refdb-users — List for RefDB users

Re: [Refdb-users] trying to get refdb up and running on Mac OSX 10.3

[Christian W. <wi...@ka...>]

"Bruce D'Arcus" <bd...@fa...> writes:

> On Jan 28, 2004, at 9:30 PM, Christian Wittern wrote:
>
>> It seems to have problems with dbi, which is the latest package from 
>> fink: 0.70-1

Argg.  Thanks for pointing this out.  I updated only three days ago and there are already new
versions.

>
>
> So, I'd try updating that first.

Which I did and compiled now without problems.  Hurra!

> I recall I had problems figuring out how to get this working with my 
> non-Fink Postgre installation (and I couldn't figure out how to get my 
> Fink Postgre installation working correctly!).

Well, it seems my pg is working fine now, the only modification I made
was grabbing a startup script from the Web and adopting that.

Thank you all for your help,

Christian 


-- 

 Christian Wittern 
 Institute for Research in Humanities, Kyoto University
 47 Higashiogura-cho, Kitashirakawa, Sakyo-ku, Kyoto 606-8265, JAPAN

Re: [Refdb-users] html output

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

On Tue, Jan 27, 2004 at 07:53:46PM -0500, Bruce D'Arcus wrote:
> 
> On Jan 27, 2004, at 7:32 PM, Marc Herbert wrote:
> 
> >>>While I've not experimented with it, I suppose it would offer even 
> >>>more
> >>>flexibility to use spans throughout, such as where you're currently
> >>>using p tags...
> >>
> >>What is the particular advantage of this approach?
> >
> >I guess the motivation here is: <span>s are completely neutral, they
> >do not carry any meaning at all, thus they offer more "flexibility":
> >the CSS can do whatever it wants with them.
> >On the other hand, <p> mean "this is a paragraph", which may not be
> >wanted. Correct me if I that was not what you meant.
> 
> It all really depends of what we feel is the best way to render web 
> output.  Using p tags suggest one way, but spans another (more suitable 
> to the example that Matt posted).  I'm working on xslt rendering of 
> this stuff, so I'd appreciate feedback on what people feel is ideal 
> here...


Just wanted to say that I've been enjoying this conversation, and to
apologize for dropping out a bit (I tend to do that, since this
project isn't something I'm SUPPOSED to be working on).  Bruce, I will
look more closely at your suggestions and try to replicate your
suggestions.  Your approach, I see, is to put as much of the
formatting information as possible into the css, to make the html
output more flexible...  Obviously this is the preferred approach.  

I had something else to say but actually I think I will go to bed
now.  More in a couple of days!  best,
m

> 
> Bruce
> 
> 
> 
> The SF.Net email is sponsored by EclipseCon 2004
> Premiere Conference on Open Tools Development and Integration
> See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
> http://www.eclipsecon.org/osdn
> _______________________________________________
> Refdb-users mailing list
> Ref...@li...
> https://lists.sourceforge.net/lists/listinfo/refdb-users
> 

Re: [Refdb-users] php frontend to refdb

[rob c. <ro...@st...>]

> would love to see the code, just to know what other php-oriented users
> are doing.  

here ya go:

    http://junk.lib.muohio.edu/refdb/refdb.tar.gz

its also linked in the top right corner of the live demo:

    http://junk.lib.muohio.edu/refdb/

strictly proof-of-concept, and nothing i'm particularly attached to, so
recommendations welcome...

Re: [Refdb-users] trying to get refdb up and running on Mac OSX 10.3

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

On Jan 28, 2004, at 9:30 PM, Christian Wittern wrote:

> It seems to have problems with dbi, which is the latest package from 
> fink: 0.70-1

On my Fink setup I have:

      libdbi           0.7.2-1      Database Independent Abstraction 
Layer for C
  i   libdbi-dev       0.7.2-1      Development package for libdbi
      libdbi-drive...  0.7.1-1      MySQL driver for libdbi
      libdbi-drive...  0.7.1-1      PostgreSQL driver for libdbi
  i   libdbi-drive...  0.7.1-1      SQLite driver for libdbi
  i   libdbi-shlibs    0.7.2-1      Shared libraries for libdbi

Probably the latest is in the unstable branch, and you have to 
configure Fink to access that (don't remember how).

