teemperor created this revision. teemperor added a reviewer: labath. Herald added subscribers: lldb-commits, JDevlieghere, abidh, mgorny. Herald added a project: LLDB. teemperor added a comment.
Btw, the current declaration of MemorySource is just what I came up with on the fly. Not sure where the documentation should go or if that class deserves its own header. As we found out already, StringPrinter has some surprising behaviour and bugs. This patch adds a (hopefully) exhaustive test of all the StringPrinter functionality for all supported encodings and API variants. Also adds a bunch of FIXME's into the test for all the unexpected behaviour I found. This patch is not fully NFC as it changes that the StringPrinter now treats both 0 and LLDB_INVALID_ADDRESS as invalid addresses (this seems like a really safe fix to me and it makes the test a bit simpler). Also adds the MemorySource abstraction class that allows us to mock a process in this unit test. It essentially has the minimal interface so that StringPrinter can do all the calls it currently does, but obviously the implementation could be more minimal. We can reduce that interface more in the future once StringPrinter doesn't use all of these functions to do its job. Repository: rLLDB LLDB https://reviews.llvm.org/D68691 Files: lldb/include/lldb/DataFormatters/StringPrinter.h lldb/include/lldb/Target/Process.h lldb/source/DataFormatters/StringPrinter.cpp lldb/unittests/CMakeLists.txt lldb/unittests/DataFormatters/CMakeLists.txt lldb/unittests/DataFormatters/StringPrinterTest.cpp
Index: lldb/unittests/DataFormatters/StringPrinterTest.cpp =================================================================== --- /dev/null +++ lldb/unittests/DataFormatters/StringPrinterTest.cpp @@ -0,0 +1,784 @@ +//===-- StringPrinterTest.cpp ---------------------------*- C++-*-===// +// +// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. +// See https://llvm.org/LICENSE.txt for license information. +// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception +// +//===----------------------------------------------------------------------===// + +#include "lldb/DataFormatters/StringPrinter.h" +#include "lldb/Target/Process.h" +#include "lldb/Utility/Endian.h" + +#include "gmock/gmock.h" +#include "gtest/gtest.h" + +using namespace lldb_private; +using namespace lldb_private::formatters; + +namespace { +/// A memory source that just stores a string and pretend it's the memory of +/// a process. +struct MockMemorySource : public MemorySource { + virtual ~MockMemorySource() = default; + std::string m_data; + lldb::ByteOrder m_byte_order; + uint32_t m_addr_size; + + MockMemorySource(const std::string &data, lldb::ByteOrder order, + unsigned addr_size) + : m_data(data), m_byte_order(order), m_addr_size(addr_size) { + // We currently only support the host byte order (as this is the one we + // encode our string literals in). StringPrinter anyway doesn't do any + // byte order handling on its own, so different byte orders aren't really + // useful here. + assert(order == endian::InlHostByteOrder() && + "only little endian should be used"); + // Address size doesn't really influence this test or the StringPrinter, + // so let's just make sure we have something reasonable. + assert(addr_size == 4U && "only 4 byte addresses should be used"); + } + + virtual size_t ReadStringFromMemory(lldb::addr_t vm_addr, char *str, + size_t max_bytes, Status &error, + size_t type_width = 1) { + std::memset(str, 0, max_bytes); + std::string terminator(type_width, '\0'); + for (size_t i = 0; i < max_bytes - type_width; ++i) { + const size_t mem_offset = i + vm_addr; + assert(mem_offset < m_data.size() && "Out of bounds read?"); + str[i] = m_data.at(mem_offset); + if (llvm::StringRef(m_data).substr(mem_offset).startswith(terminator)) + return