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

@@ -10,6 +10,7 @@ import math
 import os
 import sys
 from typing import Tuple
+import subprocess
 
 from Bio import SeqIO
 import labkey
@@ -55,6 +56,28 @@ def parse_cli_args() -> argparse.Namespace:
         metavar="STR",
     )
 
+    zarp = parser.add_argument_group("ZARP execution")
+    zarp.add_argument(
+        "--execution_env",
+        choices=['local', 'slurm', None],
+        default=None,
+        help="zarp execution wrapper",
+        metavar="STR",
+    )
+    zarp.add_argument(
+        "--cluster_json",
+        type=str,
+        default=None,
+        help="slurm specific resources config for the zarp rules",
+        metavar="STR",
+    )
+    zarp.add_argument(
+        "--snakefile_path",
+        type=str,
+        default=None,
+        help="path of the snakefile to be executed",
+        metavar="STR",
+    )
     io = parser.add_argument_group("input/output")
     io.add_argument(
         "--input-to-output-mapping",
@@ -195,6 +218,14 @@ def parse_cli_args() -> argparse.Namespace:
             "\n[ERROR] Either none or both of '--labkey-domain' and "
             "'--labkey-path' are required."
         )
+
+    if (args.execution_env and not args.cluster_json) or \
+       (args.cluster_json and not args.execution_env):
+        parser.print_help()
+        sys.exit(
+            "\n[ERROR] Either none or both of '--execution_env' and "
+            "'--cluster_json' are required."
+        )
     return args
 
 
@@ -451,6 +482,95 @@ def expand_path(
     return os.path.normpath(path)
 
 
+def run_zarp(
+        execution_env: str,
+        config: str,
+        result_dir: str,
+        cluster_json: str,
+        snakefile_path: str,
+        anchor: str = os.getcwd()):
+    """
+        trigger zarp snakemake workflow execution, according to
+        the environment chosen by the user if no environment is
+        chosen zarp workflow is not triggered
+
+    :param execution_env: environment where zarp will be executed;
+                          for now support only local and slurm
+
+    : config: config file to be executed
+    : samples: samples table for which zarp is triggered
+
+    :returns: tuple of ALFA strand name suffixes for two coverage tracks of a
+        paired-end sequencing library
+    """
+    report_path = os.path.join(result_dir, "snakemake_report.html")
+    if execution_env == "local":
+        try:
+            command = [
+                'snakemake',
+                f'--snakefile="{snakefile_path}"',
+                f'--configfile="{config}"',
+                '--cores=4',
+                '--printshellcmds',
+                '--rerun-incomplete',
+                '--use-singularity',
+                f'--singularity-args="--bind {anchor}"',
+                '--verbose']
+            subprocess.check_call(
+                command,
+                stdout=sys.stdout,
+                stderr=sys.stderr)
+        except subprocess.CalledProcessError:
+            pass
+        except OSError:
+            pass
+
+    elif execution_env == "slurm":
+        try:
+            command = [
+                'snakemake',
+                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"',
+                '--cores=256',
+                '--printshellcmds',
+                '--rerun-incomplete',
+                '--use-singularity',
+                f'--singularity-args="--bind {anchor}"',
+                '--verbose']
+            subprocess.check_call(
+                command,
+                stdout=sys.stdout,
+                stderr=sys.stderr)
+        except subprocess.CalledProcessError:
+            pass
+        except OSError:
+            pass
+    try:
+        report_command = [
+            'snakemake',
+            '--snakefile="Snakefile"',
+            f'--configfile="{config}"',
+            f'--report="{report_path}"']
+        subprocess.check_call(
+            report_command,
+            stdout=sys.stdout,
+            stderr=sys.stderr)
+    except subprocess.CalledProcessError:
+        pass
+    except OSError:
+        pass
+    return
+
+
 def main(args):
     """
     Create input table and config for ZARP.
@@ -686,6 +806,14 @@ def main(args):
     args.config_file.write(config_file_content)
     args.config_file.close()
 
+    if args.execution_env:
+        run_zarp(
+            args.execution_env,
+            args.config_file,
+            results_dir,
+            args.cluster_json,
+            args.snakefile_path)
+
 
 if __name__ == '__main__':
     args = parse_cli_args()