implementation for the new CLI PR (#811)

* implementation for the cli branch
* watch
* multigather, two test fixes, raise SystemError 1 instead of 0
* add test to make sure help is printed when cmd is called without subcmd
* add import_csv
* init ksize properly in lca index

sourmash/__init__.py

@@ -28,7 +28,9 @@
 from . import sbtmh
 from . import sbt_storage
 from . import signature
+from . import sig
 from . import cli
+from . import commands
 
 from pkg_resources import get_distribution, DistributionNotFound
 

sourmash/__main__.py

@@ -11,7 +11,7 @@ def main(arglist=None):
     else:
         mod = getattr(sourmash.cli, args.cmd)
         mainmethod = getattr(mod, 'main')
-    mainmethod(args)
+    return mainmethod(args)
 
 
 if __name__ == '__main__':

sourmash/cli/__init__.py

@@ -1,24 +1,28 @@
 from argparse import ArgumentParser
 import sys
 
+import sourmash
+
 from . import utils
 
 # Commands
+from . import categorize
 from . import compare
 from . import compute
 from . import gather
+from . import import_csv
 from . import info
+from . import index
+from . import multigather
 from . import plot
 from . import search
+from . import watch
 
 # Subcommand groups
 from . import lca
 from . import sbt
-from . import signature
-
-
-DEFAULT_LOAD_K = 31
-VERSION = '2.2.0'
+from . import sig
+from . import storage
 
 
 class SourmashParser(ArgumentParser):
@@ -36,77 +40,66 @@ def print_citation(self):
         notify("== Please cite Brown and Irber (2016), doi:10.21105/joss.00027. ==\n")
         self._citation_printed = True
 
+    def _subparser_from_name(self, name):
+        """Given a name, get the subparser instance registered with this parser."""
+        container = self._actions
+        if name is None:
+            return None
+        for action in container:
+            if action.choices is None:
+                continue
+            elif name in action.choices:
+                return action.choices[name]
+
     def parse_args(self, args=None, namespace=None):
         if (args is None and len(sys.argv) == 1) or (args is not None and len(args) == 0):
             self.print_help()
-            raise SystemExit(0)
+            raise SystemExit(1)
         args = super(SourmashParser, self).parse_args(args=args, namespace=namespace)
         if ('quiet' not in args or not args.quiet) and self.citation:
             self.print_citation()
 
+        if 'subcmd' in args and args.subcmd is None:
+            self._subparser_from_name(args.cmd).print_help()
+            raise SystemExit(1)
+
         # BEGIN: dirty hacks to simultaneously support new and previous interface
         if hasattr(args, 'subcmd') and args.subcmd == 'import':
             args.subcmd = 'ingest'
         if hasattr(args, 'cmd') and args.cmd == 'sbt_combine':
             args.cmd = 'sbt'
             args.subcmd = 'combine'
-        if hasattr(args, 'cmd') and args.cmd == 'index':
-            args.cmd = 'sbt'
-            args.subcmd = 'index'
-        if hasattr(args, 'cmd') and args.cmd == 'categorize':
+        if hasattr(args, 'cmd') and args.cmd == 'migrate':
             args.cmd = 'sbt'
-            args.subcmd = 'categorize'
-        if hasattr(args, 'cmd') and args.cmd == 'watch':
-            args.subcmd = 'sbt'
-            args.subcmd = 'watch'
+            args.subcmd = 'migrate'
         if hasattr(args, 'subcmd') and args.subcmd == 'compare_csv':
             args.subcmd = 'compare'
         # END: dirty hacks to simultaneously support new and previous interface
         return args
 
 
-def add_moltype_args(parser):
-    parser.add_argument(
-        '--protein', dest='protein', action='store_true',
-        help='choose a protein signature; by default, a nucleotide signature is used'
-    )
-    parser.add_argument(
-        '--dayhoff', dest='dayhoff', action='store_true',
-        help='build Dayhoff-encoded amino acid signatures'
-    )
-    parser.add_argument(
-        '--hp', '--hydrophobic-polar', dest='hp', action='store_true',
-        help='build hydrophobic-polar-encoded amino acid signatures'
-    )
-
-
-def add_ksize_arg(parser, default=21):
-    parser.add_argument(
-        '-k', '--ksize', metavar='K', default=None, type=int,
-        help='k-mer size; default={d}'.format(d=default)
-    )
-
-
 def get_parser():
-    commands = ['compute', 'compare', 'search', 'plot', 'gather', 'lca', 'sbt', 'info', 'signature']
+    commands = ['compute', 'compare', 'search', 'plot', 'gather', 'index',
+                'lca', 'sbt', 'info', 'sig', 'categorize', 'watch', 'storage',
+                'multigather', 'migrate', 'sbt_combine', 'import_csv']
     commandstr = ' -- '.join(sorted(commands))
 