So, I'd try updating that first.

I recall I had problems figuring out how to get this working with my 
non-Fink Postgre installation (and I couldn't figure out how to get my 
Fink Postgre installation working correctly!).

Bruce

Re: [Refdb-users] php frontend to refdb

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

On Wed, Jan 28, 2004 at 11:58:38AM -0500, rob caSSon wrote:
> >  > so, I will take another look at the getref docs and see what kind of
> >  > formats it can write to.  hopefully it will be pretty easy to pass its
> >  > output back to php as an array... or I'll see about referencing the
> >  > perclient stuff from php.  I think I would just need to write a little
> >  > perl script that accepts an array, passes it to refdb, then grabs
> >  > something else back...  shouldn't be impossible, anyway.
> 
> i should have announced this a few days ago, but now seems like a good
> time....
> 
> i've been playing with a php frontend to refdb for a couple of days
> now...nothing fancy, but its a proof of concept...
> 
> right now, its only a search interface, and it interacts with the
> command line client...here's a URL if anyone's interested....bruce has
> helped a bunch with the xslt/css stuff:
> 
>     http://junk.lib.muohio.edu/refdb/
> 
> anyway, take a whack at it, send suggestions, etc....if you want the
> source, just let me know...its pretty trivial stuff at this point, and
> still needs some authentication, input verification, entry facilities...

would love to see the code, just to know what other php-oriented users
are doing.  

thx,
matt

> 
> rob casson
> miami university
> oxford, oh
> 
> 
> 
> The SF.Net email is sponsored by EclipseCon 2004
> Premiere Conference on Open Tools Development and Integration
> See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
> http://www.eclipsecon.org/osdn
> _______________________________________________
> Refdb-users mailing list
> Ref...@li...
> https://lists.sourceforge.net/lists/listinfo/refdb-users
> 

[Refdb-users] trying to get refdb up and running on Mac OSX 10.3

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

Christian Wittern writes:
 > 
 > Hello there, 
 > 
 > I guess this is not specific to the Mac, but anyway, here is my
 > problem.  I installed refdb 093 according to the instructions, using
 > Postgresql as the database.  I figured out everything, installed the
 > database refdb1, and get to the point where I am using refdba to call
 > the database.  On a "viewstat" command, all I get is "could not
 > connect to database server".  In another terminal, where refdbd is
 > running in debug mode, I see the same message.
 > 
 > If I use pgsql on the command-line, however, I am able to connect to
 > the database and issue SQL commands without a problem, so pg_hba.conf
 > and friends should not cause problems.
 > 

Would you mind posting the refdbd log (beware of plain-text passwords
though)? This should tell us what database refdbd tries to open and
which username/password combo it uses. You should also have a look at
the pgsql log (usually /var/log/pgsql) to see whether postmaster
refuses the connection or whether refdbd can't find it. In that case
you'll have to review the settings in refdbdrc or your command line
options, especially the stuff that sets the database engine, the port,
and the IP address of the database server.

hope this helps
Markus

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

Re: [Refdb-users] php frontend to refdb

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

Er, actually, the problem seems to be in the author handling (the extra 
; in places).  Keywords are fine (though still need a css def anyway).

Bruce

Re: [Refdb-users] php frontend to refdb

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

On Jan 28, 2004, at 11:58 AM, rob caSSon wrote:

> right now, its only a search interface, and it interacts with the
> command line client...here's a URL if anyone's interested....bruce has
> helped a bunch with the xslt/css stuff..

Oops, just noticed a few problems (probably my fault):

1)  Each record needs to be wrapped in a <div class="record"> element.

