Re: [eLyXer-users] Additions to BibTeXConfig and BibStylesConfig

[Jose Ramon Alvarez Sanchez]

El dom, 10-10-2010 a las 10:59 +0200, Alex Fernandez escribió:
> Done! I have added all vowels from á to ú. After all I'm a Spaniard myself :D

The construction \'{\i} was special only for accented i, because 
the normal i letter has a dot that must be removed to put other 
decoration, so \i is a latex command meaning an i without dot.
And \'\i is another alias for \'{\i}. But \a, \e, \o, \u are 
not defined in LaTeX, so alternate forms for a, e, o, u are
not valid.

It seems that Pybliographic doesn't accept characters not 
in latin1 when trying to save a .bib file. Also, it encodes 
capital accented letters inside title or booktitle fields 
different to protect capitalization, like \'{A}, \"{A}, etc.
In the case of Ç it becomes \c{{C}} that could be also 
"\c {C}" (with a space).

I have encoded all the non-ascii letters in latin1 in the 
possible forms for BibTeX inside the attached file 
bib-escaped.txt with the format to substitute all the
[BibTeXConfig.escaped] section in file conf/base.cfg.
Symbols and accented letters are in the same order as
in latin1 set. Other symbols in the original version 
are included at the end.

Opening (inverse) question and admiration marks ("!`:¡" 
and "?`:¿") cannot be added because those sequences do 
not start with "\" (included as comments).

Also find attached file testlatin1.bib containing only an
entry with all latin1 letters in author and title fields 
for testing conversion in a LyX document.

One issue remains to see here, I am not sure about solution.
Pybliographic encodes letters or symbols that have a string 
name like "\copyright" in the form "\copyright{}" (this is 
the safest way).
So, if eLyXer find an empty "{}" in bibentry will
ignore it (no problem here), but if it were (manually)
encoded (correctly) as "\copyright ", the space(s) should be 
removed for rendering by eLyXer. Also they can be 
encoded just as "\copyright" (without space nor braces)
if they are the last letter/symbol of a field.
This can occur with letters: \AA, \AE, \DH, \O, \TH, 
\ss, \aa, \ae, \dh, \o, \th.  Really not a big problem
if the encoding is done always in the first safe form.

Thanks,
Jose R.

bib-escaped.txt
Description: Text document

testlatin1.bib
Description: Text Data