     desc = 'Compute, compare, manipulate, and analyze MinHash sketches of DNA sequences.'
     parser = SourmashParser(prog='sourmash', description=desc)
     parser._optionals.title = 'Options'
-    parser.add_argument('-v', '--version', action='version', version='sourmash '+ VERSION)
+    parser.add_argument('-v', '--version', action='version', version='sourmash '+ sourmash.VERSION)
     parser.add_argument('-q', '--quiet', action='store_true', help='don\'t print citation information')
     sub = parser.add_subparsers(
         title='Commands', dest='cmd', metavar='cmd', help=commandstr,
         description='Invoke "sourmash <cmd> --help" for more details on executing each command.'
     )
     for cmd in commands:
+        if cmd in ('migrate', 'sbt_combine'):
+            continue
         getattr(sys.modules[__name__], cmd).subparser(sub)
     # BEGIN: dirty hacks to simultaneously support new and previous interface
-    sbt.categorize.subparser(sub)
     sbt.combine.alt_subparser(sub)
-    sbt.index.subparser(sub)
-    sbt.watch.subparser(sub)
+    sbt.migrate.subparser(sub)
     # END: dirty hacks to simultaneously support new and previous interface
     parser._action_groups.reverse()
     return parser

sourmash/cli/sbt/categorize.py → sourmash/cli/categorize.py

@@ -1,5 +1,7 @@
-import sourmash
-from sourmash.cli.utils import add_ksize_arg
+import argparse
+
+from sourmash.cli.utils import add_ksize_arg, add_moltype_args
+
 
 def subparser(subparsers):
     subparser = subparsers.add_parser('categorize')
@@ -22,7 +24,13 @@ def subparser(subparsers):
         '--ignore-abundance', action='store_true',
         help='do NOT use k-mer abundances if present'
     )
+    add_moltype_args(subparser)
+
+    # TODO: help messages in these
+    subparser.add_argument('--csv', type=argparse.FileType('at'))
+    subparser.add_argument('--load-csv', default=None)
 
 
 def main(args):
-    print(args)
+    import sourmash
+    return sourmash.commands.categorize(args)

sourmash/cli/compare.py

@@ -1,7 +1,8 @@
 from argparse import FileType
-import sourmash
+
 from sourmash.cli.utils import add_ksize_arg, add_moltype_args
 
+
 def subparser(subparsers):
     subparser = subparsers.add_parser('compare')
     subparser.add_argument(
@@ -36,4 +37,5 @@ def subparser(subparsers):
 
 
 def main(args):
-    print(args)
+    import sourmash
+    return sourmash.commands.compare(args)

sourmash/cli/compute.py

@@ -1,7 +1,11 @@
 from argparse import FileType
-import sourmash
+
 from sourmash._minhash import get_minhash_default_seed
-from sourmash.cli.utils import add_ksize_arg, add_moltype_args
+from sourmash.cli.utils import (
+    add_ksize_arg,
+    add_moltype_args,
+    add_construct_moltype_args
+)
 
 
 def subparser(subparsers):
@@ -24,7 +28,7 @@ def subparser(subparsers):
         '--scaled', type=float, default=0,
         help='choose number of hashes as 1 in FRACTION of input k-mers'
     )
-    add_moltype_args(sketch_args)
+    add_construct_moltype_args(sketch_args)
     sketch_args.add_argument(
         '--input-is-protein', action='store_true',
         help='Consume protein sequences - no translation needed.'
@@ -123,4 +127,5 @@ def subparser(subparsers):
 
 
 def main(args):
-    print(args)
+    from sourmash.command_compute import compute
+    return compute(args)

sourmash/cli/gather.py

@@ -1,6 +1,8 @@
 from argparse import FileType
+
 from sourmash.cli.utils import add_ksize_arg, add_moltype_args
 
+
 def subparser(subparsers):
     subparser = subparsers.add_parser('gather')
     subparser.add_argument('query', help='query signature')
@@ -50,4 +52,5 @@ def subparser(subparsers):
 
 
 def main(args):
-    print(args)
+    import sourmash
+    return sourmash.commands.gather(args)

sourmash/cli/import_csv.py

@@ -0,0 +1,17 @@
+from argparse import FileType
+import sys
+
+
+def subparser(subparsers):
+    subparser = subparsers.add_parser('import_csv')
+    subparser.add_argument('mash_csvfile', help='CSV file with mash sketches')
+    subparser.add_argument(
+        '-o', '--output', type=FileType('wt'),
+        default=sys.stdout,
+        help='save signature generated from data here'
+    )
+
+
+def main(args):
+    import sourmash
+    return sourmash.commands.import_csv(args)

sourmash/cli/sbt/index.py → sourmash/cli/index.py

@@ -1,6 +1,6 @@
-import sourmash
 from sourmash.cli.utils import add_moltype_args, add_ksize_arg
 
