[orc] branch master updated: ORC-580:[C++] Verify ColumnEncodings in StripeFooter (#463)

[gangwu]

This is an automated email from the ASF dual-hosted git repository.

gangwu pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/orc.git


The following commit(s) were added to refs/heads/master by this push:
     new 0ba08a0  ORC-580:[C++] Verify ColumnEncodings in StripeFooter (#463)
0ba08a0 is described below

commit 0ba08a092dd5ee30a16f7d09c032f8cb2ccaf041
Author: Quanlong Huang <huangquanl...@gmail.com>
AuthorDate: Fri Jan 3 10:33:49 2020 +0800

    ORC-580:[C++] Verify ColumnEncodings in StripeFooter (#463)
    
    ORC-580: [C++] Verify ColumnEncodings in StripeFooter
    
    ColumnEncodings parsed from the StripeFooter could be corrupt. We should 
verify it before using it to avoid crash.
    
    This fixes #463
---
 c++/src/Reader.cc                                       |   7 +++++++
 examples/corrupt/stripe_footer_bad_column_encodings.orc | Bin 0 -> 780 bytes
 tools/test/TestFileScan.cc                              |  11 +++++++++++
 3 files changed, 18 insertions(+)

diff --git a/c++/src/Reader.cc b/c++/src/Reader.cc
index 7957bda..ade2c29 100644
--- a/c++/src/Reader.cc
+++ b/c++/src/Reader.cc
@@ -395,6 +395,13 @@ namespace orc {
       throw ParseError(std::string("bad StripeFooter from ") +
                        pbStream->getName());
     }
+    // Verify StripeFooter in case it's corrupt
+    if (result.columns_size() != contents.footer->types_size()) {
+      std::stringstream msg;
+      msg << "bad number of ColumnEncodings in StripeFooter: expected="
+          << contents.footer->types_size() << ", actual=" << 
result.columns_size();
+      throw ParseError(msg.str());
+    }
     return result;
   }
 
diff --git a/examples/corrupt/stripe_footer_bad_column_encodings.orc 
b/examples/corrupt/stripe_footer_bad_column_encodings.orc
new file mode 100644
index 0000000..2446623
Binary files /dev/null and 
b/examples/corrupt/stripe_footer_bad_column_encodings.orc differ
diff --git a/tools/test/TestFileScan.cc b/tools/test/TestFileScan.cc
index c8b5fe8..b4938ea 100644
--- a/tools/test/TestFileScan.cc
+++ b/tools/test/TestFileScan.cc
@@ -135,3 +135,14 @@ TEST (TestFileScan, testBadCommand) {
   EXPECT_EQ("", output);
   EXPECT_EQ("The --batch parameter requires an integer option.\n", error);
 }
+
+TEST (TestFileScan, testErrorHandling) {
+  const std::string pgm = findProgram("tools/src/orc-scan");
+  const std::string file = 
findExample("corrupt/stripe_footer_bad_column_encodings.orc");
+  std::string output;
+  std::string error;
+  EXPECT_EQ(1, runProgram({pgm, file}, output, error));
+  EXPECT_EQ("", output);
+  EXPECT_NE(std::string::npos, error.find(
+      "bad number of ColumnEncodings in StripeFooter: expected=6, actual=0"));
+}