* Correct Sniffer doc to correspond to the implementation.

Skip Montanaro · Skip Montanaro · commit 77892373313b · 2003-05-19T15:33:36.000Z
* Add optional delimiters arg to Sniffer.sniff() which restricts the set of
  candidate field delimiters.
diff --git a/Doc/lib/libcsv.tex b/Doc/lib/libcsv.tex
@@ -152,17 +152,17 @@ \subsection{Module Contents}
 \class{reader} or \class{writer} instance.
 \end{classdesc*}
 
-\begin{classdesc}{Sniffer}{\optional{sample=16384}}
-The \class{Sniffer} class is used to deduce the format of a CSV file.  The
-optional \var{sample} argument to the constructor specifies the number of
-bytes to use when determining Dialect parameters.
+\begin{classdesc}{Sniffer}{}
+The \class{Sniffer} class is used to deduce the format of a CSV file.
 \end{classdesc}
 
 The \class{Sniffer} class provides a single method:
 
-\begin{methoddesc}{sniff}{fileobj}
-Analyze the next chunk of \var{fileobj} and return a \class{Dialect} subclass
-reflecting the parameters found.
+\begin{methoddesc}{sniff}{sample\optional{,delimiters=None}}
+Analyze the given \var{sample} and return a \class{Dialect} subclass
+reflecting the parameters found.  If the optional \var{delimiters} parameter
+is given, it is interpreted as a string containing possible valid delimiter
+characters.
 \end{methoddesc}
 
 \begin{methoddesc}{has_header}{sample}
diff --git a/Lib/csv.py b/Lib/csv.py
@@ -159,15 +159,16 @@ def __init__(self):
         self.preferred = [',', '\t', ';', ' ', ':']
 
 
-    def sniff(self, sample):
+    def sniff(self, sample, delimiters=None):
         """
         Returns a dialect (or None) corresponding to the sample
         """
 
         quotechar, delimiter, skipinitialspace = \
-                   self._guess_quote_and_delimiter(sample)
+                   self._guess_quote_and_delimiter(sample, delimiters)
         if delimiter is None:
-            delimiter, skipinitialspace = self._guess_delimiter(sample)
+            delimiter, skipinitialspace = self._guess_delimiter(sample,
+                                                                delimiters)
 
         class dialect(Dialect):
             _name = "sniffed"
@@ -184,7 +185,7 @@ class dialect(Dialect):
         return dialect
 
 
-    def _guess_quote_and_delimiter(self, data):
+    def _guess_quote_and_delimiter(self, data, delimiters):
         """
         Looks for text enclosed between two identical quotes
         (the probable quotechar) which are preceded and followed
@@ -222,7 +223,7 @@ def _guess_quote_and_delimiter(self, data):
                 key = m[n]
             except KeyError:
                 continue
-            if key:
+            if key and (delimiters is None or key in delimiters):
                 delims[key] = delims.get(key, 0) + 1
             try:
                 n = regexp.groupindex['space'] - 1
@@ -248,7 +249,7 @@ def _guess_quote_and_delimiter(self, data):
         return (quotechar, delim, skipinitialspace)
 
 
-    def _guess_delimiter(self, data):
+    def _guess_delimiter(self, data, delimiters):
         """
         The delimiter /should/ occur the same number of times on
         each row. However, due to malformed data, it may not. We don't want
@@ -316,7 +317,8 @@ def _guess_delimiter(self, data):
             while len(delims) == 0 and consistency >= threshold:
                 for k, v in modeList:
                     if v[0] > 0 and v[1] > 0:
-                        if (v[1]/total) >= consistency:
+                        if ((v[1]/total) >= consistency and
+                            (delimiters is None or k in delimiters)):
                             delims[k] = v
                 consistency -= 0.01
 
diff --git a/Lib/test/test_csv.py b/Lib/test/test_csv.py
@@ -551,6 +551,12 @@ class TestSniffer(unittest.TestCase):
     header = '''\
 "venue","city","state","date","performers"
 '''
+    sample3 = '''\
+05/05/03?05/05/03?05/05/03?05/05/03?05/05/03?05/05/03
+05/05/03?05/05/03?05/05/03?05/05/03?05/05/03?05/05/03
+05/05/03?05/05/03?05/05/03?05/05/03?05/05/03?05/05/03
+'''
+
     def test_has_header(self):
         sniffer = csv.Sniffer()
         self.assertEqual(sniffer.has_header(self.sample1), False)
@@ -568,6 +574,15 @@ def test_sniff(self):
         self.assertEqual(dialect.quotechar, "'")
         self.assertEqual(dialect.skipinitialspace, False)
 
+    def test_delimiters(self):
+        sniffer = csv.Sniffer()
+        dialect = sniffer.sniff(self.sample3)
+        self.assertEqual(dialect.delimiter, "0")
+        dialect = sniffer.sniff(self.sample3, delimiters="?,")
+        self.assertEqual(dialect.delimiter, "?")
+        dialect = sniffer.sniff(self.sample3, delimiters="/,")
+        self.assertEqual(dialect.delimiter, "/")
+
 if not hasattr(sys, "gettotalrefcount"):
     if test_support.verbose: print "*** skipping leakage tests ***"
 else:
