Test LIKE with implicit \ escape

apache · Nov 7, 2024 · f33b075 · f33b075
1 parent 1094651
commit f33b075
Show file tree

Hide file tree

Showing 6 changed files with 247 additions and 2 deletions.
diff --git a/datafusion/sqllogictest/test_files/string/dictionary_utf8.slt b/datafusion/sqllogictest/test_files/string/dictionary_utf8.slt
@@ -59,6 +59,8 @@ Raphael datafusionДатаФусион false false false false
 under_score un iść core false false false false
 percent pan Tadeusz ma iść w kąt false false false false
 (empty) (empty) false false false false
+% (empty) false false false false
+_ (empty) false false false false
 NULL NULL NULL NULL NULL NULL
 NULL NULL NULL NULL NULL NULL
 

diff --git a/datafusion/sqllogictest/test_files/string/init_data.slt.part b/datafusion/sqllogictest/test_files/string/init_data.slt.part
@@ -23,6 +23,8 @@ create table test_source as values
   ('under_score', 'un_____core', 'un iść core', 'chrząszcz na łące w 東京都'),
   ('percent', 'p%t', 'pan Tadeusz ma iść w kąt', 'Pan Tadeusz ma frunąć stąd w kąt'),
   ('', '%', '', ''),
+  ('%', '\%', '', ''),
+  ('_', '\_', '', ''),
   (NULL, '%', NULL, NULL),
   (NULL, 'R', NULL, '🔥');
 

diff --git a/datafusion/sqllogictest/test_files/string/large_string.slt b/datafusion/sqllogictest/test_files/string/large_string.slt
@@ -44,6 +44,8 @@ Raphael R datafusionДатаФусион аФус
 under_score un_____core un iść core chrząszcz na łące w 東京都
 percent p%t pan Tadeusz ma iść w kąt Pan Tadeusz ma frunąć stąd w kąt
 (empty) % (empty) (empty)
+% \% (empty) (empty)
+_ \_ (empty) (empty)
 NULL % NULL NULL
 NULL R NULL 🔥
 
@@ -64,6 +66,8 @@ Raphael datafusionДатаФусион false false false false
 under_score un iść core false false false false
 percent pan Tadeusz ma iść w kąt false false false false
 (empty) (empty) false false false false
+% (empty) false false false false
+_ (empty) false false false false
 NULL NULL NULL NULL NULL NULL
 NULL NULL NULL NULL NULL NULL
 

diff --git a/datafusion/sqllogictest/test_files/string/string.slt b/datafusion/sqllogictest/test_files/string/string.slt
@@ -51,6 +51,8 @@ Raphael datafusionДатаФусион false false false false
 under_score un iść core false false false false
 percent pan Tadeusz ma iść w kąt false false false false
 (empty) (empty) false false false false
+% (empty) false false false false
+_ (empty) false false false false
 NULL NULL NULL NULL NULL NULL
 NULL NULL NULL NULL NULL NULL
 
@@ -74,13 +76,19 @@ SELECT unicode_2, 'is LIKE', ascii_2 FROM test_basic_operator WHERE unicode_2 LI
 UNION ALL
 SELECT unicode_2, 'is NOT LIKE', ascii_2 FROM test_basic_operator WHERE unicode_2 NOT LIKE ascii_2
 ----
+% is LIKE \%
 (empty) is LIKE %
 (empty) is LIKE %
 (empty) is LIKE %
+(empty) is NOT LIKE \%
+(empty) is NOT LIKE \%
+(empty) is NOT LIKE \_
+(empty) is NOT LIKE \_
 Andrew is NOT LIKE X
 Pan Tadeusz ma frunąć stąd w kąt is NOT LIKE p%t
 Raphael is NOT LIKE R
 Xiangpeng is LIKE Xiangpeng
+_ is LIKE \_
 chrząszcz na łące w 東京都 is NOT LIKE un_____core
 datafusionДатаФусион is NOT LIKE R
 datafusion数据融合 is NOT LIKE Xiangpeng
@@ -105,12 +113,14 @@ SELECT
     (unicode_2 LIKE ascii_2) AS unicode_2_like_ascii_2
 FROM test_basic_operator
 ----
+% \% (empty) (empty) true true false false
 (empty) % (empty) (empty) true false true true
 Andrew X datafusion📊🔥 🔥 false false false false
 NULL % NULL NULL NULL NULL NULL NULL
 NULL R NULL 🔥 NULL NULL NULL false
 Raphael R datafusionДатаФусион аФус false false false false
 Xiangpeng Xiangpeng datafusion数据融合 datafusion数据融合 true true false false
+_ \_ (empty) (empty) true false false false
 percent p%t pan Tadeusz ma iść w kąt Pan Tadeusz ma frunąć stąd w kąt true false true false
 under_score un_____core un iść core chrząszcz na łące w 東京都 true false true false
 
