KBibtex in Ubuntu Oneiric 11.10

KBibtex is a useful bibtex bibliography management tool for the KDE desktop.
However, it seems that the package is missing in the Ubuntu 11.10 repositories.

A quick remedy (as tweeted here) is to install the package from the Debian Sid distribution: kbibtex.
To install on the command line (check the exact name of the .deb file)
sudo dpkg -i kbibtex_0.4-1_amd64.deb

Alternatively .deb packages can be installed in the GUI (e.g. by opening the file in Dolphin).

The package should be compatible with Ubuntu 11.10, and seems to work fine.

The 0.4 version has improved a lot and has a very useful online search tool for acquiring bibtex files. Google scholar and other data bases can be used.
Other tools include a reference preview (though bibtex2html) and a PDF preview.


Autocrop images with imagemagick

Often one has to crop excess margins from an raster image. Common image graphical editors have autocrop tool, but if many files need to be processed, opening and saving files one by one quickly becomes tedious.

A better solution is to use convert tool that is a part of the powerful Imagemagick toolkit:

convert image.jpg -trim -bordercolor White -border 20x10 +repage cropped_image.jpg
This command reads the original file, removes (trims, crops) white borders, adds 20 pixel white borders horizontally and 10px vertically, and stores the file as cropped_image.jpg.

It is also possible to process all image files in the current directory with the following one line command:
for i in `ls *.jpg`; do convert $i -trim -bordercolor White -border 20x10 +repage cropped_`basename $i`; done

If you want to convert to another image format, it can be done conveniently at the same time (-quality option controls the jpg quality):
for i in `ls *.gif`; do convert $i -trim -bordercolor White -border 20x10 -quality 92 +repage cropped_`basename $i gif`jpg; done
Above basename utility is used for trimming suffix from the file name.

Extract images from a pdf file

In Linux pdfimages utility extracts raster images from a pdf file.

Extract all raster images from pdf file, save in jpg format (creates bar-000.jpg, bar-001.jpg, etc.):
pdfimages -j file.pdf image

The -j option is to convert the images to jpeg format if possible.
Other options can be found in the manual.

Use pdftk to edit pdf files

pdftk is a useful free utility for editing pdf files.

Split a pdf file to individual pages (generates: page01.pdf, page02.pdf, etc.)
pdftk original.pdf burst output page%02d.pdf

Split a pdf file to two parts:
pdftk original.pdf cat 1-3 output threeFirstPages.pdf
pdftk original.pdf cat 4-end output rest.pdf

Merge multiple pdf files to one:
pdftk first.pdf second.pdf third.pdf cat output 123.pdf

Many more examples can be found on pdflabs examples page.

Using wget to download multiple files from a website

The following command downloads all files pdf files from http://www.host.com/some/path/ to currenct directory
wget -r -l1 -nd -nc -A.pdf http://www.host.com/some/path/

The options are:
-r Makes it recursive for subfolders
-l1 set maximum recursion, 1 levels of subfolders
-nd no directories — copies all matching files to current directory, discards directory information
-nc Do not download it file already exists
-A.pdf Accept only certain files (with pdf suffix in this case). Specify comma-separated lists of file name suffixes or patterns.

Script for downloading bibtex file using DOI

Digital Object Identifier (DOI) and bibtex files are common for anyone citing scientific articles. The DOI is used to uniquely identify any object, such as an electronic document, while bibtex is used to cite an article in Latex environment.

An article can be found on the internet via the address dx.doi.org/ followed by the DOI, e.g. http://dx.doi.org/10.1016/j.ocemod.2004.08.002. However, it takes a while to browse the publishers website, locate the appropriate bibtex file and save it to disk.

Here is a very handy Python script that fetches a bibtex file from NASA’s Astrophysics Data System using the article’s DOI.

It requires Python Beautiful Soup XML/HTML parser.
In Ubuntu-like linux distributions it can be installed easily:
sudo apt-get install python-beautifulsoup


>>doi2bibtex.py 10.1016/j.ocemod.2004.08.002
author = {{Shchepetkin}, A.~F. and {McWilliams}, J.~C.},
title = "{The regional oceanic modeling system (ROMS): a split-explicit, free-surface, topography-following-coordinate oceanic model}",
journal = {Ocean Modelling},
year = 2005,
volume = 9,
pages = {347-404},
doi = {10.1016/j.ocemod.2004.08.002},
adsurl = {http://adsabs.harvard.edu/abs/2005OcMod...9..347S},
adsnote = {Provided by the SAO/NASA Astrophysics Data System}

Fixing Klipper shortcut keys in KDE4.5+

After an update to Kubuntu 11.04 Natty Narwhal (possibly also Lucid and Maverick) the commonly used Klipper shortcut Ctrl+Alt+V for invoking the pop-up menu does not work anymore. Very annoying, for browsing the clipboard history was very useful.

Here’s a fix (for Natty):

Go to System settings->Shortcuts and Gestures->Global Keyboard Shortcuts and select Plasma Desktop Shell in the KDE component box. Look for Systemtray-Klipper-10 and assign Ctrl-Alt-V to it. Most likely this will create a conflict with Klipper keyboard shortcuts, accept to force the change.

Pressing Ctrl-Alt-V should now bring up the Klipper pop-up menu. To select an item in history using only keyboard, choose with up/down arrows. The press Enter (to choose the item), press Esc (to close the pop-up). This is different than before when Enter also closed the pop-up. Note that the chosen item is not indicated in any way in the pop-up. Hopefully this will change in the future.

This hassle is all due to the fact that same shortcut is defined in (at least) two places. As plasma desktop shell is handling the shortcuts now, the old setting in Klipper section is simply ignored, but still present.

%d bloggers like this: