diff options
| author | Miss Islington (bot) <31488909+miss-islington@users.noreply.github.com> | 2021-07-30 17:30:09 (GMT) |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2021-07-30 17:30:09 (GMT) |
| commit | 440c9f772a9b66c1ea387c1c3efc9ff438880acf (patch) | |
| tree | b0383b5f00c3c76663c7947a2ce9d842d1b42aea /Lib | |
| parent | 6f950023c6a2168b229363d75f59a24ecdd66d19 (diff) | |
| download | cpython-440c9f772a9b66c1ea387c1c3efc9ff438880acf.zip cpython-440c9f772a9b66c1ea387c1c3efc9ff438880acf.tar.gz cpython-440c9f772a9b66c1ea387c1c3efc9ff438880acf.tar.bz2 | |
bpo-43625: Enhance csv sniffer has_headers() to be more accurate (GH-26939) (GH-27494)
(cherry picked from commit ceea579ccc51791f3e115155d6f27905bc7544a9)
Co-authored-by: andrei kulakov <andrei.avk@gmail.com>
Diffstat (limited to 'Lib')
| -rw-r--r-- | Lib/csv.py | 12 | ||||
| -rw-r--r-- | Lib/test/test_csv.py | 36 |
2 files changed, 40 insertions, 8 deletions
@@ -409,14 +409,10 @@ class Sniffer: continue # skip rows that have irregular number of columns for col in list(columnTypes.keys()): - - for thisType in [int, float, complex]: - try: - thisType(row[col]) - break - except (ValueError, OverflowError): - pass - else: + thisType = complex + try: + thisType(row[col]) + except (ValueError, OverflowError): # fallback to length of string thisType = len(row[col]) diff --git a/Lib/test/test_csv.py b/Lib/test/test_csv.py index 18b86aa..09e72a7 100644 --- a/Lib/test/test_csv.py +++ b/Lib/test/test_csv.py @@ -1020,6 +1020,42 @@ Stonecutters Seafood and Chop House+ Lemont+ IL+ 12/19/02+ Week Back 'Stonecutters ''Seafood'' and Chop House'+ 'Lemont'+ 'IL'+ '12/19/02'+ 'Week Back' """ + sample10 = dedent(""" + abc,def + ghijkl,mno + ghi,jkl + """) + + sample11 = dedent(""" + abc,def + ghijkl,mnop + ghi,jkl + """) + + sample12 = dedent(""""time","forces" + 1,1.5 + 0.5,5+0j + 0,0 + 1+1j,6 + """) + + sample13 = dedent(""""time","forces" + 0,0 + 1,2 + a,b + """) + + def test_issue43625(self): + sniffer = csv.Sniffer() + self.assertTrue(sniffer.has_header(self.sample12)) + self.assertFalse(sniffer.has_header(self.sample13)) + + def test_has_header_strings(self): + "More to document existing (unexpected?) behavior than anything else." + sniffer = csv.Sniffer() + self.assertFalse(sniffer.has_header(self.sample10)) + self.assertFalse(sniffer.has_header(self.sample11)) + def test_has_header(self): sniffer = csv.Sniffer() self.assertIs(sniffer.has_header(self.sample1), False) |