@@ -129,13 +139,19 @@ SELECT unicode_2, 'is ILIKE', ascii_2 FROM test_basic_operator WHERE unicode_2 I
 UNION ALL
 SELECT unicode_2, 'is NOT ILIKE', ascii_2 FROM test_basic_operator WHERE unicode_2 NOT ILIKE ascii_2
 ----
+% is ILIKE \%
 (empty) is ILIKE %
 (empty) is ILIKE %
 (empty) is ILIKE %
+(empty) is NOT ILIKE \%
+(empty) is NOT ILIKE \%
+(empty) is NOT ILIKE \_
+(empty) is NOT ILIKE \_
 Andrew is NOT ILIKE X
 Pan Tadeusz ma frunąć stąd w kąt is ILIKE p%t
 Raphael is NOT ILIKE R
 Xiangpeng is ILIKE Xiangpeng
+_ is ILIKE \_
 chrząszcz na łące w 東京都 is NOT ILIKE un_____core
 datafusionДатаФусион is NOT ILIKE R
 datafusion数据融合 is NOT ILIKE Xiangpeng
@@ -160,12 +176,14 @@ SELECT
     (unicode_2 ILIKE ascii_2) AS unicode_2_ilike_ascii_2
 FROM test_basic_operator
 ----
+% \% (empty) (empty) true true false false
 (empty) % (empty) (empty) true false true true
 Andrew X datafusion📊🔥 🔥 false false false false
 NULL % NULL NULL NULL NULL NULL NULL
 NULL R NULL 🔥 NULL NULL NULL false
 Raphael R datafusionДатаФусион аФус false false false false
 Xiangpeng Xiangpeng datafusion数据融合 datafusion数据融合 true true false false
+_ \_ (empty) (empty) true false false false
 percent p%t pan Tadeusz ma iść w kąt Pan Tadeusz ma frunąć stąd w kąt true false true true
 under_score un_____core un iść core chrząszcz na łące w 東京都 true false true false
 

diff --git a/datafusion/sqllogictest/test_files/string/string_literal.slt b/datafusion/sqllogictest/test_files/string/string_literal.slt
@@ -15,6 +15,15 @@
 # specific language governing permissions and limitations
 # under the License.
 
+# a backslash in a string literal is not a special character
+query T
+VALUES ('\'), ('\\'), ('\\\'), ('\\\\')
+----
+\
+\\
+\\\
+\\\\
+
 query T
 SELECT substr('alphabet', -3)
 ----
@@ -829,3 +838,95 @@ SELECT
     'a' LIKE '%'
 ----
 NULL true true
+
+# \ is an implicit escape character
+query BBBB
+SELECT
+    'a' LIKE '\%',
+    '\a' LIKE '\%',
+    '%' LIKE '\%',
+    '\%' LIKE '\%'
+----
+false false true false
+
+# \ is an implicit escape character
+query BBBBBB
+SELECT
+    'a' LIKE '\_',
+    '\a' LIKE '\_',
+    '_' LIKE '\_',
+    '\_' LIKE '\_',
+    'abc' LIKE 'a_c',
+    'abc' LIKE 'a\_c'
+----
+false false true false true false
+
+# \ as an explicit escape character is currently not supported
+query error DataFusion error: SQL error: ParserError\("Expected end of statement, found: \\\\"\)
+SELECT
+    'a' LIKE '\%' ESCAPE '\',
+    '\a' LIKE '\%' ESCAPE '\',
+    '%' LIKE '\%' ESCAPE '\',
+    '\%' LIKE '\% ESCAPE '\''
+
+# \ as an explicit escape character is currently not supported
+query error DataFusion error: Execution error: LIKE does not support escape_char
+SELECT
+    'a' LIKE '\_' ESCAPE '\',
+    '\a' LIKE '\_' ESCAPE '\',
+    '_' LIKE '\_' ESCAPE '\',
+    '\_' LIKE '\_' ESCAPE '\',
+    'abc' LIKE 'a_c' ESCAPE '\',
+    'abc' LIKE 'a\_c' ESCAPE '\'
+
+# a LIKE pattern containing escape can never match an empty string
+query BBBBB
+SELECT
+    '' LIKE '\',
+    '' LIKE '\\',
+    '' LIKE '\_',
+    '' LIKE '\%',
+    '' LIKE '\a'
+----
+false false false false false
+
+# escape before non-wildcard matches the escape itself
+query BBBBBBB
+SELECT
+    'a' LIKE '\a',
+    '\a' LIKE '\a',
+    '\a' LIKE '\b',
+    '\' LIKE '\',
+    '\\' LIKE '\',
+    '\' LIKE '\\',
+    '\\' LIKE '\\'
+----
+false true false true false false true
+
+# if "%%" in the pattern was simplified to "%", the pattern semantics would change
+query BBBBB
+SELECT
+    '%' LIKE '\%%',
+    '%%' LIKE '\%%',
+    '\%%' LIKE '\%%',
+    '%abc' LIKE '\%%',
+    '\%abc' LIKE '\%%'
+----
+true true false true false
+
+statement ok
+create table inputs AS SELECT * FROM (VALUES ('%'), ('%%'), ('\%%'), ('%abc'), ('\%abc')) t(a);
+
+# if "%%" in the pattern was simplified to "%", the pattern semantics would change
+# same as above query, but with data coming from a table, so that constant folding cannot kick in, but expression simplification can
+query TB
+SELECT a, a LIKE '\%%' FROM inputs
+----
+% true
+%% true
+\%% false
+%abc true
+\%abc false
+
+statement ok
+drop table inputs;