refdb-users — List for RefDB users

[Refdb-users] forwarded message from Markus Hoenicka

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

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

Re: [Refdb-users] Problems retrieving and inserting references

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

Hi,

Quoting "O. Hartmann" <oha...@ze...>:

> The server (refdbd) is running on a FreeBSD 8.0-CURRENT/AMD64 SMP box,
> another test facility is located on an older FreeBSD 7.2-STABLE/i386 box.

My laptop runs 7.2-STABLE as well. This should make it fairly easy to  
compare setups.

> I can add references from a file very easily as far as no UTF-8 or ISO
> specific encodings (Umlaute) are used.

This is unusual and probably a configuration problem. Before going  
into the details, I'd ask you to provide the following information:

- the versions of RefDB, libdbi, libdbi-drivers, and of your database engine
- the contents of your configuration files. Running "refdb-bug"  
retrieves your config files and a few other things. Please post the  
file created by this tool.
- what kind of reference data are you using? The /examples  
subdirectory of the source distribution contains some test data that  
should just work. If you're experiencing the problems with some kind  
of non-standard data (e.g. exported from other software), feel free to  
zip or tar.gz some examples and send them to me off-list.

> When using then a client, refdbc, I'm able to retrieve via 'getau' all
> auhors, but trying 'getref' results in a got-stuck process which I can
> only leave by hitting Ctrl-C.

That is, the client gets stuck? Or the server child process? If the  
client appears to be stuck, most likely the server child process has  
run into trouble. In that case, please set the debug level of refdbd  
to "7", truncate the server log (usually /var/log/refdbd.log), and  
restart the server. Repeat the queries which fail and post the log.

>
> I also installed the web based contributed interface, but with this, on
> no server I'm able to insert references or retrieve references. The only
> way to retrieve anything is searching by 'Search indices', 'authors'.
> There I can retrieve author names.
>

I don't recommend to fiddle with the web interface at this moment. The  
web interface calls the C clients, so if these don't work ok in the  
first place, the web interface won't either.

> I digged the database server and there the tables (most recent
> PostgreSQL), the references are accurate stored.
>
> Even with the supervisor/administration account for PostgreSQL I'm
> incapable of retrieving references.

Did you check the PostgreSQL error logs for anything unusual? I don't  
expect any permission problems though, as you seem to be able to  
create and fill the tables.

> Another problem concerns trouble when inserting references taken from
> our 'JabRef' output or BibTeX in gemeral. I use bib2ris-utf creating RIS
> files for insertion into RefDB, but something gets confused. In most
> cases, the end tag 'ER  - ' is corrupted that way that the trailing
> space gets lost.
> Another issue targets LaTeX/TeX tags, like {\'C}uk. As often special
> tags occur at the beginning of a line, say the first author's name, the
> whole entry gets corrupted, the end of line gets lost and the next line
> seems to merge with the first, faulty one.
>

This is probably due to limitations of the RefDB tools in question. It  
would be helpful if you could send me a couple of JabRef/BibTeX  
datasets, along with the datasets after running them through  
bib2ris-utf8, and with some comments on how the converted datasets  
should look like (I'm no BibTeX expert, so some conversion problems  
might not be obvious to me).

regards,
Markus

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

[Refdb-users] Problems retrieving and inserting references

[O. H. <oha...@ze...>]

Hello.

Since a while we check out RefDB as a poential backbone for our 
institute's literature and reference system.

The server (refdbd) is running on a FreeBSD 8.0-CURRENT/AMD64 SMP box, 
another test facility is located on an older FreeBSD 7.2-STABLE/i386 box.

Well, I ran into several severe problems, but I will target the first 
and possibly simple-to-fix one first.

I can add references from a file very easily as far as no UTF-8 or ISO 
specific encodings (Umlaute) are used.
When using then a client, refdbc, I'm able to retrieve via 'getau' all 
auhors, but trying 'getref' results in a got-stuck process which I can 
only leave by hitting Ctrl-C.

