Merge pull request #1596: Use ngettext to handle plurality

Author: victorlin

augur/filter/_run.py

@@ -18,7 +18,7 @@
 from augur.io.file import PANDAS_READ_CSV_OPTIONS, open_file
 from augur.io.metadata import InvalidDelimiter, Metadata, read_metadata
 from augur.io.sequences import read_sequences, write_sequences
-from augur.io.print import print_err
+from augur.io.print import print_err, _n
 from augur.io.vcf import is_vcf as filename_is_vcf, write_vcf
 from augur.types import EmptyOutputReportingMethod
 from . import include_exclude_rules
@@ -432,7 +432,7 @@ def run(args):
     total_strains_passed = len(valid_strains)
     total_strains_filtered = len(metadata_strains) + num_excluded_by_lack_of_metadata - total_strains_passed
 
-    print_err(f"{total_strains_filtered} {'strain was' if total_strains_filtered == 1 else 'strains were'} dropped during filtering")
+    print_err(f"{total_strains_filtered} {_n('strain was', 'strains were', total_strains_filtered)} dropped during filtering")
 
     if num_excluded_by_lack_of_metadata:
         print_err(f"\t{num_excluded_by_lack_of_metadata} had no metadata")
@@ -462,13 +462,13 @@ def run(args):
             parameters = {}
 
         parameters["count"] = count
-        parameters["were"] = "was" if count == 1 else "were"
-        parameters["they"] = "it"  if count == 1 else "they"
+        parameters["were"] = _n("was", "were", count)
+        parameters["they"] = _n("it", "they", count)
         print_err("\t" + report_template_by_filter_name[filter_name].format(**parameters))
 
     if (group_by and args.sequences_per_group) or args.subsample_max_sequences:
         seed_txt = ", using seed {}".format(args.subsample_seed) if args.subsample_seed else ""
-        print_err(f"\t{num_excluded_subsamp} {'was' if num_excluded_subsamp == 1 else 'were'} dropped because of subsampling criteria{seed_txt}")
+        print_err(f"\t{num_excluded_subsamp} {_n('was', 'were', num_excluded_subsamp)} dropped because of subsampling criteria{seed_txt}")
 
     if total_strains_passed == 0:
         empty_results_message = "All samples have been dropped! Check filter rules and metadata file format."
@@ -481,4 +481,4 @@ def run(args):
         else:
             raise ValueError(f"Encountered unhandled --empty-output-reporting method {args.empty_output_reporting!r}")
 
-    print_err(f"{total_strains_passed} {'strain' if total_strains_passed == 1 else 'strains'} passed all filters")
+    print_err(f"{total_strains_passed} {_n('strain', 'strains', total_strains_passed)} passed all filters")

augur/filter/subsample.py

@@ -10,7 +10,7 @@
 from augur.dates import get_year_month, get_year_week
 from augur.errors import AugurError
 from augur.io.metadata import METADATA_DATE_COLUMN
-from augur.io.print import print_err
+from augur.io.print import print_err, _n
 from . import constants
 from .weights_file import WEIGHTS_COLUMN, COLUMN_VALUE_FOR_DEFAULT_WEIGHT, get_default_weight, get_weighted_columns, read_weights_file
 
@@ -352,8 +352,8 @@ def get_weighted_group_sizes(
     # Warn on any under-sampled groups
     for _, row in weights.iterrows():
         if row[INPUT_SIZE_COLUMN] < row[TARGET_SIZE_COLUMN]:
-            sequences = 'sequence' if row[TARGET_SIZE_COLUMN] == 1 else 'sequences'
-            are = 'is' if row[INPUT_SIZE_COLUMN] == 1 else 'are'
+            sequences = _n('sequence', 'sequences', row[TARGET_SIZE_COLUMN])
+            are = _n('is', 'are', row[INPUT_SIZE_COLUMN])
             group = list(f'{col}={value!r}' for col, value in row[group_by].items())
             print_err(f"WARNING: Targeted {row[TARGET_SIZE_COLUMN]} {sequences} for group {group} but only {row[INPUT_SIZE_COLUMN]} {are} available.")
 
@@ -411,7 +411,7 @@ def _drop_unused_groups(
     extra_groups = set(weights.index) - valid_index
     if extra_groups:
         count = len(extra_groups)
-        unit = "group" if count == 1 else "groups"
+        unit = _n("group", "groups", count)
         print_err(f"NOTE: Skipping {count} {unit} due to lack of entries in metadata.")
         weights = weights[weights.index.isin(valid_index)]
 
@@ -427,8 +427,8 @@ def _adjust_weights_for_unweighted_columns(
     ) -> pd.DataFrame:
     """Adjust weights for unweighted columns to reflect equal weighting within each weighted group.
     """
-    columns = 'column' if len(unweighted_columns) == 1 else 'columns'
-    those = 'that' if len(unweighted_columns) == 1 else 'those'
+    columns = _n('column', 'columns', len(unweighted_columns))
+    those = _n('that', 'those', len(unweighted_columns))
     print_err(f"NOTE: Weights were not provided for the {columns} {', '.join(repr(col) for col in unweighted_columns)}. Using equal weights across values in {those} {columns}.")        
 
     weights_grouped = weights.groupby(weighted_columns)

augur/io/print.py

@@ -1,3 +1,4 @@
+import gettext
 import sys
 
 from augur.debug import DEBUGGING
@@ -13,3 +14,9 @@ def print_debug(*args):
     """Print to stderr if in debugging mode."""
     if DEBUGGING:
         print_err(*args)
+
+
+# Use ngettext() without a message catalog for its singular/plural handling so
+# we can make proper error messages.  gettext() (no "n") is conventionally
+# aliased as "_", so alias ngettext() as "_n".
+_n = gettext.NullTranslations().ngettext

augur/merge.py

@@ -35,7 +35,6 @@
 you want to use a version different from what's on PATH), set the SQLITE3
 environment variable to path of the desired sqlite3 executable.
 """
-import gettext
 import os
 import re
 import subprocess
@@ -51,19 +50,13 @@
 from augur.argparse_ import ExtendOverwriteDefault, SKIP_AUTO_DEFAULT_IN_HELP
 from augur.errors import AugurError
 from augur.io.metadata import DEFAULT_DELIMITERS, DEFAULT_ID_COLUMNS, Metadata
-from augur.io.print import print_err, print_debug
+from augur.io.print import print_err, print_debug, _n
 from augur.utils import first_line
 
 
 T = TypeVar('T')
 
 
-# Use ngettext() without a message catalog for its singular/plural handling so
-# we can make proper error messages.  gettext() (no "n") is conventionally
-# aliased as "_", so alias ngettext() as "_n".
-_n = gettext.NullTranslations().ngettext
-
-
 class NamedMetadata(Metadata):
     name: str
     """User-provided descriptive name for this metadata file."""