Patch: check and require dependencies for each sub command.

Author: skchronicles

rna-seek

@@ -21,7 +21,7 @@ import argparse  # potential python3 3rd party package, added in python/3.5
 
 # Pipeline Metadata and globals 
 __author__   = 'Skyler Kuhn'
-__version__  = 'v1.9.4'
+__version__  = 'v1.9.5'
 __email__    = '[email protected]'
 __home__     =  os.path.dirname(os.path.abspath(__file__))
 _name        = os.path.basename(sys.argv[0])
@@ -122,6 +122,34 @@ def exe_in_path(cmd, path=None):
     return False
 
 
+def require(cmds, suggestions, path=None):
+    """Enforces an executable is in $PATH
+    @param cmds list[<str>]:
+        List of executable names to check
+    @param suggestions list[<str>]:
+        Name of module to suggest loading for a given index
+        in param cmd.
+    @param path list[<str>]]:
+        Optional list of PATHs to check [default: $PATH]
+    @return None
+    """
+    error = False
+    
+    for i in range(len(cmds)):
+        available = exe_in_path(cmds[i])
+        if not available:
+            c = Colors
+            error = True
+            err(
+                "\n{0}{1}Fatal: {2} is not in $PATH and is required during runtime!{3}".format(c.bg_red, c.white, cmds[i], c.end),
+                "\n  └── Possible solution: please 'module load {0}' and run again!".format(suggestions[i])
+            )
+    
+    if error: fatal()
+    
+    return 
+
+
 def permissions(parser, filename, *args, **kwargs):
     """Checks permissions using os.access() to see the user is authorized to access
     a file/directory. Checks for existence, readability, writability and executability via:
@@ -653,7 +681,6 @@ def dryrun(outdir, config='config.json', snakefile=os.path.join('workflow', 'Sna
             '--configfile={}'.format(config)
         ], cwd = outdir,
         stderr=subprocess.STDOUT)
-
     except subprocess.CalledProcessError as e:
         # Singularity is NOT in $PATH
         # Tell user to load both main dependencies to avoid the OSError below
@@ -841,6 +868,11 @@ def run(sub_args):
     @param sub_args <parser.parse_args() object>:
         Parsed arguments for run sub-command
     """
+    # Check for required dependencies
+    # The pipelines has only two requirements:
+    # snakemake and singularity 
+    require(['snakemake', 'singularity'], ['snakemake', 'singularity'])
+
     # Get PATH to RNA-seek git repository for copying over pipeline resources
     git_repo = os.path.dirname(os.path.abspath(__file__))
 
@@ -911,6 +943,9 @@ def unlock(sub_args):
     print("Unlocking the pipeline's output directory...")
     outdir = sub_args.output
 
+    # Check for required dependencies: snakemake
+    require(['snakemake'], ['snakemake'])
+
     try:
         unlock_output = subprocess.check_output([
             'snakemake', '--unlock',
@@ -1037,8 +1072,13 @@ def build(sub_args):
     """Builds the reference files for the RNA-seek pipeline from a genomic FASTA
     file and a GTF file. Disclaimer: hybrid genomes not supported.
     @param sub_args <parser.parse_args() object>:
-        Parsed arguments for unlock sub-command
+        Parsed arguments for build sub-command
     """
+    # Check for required dependencies
+    # The pipelines has only two requirements:
+    # snakemake and singularity 
+    require(['snakemake', 'singularity'], ['snakemake', 'singularity'])
+
     # Get PATH to RNA-seek git repository 
     # for copying over pipeline resources
     git_repo = os.path.dirname(os.path.abspath(__file__))
@@ -1115,14 +1155,16 @@ def cache(sub_args):
     Local SIFs will be created from images defined in 'config/containers/images.json'.
     @TODO: add option to cache other shared S3 resources (i.e. kraken db and fqscreen indices)
     @param sub_args <parser.parse_args() object>:
-        Parsed arguments for unlock sub-command
+        Parsed arguments for cache sub-command
     """
+    # Check for required dependencies: singularity 
+    require(['singularity'], ['singularity'])
+
     sif_cache = sub_args.sif_cache
     # Get absolute PATH to templates in rna-seek git repo
     repo_path = os.path.dirname(os.path.abspath(__file__))
     images = os.path.join(repo_path, 'config','containers', 'images.json')
 
-
     # Create image cache
     if not exists(sif_cache):
         # Pipeline output directory does not exist on filesystem