2)  The citekey needs to be dropped from the container title

3)  The "Keyword: " heading is not being displayed, so likely needs a 
css definition, maybe:

span.heading {
	font-weight: bold;
}

[Refdb-users] php frontend to refdb

[rob c. <ro...@st...>]

>  > so, I will take another look at the getref docs and see what kind of
>  > formats it can write to.  hopefully it will be pretty easy to pass its
>  > output back to php as an array... or I'll see about referencing the
>  > perclient stuff from php.  I think I would just need to write a little
>  > perl script that accepts an array, passes it to refdb, then grabs
>  > something else back...  shouldn't be impossible, anyway.

i should have announced this a few days ago, but now seems like a good
time....

i've been playing with a php frontend to refdb for a couple of days
now...nothing fancy, but its a proof of concept...

right now, its only a search interface, and it interacts with the
command line client...here's a URL if anyone's interested....bruce has
helped a bunch with the xslt/css stuff:

    http://junk.lib.muohio.edu/refdb/

anyway, take a whack at it, send suggestions, etc....if you want the
source, just let me know...its pretty trivial stuff at this point, and
still needs some authentication, input verification, entry facilities...

rob casson
miami university
oxford, oh

Re: [Refdb-users] trying to get refdb up and running on Mac OSX 10.3

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

On Jan 27, 2004, at 9:27 PM, Christian Wittern wrote:

> I guess this is not specific to the Mac,

Actually, it probably is ...

> but anyway, here is my problem. I installed refdb 093 according to the 
> instructions, using Postgresql as the database.

Try using 0.9.4-pre5 instead.  Both libdbi and refdb needed a lot of 
changes to compile correctly on Mac OS X, so it's critical you're using 
the most up-to-date version.  If you have further problems, it may be 
related to libdbi, for which there is a Fink package that works 
correctly and is up-to-date.

Bruce

[Refdb-users] trying to get refdb up and running on Mac OSX 10.3

[Christian W. <wi...@ka...>]

Hello there, 

I guess this is not specific to the Mac, but anyway, here is my
problem.  I installed refdb 093 according to the instructions, using
Postgresql as the database.  I figured out everything, installed the
database refdb1, and get to the point where I am using refdba to call
the database.  On a "viewstat" command, all I get is "could not
connect to database server".  In another terminal, where refdbd is
running in debug mode, I see the same message.

If I use pgsql on the command-line, however, I am able to connect to
the database and issue SQL commands without a problem, so pg_hba.conf
and friends should not cause problems.

I am a bit at a loss at what to do next to trace the problem.  BTW,
most of the stuff is from fink, but I installed postgresql 7.4.1 from
source.

I hope somebody can help,

All the best,

Christian



-- 

 Christian Wittern 
 Institute for Research in Humanities, Kyoto University
 47 Higashiogura-cho, Kitashirakawa, Sakyo-ku, Kyoto 606-8265, JAPAN

Re: [Refdb-users] html output

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

On Jan 27, 2004, at 7:32 PM, Marc Herbert wrote:

>>> While I've not experimented with it, I suppose it would offer even 
>>> more
>>> flexibility to use spans throughout, such as where you're currently
>>> using p tags...
>>
>> What is the particular advantage of this approach?
>
> I guess the motivation here is: <span>s are completely neutral, they
> do not carry any meaning at all, thus they offer more "flexibility":
> the CSS can do whatever it wants with them.
> On the other hand, <p> mean "this is a paragraph", which may not be
> wanted. Correct me if I that was not what you meant.

It all really depends of what we feel is the best way to render web 
output.  Using p tags suggest one way, but spans another (more suitable 
to the example that Matt posted).  I'm working on xslt rendering of 
this stuff, so I'd appreciate feedback on what people feel is ideal 
here...

Bruce

Re: [Refdb-users] html output

[Marc H. <mar...@fr...>]

On Tue, 27 Jan 2004, Markus Hoenicka wrote:

>
>  > While I've not experimented with it, I suppose it would offer even more
>  > flexibility to use spans throughout, such as where you're currently
>  > using p tags...
>  >
>
> What is the particular advantage of this approach?
>

I guess the motivation here is: <span>s are completely neutral, they
do not carry any meaning at all, thus they offer more "flexibility":
the CSS can do whatever it wants with them.
On the other hand, <p> mean "this is a paragraph", which may not be
wanted. Correct me if I that was not what you meant.


