pitrou commented on code in PR #13497:
URL: https://github.com/apache/arrow/pull/13497#discussion_r917939619
##########
cpp/src/gandiva/gdv_function_stubs_test.cc:
##########
@@ -453,6 +453,81 @@ TEST(TestGdvFnStubs, TestCastVARCHARFromDouble) {
EXPECT_FALSE(ctx.has_error());
}
+TEST(TestGdvFnStubs, TestEncode) {
+ gandiva::ExecutionContext ctx;
+ int64_t ctx_ptr = reinterpret_cast<int64_t>(&ctx);
+ int32_t out_len = 0;
+
+ const char* out_str = gdv_fn_encode(ctx_ptr, "AbcDEfGh", 8, "UTF-16BE", 8,
&out_len);
+ EXPECT_EQ(std::string(out_str, out_len), "00410062006300440045006600470068");
+ EXPECT_FALSE(ctx.has_error());
+
+ out_str = gdv_fn_encode(ctx_ptr, "asdfj", 5, "UTF-16BE", 8, &out_len);
+ EXPECT_EQ(std::string(out_str, out_len), "0061007300640066006A");
+ EXPECT_FALSE(ctx.has_error());
+
+ out_str = gdv_fn_encode(ctx_ptr, "s;dcGS,jO!l", 11, "UTF-16BE", 8, &out_len);
+ EXPECT_EQ(std::string(out_str, out_len),
+ "0073003B0064006300470053002C006A004F0021006C");
+ EXPECT_FALSE(ctx.has_error());
+
+ out_str = gdv_fn_encode(ctx_ptr, "AbcDEfGh", 8, "UTF-16LE", 8, &out_len);
+ EXPECT_EQ(std::string(out_str, out_len), "41006200630044004500660047006800");
+ EXPECT_FALSE(ctx.has_error());
+
+ out_str = gdv_fn_encode(ctx_ptr, "asdfj", 5, "UTF-16LE", 8, &out_len);
+ EXPECT_EQ(std::string(out_str, out_len), "61007300640066006A00");
+ EXPECT_FALSE(ctx.has_error());
+
+ out_str = gdv_fn_encode(ctx_ptr, "s;dcGS,jO!l", 11, "UTF-16LE", 8, &out_len);
+ EXPECT_EQ(std::string(out_str, out_len),
+ "73003B0064006300470053002C006A004F0021006C00");
+ EXPECT_FALSE(ctx.has_error());
+
+ std::string d("AbOJjÜoß\xc3");
+ out_str = gdv_fn_encode(ctx_ptr, d.data(), static_cast<int>(d.length()),
"UTF-16LE", 8,
+ &out_len);
+ EXPECT_EQ(std::string(out_str, out_len), "");
+ EXPECT_THAT(ctx.get_error(),
+ ::testing::HasSubstr(
+ "unexpected byte \\c3 encountered while decoding utf8
string"));
+ ctx.Reset();
+}
+
+TEST(TestGdvFnStubs, TestDecode) {
+ gandiva::ExecutionContext ctx;
+ int64_t ctx_ptr = reinterpret_cast<int64_t>(&ctx);
+ int32_t out_len = 0;
+
+ const char* out_str = gdv_fn_decode(ctx_ptr,
"00410062006300440045006600470068", 32,
+ "UTF-16BE", 8, &out_len);
+ EXPECT_EQ(std::string(out_str, out_len), "AbcDEfGh");
+ EXPECT_FALSE(ctx.has_error());
+
+ out_str = gdv_fn_decode(ctx_ptr, "0061007300640066006A", 20, "UTF-16BE", 8,
&out_len);
+ EXPECT_EQ(std::string(out_str, out_len), "asdfj");
+ EXPECT_FALSE(ctx.has_error());
+
+ out_str = gdv_fn_decode(ctx_ptr,
"0073003B0064006300470053002C006A004F0021006C", 44,
+ "UTF-16BE", 8, &out_len);
+ EXPECT_EQ(std::string(out_str, out_len), "s;dcGS,jO!l");
+ EXPECT_FALSE(ctx.has_error());
+
+ out_str = gdv_fn_decode(ctx_ptr, "41006200630044004500660047006800", 32,
"UTF-16LE", 8,
+ &out_len);
+ EXPECT_EQ(std::string(out_str, out_len), "AbcDEfGh");
+ EXPECT_FALSE(ctx.has_error());
+
+ out_str = gdv_fn_decode(ctx_ptr, "61007300640066006A00", 20, "UTF-16LE", 8,
&out_len);
+ EXPECT_EQ(std::string(out_str, out_len), "asdfj");
+ EXPECT_FALSE(ctx.has_error());
+
+ out_str = gdv_fn_decode(ctx_ptr,
"73003B0064006300470053002C006A004F0021006C00", 44,
+ "UTF-16LE", 8, &out_len);
+ EXPECT_EQ(std::string(out_str, out_len), "s;dcGS,jO!l");
+ EXPECT_FALSE(ctx.has_error());
+}
Review Comment:
Can you add tests with non-ASCII characters? Also, what about codepoints
larger than 65535? Do they get "decoded" as surrogate pairs?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscr...@arrow.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