I also installed the web based contributed interface, but with this, on 
no server I'm able to insert references or retrieve references. The only 
way to retrieve anything is searching by 'Search indices', 'authors'. 
There I can retrieve author names.

I digged the database server and there the tables (most recent 
PostgreSQL), the references are accurate stored.

Even with the supervisor/administration account for PostgreSQL I'm 
incapable of retrieving references.

At this very moment I'm floating like a dead man in the water.

Another problem concerns trouble when inserting references taken from 
our 'JabRef' output or BibTeX in gemeral. I use bib2ris-utf creating RIS 
files for insertion into RefDB, but something gets confused. In most 
cases, the end tag 'ER  - ' is corrupted that way that the trailing 
space gets lost.
Another issue targets LaTeX/TeX tags, like {\'C}uk. As often special 
tags occur at the beginning of a line, say the first author's name, the 
whole entry gets corrupted, the end of line gets lost and the next line 
seems to merge with the first, faulty one.

It would be kind if someone could give me advice.
Thanks in advance,
Oliver

[Refdb-users] Quotes in addlink

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

Torsten Bronger writes:
 > Hallöchen!
 > 
 > If I connect to refdbd using sockets, why does the command
 > 
 >     addlink -u user -w password :NCK:=django-refdb-offprints-1 :CK:=Bld2006
 > 
 > not work, however,
 > 
 >     addlink -u user -w password ":NCK:=django-refdb-offprints-1 :CK:=Bld2006"
 > 
 > does work?
 > 

I haven't analyzed this situation in detail, but I suspect it is
caused by the use of getopt() to parse the command. Looks like getopt
attempts to interpret the characters of the query string preceded by a
dash as option characters unless they are enclosed in quotes. Let me
know if this does not explain your experiences in full, I'd then have
to have a second look.


regards,
Markus

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

[Refdb-users] Quotes in addlink

[Torsten B. <br...@ph...>]

Hallöchen!

If I connect to refdbd using sockets, why does the command

    addlink -u user -w password :NCK:=django-refdb-offprints-1 :CK:=Bld2006

not work, however,

    addlink -u user -w password ":NCK:=django-refdb-offprints-1 :CK:=Bld2006"

does work?

Tschö,
Torsten.

-- 
Torsten Bronger, aquisgrana, europa vetus
                   Jabber ID: tor...@ja...

Re: [Refdb-users] refdbctl do not stop deamon

[Wei-Wei G. <ww...@gm...>]

Markus Hoenicka 写道:

> 
> I've checked in a patch which fixes this problem, see revision 675.
> 

Thanks a lot!

Best wishes,
Wei-Wei

Re: [Refdb-users] refdbctl do not stop deamon

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

Markus Hoenicka writes:
 > Quoting Wei-Wei Guo <ww...@gm...>:
 > 
 > > After changing the variable, the refdbd can be stopped correctly. It seems
 > > refdbctl doesn't use the variable set in refdbdrc.
 > >
 > 
 > Ah, I see. I'll try to fix this asap.
 > 

I've checked in a patch which fixes this problem, see revision 675.

regards,
Markus

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

Re: [Refdb-users] refdbctl do not stop deamon

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

Quoting Wei-Wei Guo <ww...@gm...>:

> After changing the variable, the refdbd can be stopped correctly. It seems
> refdbctl doesn't use the variable set in refdbdrc.
>

Ah, I see. I'll try to fix this asap.

regards,
Markus

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

Re: [Refdb-users] refdbctl do not stop deamon

[Wei-Wei G. <ww...@gm...>]

Markus Hoenicka 写道:

> Thanks again. Both commands work as expected. Therefore the script 
> should work as well. Can you still reproduce the script failure? Also, 
> which shell did you use for these tests, i.e. what is your login shell? 
> If it is not bash, you may try repeating the above tests using bash: 
> type "bash" at your prompt, repeat the above tests, and type exit again 
> to return to your login shell. If that should give different results, 
> try editing the first line of refdbctl (something like 
> "#!/usr/bin/bash") and make it point to your login shell.
> 

I opened /usr/bin/refdbctl and happened to find the problem. the script refdbctl
set PIDFILE to:

PIDFILE='/var/lib/run/refdbd.pid'

instead of

PIDFILE='/var/run/refdbd.pid'

After changing the variable, the refdbd can be stopped correctly. It seems
refdbctl doesn't use the variable set in refdbdrc.


