[WIP] make sure query and database files exist (#440)

* make sure query and database files exist

The databases in the lca commands are loaded first which can
take a bit of time (~20s) and means that there is a delay in
finding out if one of the query files has been incorrectly typed.
This small function checks if all of the database and query files
exist before trying to load anything.

* fix merge problems; switch to using 'exists'

* make two calls to check_file_exists

python 2.7 can only expand one list per function call. For compatibility,
split the calls to check_file_exists for both the db and query. When
python 2.7 compatibility is dropped these duplicated calls can be
merged into one

* remove some unnecessary spaces

sourmash/lca/command_classify.py

@@ -11,8 +11,7 @@
 from .. import sourmash_args, load_signatures
 from ..logging import notify, error
 from . import lca_utils
-from .lca_utils import debug, set_debug
-
+from .lca_utils import debug, set_debug, check_files_exist
 
 DEFAULT_THRESHOLD=5                  # how many counts of a taxid at min
 
@@ -107,6 +106,13 @@ def classify(args):
     args.db = [item for sublist in args.db for item in sublist]
     args.query = [item for sublist in args.query for item in sublist]
 
+    # have to have two calls as python < 3.5 can only have one expanded list
+    if not check_files_exist(*args.query):
+        sys.exit(-1)
+
+    if not check_files_exist(*args.db):
+        sys.exit(-1)
+
     # load all the databases
     dblist, ksize, scaled = lca_utils.load_databases(args.db, args.scaled)
 

sourmash/lca/command_gather.py

@@ -13,10 +13,9 @@
 from .. import sourmash_args, save_signatures, SourmashSignature
 from ..logging import notify, error, print_results
 from . import lca_utils
-from .lca_utils import debug, set_debug
+from .lca_utils import debug, set_debug, check_files_exist
 from ..search import format_bp
 
-
 LCAGatherResult = namedtuple('LCAGatherResult',
                              'intersect_bp, f_unique_to_query, f_unique_weighted, average_abund, lineage, f_match, name, n_equal_matches')
 
@@ -77,7 +76,7 @@ def gather_signature(query_sig, dblist, ignore_abundance):
         orig_abunds = { k: 1 for k in query_mins }
     sum_abunds = sum(orig_abunds.values())
 
- 
+
     # collect all mentioned lineage_ids -> md5s, from across the databases
     md5_to_lineage = {}
     md5_to_name = {}
@@ -108,7 +107,6 @@ def gather_signature(query_sig, dblist, ignore_abundance):
                     md5 = lca_db.lineage_id_to_signature[lid]
                     signature_size = lca_db.lineage_id_counts[lid]
                     assignments[hashval].add((md5, signature_size))
-
         # none? quit.
         if not assignments:
             break
@@ -172,7 +170,6 @@ def gather_signature(query_sig, dblist, ignore_abundance):
         yield result, f_unassigned, est_bp, query_mins
 
     ## done.
-
 
 
 def gather_main(args):
@@ -203,6 +200,9 @@ def gather_main(args):
     if args.debug:
         set_debug(args.debug)
 
+    if not check_files_exist(args.query, *args.db):
+        sys.exit(-1)
+
     # load all the databases
     dblist, ksize, scaled = lca_utils.load_databases(args.db, None)
 

sourmash/lca/command_summarize.py

@@ -11,8 +11,7 @@
 from .. import sourmash_args, load_signatures
 from ..logging import notify, error, print_results
 from . import lca_utils
-from .lca_utils import debug, set_debug
-
+from .lca_utils import debug, set_debug, check_files_exist
 
 DEFAULT_THRESHOLD=5
 
@@ -89,6 +88,13 @@ def summarize_main(args):
     args.db = [item for sublist in args.db for item in sublist]
     args.query = [item for sublist in args.query for item in sublist]
 
+    # have to have two calls as python < 3.5 can only have one expanded list
+    if not check_files_exist(*args.query):
+        sys.exit(-1)
+
+    if not check_files_exist(*args.db):
+        sys.exit(-1)
+
     # load all the databases
     dblist, ksize, scaled = lca_utils.load_databases(args.db, args.scaled)
 

sourmash/lca/lca_utils.py

@@ -5,6 +5,7 @@
 import sys
 import json
 import gzip
+from os.path import exists
 from collections import OrderedDict, namedtuple, defaultdict, Counter
 
 try:                                      # py2/py3 compat
@@ -14,13 +15,27 @@
 import pprint
 
 from .._minhash import get_max_hash_for_scaled
-from ..logging import notify
-
+from ..logging import notify, error
 
 # type to store an element in a taxonomic lineage
 LineagePair = namedtuple('LineagePair', ['rank', 'name'])
 
 
+def check_files_exist(*files):
+    ret = True
+    not_found = []
+    for f in files:
+        if not exists(f):
+            not_found.append(f)
+            ret = False
+
+    if len(not_found):
+        error('Error! Could not find the following files.'
+              ' Make sure the file paths are specified correctly.\n{}'.format('\n'.join(not_found)))
+
+    return ret
+
+
 # ordered list of taxonomic ranks
 def taxlist(include_strain=True):
     for k in ['superkingdom', 'phylum', 'class', 'order', 'family', 'genus',