BTW: I find that <http://www.htmlhelp.org/> is kind of the ultimate
HTML reference. The precision of the standard, but translated in human
language. Featuring, among others, deep hyperlinks into the standards.
Enjoy.

[Refdb-users] html output

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

Bruce D'Arcus writes:
 > ...it would offer more CSS formatting flexibility if you had each 
 > record wrapped in a div element, e.g.:
 > 

this is no biggie. I'll do just that.

 > Also, those <em> elements probably ought to be span.  That way the CSS 
 > definition could say whether it was italicized or not.
 > 

This works with <em> just as well. Have a look at refdb.css:

em.periodical 	{	font-family: Verdana, Arial, Helvetica, sans-serif;
					font-size: 80%;
					font-style: italic;
					color: #003366;
					background-color: transparent;
				}				

em.volume 	{	font-family: Verdana, Arial, Helvetica, sans-serif;
					font-size: 80%;
					font-weight: bold;
					color: #003366;
					background-color: transparent;
				}				

 > While I've not experimented with it, I suppose it would offer even more 
 > flexibility to use spans throughout, such as where you're currently 
 > using p tags...
 > 

What is the particular advantage of this approach?

regards,
Markus

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

[Refdb-users] Re: forwarded message from Markus Hoenicka

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

Diwaker Gupta writes:
 > Hi Markus,
 > 
 > I'm running MySQL 4.0.17 right now. And I had already mentioned in my 
 > mail that I've set the fromencoding variable in refdbcrc to ISO-8859-1. 
 > But somehow getref just doesn't use it.
 > 

getref should use the toencoding variable, if at all.

 > I've also been facing some problem with xnotes. I added the first note 
 > using an xml input file. Then I made another xml file with 2 xnotes in 
 > one xnoteset. While trying to add it, I got a segmentation fault. Since 
 > then, I havent been able to add any notes. addnote does not show 
 > consistent behavior -- sometimes it segfaults, at other times, it 
 > returns saying 0 notes added.
 > 
 > I checked the refdbd logs, and they just show this: (i'll try later with 
 > more verbose logging turned on):
 > 

This is real bad. I was not able to cause any segfault with xnotes
over here, but this may depend on the kind of data. I'd greatly
appreciate if you could send both the more verbose log and example
xnotes that cause refdbd to crash.

thanks
Markus

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

[Refdb-users] forwarded message from Markus Hoenicka

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

[apologies if it's a duplicate, but the list apparently swallowed
yesterday's posts]

[Refdb-users] Re: forwarded message from Markus Hoenicka

[Diwaker G. <dg...@cs...>]

Hi Markus,

I did some experiments:

o The default encoding of MySQL databases is latin5 (ISO-8859-1). 
Apparently, there isn't any way to set the default encoding to UTF-8 -- 
atleast I couldnt find any such encoding among the charsets that ship 
with MySQL

o The default encoding of the database is also ISO-8859-1. I have also 
set the encoding and fromencoding option in refdbcrc to ISO-8859-1. 
However, without giving -E option, I still get "cannot set conversion 
descriptor"

o With the -E option, I can get the output in any encoding.

Hope that helps,
Diwaker

Markus Hoenicka wrote:
> [apologies if it's a duplicate, but the list apparently swallowed
> yesterday's posts]
> 
> 
> 


-- 
Diwaker Gupta
Graduate Student, Computer Sc. and Engg.
University of California, San Diego
<http://www.cse.ucsd.edu/users/dgupta>

[Refdb-users] Re: forwarded message from Markus Hoenicka

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

Hi Diwaker,

thanks for running the experiments. The online docs tell me that
character encoding support was greatly enhanced in MySQL 4.1, so it is
likely that there is no Unicode support yet in your version.

