On 06/18/2012 10:56 AM, Paul Wise wrote:
On Mon, Jun 18, 2012 at 4:24 PM, Dirk Hünniger wrote:Essentially the development of the main LaTeX compiler pdflatex, that is used by basically everybody has stopped a few years ago. I recently discussed the unicode issue with the german tex mailing list. http://www.listserv.dfn.de/cgi-bin/wa?A1=ind1206&L=tex-d-lSo the problem is that pdflatex is not maintained, not that the developers of it refuse to fix the issue.
If it is not maintained, it means that the set of developers is empty. Mathematically everything you say about members of the empty set true. So also your statement above. But it is practically not very helpful, since the bug wont be fixed, since there is nobody to fix it.
I want to publish the following project: http://de.wikibooks.org/wiki/Benutzer:Dirk_Huenniger/wb2pdfI note there already are several ways to convert MediWiki to LaTeX& PDF: http://www.mediawiki.org/wiki/Extension:Wiki2LaTeX http://www.mediawiki.org/wiki/Extension:WikiPDF http://www.mediawiki.org/wiki/Extension:PDF_Writer http://www.mediawiki.org/wiki/Extension:PdfBook http://www.mediawiki.org/wiki/Extension:Pdf_Export
Only the first one produces LaTeX code. But it is not installed on Wikipedia, Wikibooks etc and it has to be installed on the server side in order to run. My project works client side only. So it is currently the only practical way of getting a LaTeX version of a Wikipedia article.
The merged font is needed at runtime. So it needs to be shipped.Yeah, that isn't really a workable. How about using something other than pdflatex is so buggy that it cannot do Unicode properly? pandoc can now convert LaTeX to PDF without pdflatex.
I tried the following
\section{lorem 合}
lorem ipsum
pandoc -flatex 1.wiki -o main.pdf
it says unicode error. So not pandoc is not able to do that properly.
The strange unicode symbol I used it Japanese ai meaning harmony.
Yours Dirk