names with {\v...}

[davydden]

When renaming a bibtex entry with first author name like:

• {\v s}v
• A{\c s}k

the resulting file has an empty [auth] field. Perhaps, the best option would be to remove {} and \v \c, etc from file names. So that the two examples result in the following [auth] field:

• Sv
• Ask

p.s. JabRef 2.10, renameFile 1.5.0. p.p.s. I see that in Hystory: 1.5.0 "Remove symbols \,/,<,>,?,{,},$,",:,\n from file names", but this does not seem to work for me, at least here on OS-X.

[davydden]

apparently, the [auth] field is empty for me irrespectively of single/multiple authors, with/without special characters... my "Use name pattern" is (without extra spaces) : [type] / [auth] [year] [title]

[korv]

Could you write the complete bibtex entry and the name pattern that works incorrectly?

[korv]

Please, write also the version of Jabref that you are using.

[davydden]

sure, here is a simple example which does not work for me (JabRef 2.10, renameFile 1.5.0):

% This file was created with JabRef 2.10. % Encoding: ISO8859_1

@Article{Tanaka2014, Title = {Robust numerical calculation of tangent moduli at finite strains based on complex-step derivative approximation and its application to localization analysis}, Author = {Masato Tanaka and Masaki Fujikawa and Daniel Balzani and J{\"o}rg Schr{\"o}der}, Journal = {Comput. Methods Appl. Mech. Engrg.}, Year = {2014}, Pages = {454--470}, Volume = {269},

Doi = {doi:10.1016/j.cma.2013.11.005}, File = {:test.pdf:PDF}, Keywords = {Tangent moduli,Complex-step derivative approximation,Finite deformations,Nonlinear finite element method,Localization analysis}, Publishers = {Elsevier B.V.} }

don't forget to add a .pdf "test.pdf" in the same folder. Screen shot is attached to illustrate the problem.

[korv]

The problem with [auth] appears only in Jabref 2.10. It seems to disappear after applying the key generator one time. I have to analyze their code in order to find the problem, as my code uses the key generator functionality of Jabref. Apparently Jabref 2.10 is still somewhat raw. For example, the [auth] for Bergman returns "Bergm", as the program considers "an" to be an article. As a temporary solution you can use Jabref 2.9.2

[davydden]

interesting, thanks a lot for investigating what's wrong...

[davydden]

i tried 2.9.2., [auth] shows correct, but renaming does not work, nothing happens :( I think i did not have this problem with 1.4 version of the plugin...

[korv]

Strange, it works under Windows and Linux. Could you enter a non-empty main file directory (under options/preferences/external programs)?

[davydden]

That worked!..

I also noticed that paths which start "Users/...." can not be renamed. Although i can open the .pdf fine. Actually i do not even now how i ended up with such "un-complete" full paths...

But as soon as I do "/Users/..." or re-add them and have relative path to the .bib file (e.g. "Articles/..."), renaming works fine.

[korv]

Version 1.6 of the plugin is compatible with JabRef 2.10. It also fixes some small bugs.

[korv]

Could you please check if there is still any wrong behavior? And if yes, write a bibtex example.

[davydden]

Sure. Thanks for looking at it.

Below are two bibtex examples which would benefit from some extra substitution rules related to "special" characters in names:

@Article{svec:2011vs, Title = {{Model predicting the free surface flow of a suspension of rigid solid particles in a non-Newtonian fluid}}, Author = {{\v s}vec, Oldrich and Sko{\v c}ek, Jan}, }

and

@Article{Askar:1973hj, Title = {{A model for coupled rotation-displacement modes of certain molecular crystals. Illustration for KNO3}}, Author = {A{\c s}kar, A}, }

If there is no open-source code in Java which can do the job, I would propose APPROACH 1: to add another filter to names before any others (i think you already have one which removes thinks like "{","}","/"). And do the following:

• Look for "{\".
• If it is there, find the next closing "}".
• Keep a single character before "}" and remove everything else in between "{" and "}".
• BUT, there are exceptions (see below).

APPROACH 2: one could look for accents exactly (and update the list if needed), see examples here: http://tex.stackexchange.com/questions/57743/how-to-write-ä-and-other-umlauts-and-accented-letters-in-bibliography http://tex.stackexchange.com/tags/accents/info

and have rules to substitute them in each case. Perhaps it is more robust approach than creating a general rule (approach 1). And it is easier to handle exceptions (see example below).