Your experiments show that there is some problem in the program
logic. getref should return whatever encoding the database uses unless
you use the -E option. As a temporary workaround, set the
"fromencoding" variable in refdbcrc to your preferred output
encoding. This way you don't have to specify the -E option in every
getref call.

regards,
Markus

Diwaker Gupta writes:
 > Hi Markus,
 > 
 > I did some experiments:
 > 
 > o The default encoding of MySQL databases is latin5 (ISO-8859-1). 
 > Apparently, there isn't any way to set the default encoding to UTF-8 -- 
 > atleast I couldnt find any such encoding among the charsets that ship 
 > with MySQL
 > 
 > o The default encoding of the database is also ISO-8859-1. I have also 
 > set the encoding and fromencoding option in refdbcrc to ISO-8859-1. 
 > However, without giving -E option, I still get "cannot set conversion 
 > descriptor"
 > 
 > o With the -E option, I can get the output in any encoding.
 > 
 > Hope that helps,
 > Diwaker
 > 
 > Markus Hoenicka wrote:
 > > [apologies if it's a duplicate, but the list apparently swallowed
 > > yesterday's posts]
 > > 
 > > 
 > > 
 > 
 > 
 > -- 
 > Diwaker Gupta
 > Graduate Student, Computer Sc. and Engg.
 > University of California, San Diego
 > <http://www.cse.ucsd.edu/users/dgupta>
 > 

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

Re: [Refdb-users] Problems compiling pre4

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

Hi Mike,

you may have noticed that the SF lists are broke currently. I'm not
sure whether my previous replies ever made it to your inbox. In any
case, I had a second look at your Cygwin log output and noticed
something pretty strange:

refdbdnote.o(.text+0x5926):refdbdnote.c: undefined reference to `_dbi_conn_get_e
ncoding'
refdbdnote.o(.text+0x599c):refdbdnote.c: undefined reference to `_libiconv_close
'

The first line is ok and looks as expected if a library function is
missing (you have an old libdbi version). The second line should
complain about an undefined reference to "_iconv_close", not
"_libiconv_close". However I could not find any hint in the Cygwin
mailing lists about any oddities of libiconv on this platform. I'll
check myself on Monday when I'll get my hands on a Windoze box again.

I also ran another test on a Debian box. Linux apparently has the
iconv stuff built into libc, that's why it is not necessary to link
against libiconv. RefDB works allright on Debian 3.0 for me.

regards,
Markus

Michael Smith writes:
 > Markus,
 > 
 > I just updated by sandboxes and am now getting similar "undefined
 > reference to `_libiconv_close'" build errors on Cygwin, along with an
 > "undefined reference to `_dbi_conn_get_encoding'" on both Cygwin and
 > Debian ('make' logs attached).
 > 
 > On Cygwin, looks like the libiconv library and headers are in the normal
 > lib and include directories -
 > 
 >   /lib/libiconv.a
 >   /lib/libiconv.dll.a
 >   /lib/libiconv.la
 >   /usr/lib/libiconv.a
 >   /usr/lib/libiconv.dll.a
 >   /usr/lib/libiconv.la
 > 
 >   /usr/include/iconv.h
 > 
 > I shouldn't need to tell make to go looking for them there, right?
 > 

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

Re: [Refdb-users] ANN: refdb-0.9.4-pre4 available for testing

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

OK, below's what I get at the end of the build process.

ld: warning prebinding disabled because of undefined symbols
ld: Undefined symbols:
_dbi_conn_get_encoding
_libiconv
_libiconv_close
_libiconv_open
make[2]: *** [refdbd] Error 1
make[1]: *** [all] Error 2
make: *** [all-recursive] Error 1

Here's my config command:

./configure --with-libdbi-lib=/usr/local/lib/libdbi.0.0.5.dylib 
--with-expat-lib=/sw/lib 
--with-docbook-xsl=/Users/darcusb/Stylesheets/XSLT/docbook/docbook-xsl 
--with-db-server=sqlite 
--with-tei-xsl=/Users/darcusb/Stylesheets/XSLT/TEI CFLAGS="-L/sw/lib 
-I/sw/include"

What going on?   Maybe it has something to do with this, but I've no 
idea how to fix it:

	checking for library containing iconv... no

I have these files in /sw/lib:

