From 615f4a8c9e6679322c4b788d8bbc573e53267f05 Mon Sep 17 00:00:00 2001
From: Robin Newhouse <robinnew@amazon.com>
Date: Thu, 7 Dec 2023 00:16:28 +0000
Subject: [PATCH] MDEV-32587 Allow json exponential notation starting with zero

Modify the NS_ZERO state in the JSON number parser to allow
exponential notation with a zero coefficient (e.g. 0E-4).

The NS_ZERO state transition on 'E' was updated to move to the
NS_EX state rather than returning a syntax error. Similar change
was made for the NS_ZE1 (negative zero) starter state.

This allows accepted number grammar to include cases like:

- 0E4
- -0E-10

which were previously disallowed. Numeric parsing remains
the same for all other states.

Test cases are added to func_json.test to validate parsing for
various exponential numbers starting with zero coefficients.

All new code of the whole pull request, including one or several files
that are either new files or modified ones, are contributed under the
BSD-new license. I am contributing on behalf of my employer Amazon Web
Services.
---
 mysql-test/main/func_json.result | 33 ++++++++++++++++++++++++++++++++
 mysql-test/main/func_json.test   | 16 ++++++++++++++++
 strings/json_lib.c               |  6 +++---
 3 files changed, 52 insertions(+), 3 deletions(-)

diff --git a/mysql-test/main/func_json.result b/mysql-test/main/func_json.result
index 4b42c1def16..5d80ad2a7fc 100644
--- a/mysql-test/main/func_json.result
+++ b/mysql-test/main/func_json.result
@@ -1300,5 +1300,38 @@ f
 foo
 SET @@COLLATION_CONNECTION= @old_collation_connection;
 #
+# MDEV-32587 JSON_VALID fail to validate integer zero in scientific notation
+#
+select JSON_VALID(' {"number": 1E-4}');
+JSON_VALID(' {"number": 1E-4}')
+1
+select JSON_VALID(' {"number": 0E-4}');
+JSON_VALID(' {"number": 0E-4}')
+1
+select JSON_VALID(' {"number": 0.0}');
+JSON_VALID(' {"number": 0.0}')
+1
+select JSON_VALID(' {"number": 0.1E-4}');
+JSON_VALID(' {"number": 0.1E-4}')
+1
+select JSON_VALID(' {"number": 0e-4}');
+JSON_VALID(' {"number": 0e-4}')
+1
+select JSON_VALID(' {"number": -0E-4}');
+JSON_VALID(' {"number": -0E-4}')
+1
+select JSON_VALUE(' {"number": 0E-4}', '$.number');
+JSON_VALUE(' {"number": 0E-4}', '$.number')
+0E-4
+select JSON_VALID(' {"number": 00E-4}');
+JSON_VALID(' {"number": 00E-4}')
+0
+select JSON_VALID(' {"number": 01E-4}');
+JSON_VALID(' {"number": 01E-4}')
+0
+select JSON_VALID(' {"number": 0E-4.0}');
+JSON_VALID(' {"number": 0E-4.0}')
+0
+#
 # End of 10.4 tests
 #
diff --git a/mysql-test/main/func_json.test b/mysql-test/main/func_json.test
index eed433aa118..a271dac3dca 100644
--- a/mysql-test/main/func_json.test
+++ b/mysql-test/main/func_json.test
@@ -840,6 +840,22 @@ SELECT JSON_VALUE('["foo"]', '$**[0]') AS f;
 
 SET @@COLLATION_CONNECTION= @old_collation_connection;
 
+--echo #
+--echo # MDEV-32587 JSON_VALID fail to validate integer zero in scientific notation
+--echo #
+# Passing
+select JSON_VALID(' {"number": 1E-4}');
+select JSON_VALID(' {"number": 0E-4}');
+select JSON_VALID(' {"number": 0.0}');
+select JSON_VALID(' {"number": 0.1E-4}');
+select JSON_VALID(' {"number": 0e-4}');
+select JSON_VALID(' {"number": -0E-4}');
+select JSON_VALUE(' {"number": 0E-4}', '$.number');
+# Failing
+select JSON_VALID(' {"number": 00E-4}');
+select JSON_VALID(' {"number": 01E-4}');
+select JSON_VALID(' {"number": 0E-4.0}');
+
 --echo #
 --echo # End of 10.4 tests
 --echo #
diff --git a/strings/json_lib.c b/strings/json_lib.c
index 781172f0340..3b9b169e2d1 100644
--- a/strings/json_lib.c
+++ b/strings/json_lib.c
@@ -467,12 +467,12 @@ enum json_num_states {
 
 static int json_num_states[NS_NUM_STATES][N_NUM_CLASSES]=
 {
-/*         -        +       0        1..9    POINT    E       END_OK   ERROR */
+/*         -        +       0         1..9    POINT    E       END_OK   ERROR */
 /*OK*/   { JE_SYN,  JE_SYN, JE_SYN,   JE_SYN, JE_SYN,  JE_SYN, JE_SYN, JE_BAD_CHR },
 /*GO*/   { NS_GO1,  JE_SYN, NS_Z,     NS_INT, JE_SYN,  JE_SYN, JE_SYN, JE_BAD_CHR },
 /*GO1*/  { JE_SYN,  JE_SYN, NS_Z1,    NS_INT, JE_SYN,  JE_SYN, JE_SYN, JE_BAD_CHR },
-/*ZERO*/ { JE_SYN,  JE_SYN, JE_SYN,   JE_SYN, NS_FRAC, JE_SYN, NS_OK,  JE_BAD_CHR },
-/*ZE1*/  { JE_SYN,  JE_SYN, JE_SYN,   JE_SYN, NS_FRAC, JE_SYN, NS_OK,  JE_BAD_CHR },
+/*ZERO*/ { JE_SYN,  JE_SYN, JE_SYN,   JE_SYN, NS_FRAC, NS_EX,  NS_OK,  JE_BAD_CHR },
+/*ZE1*/  { JE_SYN,  JE_SYN, JE_SYN,   JE_SYN, NS_FRAC, NS_EX,  NS_OK,  JE_BAD_CHR },
 /*INT*/  { JE_SYN,  JE_SYN, NS_INT,   NS_INT, NS_FRAC, NS_EX,  NS_OK,  JE_BAD_CHR },
 /*FRAC*/ { JE_SYN,  JE_SYN, NS_FRAC,  NS_FRAC,JE_SYN,  NS_EX,  NS_OK,  JE_BAD_CHR },
 /*EX*/   { NS_EX,   NS_EX,  NS_EX1,   NS_EX1, JE_SYN,  JE_SYN, JE_SYN, JE_BAD_CHR },