Added/Updated tests\bugs\gh_5749_test.py: Checked on 5.0.0.742 - all OK.

tests/bugs/gh_5749_test.py

@@ -0,0 +1,63 @@
+#coding:utf-8
+
+"""
+ID:          issue-5749
+ISSUE:       5749
+TITLE:       Token unknown error on formfeed in query [CORE5479]
+DESCRIPTION:
+    Test makes temporary .sql file and writes there several queries like:
+        select 'some_text'<DELIMITER>from rdb$database;
+    - where <DELIMITER> is every character from following list:
+        findstr /i /c:"CHR_WHITE," %fb_sources_home%/master/src/dsql/chars.h
+    Output:
+        009     CHR_WHITE,  // 0x9
+        010     CHR_WHITE,  // 0xA
+        012     CHR_WHITE,  // 0xC
+        013     CHR_WHITE,  // 0xD
+        032     CHR_WHITE,  // this can be skipped from check
+
+    NB: character \u000B [currently] NOT present in this list.
+
+NOTES:
+    [18.02.2023] pzotov
+    Confirmed problem on 5.0.0.736 (18-sep-2022): literal 0xC ('\u000c') could not be used as delimiter, got:
+        Statement failed, SQLSTATE = 42000
+        Dynamic SQL Error
+        -SQL error code = -104
+        -Token unknown - line 1, column 15
+        -
+    Checked on 5.0.0.742 - all OK.
+"""
+
+import pytest
+from firebird.qa import *
+from pathlib import Path
+
+db = db_factory()
+act = python_act('db')
+
+tmp_file = temp_file('gh_5749_tmp.sql')
+
+expected_stdout = """
+    CONSTANT                        u0009
+    CONSTANT                        u000A
+    CONSTANT                        u000C
+    CONSTANT                        u000D
+"""
+
+whitespace_sql = u"""set list on;
+select 'u0009'\u0009from rdb$database;
+select 'u000A'\u000Afrom rdb$database;
+select 'u000C'\u000Cfrom rdb$database;
+select 'u000D'\u000Dfrom rdb$database;
+"""
+# NB: select 'u000B'\u000Bfrom rdb$database; -- FAILS with token unknown.
+
+@pytest.mark.version('>=5.0')
+def test_1(act: Action, tmp_file: Path):
+
+    tmp_file.write_bytes(whitespace_sql.encode('utf-8'))
+
+    act.expected_stdout = expected_stdout
+    act.isql(switches=['-q'], input_file=tmp_file, charset='utf-8', io_enc='utf-8', combine_output = True)
+    assert act.clean_stdout == act.clean_expected_stdout