My project is accepted for the Google Summer of Code project, and now it’s time to ask for money. From discussions: “Figuring out that damn tax thing appears to be the hardest part of the project.” The post with preliminary instructions.
Archive for June, 2005
A list of papers related to PreScheme and VLISP: http://www.coyotos.org/pipermail/bitc-dev/2004-October/000003.html. Should be useful for my future developments.
I don’t make any presentation at the GTTSE 2005, but I want to share my ideas and get feedback. Specially for this conference, I’ve ordered and got a T-shirt with the buzz word “Generative XPath”, this blog URL and words “Scheme” and “[La]TeX”.
I’ve spent two days understanding how to create CJK bookmarks using LaTeX. Finally, I checked the internal structure of a PDF document with a sheet of paper, a calculator and the PDF reference. The PDF document is correct. Problem is in Acrobat Reader 5 (AR5). It is said that AR5 uses the system font to display the bookmarks, and the system font on Windows 2000 Rus doesn’t have CJK symbols.
I’m going to write quite a big text in DocBook format. At work I use FrameMaker, but at home I don’t have FM, so I’ll use the universal text editor vim. DocBook is very verbose format, and it is very unefficient to edit DocBook texts without assistance from a editor. I’ve looked for how vim can help me and found a life changing solution.
Some tutorials on non-root installation of LaTeX modules rely on the environment variable TEXMFLOCAL. Unfortunately, it doesn’t work when the folder “~/texmf” contains conflicting files.
Finally, “with-param” in the Scheme function “x:apply-templates” works. It took some time because I forget an experience with making “mode” working. Again, parameter name should be in the dictionary. Unfortunately, namespaced parameters doesn’t work, probably due to a bug of libxslt. Reported, waiting.
The latest CVS version of libxml2 doesn’t build. Linking fails with messages:
../../.libs/libxml2.so: undefined reference to `pthread_getspecific' ../../.libs/libxml2.so: undefined reference to `pthread_once' ../../.libs/libxml2.so: undefined reference to `pthread_key_create' ../../.libs/libxml2.so: undefined reference to `pthread_setspecific'
The TeXML development version 1.27 brings new essential functionality: “--ascii” parameter. Now it’s possible to generate plain ASCII TeX files in a desired encoding. Non-ascii bytes are encoded as “^^XX“.
Unfortunately, I’m not giving a presentation at the «Participants Workshop of the summer school on Generative and Transformational Techniques in Software Engineering (GTTSE 2005).» The competition was a quite tight: there were 32 extended abstracts and only 14 time slots. Well, anyway, at least I again have got a valuable feedback.
Sometimes I need to look inside XML files which consist of one big line. It’s very impossible, and reformatting takes a time. Fortunately, I’ve found a good solution for vim.
I have problems with non-latin publishing through LaTeX, and I’m going to ask for help in forums. While asking, I should provide a sample LaTeX file, but how to show non-latin characters? Fortunately, I have TeX background and know some tricks. One of them is that combination “^^xx” is interpreted as a character with the hexadecimal code “xx” by the TeX reader before any other processing. So, for example, if I use the symbol “\u9009” which is encoded in utf8 by bytes “e9”, “80” and “89”, I can write in the source LaTeX file: “^^e9^^80^^89”.
I often need to know the code of a symbol, and I always forget how to reveal the code in vim. In order to remember, I’m saving it in the blog: “ga”. Meanwhile, a useful list of vim command is found: “Tip #305: Best of VIM Tips (VIM’s best Features)“.
From the mail discussions: www.askemos.org. “This is actually a Scheme VM in byzantine synchronisation over a P2P network with some administration free (root-less) permission control system.” The point of interest for me is “Scheme VM”.
Without any knowledge of Chinese, I have to publish a Chinese document. Fortunately, I have a source as an utf8-encoded XML.
I’ve published the article “Reusing XML Processing Code in non-XML Applications” ([PDF]) version 1.0 and posted announces to different forums.
On the 1st April (fool’s day; in our culture a day for jokes) I got a letter from a customer with a specification and the words: “take some time to do it”. Unintentional joke of “take some time” is that I have finished coding only now, 7 June.
I’m visiting Summer School on Generative and Transformational Techniques in Software Engineering (GTTSE 2005) and hopefully give a presentation.