[ refdb-Feature Requests-3000131 ] "addword" - make new reserved words available

Feature Requests item #3000131, was opened at 2010-05-11 19:26
Message generated for change (Comment added) made by akusmin
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=385994&aid=3000131&group_id=26091

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: refdbd
Group: None
Status: Open
Priority: 5
Private: No
Submitted By: akusmin (akusmin)
Assigned to: Markus Hoenicka (mhoenicka)
Summary: "addword" - make new reserved words available

Initial Comment:
Hello!
I had to add some reserved words using the addword command. 
The problem: after I added  new reserved words (e.g. Solid, State, and Ionics), even after I used the "updateref" commando,
the latex  bibliography is still created with "Solid.State.Ionics" as journal abbreviation.
The only solution I found so far is to delete the database and recreate it. 

Request/Question Could you add / Is there any simpler way to take into account new reserved words? 
If there is already such way, could add this information to the manual?

Thank you!

P.S.
please consider adding the following words into the default list (some of them may be in this list of the development version I guess)

MACROMOLECULES
MATTER
SOLID
STATE
IONICS


----------------------------------------------------------------------

>Comment By: akusmin (akusmin)
Date: 2010-06-04 09:27

Message:
1) I checked again whether with refdb-0.9.9-1.tar.gz doing "updateref"
makes an user-added reserved word  without a dot in the bibliography. Yes,
it does. So, I was wrong when in the first post  I said updateref did not
help.

2)  Actually, from the practical point of view it is not a big issue: the
list of reserved words is indeed rather complete.  On a rare occasion when
one has to add a new reserved words one could just use updateref on all
citations. (Or delete databases and recreate them, that's what I did).  
It's just the necessity to run updateref must be mentioned/stressed in
docs.

----------------------------------------------------------------------

Comment By: Markus Hoenicka (mhoenicka)
Date: 2010-06-04 08:48

Message:
I see the problem. There was no way for you to know that you need to run
updateref after adding reserved words as I can't expect you to know the
innards of RefDB. However, the current behaviour is pretty much a design
issue. In terms of computing time, it is prudent to perform time-consuming
tasks once (when you add a reference) instead of multiple times (when you
retrieve references or build bibliographies). I do understand that it is
inconvenient if you add several references using words which are not yet
known to the database without realizing this fact after the first
reference. I personally run getref after each addref just to make sure the
datasets arrived ok. I'll try to figure out a better way to deal with this.
I could make refdbd run a background task to automatically update affected
journal titles after adding reserved words, just like it runs a background
task to check titles and abstracts for existing keywords after each addref
command.

----------------------------------------------------------------------

Comment By: Nobody/Anonymous (nobody)
Date: 2010-06-03 19:57

Message:
Hi. I found out why the bibliography generated after new reserved words
were added via "addword"  was exactly the same as before.
You said that you did "updateref" after you added a new reserved word.  I
did not.
At least in the manual for 0.9.9-1 there is 
nothing said about the necessity to run "updateref" whenever you use
addword/deleteword.

See, I thought that  in the process of generation of the bibliography
every citation is checked against the list of the reserved words.
Therefore, any change in the list of the reserved words does not require
updating citations. 
>From the fact that "updateref" is required I conclude that  1) the form in
which the citation is stored internally *depends* on the list of the
reserved words 2) hence, generation of the bibliography does not involve
consulting the list of reserved words.

In my opinion the necessity to use updateref is not convenient,   (I *have
to know* in which file I have the citation  to make the new reserved word
to have an effect on the freshly produced latex article). 


Anyway, revision 718 works for me on Slackware. Thank you) 


----------------------------------------------------------------------

Comment By: akusmin (akusmin)
Date: 2010-05-26 07:43

Message:
Thank you, Markus. In the next several days I'll try to reproduce the
problem.  

Well, I use refdb-0.9.9-1.tar.gz , which is about  3 years old. On the
other hand, perhaps I could have introduced the problem myself by manually
changing permissions for some refDB directories in /var.  
OK, I'll write again when I have more info.

----------------------------------------------------------------------

Comment By: Markus Hoenicka (mhoenicka)
Date: 2010-05-25 22:07

Message:
I've checked the code again - there is nothing wrong with updating the
journal synonyms. The current (and correct) behaviour has been around for
at least three years. 

I've tried to reproduce your problem to no avail. I added a random journal
article containing "Solid" in the fake journal abbreviation. Next I've
added "SOLID" to the list of abbreviated words. I've then updated the
article without changing anything in the RIS dataset. Next time I retrieved
it, "Solid" was written without a trailing dot, as expected.

Could you please create a test kit with some sample data and an exact list
of commands which reproduces your problem. I'll be happy to try again.

regards,
Markus

----------------------------------------------------------------------

Comment By: Markus Hoenicka (mhoenicka)
Date: 2010-05-17 09:56

Message:
I'll have to run a few tests, but it looks like there is a bug in the
insert_periodical() function. As far as I understand the source code, the
function adds additional synonyms when updating, but it won't touch
existing synonyms even if a corrected version is provided. Obviously,
updateref should update synonyms if they need to be changed. I'll look into
this. Also, adding reserved words is no big deal. Feel free to throw
further words my way .

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=385994&aid=3000131&group_id=26091

------------------------------------------------------------------------------
ThinkGeek and WIRED's GeekDad team up for the Ultimate 
GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the 
lucky parental unit.  See the prize list and enter to win: 
http://p.sf.net/sfu/thinkgeek-promo