i; + } + return max_bytes; + } + + virtual size_t ReadCStringFromMemory(lldb::addr_t vm_addr, char *cstr, + size_t cstr_max_len, Status &error) { + for (size_t i = 0; i < cstr_max_len; ++i) { + char c = m_data.at(i + vm_addr); + if (c == '\0') + return i; + if (i == cstr_max_len) + break; + cstr[i] = c; + } + return cstr_max_len; + } + + virtual size_t ReadMemoryFromInferior(lldb::addr_t vm_addr, void *buf, + size_t size, Status &error) { + assert(size + vm_addr < m_data.size()); + std::memcpy(buf, m_data.data() + vm_addr, size); + return size; + } + + virtual lldb::ByteOrder GetByteOrder() const { return m_byte_order; } + + virtual uint32_t GetAddressByteSize() const { return m_addr_size; } +}; +} // namespace + +#define MEM_STR(DATA) (std::string(DATA, sizeof(DATA))) + +namespace { +struct StringPrinterTest : public ::testing::Test { + /// The string we read into. + StreamString string; + /// By default the test sets the location to character position 1 (as 0 + /// is always an invalid location). To disable this for a test case (e.g., + /// because the test is testing other locations than 1) set this to false. + /// This will disable any future updates to the character_pos setting and + /// disable the automatic padding of the input string to move it to + /// character position one. + bool enable_auto_setting_location = true; + llvm::Optional<size_t> character_pos; + /// Sets the location of the string start to the given *character* position. + /// This position is translated into the final byte position by runStringTest + /// by multiplying it with the size of the character data type (e.g., 1 byte + /// for ASCII/UTF8, 2 byte for UTF16 and 4 for UTF32). + void SetCharLocation(size_t pos) { + if (enable_auto_setting_location) + character_pos = pos; + } + + /// Sets up StringPrinter options that we share between buffer and string + /// reading. + void initOpts(StringPrinter::DumpToStreamOptions &opts) { + string.Clear(); + opts.SetStream(&string); + } + + std::string GetReadData() const { return string.GetString().str(); } + + const lldb::ByteOrder byte_order = endian::InlHostByteOrder(); + const uint32_t addr_size = 4U; + /// The max size we allow reading from the memory source. Has to be set + /// alongside the memory source, so we store this here that we can pass it + /// along when we attach our MockMemorySource. + size_t max_read_size = 1024; + + /// True iff we expect that the StringPrinter functions return 'true' for + /// success. + bool expect_success = true; + + /// The options we use for tringPrinter::ReadStringAndDumpToStream. + StringPrinter::ReadStringAndDumpToStreamOptions string_opts; + /// The options we use for tringPrinter::ReadBufferAndDumpToStream. + StringPrinter::ReadBufferAndDumpToStreamOptions buffer_opts; + + /// Generic function that takes a string-like class and returns its bytes + /// as a std::string. + template <class T> std::string toCharString(const T &mem) { + // Explicitly pass the length in case of embedded \0 bytes in mem. + return std::string(reinterpret_cast<const char *>(mem.data()), + mem.size() * sizeof(typename T::value_type)); + } + + void runUTF8StringAndBufferTest(std::string input, std::string output) { + SetCharLocation(1); + runStringTest<std::string, StringPrinter::StringElementType::UTF8>( + input, output, "u8"); + runBufferTest<std::string, StringPrinter::StringElementType::UTF8>( + input, output, "u8"); + } + + /// Tries to read the string with the ReadStringAndDumpToStream function. + /// \param input The input string as any kind of string type. + /// \param output The expected output as UTF8. + /// \param CharKindStr A string describing the character kind this test is + /// testing. Used to make any gtest assertion failures + /// more informative. + template <typename T, StringPrinter::StringElementType