From b33d91412db32a9352c481f2e7d6b10f79bdb6c9 Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@yandex-team.ru>
Date: Fri, 2 Jul 2021 03:07:23 +0300
Subject: [PATCH] Correctly throw exception on invalid dates

---
 src/IO/ReadHelpers.cpp                        |  2 +-
 src/IO/ReadHelpers.h                          | 40 +++++++++++++------
 .../0_stateless/01933_invalid_date.reference  |  1 +
 .../0_stateless/01933_invalid_date.sql        | 10 +++++
 4 files changed, 40 insertions(+), 13 deletions(-)
 create mode 100644 tests/queries/0_stateless/01933_invalid_date.reference
 create mode 100644 tests/queries/0_stateless/01933_invalid_date.sql

diff --git a/src/IO/ReadHelpers.cpp b/src/IO/ReadHelpers.cpp
index 8e9a14a20fb..2a5594a6866 100644
--- a/src/IO/ReadHelpers.cpp
+++ b/src/IO/ReadHelpers.cpp
@@ -765,7 +765,7 @@ ReturnType readDateTextFallback(LocalDate & date, ReadBuffer & buf)
 
     auto ignore_delimiter = [&]
     {
-        if (!buf.eof())
+        if (!buf.eof() && !isNumericASCII(*buf.position()))
         {
             ++buf.position();
             return true;
diff --git a/src/IO/ReadHelpers.h b/src/IO/ReadHelpers.h
index d4e2db0b553..4e101aaaf63 100644
--- a/src/IO/ReadHelpers.h
+++ b/src/IO/ReadHelpers.h
@@ -572,27 +572,43 @@ inline ReturnType readDateTextImpl(LocalDate & date, ReadBuffer & buf)
     /// Optimistic path, when whole value is in buffer.
     if (!buf.eof() && buf.position() + 10 <= buf.buffer().end())
     {
-        UInt16 year = (buf.position()[0] - '0') * 1000 + (buf.position()[1] - '0') * 100 + (buf.position()[2] - '0') * 10 + (buf.position()[3] - '0');
-        buf.position() += 5;
+        char * pos = buf.position();
 
-        UInt8 month = buf.position()[0] - '0';
-        if (isNumericASCII(buf.position()[1]))
+        /// YYYY-MM-DD
+        /// YYYY-MM-D
+        /// YYYY-M-DD
+        /// YYYY-M-D
+
+        /// The delimiters can be arbitrary characters, like YYYY/MM!DD, but obviously not digits.
+
+        UInt16 year = (pos[0] - '0') * 1000 + (pos[1] - '0') * 100 + (pos[2] - '0') * 10 + (pos[3] - '0');
+        pos += 5;
+
+        if (isNumericASCII(pos[-1]))
+            return ReturnType(false);
+
+        UInt8 month = pos[0] - '0';
+        if (isNumericASCII(pos[1]))
         {
-            month = month * 10 + buf.position()[1] - '0';
-            buf.position() += 3;
+            month = month * 10 + pos[1] - '0';
+            pos += 3;
         }
         else
-            buf.position() += 2;
+            pos += 2;
 
-        UInt8 day = buf.position()[0] - '0';
-        if (isNumericASCII(buf.position()[1]))
+        if (isNumericASCII(pos[-1]))
+            return ReturnType(false);
+
+        UInt8 day = pos[0] - '0';
+        if (isNumericASCII(pos[1]))
         {
-            day = day * 10 + buf.position()[1] - '0';
-            buf.position() += 2;
+            day = day * 10 + pos[1] - '0';
+            pos += 2;
         }
         else
-            buf.position() += 1;
+            pos += 1;
 
+        buf.position() = pos;
         date = LocalDate(year, month, day);
         return ReturnType(true);
     }
diff --git a/tests/queries/0_stateless/01933_invalid_date.reference b/tests/queries/0_stateless/01933_invalid_date.reference
new file mode 100644
index 00000000000..829e7e8c420
--- /dev/null
+++ b/tests/queries/0_stateless/01933_invalid_date.reference
@@ -0,0 +1 @@
+2019-07-08
diff --git a/tests/queries/0_stateless/01933_invalid_date.sql b/tests/queries/0_stateless/01933_invalid_date.sql
new file mode 100644
index 00000000000..aac09c99e60
--- /dev/null
+++ b/tests/queries/0_stateless/01933_invalid_date.sql
@@ -0,0 +1,10 @@
+SELECT toDate('07-08-2019'); -- { serverError 6 }
+SELECT toDate('2019-0708'); -- { serverError 38 }
+SELECT toDate('201907-08'); -- { serverError 38 }
+SELECT toDate('2019^7^8');
+
+CREATE TEMPORARY TABLE test (d Date);
+INSERT INTO test VALUES ('2018-01-01');
+
+SELECT * FROM test WHERE d >= '07-08-2019'; -- { serverError 53 }
+SELECT * FROM test WHERE d >= '2019-07-08';