wrapper for zarp execution

wrapper in the prepare_inputs.py for executing zarp additional options added: --execution_env (allowed values local or slurm or none for now) --snakefile_path path to the snake file --bindings paths to be included in the singularity --cluster_json path to the cluster.json (used in the slurm run)

Fixed also a test_zarp_wrapper test (this needs to be refactored and be merged with existing tests) Fixes #130 (closed)

scripts/prepare_inputs.py

@@ -78,6 +78,13 @@ def parse_cli_args() -> argparse.Namespace:
         help="path of the snakefile to be executed",
         metavar="STR",
     )
+    zarp.add_argument(
+        "--bindings",
+        type=str,
+        default=None,
+        help="bindings for singularity",
+        metavar="STR",
+    )
     io = parser.add_argument_group("input/output")
     io.add_argument(
         "--input-to-output-mapping",
@@ -219,12 +226,10 @@ def parse_cli_args() -> argparse.Namespace:
             "'--labkey-path' are required."
         )
 
-    if (args.execution_env and not args.cluster_json) or \
-       (args.cluster_json and not args.execution_env):
+    if (args.execution_env == 'slurm' and not args.cluster_json):
         parser.print_help()
         sys.exit(
-            "\n[ERROR] Either none or both of '--execution_env' and "
-            "'--cluster_json' are required."
+            "\n[ERROR] '--cluster_json' is required."
         )
     return args
 
@@ -488,7 +493,7 @@ def run_zarp(
         result_dir: str,
         cluster_json: str,
         snakefile_path: str,
-        anchor: str = os.getcwd()):
+        bindings):
     """
         trigger zarp snakemake workflow execution, according to
         the environment chosen by the user if no environment is
@@ -508,13 +513,13 @@ def run_zarp(
         try:
             command = [
                 'snakemake',
-                f'--snakefile="{snakefile_path}"',
-                f'--configfile="{config}"',
+                f'--snakefile={snakefile_path}',
+                f'--configfile={config}',
                 '--cores=4',
                 '--printshellcmds',
                 '--rerun-incomplete',
                 '--use-singularity',
-                f'--singularity-args="--bind {anchor}"',
+                f'--singularity-args=--bind {bindings}',
                 '--verbose']
             subprocess.check_call(
                 command,
@@ -529,22 +534,23 @@ def run_zarp(
         try:
             command = [
                 'snakemake',
-                f'--snakefile="{snakefile_path}"',
-                f'--configfile="{config}"',
-                f'--cluster-config="{cluster_json}"',
-                '--cluster="sbatch \
+                f'--snakefile={snakefile_path}',
+                f'--configfile={config}',
+                f'--cluster-config={cluster_json}',
+                '''--cluster=sbatch \
                     --cpus-per-task={cluster.threads} \
                     --mem={cluster.mem} \
                     --qos={cluster.queue} \
                     --time={cluster.time} \
                     --job-name={cluster.name} \
                     -o {cluster.out} \
-                    -p scicore"',
+                    -p scicore''',
                 '--cores=256',
                 '--printshellcmds',
                 '--rerun-incomplete',
                 '--use-singularity',
-                f'--singularity-args="--bind {anchor}"',
+                '--latency-wait=60',
+                f'--singularity-args=--bind {bindings}',
                 '--verbose']
             subprocess.check_call(
                 command,
@@ -557,9 +563,9 @@ def run_zarp(
     try:
         report_command = [
             'snakemake',
-            '--snakefile="Snakefile"',
-            f'--configfile="{config}"',
-            f'--report="{report_path}"']
+            f'--snakefile={snakefile_path}',
+            f'--configfile={config}',
+            f'--report={report_path}']
         subprocess.check_call(
             report_command,
             stdout=sys.stdout,
@@ -809,10 +815,11 @@ def main(args):
     if args.execution_env:
         run_zarp(
             args.execution_env,
-            args.config_file,
+            args.config_file.name,
             results_dir,
             args.cluster_json,
-            args.snakefile_path)
+            args.snakefile_path,
+            args.bindings)
 
 
 if __name__ == '__main__':