MEGAcmd provides non UI access to MEGA services. It intends to offer all the functionality with your MEGA account via commands. It features synchronization , backup of local folders into your MEGA account and a webdav/streaming server. See Usage Examples.

Available packages for MEGAcmd in all supported platforms should be found here.

It supports 2 modes of interaction:

• interactive. A shell to query your actions
• scriptable. A way to execute commands from a shell/a script/another program.

In order to provide those 2 modes, it features one server (MEGAcmdServer), an interactive shell (MEGAcmdShell) and several commands that will launch the non-interactive client (MEGAcmdClient). See Usage and Platform to understand how to use it in your particular system.

If you wish to build MEGAcmd using this repository, here are a list of requirements and building instructions.

The same as those for the sdk (cryptopp, zlib, sqlite3, cares, libuv, ssl, curl, sodium) and of course readline. Also, it is recommended to include pcre to have support for regular expressions.

Also, in order to have support for thumbnails and previews, it is highly recommended to have ffmpeg (libavcodec-dev libavutil-dev libavformat-dev libswscale-dev) and mediainfo(libmediainfo-dev + libzen-dev) for media file attributes.

• For convenience here is a list of packages for ubuntu 18.04: autoconf libtool g++ libcrypto++-dev libz-dev libsqlite3-dev libssl-dev libcurl4-gnutls-dev libreadline-dev libpcre++-dev libsodium-dev libc-ares-dev libfreeimage-dev libavcodec-dev libavutil-dev libavformat-dev libswscale-dev libmediainfo-dev libzen-dev

and for ubuntu 16.04: autoconf libtool g++ libcrypto++-dev libz-dev libsqlite3-dev libssl-dev libcurl4-openssl-dev libreadline-dev libpcre++-dev libsodium-dev libc-ares-dev libfreeimage-dev libavcodec-dev libavutil-dev libavformat-dev libswscale-dev libmediainfo-dev libzen-dev

Ensure you obtain the repository recursively.

git clone https://github.com/meganz/MEGAcmd.git
git submodule update --init --recursive

For platforms with Autotools, MEGAcmd can be built and installed with:

sh autogen.sh
./configure
make
make install

• You will need to run make install as root

Note: if you use a prefix in configure, autocompletion from non-interactive usage won't work. You would need to source /YOUR/PREFIX/etc/bash_completion.d/megacmd_completion.sh (or link it at /etc/bash_completion.d)

Before explaining the two ways of interaction, it is important to understand how MEGAcmd works. When you login with MEGAcmd, your session, the list of synced folders, and some cache database are stored in your local home folder. MEGAcmd also stores some other configuration in that folder. Closing it does not delete those and restarting your computer will restore your previous session (the same as megasync won't ask for user/password once you restart your computer). You will need to logout properly in order to clean your data.

Now let's get into details of the two usage modes. Both modes require that MEGAcmdServer is running. You can manually launch it. Fortunately, you can also open the interactive shell or execute any command and the server will start automatically.

Execute MEGAcmd shell. Platform section explains how to do that in the different supported systems. You should be facing an interactive shell where you can start typing your commands, with their arguments and flags. You can list all the available commands with help. And obtain useful information about a command with command --help

First you would like to log in into your account. Again: notice that doing this stores the session and other stuff in your home folder. A complete logout is required if you want to end you session permanently and clean any traces (see logout --help for further info).

When MEGAcmd server is running, it will be listening for client commands. Use the different mega-* commands available. mega-help will list all these commands (you will need to prepend "mega-" to the commands listed there). To obtain further info use mega-command --help

Those commands will have an output value != 0 in case of failure. See megacmd.h to view the existing error codes.

Ideally, you would like to have these commands in your PATH (See Platform for more info). For further info use mega-help --non-interactive.

• A synchronization can be stablished simply by typing:

sync /path/to/local/folder /folder/in/mega

This will synchronize the contents in your local and your mega folder both ways.

• You can also set remote backups of a local folder to keep historical snapshots of your files. So simple as:

backup /path/to/myfolder /remote/path --period="0 0 4 * * *" --num-backups=10

This will configure a backup of "myfolder" into /remote/path that will be carried out at 4:00 A.M. (UTC) every day. It will store the last 10 copies. Further info on backups here.

• You serve a location via webdav:

webdav /path/to/myfolder

• Or stream a file in your MEGA account:

webdav /path/to/myfile.mp4

Further info on webdav and streaming here.

