Xnat-utils is a collection of scripts for conveniently up/downloading and listing data on/from XNAT based on the XnatPy package.
The following converters are required for automatic conversions of downloaded images (using the '--convert_to' and '--converter' options)
- dcm2niix (https://github.com/rordenlab/dcm2niix)
- MRtrix 3 (http://mrtrix.readthedocs.io/en/latest/)
While many systems (particularly in research contexts) will already have Python 3 installed (note that Python 2 is not sufficient), if your workstation doesn't here are some basic instructions on how to install it.
macOS ships with it's own, slightly modified, version of Python, which it uses in some applications/services. For the most part it is okay for general use but in some cases, such as with xnat-utils, the modifications can cause problems. To avoid these I recommend installing an unmodified version of Python for use in your scientific programs using Homebrew (http://brew.sh). To do this first install Homebrew:
/usr/bin/ruby -e "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install)"
then install Python with:
brew install python3
If everything has gone well, when you type:
which python3
it should come back with:
/usr/local/bin/python3
If it doesn't or your run into any problems follow the instructions you receive when you run:
brew doctor
Note that these instructions are just recommendations so you don't have to follow all of them, just the ones that are likely to be related to your problem.
Download the version of Python for Windows using the most appropriate installer for Python (>=3.4), here https://www.python.org/downloads/windows/.
Python3 is most likely already installed but if it isn't it is best to install it using your package manager.
Pip is probably already be installed by default with your Python package so check whether it is installed first:
pip3 --version
Noting that it should be in /usr/local/bin if you are using Homebrew on macOS.
If pip is not installed you can install it by downloading the following script, https://bootstrap.pypa.io/get-pip.py and:
python3 <path-to-downloaded-file>
The XnatUtils source code can be downloaded (or cloned using git) from
https://github.com/MonashBI/xnatutils.git. To install it
cd to to the directory you have downloaded and run:
pip3 install xnatutils
If you get permission denied errors you may need to use sudo,
or if you don't have admin access then you can install it in your
user directory with the --user flag.:
pip3 install --user xnatutils
I have had some difficulty with the installation of progressbar2 as there is a
conflict with the progressbar package (they both produce packages called
progressbar). In this case it is probably a good idea to install xnat-utils
in a virtual environment (https://virtualenv.readthedocs.io/en/latest/).
The first time you use one of the utilities you will be prompted for the address of the server would like to connect to, in addition to your username and password. By default a alias token for these credentials will be stored in a ~/.netrc file with the following format (with permissions set to 600 on the file):
machine <your-server-url> user <your-alias-token> password <your-alias-secret>
If you don't want these credentials stored, then pass the '--no_netrc' (or '-n') option.
If you have saved your credentials in the ~/.netrc file, subsequent calls won't require
you to provide the server address or username/password until the token
expires (if you don't want deal with expiring tokens you can just save your username/password
in the ~/.netrc file instead, however, please be careful with important passwords). To reset
the saved credentials provide --server option again with the full server address
including the protocol (e.g. 'https://') or edit the ~/.netrc file directly.
To connect to an additional XNAT server, provide the new server address via the --server option.
Credentials for this server will be saved alongside the credentials for your previously saved
servers. If the --server option is not provided the first server in the file will be used. To
used the save credentials for a secondary server you only need to provide as of the secondary server
address to --server to distinguish it from the other saved servers. For example given the following
saved credentials in a ~/.netrc file:
machine xnat.myuni.edu user myusername password mypassword machine xnat-dev.myuni.edu user mydevusername password mydevpassword
then:
$ xnat-ls -s dev MYPROJECT
will be enough to select the development server from the saved credentials list.
Six commands will be installed
- xnat-get - download scans and resources
- xnat-put - upload scans and resources (requires write privileges to project)
- xnat-ls - list projects/subjects/sessions/scans
- xnat-rename - renames an XNAT session
- xnat-varget - retrieve a metadata field (including "custom variables")
- xnat-varput - set a metadata field (including "custom variables")
Please see the help for each tool by passing it the '-h' or '--help' option.
The regular expression syntax used by xnat-get and xnat-ls is fully defined
here, https://docs.python.org/2/library/re.html. However, for most basic use
cases you will probably only need to use the '.' and '*' operators.
'.' matches any character so the pattern:
MRH060_00._MR01
will match
MRH060_001_MR01 MRH060_002_MR01 MRH060_003_MR01 MRH060_004_MR01 MRH060_005_MR01 MRH060_006_MR01 MRH060_007_MR01 MRH060_008_MR01 MRH060_009_MR01
The '*' matches 0 or more repeats of the previous character, which is most useful in conjunction with the '.' character to match string of wildcard characters, e.g.:
MRH060.*
will match all subjects/sessions in the MRH060 project.
Note, that when using regular expressions that use '*' on the command line you will need to enclose them in single quotes to avoid the default wilcard file search, e.g.:
$ xnat-ls 'MRH099.*'
Probably the only other syntax that will prove useful is the '(option1|option2|...)'. For example:
MRH060_00(1|2|3)_MR01
will match
MRH060_001_MR01 MRH060_002_MR01 MRH060_003_MR01
For more advanced syntax please refer to the numerous tutorials on regular expressions online.