New server, new MT version, new theme (which I don't like as much as the old one, but I didn't find the old one ;) and one new entry to tell you this :P
I had to fix some issues of my initial 0.1 release that bugged me. First off there is now the possibility to start pdf.quicksplit without a filename. It then shows a FileDialog in which you can specify a PDF file to split. It's also possible to right-click on a PDF file an select "Open with" -> Browse -> (select the pdf.quicksplit.exe). You can see this on the new screenshots.
There are two other fixes:
- if there are already temporary files: ask the user if we should delete them
- show error dialog if user selects a pdf file with only 1 page (or if there was an error while extracting pages - e.g. corrupt PDF file)
And there's one annoying bug (under windoze) left which I don't know how to fix: if pdf.quicksplit.exe is put in a path which contains spaces it doesn't find the other exe-tools (like pdftk.exe, ...).
The new version can be downloaded here:
PDF QuickSplit v0.2 Source Code
PDF QuickSplit v0.2 Linux Version
PDF QuickSplit v0.2 Windoze Version
Btw: if you want to help improve pdf.quicksplit please contact me (email is in the source and perhaps it's best to transfer the source to sourceforge or sth. similar before we start coding in different directions ;).
Some days ago I wanted to finish (read: "bring it in an usable state ;) a project which I started some months ago: a program for splitting PDF files. I know that there are many programs out there which can do this but I searched hours for a program which serves the purpose of splitting a PDF in a fast way by showing you a preview of the pages. So PDF QuickSplit isn't able to automatically split a PDF file into single pages or on every N pages (you can accomplish this much easier by using something like
for i in `seq 1 1 15`; do pdftk input.pdf cat ${i}-${i} output output.${i}-${i}.pdf; done;or using GUI for PDFTK).
PDF QuickSplit is for users who have one of those fancy ADF scanners and just put a bunch of documents on it which get put into a single PDF file. The GUIs I saw are very discomforting when I try to split the PDF file into the single documents because some of them only have one page, some have a double page and finally there are multiple pages which you want to keep into one file. So you need an easy tool to set cut-markers. For that you need a window with a preview of the PDF pages. And that's exactly what PDF QuickSplit does (under Windoze and Linux):
A final word: the state of the program is something like an alpha1 (or as KDE developers would call it: v4.2.1 rock-stable without bugs... - but that's another story...). The code is pretty ugly (static char[] stuff etc) - so it's main purpose is to show programmers how _not_ to code ;) But you are free to send me patches... The following downloads include the source code which can be compiled on linux and windoze (dependencies: gtk, pdftk, ghostscript, imagemagick - for windoze you need the mingw-versions), the linux binary and the windoze binary with it's dependencies. So finally here are the downloads:
PDF QuickSplit v0.1 Source Code
PDF QuickSplit v0.1 Linux Version
PDF QuickSplit v0.1 Windoze Version (with all necessary dlls)
And if you like my program soooo much that you need to buy an ADF scanner immediately (or something else ;) from Amazon please click on the following link to support me :)
I had to struggle to get my Oracle ODBC connection to work with the Oracle Instant Client - the full-install worked out-of-the-box, but who the faq wants to install 700MB of unneeded software if there's an "only-what-I-need" Oracle Instant Client available ;). I created a complete howto which is online now as OpenOffice Writer (odt) document here and as PDF file here.
After some days of struggling with a VMware VirtualCenter installation I wanted to share some information on the Error-Message 25003 which the VC-setup gave me while trying to recreate/update the VC-Database. The solution in my environment (VC-2.5.0u3 + Oracle 10.2.0.3 with Oracle Instant Client 10.2.0.4 + Oracle Instant Client ODBC Addon 10.2.0.4) was simple (but very difficult to find ;):
I only got the problem if there was still an active VC connected to my OracleDB while trying to recreate/update the VC-Database. This happened because I tried to reproduce another error. So I still had a VM with VC running while I tried to install a new VC with the same DB in a different VM. Simply shutting down the VM with the old (outdated) VC solved the error.
So in short: if you get the Error 25003 while trying to install VMware VirtualCenter check if there's still another connection to the database in which you try to install VC.
For some time I wanted to write a quick reference card for the service console commands of VMware's ESX Server Service Console. While I was about finishing it VMware released ESX3.5.0-Update2. So this may be already outdated ;) Nevertheless you can download my QuickRef card here. The latex source code is also available here. If you want to use it (e.g. to create a quickref card sorted by topic rather than alphabetically) you can use if you keep me as origin somewhere in the fine-print ;)
After struggling with several problems when installing a 64Bit Linux for the first time I want to write down the problems for which I didn't find a solution by using google. Two problems were related to Matrica Moneyplex. The first problem was that it didn't find my kobil chipcard-reader.
The solution in short is: follow the official docs from matrica on how to setup the necessary stuff (apt-get install pcscd pcsc-tools). Do NOT install the openct package! Then install the driver for your chipcardreader by downloading the necessary libct.zip from www.kobil.de and run the install64-script as root AND the installuser64-script as normal user. In the moneyplex settings select the libproxy-kobil.so as driver for the chipcard-reader.
Now the problem started: I wasn't able to access my reader. After some investigation I saw the following error messages when starting moneyplex in the console:
fejf@fejfs:[~/moneyplex] > ./start
=================================================
32<->64 bit CTAPI-Proxy library.
(C) 2007 Martin Preuss
=================================================
*** glibc detected ***
0x0000000000604e50 ***
[...]
This took me some time but the solution is simple: add a export MALLOC_CHECK_=0 to the moneyplex start-script (normally ~/moneyplex/start) right after the other export-statement(s).
After that my chipcard-reader worked as user root. But when I tried to access it as normal user I still got error messages. The solution was simple again (after hours of finding it ;): I didn't have the right .CT_devices file in my home. I don't know why I got the wrong version (the installuser64-script from the kobil-installer should copy a new version!) but I could fix this problem by copying the file from root's home to mine (replace fejf with your username...):
cp /root/.CT_devices ~fejf
chown fejf:fejf ~fejf/.CT_devices
FYI: the .CT_devices file should contain only ONE Port#-Line which should look like this one:
Port1=PCSC;[IDX="1"];3;0;reader via PCSC [IDX="1"]: transparent
Starting with linux kernel 2.6.22 Kobil doesn't support the Kobil KAAN Standard Plus USB Card Reader anymore. That means that i couldn't do any bank transfers if i wanted to upgrade to a post 2.6.21 kernel. The only way to fix this remedy is to get a new card reader - the Kobil KAAN Advanced USB Card Reader ("they know how to make money" crosses my mind...). Anyway, they changed the usage of the new card reader and i counldn't access the card reader from within moneyplex. After contacting Matrica they instantly tried to support me and sent a howto for OpenSuSE which was a good starting point - but i still use Gentoo ;) So to give something back: here's the Gentoo Version.
As requested: my .vimrc files. If you use a light background then edit the .vimrc file and change the set background setting line to:
set background=lightDownload hier.
Einige Typos gefixed - Download hier.
