[Gchemutils-main] Re: New database

[Jean Bréfort]

Le dim 19/10/2003 à 11:05, Kostantino a écrit :
> Dear Jean,
> I write you my job for the new element database. I got a look to the 
> your and below I report both your original and my new version for one 
> element: hydrogen.
> As I wrote you in the past, the idea of a long and single element file, 
> even written in xml, is usefull if we don't write too many informations 
> and specifications. But, I'm not sure this idea is good if we have many 
> specifications, many characteristics and, of course, their translations 
> in several languages.
> Do you remember my idea of a file for every element, written in xml format?
> It will be very easy to update and correct instead of a longer unique file.
> It's hard to pay attention on single mistake if we work on a single and 
> very long file.
> My proposal: if you can (if you want / if you have time) could you 
> develop a non-public version to test my idea and so decide if it's good 
> or not? Every element file could be named with the relative atomic 
> number to find and recall it easier.
> 
> We must decide where find the correct data: I used a periodic table of 
> 1993 reporting both the first and the second IUPAC proposal about the 
> name of the groups. Example: IIIB ---> 13.
> Are you agree? Any suggestions about handbook?
> To finish: the attached file is an artical about a new icosaedric order 
> of matter, written by italian scientist.
> I hope it could be interesting for you: write me if you don't want I 
> send you any other article in future.
> I awake your answer.
> 
>          best regards
> 
> Costantino

I think that this discussion should be continued on the gchemutils list,
so I add the list int the cc field.

I have comments about your work. First of all, the translations should
not be written directly in the xml file. In my version, they are in the
po files, for example the name of the elements is in the xml.in file:
<_name>Hydrogen</_name>
Translations are merged at build time. So you should only edit the
elements.xml.in file.

I agree with you that one huge xml file is not a good idea. There are
several possibilities to avoid this. We can have, as you say, one file
per element, or have several files for different properties. My idea was
for the >=0.3 versions to be able to merge at run time several
databases, even user defined ones. Of course this can work with
supplementary files dedicated to single elements.

Otherwise, the way you treat things is far from perfect. Let's take the
symbol example. For hydrogen, it is just H, the string "Symbol: " should
NOT be there as it is presentation stuff, not data. The way data are
presented should appear in the program used for the presentation, or it
is possible to use an xsl stylesheet to get a more readable document
(for example an html page). The same applies to the other lines.

My version of the database conforms to a dtd I wrote for gperiodic some
times ago. It probably have never been released. We should extend it as
needed to add new data, but many things are already there. As I probably
never published it, I join the file (also, it should be replaced by an
XML schema).

The max_bond field is a good candidate for addition. I already use such
a thing in GChemPaint and moving that to the gchemutils is already
planed. The values used must be choosed very carefully. I had a problem
with the value found in OpenBabel for phosphorus (5); the use of an xyz
file for the hexafluorophosphate ion failed, a bond was missing.

> 
> ************************************************************************
>                According with Costantino
> 
> <element>
> <color red="1.00" green="1.00" blue="1.00"/>
> <name>Hydrogen</name>
> <name xml:lang="fr">Hydrogène</name>
> <name xml:lang="it">Idrogeno</name>
> <name xml:lang="pl">Wodór</name>
> <symbol>Symbol: H</symbol>
> <symbol xml:lang="it">Simbolo: H</symbol>
> <atomic_number>Atomic number (Z): 1</atomic_number>
> <atomic_number xml:lang="it">Numero atomico (Z): 1</atomic_number>
> <atomic_weight>Atomic weight (amu): 1.00794</atomic_weight>
> <atomic_weight xml:lang="it">Peso atomico (uma): 1.00794</atomic_weight>
> <max_bonds>Max bonds number: 1</max_bonds>
> <max_bonds name xml:lang="it">Massimo numero di legami: 1</max_bonds>
> <configuration>Electronic Configuration: 1s1</configuration>
> <configuration lang="it">Configurazione Elettronica: 1s1</configuration>
> <en_scale>Electro negativity (Pauling): 2.20</en_scale>
> <en_scale xml:lang="it"> Elettronegatività (secondo Pauling): 
> 2.20</en_scale>
> <covalent_radius>Covalent radius: 37</covalent_radius>
> <covalent_radius xml:lang="it"> Raggio covalente: 37</covalent_radius>
> <vdW_radius>VdW radius: 120</vdW_radius>
> <vdW_radius xml:lang="it"> Raggio vdW: 37</vsW_radius>
> <ionic_radius0>Ionic radius: 154</ionic_radius0>
> <ionic_radius0 xml:lang="it"> Raggio ionico: 154</ionic_radius0>
> <ionic_radius1>Ionic charge: +1, cn: 1, Shannon's value: -38</ionic_radius1>
> <ionic_radius1 xml:lang="it"> Carica ionica: +1, cn: 1, valore di 
> Shannon: -38</ionic_radius1>
> <ionic_radius2>Ionic charge: +1, cn: 2, Shannon's value: -18</ionic_radius2>
> <ionic_charge1 xml:lang="it"> Carica ionica: +1, cn: 2, valore di 
> Shannon: -18</ionic_radius2>
> </element>
> 
> *********************************************************************
> 
>                        According with Jean
> 
> <element Z="1" symbol="H" max_bonds="1">
>       <name>Hydrogen</name>
>       <name xml:lang="fr">Hydrogène</name>
>       <name xml:lang="it">Idrogeno</name>
>       <name xml:lang="pl">Wodór</name>
>       <color red="1.00" green="1.00" blue="1.00"/>
>       <en scale="Pauling" value="2.20"/>
>       <radius type="covalent" value="37"/>
>       <radius type="vdW" value="120"/>
>       <radius type="ionic" charge="+1" cn="1" scale="Shannon" value="-38"/>
>       <radius type="ionic" charge="+1" cn="2" scale="Shannon" value="-18"/>
> </element>

Best regards,

Jean

gperiodic.dtd
Description: Text Data