Thanks for help!


Best wishes,
Wei-Wei

Re: [Refdb-users] refdbctl do not stop deamon

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

Quoting Wei-Wei Guo <ww...@gm...>:

> Is this what you mean?
>
> $ ls /var/run/refdb*
> /var/run/refdbd.pid
>
> $ ps -C refdbd
>   PID TTY          TIME CMD
>  2904 ?        00:00:00 refdbd
>
> $ cat /var/run/refdbd.pid|sed 's/^ *\([0-9]*\).*/\1/'
> 2904
>
> $ ps ax | grep -i "^ *2904.*[r]efdbd" | sed 's/^ *\([0-9]*\).*/ 1/'
> 2904
>

Thanks again. Both commands work as expected. Therefore the script  
should work as well. Can you still reproduce the script failure? Also,  
which shell did you use for these tests, i.e. what is your login  
shell? If it is not bash, you may try repeating the above tests using  
bash: type "bash" at your prompt, repeat the above tests, and type  
exit again to return to your login shell. If that should give  
different results, try editing the first line of refdbctl (something  
like "#!/usr/bin/bash") and make it point to your login shell.

regards,
Markus



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

Re: [Refdb-users] refdbctl do not stop deamon

[Wei-Wei G. <ww...@gm...>]

Markus Hoenicka 写道:

> Thanks for running these tests. Both look ok, so I'm entirely clueless 
> why the script fails on your box. refdbctl uses two shell commands to 
> isolate the process ID from the ps output and from the PID file. Maybe 
> one of these commands fails on your box. Could you please test those 
> shell commands manually? You'll have to replace two variables with their 
> real values, like this (using the values "2908" as the PID and 
> "/var/run/refdbd.pid" as the full path of the PID file in this example):
> 
> cat /var/run/refdbd.pid|sed 's/^ *\([0-9]*\).*/\1/'
> 
> and
> 
> ps ax | grep -i "^ *2908.*[r]efdbd" | sed 's/^ *\([0-9]*\).*/\1/'
> 
> Both commands should return the process ID, i.e. "2908" in this example.
> 

Is this what you mean?

$ ls /var/run/refdb*
/var/run/refdbd.pid

$ ps -C refdbd
   PID TTY          TIME CMD
  2904 ?        00:00:00 refdbd

$ cat /var/run/refdbd.pid|sed 's/^ *\([0-9]*\).*/\1/'
2904

$ ps ax | grep -i "^ *2904.*[r]efdbd" | sed 's/^ *\([0-9]*\).*/ 1/'
2904


Best wishes,
Wei-Wei

Re: [Refdb-users] refdbctl do not stop deamon

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

Quoting Wei-Wei Guo <ww...@gm...>:

> There is only one line in /var/run/refdbd.pid:
>
> 2908
>
> It is the PID of refdbd.
>
> $ ps -C refdbd
>   PID TTY          TIME CMD
>  2908 ?        00:00:00 refdbd
>
>
>> Please also check the setting of the "pidfile" variable in your   
>> refdbdrc. It should have the value "/var/run/refdbd.pid" as well.
>>
>
> Here is the 'pidfile' line in refdbdrc:
>
> pidfile         /var/run/refdbd.pid
>
>

Thanks for running these tests. Both look ok, so I'm entirely clueless  
why the script fails on your box. refdbctl uses two shell commands to  
isolate the process ID from the ps output and from the PID file. Maybe  
one of these commands fails on your box. Could you please test those  
shell commands manually? You'll have to replace two variables with  
their real values, like this (using the values "2908" as the PID and  
"/var/run/refdbd.pid" as the full path of the PID file in this example):

cat /var/run/refdbd.pid|sed 's/^ *\([0-9]*\).*/\1/'

and

ps ax | grep -i "^ *2908.*[r]efdbd" | sed 's/^ *\([0-9]*\).*/\1/'