-rw-r--r--  1 root  admin  1112828  3 Dec 10:03 libiconv.2.2.0.dylib
-rw-r--r--  1 root  admin  1112828  3 Dec 10:01 libiconv.2.dylib
lrwxr-xr-x  1 root  admin       20  3 Dec 09:59 libiconv.dylib -> 
libiconv.2.2.0.dylib
-rw-r--r--  1 root  admin      785 13 Nov 16:58 libiconv.la

Bruce

Re: [Refdb-users] Problems compiling pre4

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

Hi Mike,

you're facing two problems:

- you're using an outdated libdbi version. The whole character
  conversion stuff relies on a libdbi function that I've added to the
  latest release only. You'll have to install libdbi-0.7.2 and
  libdbi-drivers-0.7.1 before RefDB will compile.

- for some reason, configure does not perform the libiconv test
  properly. The ld commands show that there's no attempt to link
  against libiconv, hence the unresolved iconv* function calls. Could
  you please check what configure says about libiconv?

I've tested the latest prerelease on FreeBSD 4.7 and Debian 3.0
without any problems. I hope we can resolve these issues on the other
platforms asap.

regards,
Markus


Michael Smith writes:
 > Markus,
 > 
 > I just updated by sandboxes and am now getting similar "undefined
 > reference to `_libiconv_close'" build errors on Cygwin, along with an
 > "undefined reference to `_dbi_conn_get_encoding'" on both Cygwin and
 > Debian ('make' logs attached).
 > 
 > On Cygwin, looks like the libiconv library and headers are in the normal
 > lib and include directories -
 > 
 >   /lib/libiconv.a
 >   /lib/libiconv.dll.a
 >   /lib/libiconv.la
 >   /usr/lib/libiconv.a
 >   /usr/lib/libiconv.dll.a
 >   /usr/lib/libiconv.la
 > 
 >   /usr/include/iconv.h
 > 
 > I shouldn't need to tell make to go looking for them there, right?
 > 

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

Re: [Refdb-users] cannot add records!

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

Matt Price writes:
 > so, I will take another look at the getref docs and see what kind of
 > formats it can write to.  hopefully it will be pretty easy to pass its
 > output back to php as an array... or I'll see about referencing the
 > perclient stuff from php.  I think I would just need to write a little
 > perl script that accepts an array, passes it to refdb, then grabs
 > something else back...  shouldn't be impossible, anyway.
 > 

The perlclient module actually returns just about everything as an
array, ready for further processing. However, I still don't know
whether you can weld Perl modules onto your PHP code, or whether it is
trivial to convert Perl modules to something PHP can grok.

regards,
Markus

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

[Refdb-users] Re: forwarded message from Markus Hoenicka

[Diwaker G. <dg...@cs...>]

Hi Markus,

I'm running MySQL 4.0.17 right now. And I had already mentioned in my 
mail that I've set the fromencoding variable in refdbcrc to ISO-8859-1. 
But somehow getref just doesn't use it.

I've also been facing some problem with xnotes. I added the first note 
using an xml input file. Then I made another xml file with 2 xnotes in 
one xnoteset. While trying to add it, I got a segmentation fault. Since 
then, I havent been able to add any notes. addnote does not show 
consistent behavior -- sometimes it segfaults, at other times, it 
returns saying 0 notes added.

I checked the refdbd logs, and they just show this: (i'll try later with 
more verbose logging turned on):

6:pid=19043:Sat Jan 24 08:10:48 2004:serving client on fd 6 with 
protocol version 1
6:pid=19043:Sat Jan 24 08:10:48 2004:dbi is up
4:pid=19043:Sat Jan 24 08:10:48 2004:insert into t_note failed
6:pid=19043:Sat Jan 24 08:10:48 2004:chunk added successfully
6:pid=19043:Sat Jan 24 08:10:48 2004:no IDs found for keyword scan
6:pid=19043:Sat Jan 24 08:10:48 2004:child finished client on fd 6
6:pid=18017:Sat Jan 24 08:10:48 2004:parent removing client on fd 6
6:pid=18017:Sat Jan 24 08:10:48 2004:server waiting n_max_fd=5
6:pid=18017:Sat Jan 24 08:10:48 2004:child exited with code 0
6:pid=18017:Sat Jan 24 08:10:48 2004:server waiting n_max_fd=5
6:pid=18017:Sat Jan 24 08:10:57 2004:server exited gracefully

I dont know if thats any help though.