I would say for the two examples above, "good" names would be: {\v s}vec -> vec A{\c s}kar -> Askar

There are also cases like:

@Booklet{Csanyi:2006tw, Title = {{Multiscale modeling of defects in semiconductors: A novel molecular-dynamics scheme}}, Author = {Cs{\'a}nyi, G and Moras, G and Kermode, J R and Payne, M C and Mainwood, A and De Vita, A} }

Cs{\'a}nyi -> Csanyi (?)

@Article{Diuzewski:2000fc, Title = {{Anisotropic Hyperelasticity Based Upon General Strain Measures}}, Author = {D{\l}u{.{z}}ewski, Pawe{\l}} }

D{\l}u{.{z}}ewski, Pawe{\l} -> Dluzewski, Pawel

@InCollection{davydov2009modeling, Title = {Modeling of nanoindentation by a visco-elastic porous model with application to cement paste}, Author = {Davydov, Denis and Jir{\'a}sek, Milan} }

Jir{\'a}sek - > Jirasek

@Article{Espanol1995, Title = {{Statistical Mechanics of Dissipative Particle Dynamics}}, Author = {Espa{~n}ol, P and Warren, Patrick B} }

Espa{~n}ol -> Espanol

@Article{FernandezMendez:2004ug, Title = {{Imposing essential boundary conditions in mesh-free methods}}, Author = {Fern{\'a}ndez-M{\'e}ndez, Sonia and Huerta, Antonio} }

Fern{\'a}ndez-M{\'e}ndez -> Fernandez-Mendez

@Article{Mueller-Plathe2002, Title = {{Coarse-Graining in Polymer Simulation: From the Atomistic to the Mesoscopic Scale and Back}}, Author = {M{\"u}ller-Plathe, Florian} }

M{\"u}ller-Plathe -> Muller-Plathe

@Article{Soderhjelm:2009dn, Title = {{How Accurate Can a Force Field Become? A Polarizable Multipole Model Combined with Fragment-wise Quantum-Mechanical Calculations}}, Author = {S{\"o}derhjelm, P{\"a}r and Ryde, Ulf} }

S{\"o}derhjelm, P{\"a}r -> Soderhjelm, Par

EXCEPTIONS:

@Article{OKeeffe:2002fv, Title = {{Bandstructure modulation for carbon nanotubes in a uniform electric field}}, Author = {O{\textquoteright}Keeffe, James and Wei, Chengyu and Cho, Kyeongjae} }

O{\textquoteright}Keeffe -> OKeeffe

[korv]

Please test v1.6.2. It should handle all accents and latex commands.

[davydden]

I went through my library and everything works fine with accents.

The only problem I encountered are with:

@Article{OKeeffe:2002fv, Title = {{Bandstructure modulation for carbon nanotubes in a uniform electric field}}, Author = {O{\textquoteright}Keeffe, James and Wei, Chengyu and Cho, Kyeongjae} }

and

@Article{Sousedik2014, Title = {{Hierarchical Schur complement preconditioner for the stochastic Galerkin finite element methods}}, Author = {Soused{\'\i}k, Bed{\v{r}}ich and Ghanem, Roger G and Phipps, Eric T} }

which could also be written as

@ARTICLE{2012arXiv1205.1864S, author = {{Soused{\'{\i}}k}, B. and {Ghanem}, R.~G. and {Phipps}, E.~T. }, title = "{Hierarchical Schur complement preconditioner for the stochastic Galerkin finite element methods}" }

perhaps another substitution rules could be added for those cases.

[korv]

Try v1.6.3

[davydden]

"{\'\i}" works fine, but "{\textquoteright}" has not been fixed.

[korv]

This is a feature and not a bug. \textquoteright is substituted by ' I think that the file name should be O'Reilly and not OReilly. But you can anyway change this behavior by using the modifier: [author:regex:textquoteright:]

[davydden]

that's not what i mean, it is not substituted at all, what i see in filename is: OtextquoterightKeeffe and that's how it got renamed.

[korv]

Strange, I just downloaded v1.6.3 and it works fine. Could you download it once again and check that it is indeed v1.6.3?

[davydden]

it is 1.6.3, i am sure. JabRef 2.10. See the attached screenshot. Filename is the one after renaming.

[korv]

I see, I worked with JabRef 2.9.2. I tested now JabRef 2.10 and the problem is actually in their label creator. It transforms \textquoteright to extquoteright so, this is a bug on their side. As a workaround you can use the modifier :regex:extquoteright: this problem seems to be too specific to code it into the plugin.

[davydden]

sure, i did not notice "extquoteright", good catch!.. thank you!