Add explicit limits to backend document selection parsing

Adds the following (very generous) limits:
- Max AST depth of 1024
- Max input selection string size of 1 MiB

Have to track AST depth manually, as there is no exposed way of
doing this natively via Bison.

Also removed a regex that had the potential of catastrophic
backtracking in case of massive inputs. It wasn't removed during
the previous purge due to being used with capture groups, which are
not supported by our current vespalib regex wrapper.

1 files changed, 59 insertions, 9 deletions

diff --git a/document/src/tests/documentselectparsertest.cpp b/document/src/tests/documentselectparsertest.cpp
index 110153954af..c8cbd1be9c2 100644
--- a/document/src/tests/documentselectparsertest.cpp
+++ b/document/src/tests/documentselectparsertest.cpp
@@ -18,6 +18,7 @@
 #include <vespa/document/select/compare.h>
 #include <vespa/document/select/operator.h>
 #include <vespa/document/select/parse_utils.h>
+#include <vespa/document/select/parser_limits.h>
 #include <vespa/vespalib/util/exceptions.h>
 #include <limits>
 #include <gtest/gtest.h>
@@ -1247,17 +1248,17 @@ TEST_F(DocumentSelectParserTest, testThatSimpleFieldValuesHaveCorrectFieldName)
 
 TEST_F(DocumentSelectParserTest, testThatComplexFieldValuesHaveCorrectFieldNames)
 {
-    EXPECT_EQ(
-        vespalib::string("headerval"),
-        parseFieldValue("testdoctype1.headerval{test}")->getRealFieldName());
+    EXPECT_EQ(vespalib::string("headerval"),
+              parseFieldValue("testdoctype1.headerval{test}")->getRealFieldName());
 
-    EXPECT_EQ(
-        vespalib::string("headerval"),
-        parseFieldValue("testdoctype1.headerval[42]")->getRealFieldName());
+    EXPECT_EQ(vespalib::string("headerval"),
+              parseFieldValue("testdoctype1.headerval[42]")->getRealFieldName());
 
-    EXPECT_EQ(
-        vespalib::string("headerval"),
-        parseFieldValue("testdoctype1.headerval.meow.meow{test}")->getRealFieldName());
+    EXPECT_EQ(vespalib::string("headerval"),
+              parseFieldValue("testdoctype1.headerval.meow.meow{test}")->getRealFieldName());
+
+    EXPECT_EQ(vespalib::string("headerval"),
+              parseFieldValue("testdoctype1.headerval .meow.meow{test}")->getRealFieldName());
 }
 
 namespace {
@@ -1603,4 +1604,53 @@ TEST_F(DocumentSelectParserTest, redundant_glob_wildcards_are_collapsed_into_min
     EXPECT_EQ(GlobOperator::convertToRegex("*?*?*?*"), "..*..*."); // Don't try this at home, kids!
 }
 
+TEST_F(DocumentSelectParserTest, recursion_depth_is_bounded_for_field_exprs) {
+    createDocs();
+    std::string expr = "testdoctype1";
+    for (size_t i = 0; i < 50000; ++i) {
+        expr += ".foo";
+    }
+    expr += ".hash() != 0";
+    verifyFailedParse(expr, "ParsingFailedException: expression is too deeply nested");
+}
+
+TEST_F(DocumentSelectParserTest, recursion_depth_is_bounded_for_arithmetic_exprs) {
+    createDocs();
+    std::string expr = "1";
+    for (size_t i = 0; i < 50000; ++i) {
+        expr += "+1";
+    }
+    expr += " != 0";
+    verifyFailedParse(expr, "ParsingFailedException: expression is too deeply nested");
+}
+
+TEST_F(DocumentSelectParserTest, recursion_depth_is_bounded_for_binary_logical_exprs) {
+    createDocs();
+    // Also throw in some comparisons to ensure they carry over the max depth.
+    std::string expr = "1 == 2";
+    std::string cmp_subexpr = "3 != 4";
+    for (size_t i = 0; i < 10000; ++i) {
+        expr += (i % 2 == 0 ? " and " : " or ") + cmp_subexpr;
+    }
+    verifyFailedParse(expr, "ParsingFailedException: expression is too deeply nested");
+}
+
+TEST_F(DocumentSelectParserTest, recursion_depth_is_bounded_for_unary_logical_exprs) {
+    createDocs();
+    std::string expr;
+    for (size_t i = 0; i < 10000; ++i) {
+        expr += "not ";
+    }
+    expr += "true";
+    verifyFailedParse(expr, "ParsingFailedException: expression is too deeply nested");
+}
+
+TEST_F(DocumentSelectParserTest, selection_has_upper_limit_on_input_size) {
+    createDocs();
+    std::string expr = ("testdoctype1.a_biii"
+                        + std::string(select::ParserLimits::MaxSelectionByteSize, 'i')
+                        + "iiig_identifier");
+    verifyFailedParse(expr, "ParsingFailedException: expression is too large to be parsed");
+}
+
 } // document