Add support for running detectors from command line

comparison/detector_clevercsv.py

@@ -9,9 +9,8 @@
 """
 
 import clevercsv
-import sys
 
-from utils import DetectionError, get_sample
+from utils import DetectionError, get_sample, parse_args
 
 
 def detector(gz_filename, encoding, n_lines=None):
@@ -31,10 +30,8 @@ def detector(gz_filename, encoding, n_lines=None):
         escapechar=dialect.escapechar,
     )
 
+
 if __name__ == "__main__":
-    if len(sys.argv) > 1:
-        filename = sys.argv[1]
-        encoding = clevercsv.utils.get_encoding(filename)
-        print(detector(filename, encoding))
-    else:
-        print(f"Usage: {sys.argv[0]} filename", file=sys.stderr)
+    args = parse_args()
+    encoding = clevercsv.utils.get_encoding(args.filename)
+    print(detector(args.filename, encoding, n_lines=args.n))

comparison/detector_clevercsv_grow.py

@@ -9,9 +9,8 @@
 """
 
 import clevercsv
-import sys
 
-from utils import DetectionError, get_sample, count_lines
+from utils import DetectionError, get_sample, count_lines, parse_args
 
 
 def trailing_equality(l, k=4):
@@ -73,9 +72,6 @@ def detector(filename, encoding, n_lines=None, lines_start=100, n_equal=5):
 
 
 if __name__ == "__main__":
-    if len(sys.argv) > 1:
-        filename = sys.argv[1]
-        encoding = clevercsv.utils.get_encoding(filename)
-        print(detector(filename, encoding))
-    else:
-        print(f"Usage: {sys.argv[0]} filename", file=sys.stderr)
+    args = parse_args()
+    encoding = clevercsv.utils.get_encoding(args.filename)
+    print(detector(args.filename, encoding, n_lines=args.n))

comparison/detector_simple.py

@@ -8,7 +8,8 @@
 
 """
 
-from utils import get_sample
+
+from utils import get_sample, parse_args
 
 
 def detector(gz_filename, encoding, n_lines=None):
@@ -18,3 +19,11 @@ def detector(gz_filename, encoding, n_lines=None):
         quotechar = '"'
 
     return dict(delimiter=",", quotechar=quotechar, escapechar="")
+
+
+if __name__ == "__main__":
+    from clevercsv.utils import get_encoding
+
+    args = parse_args()
+    encoding = get_encoding(args.filename)
+    print(detector(args.filename, encoding, n_lines=args.n))

comparison/detector_sniffer.py

@@ -10,7 +10,7 @@
 
 import csv
 
-from utils import DetectionError, get_sample
+from utils import DetectionError, get_sample, parse_args
 
 
 def detector(gz_filename, encoding, n_lines=None):
@@ -33,3 +33,10 @@ def detector(gz_filename, encoding, n_lines=None):
             return None
     dialect = dict(delimiter=delimiter, quotechar=quotechar, escapechar="")
     return dialect
+
+if __name__ == "__main__":
+    from clevercsv.utils import get_encoding
+
+    args = parse_args()
+    encoding = get_encoding(args.filename)
+    print(detector(args.filename, encoding, n_lines=args.n))

comparison/utils.py

@@ -7,8 +7,9 @@
 Author: Gertjan van den Burg
 """
 
-import gzip
+import argparse
 import chardet
+import gzip
 
 
 class DetectionError(Exception):
@@ -63,3 +64,17 @@ def count_lines(filename, encoding):
 
     fp.close()
     return n_lines
+
+
+def parse_args():
+    parser = argparse.ArgumentParser()
+    parser.add_argument(
+        "-n",
+        help="Number of lines to use for detection",
+        default=None,
+        type=int,
+    )
+    parser.add_argument(
+        "filename", help="File to detect dialect for (.csv or .csv.gz)"
+    )
+    return parser.parse_args()