[Python-checkins] cpython (2.7): #15545: sort iterdump via SQL instead of in python code

[r.david.murray]

http://hg.python.org/cpython/rev/bb4e4f0cec2e
changeset:   81370:bb4e4f0cec2e
branch:      2.7
parent:      81358:ae1d34142a2a
user:        R David Murray <rdmurray at bitdance.com>
date:        Thu Jan 10 11:30:51 2013 -0500
summary:
  #15545: sort iterdump via SQL instead of in python code

Although there is not a regression in Python2, we make the same update here to
keep the code bases in sync.

(The fix for issue 9750 introduced a regression in Python 3 by sorting the row
objects returned by fetchall.  But if a row_factory such as sqlite3.Row is
used, the rows may not be sortable (in Python3), which leads to an exception.
The sorting is still a nice idea, so the patch moves the sort into the sql.)

Fix and test by Peter Otten.

files:
  Lib/sqlite3/dump.py      |   3 ++-
  Lib/sqlite3/test/dump.py |  21 +++++++++++++++++++++
  Misc/NEWS                |   4 ++++
  3 files changed, 27 insertions(+), 1 deletions(-)


diff --git a/Lib/sqlite3/dump.py b/Lib/sqlite3/dump.py
--- a/Lib/sqlite3/dump.py
+++ b/Lib/sqlite3/dump.py
@@ -25,9 +25,10 @@
         FROM "sqlite_master"
             WHERE "sql" NOT NULL AND
             "type" == 'table'
+            ORDER BY "name"
         """
     schema_res = cu.execute(q)
-    for table_name, type, sql in sorted(schema_res.fetchall()):
+    for table_name, type, sql in schema_res.fetchall():
         if table_name == 'sqlite_sequence':
             yield('DELETE FROM "sqlite_sequence";')
         elif table_name == 'sqlite_stat1':
diff --git a/Lib/sqlite3/test/dump.py b/Lib/sqlite3/test/dump.py
--- a/Lib/sqlite3/test/dump.py
+++ b/Lib/sqlite3/test/dump.py
@@ -49,6 +49,27 @@
         [self.assertEqual(expected_sqls[i], actual_sqls[i])
             for i in xrange(len(expected_sqls))]
 
+    def CheckUnorderableRow(self):
+        # iterdump() should be able to cope with unorderable row types (issue #15545)
+        class UnorderableRow:
+            def __init__(self, cursor, row):
+                self.row = row
+            def __getitem__(self, index):
+                return self.row[index]
+        self.cx.row_factory = UnorderableRow
+        CREATE_ALPHA = """CREATE TABLE "alpha" ("one");"""
+        CREATE_BETA = """CREATE TABLE "beta" ("two");"""
+        expected = [
+            "BEGIN TRANSACTION;",
+            CREATE_ALPHA,
+            CREATE_BETA,
+            "COMMIT;"
+            ]
+        self.cu.execute(CREATE_BETA)
+        self.cu.execute(CREATE_ALPHA)
+        got = list(self.cx.iterdump())
+        self.assertEqual(expected, got)
+
 def suite():
     return unittest.TestSuite(unittest.makeSuite(DumpTests, "Check"))
 
diff --git a/Misc/NEWS b/Misc/NEWS
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -186,6 +186,10 @@
 Library
 -------
 
+- Issue #15545: Fix regression in sqlite3's iterdump method where it was
+  failing if the connection used a row factory (such as sqlite3.Row) that
+  produced unsortable objects. (Regression was introduced by fix for 9750).
+
 - Issue #16828: Fix error incorrectly raised by bz2.compress(''). Patch by
   Martin Packman.
 

-- 
Repository URL: http://hg.python.org/cpython