• Download the contents of a shared link:

get https://mega.nz/#F!ABcD1E2F!gHiJ23k-LMno45PqrSTUvw /path/to/local/folder 

Now let's do something more complicated with non-interactive usage using some GNU tools (similar stuff can be easily done in Windows as well):

• We wants to provide something crypto secured with only 10 minutes of access:

mega-put /path/to/my/temporary_resource /exportedstuff/
mega-export -a  /exportedstuff/temporary_resource --expire=10M | awk '{print $4}'

• Or imagine we'd like to public the enterprise promotional videos of May 2015 that we have previously stored in MEGA:

for i in $(mega-find /enterprise/video/promotional2015/may --pattern="*mpeg"); do 
mega-export -a $i | awk '{print $4}'; 
done

If you have installed MEGAcmd using one of the available packages at here, or have it built without --prefix, both the server (mega-cmd-server), the shell (mega-cmd) and the different client commands (mega-*) will be in your PATH (on a fresh install, you might need to open your terminal again). If you are using bash, you should also have autocompletion for client commands working. If that is not you case, include the location for the binaries in your path.

Once you have MEGAcmd installed, you just need to execute it (via Desktop icon or Start Menu) to open the shell. This shall open MEGAcmdServer in the background (a process named MEGAcmdServer.exe).

For a better user experience (specially in WINDOWS 7) we recommend executing MEGAcmd from PowerShell: Open PowerShell and execute:

$env:PATH += ";$env:LOCALAPPDATA\MEGAcmd"
MEGAcmdShell

For non-interactive usage, there are several mega-*.bat client commands you can use writting their absolute paths, or including their location into your environment PATH and execute them normally (mega-*). If you use PowerShell and you have installed the official MEGAcmd, you can do that simply with:

$env:PATH += ";$env:LOCALAPPDATA\MEGAcmd"

Client commands completion requires bash, hence, it is not available for Windows.

Although there have been several efforts in having non-ASCII unicode characters supported in Windows, there still may be some issues. Pay special attention if you are willing to use pipes or send the output of a command into a file from your client commands. See help --unicode for further info regarding that.

For MacOS, after installing the dmg, you can launch the server using MEGAcmd in Applications. If you wish to use the client commands from MacOS Terminal, open the Terminal and include the installation folder in the PATH. Typically:

export PATH=/Applications/MEGAcmd.app/Contents/MacOS:$PATH

And for bash completion, source megacmd_completion.sh :

source /Applications/MEGAcmd.app/Contents/MacOS/megacmd_completion.sh

Currently we have build scripts for Synology and QNAP, which can be found in the build/ folder along with instructions on how to set up the build. Typically this results in a 'package' which can then be manually installed in the NAS. To use MEGAcmd on those systems, ssh into the device and run the commands as normal (having first added their folder to your PATH variable).

MEGAcmd features autocompletion in both interactive and non-interactive (only for bash) mode. It will help completing both local and remote (Mega Cloud) files, flags for commands, values for flags/access levels, even contacts.

There are two different kinds of logging messages:

• SDK based: those messages reported by the sdk and dependent libraries.
• MEGAcmd based: those messages reported by MEGAcmd itself.

You can adjust the level of logging for those kinds with log command. However, for non interactive commands, passing -v (-vv, -vvv, and so on for a more verbose output) will use higher level of verbosity to an specific command.

If you have compiled MEGAcmd with PCRE (enabled by default), you can use PCRE compatible expressions in certain commands with the flag --use-pcre. Otherwise, if compiled with c++11, c++11 regular expressions will be used. If non of the above is the case, you can only use wildcards: "*" for any number of characters or "?" for a single unknown character. You can check the regular expressions compatibility with find --help. e.g:

find --help
...
Options:
 --pattern=PATTERN	Pattern to match (Perl Compatible Regular Expressions)

Notice: if you use MEGAcmd in non interactive mode, notice that shell pattern will take precedence. You will need to either escape symbols like * (\*) or surround them between quotes (e.g: "*.txt")

• Currently there are certain discrepancies with PATHS when loggin into a public folder. For instance, imagine a folder named toshare with a subfolder named x. If we login into toshare and execute find /x, this will be the output:

/toshare/x

Whereas if we execute find /toshare/x, we receive an error, since folder absolute path refers to / as root path.

[err: 12:21:51] Couldn't find /toshare/x

It might better be referred as /toshare/x