What are the differences between -v an -V to get the Version of software in Ubuntu?

Question

I'm always confused to get the version of software installed in Ubuntu. To prevent from full typing to get the version like <software> --version instead I always use something like <software> -V.

But the problem is that not for all software it works. For some I've to use <software> -v and for some, I've to use full --version to get the version.

For example

wget, gedit, nano, mysql`, etc all work with -V (Capital V)

but Php, Skype and may be others never worked with -V instead I've to use -v (small v) to get the version:

php -V
Usage: php [options] [-f] <file> [--] [args...]
   php [options] -r <code> [--] [args...]
   php [options] [-B <begin_code>] -R <code> [-E <end_code>] [--] [args...]
   php [options] [-B <begin_code>] -F <file> [-E <end_code>] [--] [args...]
   ...

php -v
PHP 5.3.10-1ubuntu3.9 with Suhosin-Patch (cli) (built: Dec 12 2013 04:27:25) 
...

Some work with both -v and -V like firefox. And some even don't work with either of -v or -V like totem, wine and google-chrome.

• Why there is this much difference?
• Since -V is always preferred to get the version of the software, why there is no any standard? or is there any standard that I don't know?

Answer 1

The correct way to get the version of system-installed software is to use your package manager tools!

--version is not a reliable way to do so, for several reasons:

• Not all programs have executables. Libraries are a good example. You can't check your kernel version using linux --version, as there is no such command. Or your video driver version.

• Not all executables have command-line arguments. Most do, but GUI programs don't need to, and some don't.

• --version, as any command-line argument, is application-dependent. It depends on the developer to implement it, and there is no "standard" per-se, merely a convention. As you've noticed with -v|-V, it is not consistent. Even --help is not universal.

• The format output of --version is not consistent either. Some print a single line, some print several. Some print the version number only, some print the program name too.

That said, there is a standard, consistent way to get the installed version of any software in your system: ask the system, not the sofware!

Thanks to its Debian heritage, Ubuntu has a powerful package management system called apt (actually, dpkg). It controls installed packages, its dependencies, available repositories, and versions.

There are several package-management tools and front-ends you can use to query your installed packages. Here are some that display the version:

• apt-cache policy <package>

• dpkg --list <package> (you can use wildcards!)

And if you don't know what package a given command belongs to, you can find out in several ways:

• apt-cache search '<name>'

• apt-file search '<path>'

And the output is always consistent, reliable, standard, because you're not asking individual software made by distinct developers, you're querying your system about its status.

As an example, here's the result of a search of all the commands your mentioned in a single output:

$ dpkg --list wget gedit nano mysql-server skype php? firefox totem wine google-chrome*

Desired=Unknown/Install/Remove/Purge/Hold
| Status=Not/Inst/Conf-files/Unpacked/halF-conf/Half-inst/trig-aWait/Trig-pend
|/ Err?=(none)/Reinst-required (Status,Err: uppercase=bad)
||/ Name                       Version                    Description
+++-==========================-==========================-====================================================================
ii  firefox                    42.0+build2-0ubuntu0.12.04 Safe and easy web browser from Mozilla
ii  gedit                      3.4.1-0ubuntu1             official text editor of the GNOME desktop environment
ii  google-chrome-stable       46.0.2490.80-1             The web browser from Google
ii  mysql-server               5.5.46-0ubuntu0.12.04.2    MySQL database server (metapackage depending on the latest version)
ii  nano                       2.2.6-1                    small, friendly text editor inspired by Pico
ii  php5                       5.3.10-1ubuntu3.21         server-side, HTML-embedded scripting language (metapackage)
ii  skype                      4.3.0.37-0ubuntu0.12.04.1  client for Skype VOIP and instant messaging service
ii  totem                      3.0.1-0ubuntu21.1          Simple media player for the GNOME desktop based on GStreamer
ii  wget                       1.13.4-2ubuntu1.2          retrieves files from the web
ii  wine                       1.4-0ubuntu4.1             Microsoft Windows Compatibility Layer (meta-package)

Answer 2

Those are verbose options related to each application so some apps used the -V others use -v others both or just --version. There is no general rule for naming convention.

what i mean to say is those options don't have a standard so you may find option -X in some app do the same as -R in other ...

Answer 3

The standard is:

app --version

-v or -V is only (not generally speaking) the abbreviated form for --version. You will never see in a man page something like:

-v
    Print version...

or:

-V
    Print version...

but you will see all the time these two options, the abbreviated form (if this exists) and the standard form, together. Something like:

-v, --version
    Print version...

or:

-V, --version
    Print version...

But this depends only by developers how they want to implement their applications. For example -v is used in some cases as the abbreviated form for --verbose (see man wget), or for --invert-match (see man grep) a.o., or in other cases as stand alone (see man awk or man ps).

Answer 4

It typically ends up being caused by an option called "verbosity". Verbosity runs a program and prints as much information as possible to the terminal from which it was called.

Some programs, however, don't support verbose mode or don't run in a way that would require any form of verbosity, so they will spit out the help string. Others will treat -v and -V equally.