Re: No bibliography in the output, depends on file name

[Hans Hagen <pragma@wxs.nl>]

On 1/17/2014 1:15 PM, Marco Patzer wrote:
> On 2014–01–15 Nicola wrote:
>
>> maybe this is a known issue,
>
> It's a known fact that context has a different notion of valid file
> names than your operating system does. This is by design. To quote
> the manual:
>
>    “It is highly recommended, that all input files, i.e. the ConTEXt
>    source and other included files such as image files, have only the
>    letters a–z, digits and dashes in their names, that is in the
>    names of their full paths, otherwise you can easily get into
>    problems.”
>
>> I'm writing a ConTeXt document called modern-c++.tex (in OS X 10.7.5). The
>> content of the file is:
>>
>>      \setupbibtex[database={modern-c++}, sort=author]
>>      \setuppublications[numbering=yes]
>>      \starttext
>>      \completepublications[criterium=all]
>>      \stoptext
>
> Especially since c++ didn't work out, I expected dropping the “++”
> would work (“modern-c”), but it didn't. So I ran some tracing:
>
>    \enabletrackers [resolvers.readfile]
>    \starttext
>      \readfile{file++.ext}{}{}
>    \stoptext
>
> This reports:
>
>    files > readfile > not found by tree lookup: file  .ext
>
> Which means the “++” is replaced by two spaces, instead of searching
> for “file++.ext” or “file.ext” which is what I had expected. I
> didn't dig into the code to check where the spaces creep in.
>
> Regardless if this particular issue gets fixed or not, I doubt that
> Hans will put much effort into general support for “esoteric” file
> names. So, it's best to avoid plus signs in file names.

indeed. names are parsed as url's (so + become space) so a possible fix is:

function getreadfilename(scheme,path,name)
     local fullname
     if hasscheme(name) or is_qualified_path(name) then
         fullname = name
     else
         name = url.escape(name) -- yes or no ?
         fullname = ((path == "") and format("%s:///%s",scheme,name)) or 
format("%s:///%s/%s",scheme,path,name)
     end
     return resolvers.findtexfile(fullname) or "" -- can be more direct
end

but one cannot predict how this passes further on through the system

also, because one can say:

\readfile{file\letterpercent2B\letterpercent2B.ext}{}{}

the hack in fact should be:

    if not string.find(name,"%%") then
        name = url.escape(name) -- if no % in names
    end

which then handles both

\readfile{file\letterpercent2B\letterpercent2B.ext}{}{}

\readfile{file++.ext}{}{}

ok. Of course, when moving from c++ to c# one gets things like

     \readfile{file\letterhash.ext}{}{}

where the # will sometimes confuses macros later on.

Hans


-----------------------------------------------------------------
                                           Hans Hagen | PRAGMA ADE
               Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
     tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com
                                              | www.pragma-pod.nl
-----------------------------------------------------------------
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://tex.aanhet.net
archive  : http://foundry.supelec.fr/projects/contextrev/
wiki     : http://contextgarden.net
___________________________________________________________________________________