@@ -250,6 +250,9 @@ If different functionality and thus outputs need to be tested, a second test sho
...
@@ -250,6 +250,9 @@ If different functionality and thus outputs need to be tested, a second test sho
Reference datasets for testing are centrally stored on EOS at \dir{/eos/project/c/corryvreckan/www/data/} and are accessible over the internet. The \file{download_data.py} tool provided in the \dir{testing/} directory of the framework is capable of downloading individual data files, checking their integrity via an SHA256 hash and decompressing the tar archives.
Reference datasets for testing are centrally stored on EOS at \dir{/eos/project/c/corryvreckan/www/data/} and are accessible over the internet. The \file{download_data.py} tool provided in the \dir{testing/} directory of the framework is capable of downloading individual data files, checking their integrity via an SHA256 hash and decompressing the tar archives.
Each test can specify one or several data sets it requires to be present in order to successfully run via the \parameter{#DATASET} tag in the configuration file. The datasets have to be made known to the download script together with their calculated SHA256 hashes when adding a new test to the repository. The file name and hash have to be added to the Python \parameter{DATASETS} array, and the file of the dataset has to be uploaded manually to EOS by one of the framework maintainers.
Each test can specify one or several data sets it requires to be present in order to successfully run via the \parameter{#DATASET} tag in the configuration file. The datasets have to be made known to the download script together with their calculated SHA256 hashes when adding a new test to the repository. The file name and hash have to be added to the Python \parameter{DATASETS} array, and the file of the dataset has to be uploaded manually to EOS by one of the framework maintainers. The SHA256 hash can be generated by executing:
\begin{verbatim}
openssl sha256 <dataset>
\end{verbatim}
Paths in the test configuration files should be provided relative to the \dir{testing/} directory, all downloaded data will be stored in individual subdirectories per dataset following the naming scheme \dir{testing/data/<dataset>}.
Paths in the test configuration files should be provided relative to the \dir{testing/} directory, all downloaded data will be stored in individual subdirectories per dataset following the naming scheme \dir{testing/data/<dataset>}.