From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Message-ID: <4F63B303.8050004@thax.hardliners.org> Date: Fri, 16 Mar 2012 22:39:15 +0100 From: Tobias Hoffmann MIME-Version: 1.0 References: <4F639287.1040000@gmail.com> In-Reply-To: <4F639287.1040000@gmail.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Subject: Re: [Printing-architecture] Google Summer of Code 2012 - The Linux Foundation accepted! List-Id: Printing architecture under linux List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Till Kamppeter Cc: Open Printing Till Kamppeter wrote: > Google has accepted The Linux Foundation again as mentoring organization! I've been interested in working on the pdf manipulation tools already back in 2008, when I implemented the pdftoijs and texttopdf filters as GSoC-Project. So I'm quite thrilled to hear, that now, nearly having completed university, I might have the ability to work on them, and participate in GSoC again. Through work on my low-level pdf library unpdf ( http://code.google.com/p/unpdf/ ) I have gained much knowledge on PDF internals (Adobe's PDF Reference, different PDF versions, etc) and implementation caveats. Although the current unpdf code-base implements enough of the low-level handling (also e.g. including support for encrypted pdfs) to build the desired tools upon, there are a significant number of areas in need of work; another possibility would be to either reimplement only the bare minimum of functionality (i.e. use unpdf, and maybe other pdf libs like iText only as guideline) or use another pdf i/o library as basis (podofo comes to mind, http://podofo.sourceforge.net/ ; a pure C choice with good-looking codebase could be MuPDF, http://mupdf.com/ ). But theses details can be worked out in time. FWIW, I personally would strongly prefer C++0x/C++11 (esp. auto, std::unique_ptr, rvalue references / std::move) for this project, instead of trying to work around language limitations that will not gain us anything in the long term -- if this is possible. Tobias