 3.23.52, for suse-linux (i686) installed 
 > and  sqlite   2.8.5.  I tried  sqlite first,  thought it would be 
 > simpler to use. 
 > Any hint is welcome.
 > Regards,
 > Philipp

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

[Refdb-users] Re: refdb

[Philipp v. H. <p.v...@rz...>]

Markus Hoenicka wrote:

>(cc'ing to the list to keep things archived)
>Hi Phillip,
>
>this reminds me of a problem with MySQL that was reported lately. Are
>you using MySQL as the database server, and if yes, which version? The
>problem could be avoided previously by switching to an older MySQL
>release. If it is the same problem in your case, it is a libdbi issue,
>not a RefDB issue. If you can confirm this, I'll follow up on this on
>the libdbi list.
>
>regards,
>Markus
>
>Philipp v. Hilgers writes:
> > Hi Markus,
> > I was goimg to install refdb, but something seems broken. I checked user 
> > previlegs, correct database configuration and so on, but
> > while refdbd is coming uo  like :
> > dbi_driver_dir went to:
> > 
> > dbi is up using default driver dir
> > Available libdbi database drivers:
> > mysql
> > sqlite
> > application server started
> > use /tmp/refdbd_fifo21264 as fifo
> > server waiting n_max_fd=4
> > 
> > refdba crashs refdbd  and returns:
> > 
> > server error: incorrect scramble string
> > 
> > 
> > What is going wrong?
> > 
> > Cheers,
> > 
> > 
> > Philipp
> > 
> > 
> > 
>
>  
>
I have mysql  Ver 11.18 Distrib 3.23.52, for suse-linux (i686) installed 
and  sqlite   2.8.5.  I tried  sqlite first,  thought it would be 
simpler to use. 
Any hint is welcome.
Regards,
Philipp

[Refdb-users] refdb

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

(cc'ing to the list to keep things archived)
Hi Phillip,

this reminds me of a problem with MySQL that was reported lately. Are
you using MySQL as the database server, and if yes, which version? The
problem could be avoided previously by switching to an older MySQL
release. If it is the same problem in your case, it is a libdbi issue,
not a RefDB issue. If you can confirm this, I'll follow up on this on
the libdbi list.

regards,
Markus

Philipp v. Hilgers writes:
 > Hi Markus,
 > I was goimg to install refdb, but something seems broken. I checked user 
 > previlegs, correct database configuration and so on, but
 > while refdbd is coming uo  like :
 > dbi_driver_dir went to:
 > 
 > dbi is up using default driver dir
 > Available libdbi database drivers:
 > mysql
 > sqlite
 > application server started
 > use /tmp/refdbd_fifo21264 as fifo
 > server waiting n_max_fd=4
 > 
 > refdba crashs refdbd  and returns:
 > 
 > server error: incorrect scramble string
 > 
 > 
 > What is going wrong?
 > 
 > Cheers,
 > 
 > 
 > Philipp
 > 
 > 
 > 

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

Re: [Refdb-users] Issues and Observations

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

"Markus Hoenicka" <mar...@mh...> wrote on Thu, 28 Aug
2003 21:35:52 +0200:

>  > UTF-8 would be a sensible default in conjunction with XML,
>  > as that is the default encoding of XML. Ideally, however, the output
>  > should be convertible to a desired target format (e.g., entities for
>  > US-ASCII-HTML).
>  >=20
> Fine. It'll be a little tricky though as e.g. PostgreSQL allows to
> select the encoding individually for each database, so neither a
> server-based config option nor a user-based one will work
> reliably. There's apparently no built-in function to find out about
> the encoding of the current database.

Aargh. I'm out of ideas for now. We'll have to rely on some discipline
on the part of the user, I guess.

> This was the missing piece of information. I've never had any reason
> to look at XHTML and thus never noticed that DocBook has stylesheets
> for this. Supporting this should indeed be easy. Would it make sense
> to extend the HTML backend to allow XHTML output for the getref
> command as well?

Yes, absolutely! This is likely to be the level at which my Web site
will interact with RefDB.