+
 def subparser(subparsers):
     subparser = subparsers.add_parser('index')
     subparser.add_argument('sbt_name', help='name to save SBT into')
@@ -46,4 +46,5 @@ def subparser(subparsers):
 
 
 def main(args):
-    print(args)
+    import sourmash
+    return sourmash.commands.index(args)

sourmash/cli/info.py

@@ -7,7 +7,7 @@
 def subparser(subparsers):
     subparser = subparsers.add_parser('info')
     subparser.add_argument(
-        '--verbose', action='store_true',
+        '-v', '--verbose', action='store_true',
         help='report versions of khmer and screed'
     )
 

sourmash/cli/lca/classify.py

@@ -1,5 +1,6 @@
 from argparse import FileType
 
+
 def subparser(subparsers):
     subparser = subparsers.add_parser('classify')
     subparser.add_argument('--db', nargs='+', action='append')
@@ -26,4 +27,5 @@ def subparser(subparsers):
 
 
 def main(args):
-    print(args)
+    import sourmash
+    return sourmash.lca.command_classify.classify(args)

sourmash/cli/lca/compare.py

@@ -1,56 +1,33 @@
 def subparser(subparsers):
-    subparser = subparsers.add_parser('compare')
-    subparser.add_argument('csv1', help='taxonomy spreadsheet output by classify')
-    subparser.add_argument('csv2', help='custom taxonomy spreadsheet')
-    subparser.add_argument(
-        '-q', '--quiet', action='store_true',
-        help='suppress non-error output'
-    )
-    subparser.add_argument(
-        '-d', '--debug', action='store_true',
-        help='output debugging output'
-    )
-    subparser.add_argument(
-        '-C', '--start-column', metavar='C', default=2, type=int,
-        help='column at which taxonomic assignments start; default=2'
-    )
-    subparser.add_argument(
-        '--tabs', action='store_true',
-        help='input spreadsheet is tab-delimited; default is commas'
-    )
-    subparser.add_argument(
-        '--no-headers', action='store_true',
-        help='no headers present in taxonomy spreadsheet'
-    )
-    subparser.add_argument('-f', '--force', action='store_true')
-
     # Dirty hack to simultaneously support new and previous interface
     # If desired, this function can be removed with a major version bump.
-    subparser = subparsers.add_parser('compare_csv')
-    subparser.add_argument('csv1', help='taxonomy spreadsheet output by classify')
-    subparser.add_argument('csv2', help='custom taxonomy spreadsheet')
-    subparser.add_argument(
-        '-q', '--quiet', action='store_true',
-        help='suppress non-error output'
-    )
-    subparser.add_argument(
-        '-d', '--debug', action='store_true',
-        help='output debugging output'
-    )
-    subparser.add_argument(
-        '-C', '--start-column', metavar='C', default=2, type=int,
-        help='column at which taxonomic assignments start; default=2'
-    )
-    subparser.add_argument(
-        '--tabs', action='store_true',
-        help='input spreadsheet is tab-delimited; default is commas'
-    )
-    subparser.add_argument(
-        '--no-headers', action='store_true',
-        help='no headers present in taxonomy spreadsheet'
-    )
-    subparser.add_argument('-f', '--force', action='store_true')
+    for cmd in ('compare', 'compare_csv'):
+        subparser = subparsers.add_parser(cmd)
+        subparser.add_argument('csv1', help='taxonomy spreadsheet output by classify')
+        subparser.add_argument('csv2', help='custom taxonomy spreadsheet')
+        subparser.add_argument(
+            '-q', '--quiet', action='store_true',
+            help='suppress non-error output'
+        )
+        subparser.add_argument(
+            '-d', '--debug', action='store_true',
+            help='output debugging output'
+        )
+        subparser.add_argument(
+            '-C', '--start-column', metavar='C', default=2, type=int,
+            help='column at which taxonomic assignments start; default=2'
+        )
+        subparser.add_argument(
+            '--tabs', action='store_true',
+            help='input spreadsheet is tab-delimited; default is commas'
+        )
+        subparser.add_argument(
+            '--no-headers', action='store_true',
+            help='no headers present in taxonomy spreadsheet'
+        )
+        subparser.add_argument('-f', '--force', action='store_true')
 
 
 def main(args):
-    print(args)
+    import sourmash
+    return sourmash.lca.command_compare_csv.compare_csv(args)

sourmash/cli/lca/gather.py

@@ -1,5 +1,6 @@
 from argparse import FileType
 
+
 def subparser(subparsers):
     subparser = subparsers.add_parser('gather')
     subparser.add_argument('query')
@@ -27,4 +28,5 @@ def subparser(subparsers):
 
 
 def main(args):
-    print(args)
+    import sourmash
+    return sourmash.lca.gather_main(args)