Both commands should return the process ID, i.e. "2908" in this example.

regards,
Markus

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

Re: [Refdb-users] refdbctl do not stop deamon

[Wei-Wei G. <ww...@gm...>]

Markus Hoenicka 写道:

>> # refdbctl stop
>> /usr/bin/refdbctl stop: /usr/bin/refdbd is not running
>>
>> Am I do anything wrong?
>>
> 
> Not necessarily. It might be a permission problem or a parsing problem 
> of the grep output as well. After starting refdbd using refdbctl, do you 
> see a file called "refdbd.pid" in /var/run/? If yes, could you please 
> post the contents? 

There is only one line in /var/run/refdbd.pid:

2908

It is the PID of refdbd.

$ ps -C refdbd
   PID TTY          TIME CMD
  2908 ?        00:00:00 refdbd


> Please also check the setting of the "pidfile" 
> variable in your refdbdrc. It should have the value 
> "/var/run/refdbd.pid" as well.
> 

Here is the 'pidfile' line in refdbdrc:

pidfile         /var/run/refdbd.pid


Best wishes,
Wei-Wei

Re: [Refdb-users] refdbctl do not stop deamon

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

Quoting Wei-Wei Guo <ww...@gm...>:

> # refdbctl stop
> /usr/bin/refdbctl stop: /usr/bin/refdbd is not running
>
> Am I do anything wrong?
>

Not necessarily. It might be a permission problem or a parsing problem  
of the grep output as well. After starting refdbd using refdbctl, do  
you see a file called "refdbd.pid" in /var/run/? If yes, could you  
please post the contents? Please also check the setting of the  
"pidfile" variable in your refdbdrc. It should have the value  
"/var/run/refdbd.pid" as well.

regards,
Markus

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

[Refdb-users] refdbctl do not stop deamon

[Wei-Wei G. <ww...@gm...>]

Hi Markus,

I got a new problem. refdbctl can start daemon, but cannot stop it.

# ps -C refdbd
   PID TTY          TIME CMD
  2860 ?        00:00:00 refdbd
  3255 ?        00:00:00 refdbd
# refdbctl stop
/usr/bin/refdbctl stop: /usr/bin/refdbd is not running

Am I do anything wrong?

Best wishes,
Wei-Wei

Re: [Refdb-users] Cannot add reference.

[Wei-Wei G. <ww...@gm...>]

Markus Hoenicka 写道:

> It turned out that this was not a driver problem, but a bug in the
> libdbi library itself. I've checked in a fix today. I assume you won't
> be able to wait until the next libdbi release, which may take some
> time to be picked up by package maintainers. If you feel comfortable
> with installing libdbi from the sources, please use the appended patch
> for src/dbi_main.c.
> 

Thanks, it solves the problem.

Best wishes,
Wei-Wei

Re: [Refdb-users] Cannot add reference.

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

Wei-Wei Guo writes:
 > Thanks. I'd better wait since PG or MySql is kind of heavy for me. Hope the
 > driver problem could be solved soon :>
 > 

It turned out that this was not a driver problem, but a bug in the
libdbi library itself. I've checked in a fix today. I assume you won't
be able to wait until the next libdbi release, which may take some
time to be picked up by package maintainers. If you feel comfortable
with installing libdbi from the sources, please use the appended patch
for src/dbi_main.c.

regards,
Markus

--- dbi_main.c	2008/11/28 22:28:51	1.95
+++ dbi_main.c	2009/05/08 23:22:33	1.96
@@ -1570,8 +1574,19 @@
 	/* hahaha! who woulda ever thunk strawberry's code would come in handy? */
 	// find first (highest) bit set; methods not using FP can be found at 
 	// http://graphics.stanford.edu/~seander/bithacks.html#IntegerLogObvious
-	unsigned startbit = log(rangemin)/log(2);
-	unsigned endbit = log(rangemax)/log(2);
+/* 	unsigned startbit = log(rangemin)/log(2); */
+/* 	unsigned endbit = log(rangemax)/log(2); */
+  unsigned int startbit = 0;
+  unsigned int endbit = 0;
+
+  while (rangemin >>= 1) {
+    startbit++;
+  }
+
+  while (rangemax >>= 1) {
+    endbit++;
+  }
+
 	// construct mask from startbit to endbit inclusive
 	unsigned attrib_mask = ((1<<(endbit+1))-1) ^ ((1<<startbit)-1);
 	return attribs & attrib_mask;

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

Re: [Refdb-users] Cannot add reference.