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] Issues and Observations

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

Hi,

Justus H. Piater writes:
 > > Well, some DB servers support other encodings as well. Does it really
 > > make sense to use UTF-8 unconditionally?
 > 
 > No, I guess not. It should probably be configurable, as the RefDB
 > system in general has no way to recognize the coding system of the DB
 > internals. UTF-8 would be a sensible default in conjunction with XML,
 > as that is the default encoding of XML. Ideally, however, the output
 > should be convertible to a desired target format (e.g., entities for
 > US-ASCII-HTML).
 > 
Fine. It'll be a little tricky though as e.g. PostgreSQL allows to
select the encoding individually for each database, so neither a
server-based config option nor a user-based one will work
reliably. There's apparently no built-in function to find out about
the encoding of the current database.

 > > What should the XHTML support look like? Could you provide examples?
 > 
[...]
 > From your end, it should be very easy to do, at least for the
 > XML/xsltproc track, as XHTML is supported natively by DocBook XML.
 > 

This was the missing piece of information. I've never had any reason
to look at XHTML and thus never noticed that DocBook has stylesheets
for this. Supporting this should indeed be easy. Would it make sense
to extend the HTML backend to allow XHTML output for the getref
command as well?

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 Wednesday, August 27, 2003, at 08:24  PM, Markus Hoenicka wrote:

> What should the XHTML support look like? Could you provide examples?

Here's sample from the website, and below it as it's converted to xhtml 
by tidy (actually, I first used tidy to indent the html of the sample, 
but it also lower-cased the elements and attributes along the way).  As 
you can see, there's very little difference (since you don't need to 
write the headers, there is zero difference really, aside from the need 
for lower case elements), and I just want to be able to use the xhtml 
DocBook stylesheets.

Bruce

========

<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
   <head>
     <title>
       Some example citations
     </title>
     <meta name="GENERATOR" content="Modular DocBook HTML Stylesheet 
Version 1.64">
     <link rel="HOME" title="RefDB bibliography test document" 
href="jbcbook1.html">
     <link rel="PREVIOUS" title="RefDB bibliography test document" 
href="jbcbook1.html">
     <link rel="NEXT" title="Reference List" href="jbcb24.html">
   </head>
   <body class="CHAPTER" bgcolor="#FFFFFF" text="#000000" link="#0000FF" 
vlink="#840084" alink="#0000FF">
     <div class="NAVHEADER">
       <table width="100%" border="0" cellpadding="0" cellspacing="0">
         <tr>
           <th colspan="3" align="center">
             RefDB bibliography test document
           </th>
         </tr>
         <tr>
           <td width="10%" align="left" valign="bottom">
             <a href="jbcbook1.html">Prev</a>
           </td>
           <td width="80%" align="center" valign="bottom">
           </td>
           <td width="10%" align="right" valign="bottom">
             <a href="jbcb24.html">Next</a>
           </td>
         </tr>
       </table>
       <hr align="left" width="100%">
     </div>
     <div class="CHAPTER">
       <h1>
         <a name="AEN3">Chapter 1. Some example citations</a>
       </h1>
       <p>
         This is a citation <span class="CITATION"><a 
href="jbcb24.html#ID2">(1)</a></span>.
       </p>
       <p>
         This is a multiple citation <a href="jbcb24.html#ID2">(1-3)</a>.
       </p>
       <p>
         This is a book <span class="CITATION"><a 
href="jbcb24.html#LALL-ID23">(4)</a></span> and this is a book chapter 
<span class="CITATION"><a href="jbcb24.html#LALL-ID22">(5)</a></span>.
       </p>
       <p>
         <span class="CITATION"><a href="jbcb24.html#ID2">Graves, 
Bartleson, Biorn, and Pete</a></span> reported recently <span 
class="CITATION"><a href="jbcb24.html#ID2">(1)</a></span>, that textual 
citations also work.
       </p>
     </div>
     <div class="NAVFOOTER">
       <hr align="left" width="100%">
       <table width="100%" border="0" cellpadding="0" cellspacing="0">
         <tr>
           <td width="33%" align="left" valign="top">
             <a href="jbcbook1.html">Prev</a>
           </td>
           <td width="34%" align="center" valign="top">
             <a href="jbcbook1.html">Home</a>
           </td>
           <td width="33%" align="right" valign="top">
             <a href="jbcb24.html">Next</a>
           </td>
         </tr>
         <tr>
           <td width="33%" align="left" valign="top">
             RefDB bibliography test document
           </td>
           <td width="34%" align="center" valign="top">
             &nbsp;
           </td>
           <td width="33%" align="right" valign="top">
             Reference List
           </td>
         </tr>
       </table>
     </div>
   </body>
</html>

========

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
     "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
   <head>
     <title>
       Some example citations
     </title>
     <meta name="GENERATOR" content="Modular DocBook HTML Stylesheet 
Version 1.64" />
     <link rel="HOME" title="RefDB bibliography test document" 
href="jbcbook1.html" />
     <link rel="PREVIOUS" title="RefDB bibliography test document" 
href="jbcbook1.html" />
     <link rel="NEXT" title="Reference List" href="jbcb24.html" />
   </head>
   <body class="CHAPTER" bgcolor="#FFFFFF" text="#000000" link="#0000FF" 
vlink="#840084" alink="#0000FF">
     <div class="NAVHEADER">
       <table width="100%" border="0" cellpadding="0" cellspacing="0">
         <tr>
           <th colspan="3" align="center">
             RefDB bibliography test document
           </th>
         </tr>
         <tr>
           <td width="10%" align="left" valign="bottom">
             <a href="jbcbook1.html">Prev</a>
           </td>
           <td width="80%" align="center" valign="bottom">
           </td>
           <td width="10%" align="right" valign="bottom">
             <a href="jbcb24.html">Next</a>
           </td>
         </tr>
       </table>
       <hr align="left" width="100%" />
     </div>
     <div class="CHAPTER">
       <h1>
         <a id="AEN3" name="AEN3">Chapter 1. Some example citations</a>
       </h1>
       <p>
         This is a citation <span class="CITATION"><a 
href="jbcb24.html#ID2">(1)</a></span>.
       </p>
       <p>
         This is a multiple citation <a href="jbcb24.html#ID2">(1-3)</a>.
       </p>
       <p>
         This is a book <span class="CITATION"><a 
href="jbcb24.html#LALL-ID23">(4)</a></span> and this is a book chapter 
<span class="CITATION"><a href="jbcb24.html#LALL-ID22">(5)</a></span>.
       </p>
       <p>
         <span class="CITATION"><a href="jbcb24.html#ID2">Graves, 
Bartleson, Biorn, and Pete</a></span> reported recently <span 
class="CITATION"><a href="jbcb24.html#ID2">(1)</a></span>, that textual 
citations also work.
       </p>
     </div>
     <div class="NAVFOOTER">
       <hr align="left" width="100%" />
       <table width="100%" border="0" cellpadding="0" cellspacing="0">
         <tr>
           <td width="33%" align="left" valign="top">
             <a href="jbcbook1.html">Prev</a>
           </td>
           <td width="34%" align="center" valign="top">
             <a href="jbcbook1.html">Home</a>
           </td>
           <td width="33%" align="right" valign="top">
             <a href="jbcb24.html">Next</a>
           </td>
         </tr>
         <tr>
           <td width="33%" align="left" valign="top">
             RefDB bibliography test document
           </td>
           <td width="34%" align="center" valign="top">
             &nbsp;
           </td>
           <td width="33%" align="right" valign="top">
             Reference List
           </td>
         </tr>
       </table>
     </div>
   </body>
</html>

Re: [Refdb-users] Issues and Observations

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

"Markus Hoenicka" <mar...@mh...> wrote on Thu, 28 Aug
2003 02:24:00 +0200:

>  > Oh, I see... In my case, however, by far most proceedings/journals
>  > sort the references alphabetically, so ranges like this are extremely
>  > rare. Is it really that much harder to implement?
>  >=20
> Well, it bloats the bibliography source considerably. But I'll rethink
> it if it is important to you.

Thanks, I don't think I'm the only one who thinks this way ;-)

> Well, some DB servers support other encodings as well. Does it really
> make sense to use UTF-8 unconditionally?

