refdb-users

[Refdb-users] difficulty with refdba

[Matt P. <mat...@ut...>]

Hi there,

u, I'm having a little trouble getting refdba to work with the current
refdb prerelease.  I'm using mysql and and had no trouble creating the
refdb1 database; refdbd starts up fine; and I can connect just fine to
refdb1 using the mysql command-line:

mysql -u matt refdb1 -p

I've edited /etc/refdbarc to include my user name and password, and
I've looked at refdbdrc and everything looks ok there (though I'm not
absolutely sure which port my mysql server runs on, I haven't messed
with it so I assume that's not a problem.  

refdbd starts up fine, fand using the switches suggested in the manual
for installaiton testing, I get the following output: 

[root@anarres<520>/etc/refdb]$ refdbd -s -e 0 -l 7
dbi_driver_dir went to:

dbi is up using default driver dir
Available libdbi database drivers:
sqlite
mysql
pgsql
application server started
use /tmp/refdbd_fifo15971 as fifo
server waiting n_max_fd=4

---------------------
I then switch to another terminal to run refdba

 refdba -u matt
Please enter your password: 
refdba: set
serverip 127.0.0.1
port 9734
verbose f
pager stdout
username matt
timeout 60
logfile /var/log/refdba.log
logdest 2
loglevel 6
refdblib 
refdba: viewstat
could not connect to database server

-----------------------
looking in the refdbd window, I see: 

parent removing client on fd 5
server waiting n_max_fd=4
try to read from client
serving client on fd 5 with protocol version 1
201-89-34-90
send pseudo-random string to client
viewstat  -u matt -w 093092109054059125064105
dbi is up
localhost
matt
<password deleted>

3306
mysql

refdb
failed to connect to database server
command processing done, finish dialog now
child finished client on fd 5
child exited with code 0
server waiting n_max_fd=4
-------------------------

obviously there's some problem with the connectin, but I'm not sure
how to figure out what it is.  mysql seems to be up and accepting tcp
requests:

$ netstat -tlnp  | grep -i mysql 
tcp        0      0 0.0.0.0:3306 0.0.0.0:* LISTEN     1486/mysqld      

... so I assume the problem is somewhere in my refdb set up.  I know
this should be simple, but if someone can help me with debugging &c
I'd be grateful!  thanks,

matt

[Refdb-users] difficulty with refdba

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

Matt Price writes:
 > mysql -u matt refdb1 -p
[...]
 > ... so I assume the problem is somewhere in my refdb set up.  I know
 > this should be simple, but if someone can help me with debugging &c
 > I'd be grateful!  thanks,

Your analysis is impressive. I hope you won't bang your head against
your monitor after you find out the reason. Due to a change in the
database schema of the system database I had to rename it to
refdb. This way you can run a release version and a prerelease in
parallel if you need to. All you need to do is to create the system
database as mentioned in the manual, but make sure to name it "refdb"
(this was announced on the list and is mentioned in the file
UPGRADING). I apologize for the inconvenience.

regards,
Markus

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

Re: [Refdb-users] difficulty with refdba

[Matt P. <mat...@ut...>]

On Tue, Jan 20, 2004 at 12:23:39AM +0100, Markus Hoenicka wrote:
> Matt Price writes:
>  > mysql -u matt refdb1 -p
> [...]
>  > ... so I assume the problem is somewhere in my refdb set up.  I know
>  > this should be simple, but if someone can help me with debugging &c
>  > I'd be grateful!  thanks,
> 
> Your analysis is impressive. I hope you won't bang your head against
> your monitor after you find out the reason. Due to a change in the
> database schema of the system database I had to rename it to
> refdb. This way you can run a release version and a prerelease in
> parallel if you need to. All you need to do is to create the system
> database as mentioned in the manual, but make sure to name it "refdb"
> (this was announced on the list and is mentioned in the file
> UPGRADING). I apologize for the inconvenience.

gaah, sorry, didn't read upgrading so carefully as I haven't REALLY 
been using refdb the last little while...   
 
So about 9 months ago I was hoping to start a project to make a
web-accessible shared bibliography for use in my classes.  Now, with
the classes mostly already finished...  I'm trying to start
again.  I I've got refdb running now, so I have a somewhat better idea of
how it works...  very impressive, very sophisticated.  I was wondering
whether it's possible to modify the output of getref -t html... I've
looked at the css, that does help.  But mostly I'd like to be able to
remove the ID number & some of the paragraph markings, and to put the
output into table form (each reference in its own row table cell, &
followed by a couple of buttons "edit this reference" "comment on this
reference").  I've written some clumsy php which does this already,
but I haven't got the energy to deal with complex reference types so
it would be way better to use refdb if I can.  

  I just wondered whether there was a straightforward way to do this
  or not...  I suppose it would be possible to parse the output using
  perl or something, but it somehow seems cleaner to be able to choose
  the fields that actually print out...

anyway, hope this question is cohrent.  getting late here!  alles
beste,
matt

> 
> regards,
> Markus
> 

Re: [Refdb-users] difficulty with refdba

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

Matt Price writes:
 > how it works...  very impressive, very sophisticated.  I was wondering
 > whether it's possible to modify the output of getref -t html... I've
 > looked at the css, that does help.  But mostly I'd like to be able to
 > remove the ID number & some of the paragraph markings, and to put the
 > output into table form (each reference in its own row table cell, &
 > followed by a couple of buttons "edit this reference" "comment on this
 > reference").  I've written some clumsy php which does this already,
 > but I haven't got the energy to deal with complex reference types so
 > it would be way better to use refdb if I can.  
 > 
 >   I just wondered whether there was a straightforward way to do this
 >   or not...  I suppose it would be possible to parse the output using
 >   perl or something, but it somehow seems cleaner to be able to choose
 >   the fields that actually print out...
 > 

There's a couple of options, actually:

- customize (or copy+edit) backend-html.c until it does what you
  need. This is fairly easy, but you'll have to apply your patches to
  each new release (and pray the patches *do* apply...)

- use refdbc to retrieve the datasets as db31x, teix, risx, or
  xhtml. All of these XML output formats are suitable to be
  transformed by a nice XSLT stylesheet to whatever format you desire,
  including straight HTML.

- use the Perl client module and write your custom client in
  Perl. This could use e.g. the risx output and mangle it until you
  like the results. Perl should have an interface to libxslt, so you
  could do the XML->HTML transformation transparently to the user.

If you go down any of these paths, I'm sure the readers of this list
will be interested in the results.

hope this helps
Markus

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