[Wei-Wei G. <ww...@gm...>]

Markus Hoenicka 写道:

Hi Markus,

Sorry for late rely. I just settled down.

> I'll investigate this driver problem in the next days. I'll let you
> know as soon as there are good news. Also, feel free to watch the
> libdbi-drivers-devel list for further news
> (http://sourceforge.net/mailarchive/forum.php?forum_name=libdbi-drivers-devel).
> 
> Unfortunately, at this time there are only two options to get RefDB up
> and running: use FreeBSD (or any other OS not affected by this bug),
> or use a different database engine.
> 

Thanks. I'd better wait since PG or MySql is kind of heavy for me. Hope the
driver problem could be solved soon :>


Best wishes,
Wei-Wei

Re: [Refdb-users] Cannot add reference.

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

Wei-Wei Guo writes:
 > The new refdbd.log has 'ulonglong' now. I should restart refdbd before run
 > the test procedure.

Yes, it is all too easy to forget this. Maybe I should have mentioned
it.

Your log indicates that this is apparently not a RefDB problem, but a
problem of the libdbi driver which you use to access the SQLite
database. We (that is, a few libdbi developers and myself) came across
a platform-dependent problem with the sqlite and sqlite3 drivers just
the other day. In a nutshell, these drivers do not handle 8 byte
integers correctly on Linux, returning 0 instead of the real
value. This is what your logs indicate as well. The id value returned
by this query:

7:pid=3251:Wed May 06 10:55:45 2009:SELECT note_id, note_user_id FROM t_note WHERE note_key='wwguo-wwguo'
7:pid=3251:Wed May 06 10:55:45 2009:ulonglong: u 0<< lu 0<< llu 0

ought to be 1, which you already confirmed by running this query in
the sqlite command line app. However, the driver returns 0 as
indicated by the debug log output.

I'll investigate this driver problem in the next days. I'll let you
know as soon as there are good news. Also, feel free to watch the
libdbi-drivers-devel list for further news
(http://sourceforge.net/mailarchive/forum.php?forum_name=libdbi-drivers-devel).

Unfortunately, at this time there are only two options to get RefDB up
and running: use FreeBSD (or any other OS not affected by this bug),
or use a different database engine.

regards,
Markus


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

Re: [Refdb-users] Cannot add reference.

[Wei-Wei G. <ww...@gm...>]

Markus Hoenicka 写道:
> 
> Thanks for repeating the test. Unfortunately I can't find the additional 
> log output from the latest patch. There should be couple of lines 
> starting with either "uint:" or "ulonglong:" providing additional clues. 
> Could you please retry patching the sources with dbfncs.c.patch?
> 

The new refdbd.log has 'ulonglong' now. I should restart refdbd before run
the test procedure.


Best wishes,
Wei-Wei

Re: [Refdb-users] Cannot add reference.

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

Quoting Wei-Wei Guo <ww...@gm...>:

> Markus Hoenicka 写道:
>> I've checked your log output. I've designed another small patch to
>> drill a bit deeper. Please apply it in addition to the previous patch
>> and rebuild RefDB again. The procedure you used last time was
>> apparently successful.
>>
>
> I attached the newly produced refdbd.log with the same procedure.

Thanks for repeating the test. Unfortunately I can't find the  
additional log output from the latest patch. There should be couple of  
lines starting with either "uint:" or "ulonglong:" providing  
additional clues. Could you please retry patching the sources with  
dbfncs.c.patch?

thanks
Markus

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

Re: [Refdb-users] Cannot add reference.