No, I guess not. It should probably be configurable, as the RefDB
system in general has no way to recognize the coding system of the DB
internals. UTF-8 would be a sensible default in conjunction with XML,
as that is the default encoding of XML. Ideally, however, the output
should be convertible to a desired target format (e.g., entities for
US-ASCII-HTML).

> What should the XHTML support look like? Could you provide examples?

I'm not sure what you're asking here. There's lots of docs about XHTML
out there. It is much like HTML 4 strict, but is true XML. In
practice, this means mainly that the header organisation is somewhat
different and stricter, and that there are no open tags. E.g., all
<td> must be closed by </td>, empty tags are written as <br />,
etc. It is relatively easy to write it such that it is interpreted
correctly by legacy HTML parsers (which are still used 99% of the
time, even for XHTML pages), but parsing by XML parsers opens up new
possiblities such as MathML and SVG integration.

I plan to install on-line publications pages based on RefDB soon. They
will be part of a site based on XHTML, and being able to include
RefDB-generated XHTML code directly would just make my task that much
easier :-).

From your end, it should be very easy to do, at least for the
XML/xsltproc track, as XHTML is supported natively by DocBook XML.

> Not yet. You can of course contribute anything you come up with. Or
> we'll rewarm Bruce's suggestion to create some online "store" for
> style files.

Yes! But see also his reference to Bibliophile. (Oops, Freudian typo)

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

bibliofile (was Re: [Refdb-users] Issues and Observations)

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

This sounds very interesting, it may be a good decision to concentrate
the bib-style effort in this direction.

Bruce D'Arcus <bd...@fa...> wrote on Wed, 27 Aug 2003 08:00:25
-0400:

> BTW, one possibility that I hope will work is to move to all-xslt=20
> formatting based on this framework that Peter Flynn (with comments and=20
> advice from Markus and I) has been working on.
>
> http://silmaril.ie/bibliofile/
>
> I'm really optimistic about this, and the possibility that we'll be=20
> able to process MODS records (as an example) with default DocBook (and=20
> hopefully TEI) stylesheets, complete with support for extended and=20
> footnote citations.  The idea is that just as one has a choice of table=20
> models to use in DocBook, so too would one have a choice of=20
> bibliographic support.

--=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] Issues and Observations

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

Justus H. Piater writes:
 > Oh, I see... In my case, however, by far most proceedings/journals
 > sort the references alphabetically, so ranges like this are extremely
 > rare. Is it really that much harder to implement?
 > 
Well, it bloats the bibliography source considerably. But I'll rethink
it if it is important to you.

 > Yes, it would be nice to specify a
 >   <META http-equiv="Content-Type" content="text/html; charset=UTF-8">
 > header. BTW, I would love XHTML support...
 > 

Well, some DB servers support other encodings as well. Does it really
make sense to use UTF-8 unconditionally?

What should the XHTML support look like? Could you provide examples?

 > I'm demanding lots of things here. I'll most likely be using XML/RefDB
 > more and more seriously, which will force me to tweak/write
 > bibliography styles that could be of use for others. Are there any
 > serious user-contributed styles out there?
 > 

Not yet. You can of course contribute anything you come up with. Or
we'll rewarm Bruce's suggestion to create some online "store" for
style files.

regards,
Markus

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

Re: [Refdb-users] Very strange refdb behaviour - update citekey failed (sometimes)

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

Marc Baaden writes:
 > Yes I do use this. But I have not yet managed to use emacs to edit
 > UTF-8, so I switch to another editor now and then, and that one even
 > does trim some of the existing spaces, which is annoying.
 > 

Something like Ctrl-x Ret f utf-8 should do the trick. You may then
have to select an appropriate input method to enter non-ASCII letters.

regards,
Markus

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

Re: [Refdb-users] data read error?

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

On Wednesday, August 27, 2003, at 02:32  PM, Bruce D'Arcus wrote:

> Question: Why might I get the following error?
>
> refdbc: addref -k out1.ris
> data read error. Stop.

Answer:

When input file does not start with a newline :-)

Bruce

PS - I think refdbc might have just choked on this file though.

[Refdb-users] data read error?

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

Question: Why might I get the following error?

refdbc: addref -k out1.ris
data read error. Stop.

Bruce