LinuxQuestions.org - My Latest Opensource.com Column: Small, Useful Linux Utilities

- Linux - News (https://www.linuxquestions.org/questions/linux-news-59/)

- - My Latest Opensource.com Column: Small, Useful Linux Utilities - My Pick is a Clipboard Manager (https://www.linuxquestions.org/questions/linux-news-59/my-latest-opensource-com-column-small-useful-linux-utilities-my-pick-is-a-clipboard-manager-4175608793/)

AutoKey - I find it indispensable.

Guake is another shell console that runs in the background of your gui and drops down from the top of the screen by pressing F12. It is customizable and mimicks the console from Quake, hence the name. For me, it is more convenient to pull a shell up with a single keystroke than to double click a launcher or have open terminals cluttering my desktop space. It can also hold multiple tabbed sessions for you which is another nice feature.

Quote:

Originally Posted by powderburns (Post 5737139)

+1 Guake

For disk management/info LSBLK saved my bacon a few times.

The clipboard utility - Clipman for me with XFCE - and the duplex printing utility seem tempting. But where would I find them,and how would I integrate them into my machine?

This is not real small but invaluable for cleaning the mess on your drive. Similar to Windows WinDirStat, QDirStat for Gnome and KDirStat for KDE perform the same and show a graphical display on the bottom and a more normal file command type of display on the top. You can right click on the graphics to ID the files that are hugs and see what they are and permanently delete them from top or bottom. You really need to see any of the three different dirtrees as they all look a lot alike although I think the Windows one was around much longer.

I've used Beyond Compare for years

Quote:

Originally Posted by khentiamentiu (Post 5736903)

A utility I use practically every day when I'm coding is kdiff3. It's free, and powerful, a solid engine for comparing files or directories, whether they be in different code repositories, or different versions in one repository. It fits neatly into both git and svn command lines, giving you a gui when you want it for comparison. It's also good for comparing notes, logs, and ascii dumps.

Graphical comparisons all of the above. Cross platform. Dropbox profiles. Not free but nominal. Extensive filtering. The list goes on and on

Climate !
https://github.com/adtac/climate

It allows anyone with it installed to due lotsa things actually, like checking your battery status, the weather, and pretty much everything actually !

んー, this is very difficult to answer because there are tons of small but powerful and efficient utilities. Benefit for everyone... zsh built-ins, less, parallel,,,. んー,,,,, for everyone.....

How about Hostsblock, https://github.com/gaenserich/hostsblock ?

rhamel, I like ncdu, and I'ma let you finish, but xdiskusage is one of the best disk usage visualizers out there (especially if you want a GUI).

Xdiskusage is extremely small, simple, and light, but its world-beating feature is that it can read the plaintext output of "du". This means you can do:

Code:

ssh root@myserver "du -ax /" > /path/to/file

and you will have a stored, visualizable readout of the server's root filesystem usage, sitting on your local workstation. Then just visualize it with:

Code:

xdiskusage /path/to/file

Note that the role of xdiskusage here is purely visualization -- you don't need ANY special tools to collect and save the data! Just coreutils "du"! No need to install extra programs on your servers :)

Ncdu has something similar, but it seems like it can only read output generated by itself, so it would appear to require installing ncdu on the machine you want to analyze. This is not a huge drawback, but for me it's less than ideal. Plus, I like having a GUI frontend (that I only have to install on my workstation!).

I am continually mystified as to why all Linux disk usage visualizers don't ship this simple, elegant feature. Maybe I am just missing the other ones that do.

Not specifically a Linux utility, but nvm (Node Version Manager - https://github.com/creationix/nvm).

Allows you to install multiple versions of Node + npm and switch between them with a simple command.
Fantastic for developers switching between projects and easier than messing around with PPAs even if you just want to install the latest version.

Code:

# list available versions

nvm ls-remote



# install as many as you want

nvm install 8.2.0

nvm install 6.9.1



# list installed versions

nvm ls



# aliases (default does what it says on the tin)

nvm alias default 6.9.1



# switch version

nvm use 6.9.1

Linux Mint Software Installs

Hi
Would love to see a one click or two install new software.

It would ask for location of software open terminal and for whatever file type(which it establishes itself)(checks software manager etc) and would install the software. (dare I say it, as easy as .exe in that other software.

Would be great for the older generation like me to run Linux. User friendly. I find LM excellent and the only problem is with loading new software.

There maybe something out there. Let me know.

Thanks

Charles Harris

Ok, not open source but really useful - http://www.jgoodies.com/downloads/jdiskreport/

It's a java thingy I found years ago.

It sorts files by filetype, size. It identifies the top 50 biggest sizes.

Last modified, size disribution, useful stuff.

All kinds of surprises about what's on my hard drives.

Requires Java, so runs on Windows and Mac OSx as well.

Just saying.

@linustalman - Now you've embarrassed me! ;) I manage the original branch of AutoKey and forgot to mention it. I support it because it's so indispensable to me.

@Alok Rai - duplex is available here.

I don't know how to make an installer for it (If anybody wants to show me how, that would be great!), but it comes with very detailed installation instructions (and support from me if you ask for it.) If you have a very basic familiarity with the Linux command line, it's very easy to install. It's just a bunch of bash scripts, so there's not a lot to do to install it.

@sswcharlie - That's what Ubuntu is moving toward with the snap installation methodology. [ducking and covering]
However, making installs zipless comes with a lot of questionable tradeoffs - a whole other topic.

Using any modern package manager (pacman, yum, apt, ...) makes the installation process relatively painless these days - at least for any package which has been packaged in a repository for your distro (which is the vast majority of things most people want).

Which brings me to another cool utility: checkinstall.

Quote:

When make install is done, CheckInstall will create a Slackware, RPM or Debian compatible package and install it with Slackware's installpkg, "rpm -i" or Debian's "dpkg -i" as appropriate, so you can view it's contents with pkgtool ("rpm -ql" for RPM users or "dpkg -l" for Debian) or remove it with removepkg ("rpm -e"|"dpkg -r"). Aditionally, this script will leave you a copy of the installed package in the source directory so you can install it wherever you want,

So, when you have to manually install something on a system which uses one of the supported packaging systems, it will build it into a package so you can deal with it just like a package you got from a repository - which makes uninstalling it a lot easier.

The only thing I don't like about it is that I always remember to use it right after I've finished manually installing something. ;)

The gvim, 'split diff with' option is fantastic. It gives you side by side editing windows, update content from one to the other with diffget / diffput. (Can also be invoked with gvimdiff file 1 file2). Nothing better for applying customisations to an updated config file.

So many people overlook the wonders of sed. Some time back I had a problem where several hundred small files, originally produced with Microsoft Word and saved as html snippets, had to be processed for a website.

In it's own inimitable way, Word had rendered accented characters in two or three incompatible formats, rendered as strange codes that failed to display properly. We also had the so-called 'smart quotes' to unravel. On a very old, Pentium II with limited RAM sed chewed through 800+ files quicker than we could have hand edited more than half a dozen.

Even if you only need sed once, it will more than repay the time taken to learn to use it.