Diwaker

Markus Hoenicka wrote:
> Hi Diwaker,
> 
> thanks for running the experiments. The online docs tell me that
> character encoding support was greatly enhanced in MySQL 4.1, so it is
> likely that there is no Unicode support yet in your version.
> 
> Your experiments show that there is some problem in the program
> logic. getref should return whatever encoding the database uses unless
> you use the -E option. As a temporary workaround, set the
> "fromencoding" variable in refdbcrc to your preferred output
> encoding. This way you don't have to specify the -E option in every
> getref call.
> 
> regards,
> Markus

-- 
Diwaker Gupta
Graduate Student, Computer Sc. and Engg.
University of California, San Diego
<http://www.cse.ucsd.edu/users/dgupta>

Re: [Refdb-users] cannot add records!

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

Bruce D'Arcus writes:
 > The underling HTML code is a little ugly, but the outward appearance is 
 > nice enough.  All you really need is a way to get back some (X)HTML 
 > code that has formatted references, each of them wrapped in a div 
 > element with a class attribute.  That way you can do all the styling 
 > with CSS.  I'm even willing to help there if need be.  I've actually 

Could you tell me what the HTML output should look like to improve the
styling with CSS? It should not be too hard to fix backend-html then.

regards,
Markus

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

[Refdb-users] html output

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

On Jan 24, 2004, at 3:03 PM, Markus Hoenicka wrote:

> Bruce D'Arcus writes:
>> The underling HTML code is a little ugly, but the outward appearance 
>> is
>> nice enough.  All you really need is a way to get back some (X)HTML
>> code that has formatted references, each of them wrapped in a div
>> element with a class attribute.  That way you can do all the styling
>> with CSS.  I'm even willing to help there if need be.  I've actually
>
> Could you tell me what the HTML output should look like to improve the
> styling with CSS? It should not be too hard to fix backend-html then.

I'm not sure what it is that Matt likes about the interface he posted 
the url to, but one thing that's different is that the references are 
formatted like in a reference list.  Other than that, you've got things 
working right because you're attaching class attributes to most of the 
code.

However, if you look at this example:

<h1 class='h1'>refdb reference list</h1>
<h2 class='ID'>ID*:3 (2002)</h2>
<p class='authors'>Russwurm,M.; Mergia,E.; Mullershausen,F.; 
Koesling,D.</p>
<p class='title'>Inhibition of deactivation of NO sensitive guanylyl 
cyclase accounts for the sensitizing effect of YC-1</p>
<p><em class='periodical'>J.Biol.Chem.</em> <em 
class='volume'>277</em><em class='issue'>(28)</em>:<em 
class='page'>24883-8</em><p class='citekey'>CITEKEY:Russwurm2002</p>
<hr />

...it would offer more CSS formatting flexibility if you had each 
record wrapped in a div element, e.g.:

<h1 class='h1'>refdb reference list</h1>
<div class="record">
	<h2 class='ID'>ID*:3 (2002)</h2>
	<p class='authors'>Russwurm,M.; Mergia,E.; Mullershausen,F.; 
Koesling,D.</p>
	<p class='title'>Inhibition of deactivation of NO...</p>
	<p><em class='periodical'>J.Biol.Chem.</em> <em 
class='volume'>277</em><em class='issue'>(28)</em>:	<em 
class='page'>24883-8</em><p class='citekey'>CITEKEY:Russwurm2002</p>
</div>

That would allow the top-and-bottom borders that in Matt's interface 
separate each reference.

Also, those <em> elements probably ought to be span.  That way the CSS 
definition could say whether it was italicized or not.

While I've not experimented with it, I suppose it would offer even more 
flexibility to use spans throughout, such as where you're currently 
using p tags...

Bruce