[Wei-Wei G. <ww...@gm...>]

Markus Hoenicka 写道:
> I've checked your log output. I've designed another small patch to
> drill a bit deeper. Please apply it in addition to the previous patch
> and rebuild RefDB again. The procedure you used last time was
> apparently successful.
> 

I attached the newly produced refdbd.log with the same procedure.


> BTW which type of processor do you use? Is it a non-x86, by any
> chance?
> 

My processor is Intel Pentium 4.


Best wishes,
Wei-Wei

Re: [Refdb-users] RISX vs. RIS

[Torsten B. <br...@ph...>]

Hallöchen!

Markus Hoenicka writes:

> [...]
>
> The mapping is not documented at this time except in the source
> code. The tricky thing to understand is that RIS uses weird rules
> to assign authors and titles to the A1-A3 and T1-T3 fields,
> whereas risx attempts to use a much saner analytical=1,
> monographic=2, series=3 approach. The other risx elements map to
> RIS fields according to the RefMan table in a fairly
> straightforward fashion.

I think I can cope with RISX' semantics quite well now.  I just want
to grey out some fields sometimes and rename them (e.g. "article
title" -- "journal title" for journal articles).  I think this helps
my collegues a lot because analytical vs. monographic must be learnt
by them first.  For this renaming, the tables in the RIS
specification were helpful.

What's missing is "edition" in RISX.  One can use the "otherinfo"
for the publishing date but this doesn't survive in BibTeX output.
If you use "edition" with RIS however, RefDB creates an "edition"
field for BibTeX.

Additionally, the explanation of "city" should be expanded to
"publication place" and not just the name of a town.  In case of
patents, for example, it's the name of a country.  (The tag name may
stay, however.)

Tschö,
Torsten.

-- 
Torsten Bronger, aquisgrana, europa vetus
                   Jabber ID: tor...@ja...

Re: [Refdb-users] RISX vs. RIS

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

Torsten Bronger writes:
 > I investigated further and now I'm slightly confused.  The RefDB
 > manual says that RISX should be prefered for communicating with the
 > database, and even that it should be used for backup dumps.
 > However, it can't hold all the data in the database.  For example,
 > if I store a book reference, there is a "volume" and an "edition"
 > field, however in RISX, only "volume" survives.  In other words, the
 > RISX dump is not lossless.  An RIS dump contains all fields, though.
 > 

If a risx dump does not contain all data, the risx backend must be
considered broken. Could you please send some test data which help me
reproduce the problem?

thanks
Markus

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

[Refdb-users] RISX vs. RIS

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

Torsten Bronger writes:
 > The data model of my front-end is RISX.  Where can I get information
 > about how to interpret the RISX fields for the different types?
 > http://refdb.sourceforge.net/risx/index.html is of some help but it
 > is rather limited.  The most comprehensive tables can be found in
 > <http://www.referencemanager.com/support/docs/ReferenceManager11.pdf>,
 > pp 525 ff.  But this makes is necessary to map RISX to RIS.  Is
 > there information about this?
 > 

The mapping is not documented at this time except in the source
code. The tricky thing to understand is that RIS uses weird rules to
assign authors and titles to the A1-A3 and T1-T3 fields, whereas risx
attempts to use a much saner analytical=1, monographic=2, series=3
approach. The other risx elements map to RIS fields according to the
RefMan table in a fairly straightforward fashion.

I guess you're asking for an even larger table which shows the mapping
between real world items, RIS fields, and risx elements. I realize
that this information is helpful, but I won't be able to cough it up
in a matter of days.

 > By the way, RIS seems to have four titles (primary, secondary,
 > series, periodical), whereas RISX has only three of them.  In case
 > of conferences, RIS uses all four (see page 527).  How is this
 > mapped to RISX?
 > 

risx currently does not handle four titles, as I was not able to
figure out the difference between "series title" and "proceedings
title". With my limited knowledge about conference proceedings I
concluded these are the same, but I'm willing to learn.

regards,
Markus

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