Fix regex octal parsing to limit to 3 characters

rapidsai · Feb 7, 2022 · 13e7093 · 13e7093
1 parent 2e458b9
commit 13e7093
Show file tree

Hide file tree

Showing 2 changed files with 8 additions and 6 deletions.
diff --git a/cpp/src/strings/regex/regcomp.cpp b/cpp/src/strings/regex/regcomp.cpp
@@ -257,11 +257,13 @@ class regex_parser {
     if (quoted) {
       // treating all quoted numbers as Octal, since we are not supporting backreferences
       if (yy >= '0' && yy <= '7') {
-        yy         = yy - '0';
-        char32_t c = *exprp;
-        while (c >= '0' && c <= '7') {
+        yy          = yy - '0';
+        auto c      = *exprp;
+        auto digits = 1;
+        while (c >= '0' && c <= '7' && digits < 3) {
           yy = (yy << 3) | (c - '0');
           c  = *(++exprp);
+          ++digits;
         }
         return CHAR;
       } else {

diff --git a/cpp/tests/strings/contains_tests.cpp b/cpp/tests/strings/contains_tests.cpp
@@ -239,14 +239,14 @@ TEST_F(StringsContainsTests, MatchesIPV4Test)
 
 TEST_F(StringsContainsTests, OctalTest)
 {
-  cudf::test::strings_column_wrapper strings({"AZ", "B", "CDAZEY", ""});
+  cudf::test::strings_column_wrapper strings({"A3", "B", "CDA3EY", ""});
   auto strings_view = cudf::strings_column_view(strings);
   cudf::test::fixed_width_column_wrapper<bool> expected({1, 0, 1, 0});
   auto results = cudf::strings::contains_re(strings_view, "\\101");
   CUDF_TEST_EXPECT_COLUMNS_EQUAL(*results, expected);
-  results = cudf::strings::contains_re(strings_view, "\\101Z");
+  results = cudf::strings::contains_re(strings_view, "\\1013");
   CUDF_TEST_EXPECT_COLUMNS_EQUAL(*results, expected);
-  results = cudf::strings::contains_re(strings_view, "D*\\101\\132");
+  results = cudf::strings::contains_re(strings_view, "D*\\101\\063");
   CUDF_TEST_EXPECT_COLUMNS_EQUAL(*results, expected);
 }