CharKind> + void runStringTest(T input, std::string output, const char *CharKindStr) { + size_t TypeWidth = sizeof(typename T::value_type); + initOpts(string_opts); + + std::string mem_bytes = toCharString(input); + + // If the test has set a character positon, translate this now to an + // actual byte position in memory. + if (character_pos) { + string_opts.SetLocation(*character_pos * TypeWidth); + // Automatically add padding the input so that it starts at character + // position 1 if we have automatically set the character position to 1. + // Without this character positon 1 would point to the second character + // in the input string. + // See the enable_auto_setting_location flag. + if (enable_auto_setting_location) + mem_bytes = std::string(TypeWidth, '_') + mem_bytes; + } + // Add enough trailing 0 bytes we need for reading C strings. + mem_bytes += std::string(TypeWidth, '\0'); + + std::unique_ptr<MockMemorySource> src; + src.reset(new MockMemorySource(mem_bytes, byte_order, addr_size)); + string_opts.setMemorySourceAndMaxSize(src.get(), max_read_size); + + bool b = StringPrinter::ReadStringAndDumpToStream<CharKind>(string_opts); + EXPECT_EQ(expect_success, b); + EXPECT_EQ(output, GetReadData()) << "Failed to correctly read string " + "data for " + << CharKindStr; + } + + /// Tries to read the string with the ReadBufferAndDumpToStream function. + /// \param input The input string as any kind of string type. + /// \param output The expected output as UTF8. + /// \param CharKindStr A string describing the character kind this test is + /// testing. Used to make any gtest assertion failures + /// more informative. + template <typename T, StringPrinter::StringElementType CharKind> + void runBufferTest(T input, std::string output, const char *CharKindStr) { + initOpts(buffer_opts); + + std::string mem = toCharString(input); + // FIXME: Somehow the StringPrinter only reads the first half of the + // buffer, so we need to make the read size twice its buffer size to read + // the whole buffer. + size_t length_in_bytes = mem.size() * 2U; + DataExtractor extractor(mem.data(), length_in_bytes, byte_order, addr_size); + buffer_opts.SetData(extractor); + bool b = StringPrinter::ReadBufferAndDumpToStream<CharKind>(buffer_opts); + EXPECT_EQ(expect_success, b); + EXPECT_EQ(output, GetReadData()) << "Failed to correctly read buffer " + "data for " + << CharKindStr; + } +}; +} // namespace + +//------------------------------------------------------------------------------ +// Utility macros. +// We have a long range of similar functionality that needs to be tested in this +// test. To remove all the redundant typing of string literals with different +// prefixes, these macros transform a normal string literal to a series of +// test calls with different kinds of unicode/ascii string literals. +// We have to use macros here as we rely on the preprocessor concatenating +// unicode prefixes like u8"" and normal string literals like "foo" into +// a single unicode literal that we can pass to the actual test functions. +// By doing that we can automatically transform "foo" into different encodings. +// E.g. u8"" "foo" -> UTF8 +// u16"" "foo" -> UTF16 +// "foo" -> ASCII + +// Transforms the given literal to different unicode variants (UTF8, UTF16, +// UTF32) and also normal ASCII and then tries to read it back via the +// StringPrinter::ReadBufferAndDumpToStream. The result is compared to +// the given UTF8 result. +#define RUN_BUFFER_TEST(literal, result) \ + do { \ + runBufferTest<std::string, StringPrinter::StringElementType::ASCII>( \ + std::string(literal, sizeof(literal)), result, "ascii"); \ + runBufferTest<std::string, StringPrinter::StringElementType::UTF8>( \ + std::string(u8"" literal, sizeof(u8"" literal)), result, "u8"); \ + runBufferTest<std::u16string, StringPrinter::StringElementType::UTF16>( \ + std::u16string(u"" literal, sizeof(u"" literal)), result, "u16"); \ + runBufferTest<std::u32string, StringPrinter::StringElementType::UTF32>( \ + std::u32string(U"" literal, sizeof(U"" literal)), result, "u32"); \ + } while (false) + +#define RUN_USTRING_TEST_IMPL(literal, result) \ + runStringTest<std::string, StringPrinter::StringElementType::UTF8>( \ + std::string(u8"" literal, sizeof(u8"" literal)), result, "u8"); \ + runStringTest<std::u16string, StringPrinter::StringElementType::UTF16>( \ + std::u16string(u"" literal, sizeof(u"" literal)), result, "u16"); \ + runStringTest<std::u32string, StringPrinter::StringElementType::UTF32>( \ + std::u32string(U"" literal, sizeof(U"" literal)), result, "u32"); + +// Transforms the given literal to different unicode variants (UTF8, UTF16, +// UTF32) and then tries to read it back via the +// StringPrinter::ReadStringAndDumpToStream. The result is compared to +// the given UTF8 result. +#define RUN_USTRING_TEST(literal, result) \ + do { \ + SetCharLocation(1); \ + RUN_USTRING_TEST_IMPL(literal, result) \ + } while (false) + +#define RUN_ASCII_STRING_TEST(literal, result) \ + do { \ + SetCharLocation(1); \ + runStringTest<std::string, StringPrinter::StringElementType::ASCII>( \ + std::string(literal, sizeof(literal)), result, "ascii"); \ + } while (false) + +// Same as RUN_USTRING_TEST but also tests the literal as plain ASCII. +#define RUN_STRING_TEST(literal, result) \ + do { \ + SetCharLocation(1); \ + RUN_ASCII_STRING_TEST(literal, result); \ + RUN_USTRING_TEST_IMPL(literal, result); \ + } while (false) + +// Runs both RUN_STRING_TEST and RUN_BUFFER_TEST with the given arguments. +#define RUN_STRING_AND_BUFFER_TEST(literal, result) \ + do { \ + RUN_STRING_TEST(literal, result); \ + RUN_BUFFER_TEST(literal, result); \ + } while (false) + +//------------------------------------------------------------------------------ +// Location settings tests. + +TEST_F(StringPrinterTest, ZeroPosition) { + // Disable that we automatically set the location to character position 1. + enable_auto_setting_location = false; + string_opts.SetLocation(0U); + expect_success = false; + RUN_STRING_TEST("A", ""); +} + +TEST_F(StringPrinterTest, InvalidPosition) { + // Disable that we automatically set the location to character position 1. + enable_auto_setting_location = false; + string_opts.SetLocation(LLDB_INVALID_ADDRESS); + expect_success = false; + RUN_STRING_TEST("C", ""); +} + +TEST_F(StringPrinterTest, OtherPosition) { + // All tests use character position 1 by default, this just tests that + // character position 2 is also working. See enable_auto_setting_location. + SetCharLocation(2U); + // Disable that we automatically set the location to character position 1. + // We do this after setting our character position to prevent any future + // changes to our set character position. + enable_auto_setting_location = false; + RUN_STRING_TEST("ABC", "\"C\""); +} + +//------------------------------------------------------------------------------ +// Generic tests with a few strings. + +TEST_F(StringPrinterTest, NormalString) { + RUN_STRING_AND_BUFFER_TEST("foobar", "\"foobar\""); +} + +TEST_F(StringPrinterTest, LatinAlphabetSmallLetters) { + RUN_STRING_AND_BUFFER_TEST("abcdefghijklmnopqrstuvwxyz", // + "\"abcdefghijklmnopqrstuvwxyz\""); +} + +TEST_F(StringPrinterTest, LatinAlphabetCapitalLetters) { + RUN_STRING_AND_BUFFER_TEST("ABCDEFGHIJKLMNOPQRSTUVWXYZ", // + "\"ABCDEFGHIJKLMNOPQRSTUVWXYZ\""); +} + +TEST_F(StringPrinterTest, Numbers) { + RUN_STRING_AND_BUFFER_TEST("1234567890", "\"1234567890\""); +} + +TEST_F(StringPrinterTest, EmptyString) { + RUN_STRING_AND_BUFFER_TEST("", "\"\""); +} + +//------------------------------------------------------------------------------ +// Unicode tests + +TEST_F(StringPrinterTest, UnicodeLetters) { + RUN_USTRING_TEST("\u01AA", // + u8"\"\xC6\xAA\""); + + // Test a charaters that are corner cases of our printable/nonprintable + // checks. + RUN_USTRING_TEST("\u2027", u8"\"\xE2\x80\xA7\""); + + RUN_USTRING_TEST("\u2028", u8"\"\\U00002028\""); + + RUN_USTRING_TEST("\u2029", u8"\"\\U00002029\""); + + RUN_USTRING_TEST("\u202A", u8"\"\\U0000202a\""); + + RUN_USTRING_TEST("\uFFFE", u8"\"\\U0000fffe\""); +} + +TEST_F(StringPrinterTest, UTF8CornerCases) { + // Tests the corner cases of UTF8 codepoint decoding. + // Max codepoint in 1 bytes. + RUN_USTRING_TEST("\u007F", u8"\"\\x7f\""); + // Min codepoint in 2 bytes. + RUN_USTRING_TEST("\u0080", u8"\"\\U00000080\""); + // Max codepoint in 2 bytes. + RUN_USTRING_TEST("\u07FF", u8"\"\xDF\xBF\""); + // Min codepoint in 3 bytes. + RUN_USTRING_TEST("\u0800", u8"\"\xE0\xA0\x80\""); + // Max codepoint in 3 bytes. + RUN_USTRING_TEST("\uFFFF", u8"\"\\U0000ffff\""); + // Min codepoint in 4 bytes. + RUN_USTRING_TEST("\U00010000", "\"\xF0\\x90\\x80\\x80\""); + // Max codepoint in 4 bytes. + RUN_USTRING_TEST("\U0010FFFF", "\"\xF4\x8F\xBF\xBF\""); +} + +//------------------------------------------------------------------------------ +// Corrupted UTF8 tests. +// These tests take valid UTF8 and then remove bytes from the start or end +// to generate corrupted UTF8 strings. We mostly care about the tests that +// remove trailing bytes as this tests that we not just blindly trust the +// UTF8 prefix code when reading from memory. + +// Allow converting literals to std::string even if they have embedded zeroes. +// FIXME: Remove this, only needed to provide the result strings for currently +// broken UTF8 test cases. +#define STR_WITH_ZEROES(literal) (std::string(literal, sizeof(literal) - 1U)) + +TEST_F(StringPrinterTest, UTF8MissingStart4Bytes) { + std::string utf8 = u8"\U0010FFFE"; + assert(utf8.size() == 4); + utf8.erase(0, 1); + runUTF8StringAndBufferTest(utf8, "\"\\x8f\\xbf\\xbe\""); + utf8.erase(0, 1); + runUTF8StringAndBufferTest(utf8, "\"\\xbf\\xbe\""); + utf8.erase(0, 1); + runUTF8StringAndBufferTest(utf8, "\"\\xbe\""); +} + +TEST_F(StringPrinterTest, UTF8MissingEnd4Bytes) { + std::string utf8 = u8"\U0010FFFE"; + assert(utf8.size() == 4); + // FIXME: This shouldn't add the found 0 terminator to the output. + utf8.erase(3, 1); + runUTF8StringAndBufferTest(utf8, STR_WITH_ZEROES("\"\xF4\x8F\xBF\0\"")); + utf8.erase(2, 1); + runUTF8StringAndBufferTest(utf8, STR_WITH_ZEROES("\"\xF4\x8F\0\"")); + utf8.erase(1, 1); + runUTF8StringAndBufferTest(utf8, STR_WITH_ZEROES("\"\xF4\0\"")); +} + +TEST_F(StringPrinterTest, UTF8MissingStart3Bytes) { + std::string utf8 = u8"\u0800"; + assert(utf8.size() == 3); + utf8.erase(0, 1); + runUTF8StringAndBufferTest(utf8, "\"\\xa0\\x80\""); + utf8.erase(0, 1); + runUTF8StringAndBufferTest(utf8, "\"\\x80\""); +} + +TEST_F(StringPrinterTest, UTF8MissingEnd3Bytes) { + std::string utf8 = u8"\u0800"; + assert(utf8.size() == 3); + // FIXME: This shouldn't add the unexpected 0 terminator to the output. + utf8.erase(2, 1); + runUTF8StringAndBufferTest(utf8, STR_WITH_ZEROES("\"\xE0\xA0\0\"")); + utf8.erase(1, 1); + runUTF8StringAndBufferTest(utf8, STR_WITH_ZEROES("\"\xE0\0\"")); +} + +TEST_F(StringPrinterTest, UTF8MissingStart2Bytes) { + std::string utf8 = u8"\u0080"; + assert(utf8.size() == 2); + utf8.erase(0, 1); + runUTF8StringAndBufferTest(utf8, "\"\\x80\""); +} + +TEST_F(StringPrinterTest, UTF8MissingEnd2Bytes) { + std::string utf8 = u8"\u0080"; + assert(utf8.size() == 2); + utf8.erase(1, 1); + runUTF8StringAndBufferTest(utf8, "\"\xC2\""); +} + +//------------------------------------------------------------------------------ +// Source size and zero terminated result tests. + +TEST_F(StringPrinterTest, SourceSizeZeroWithoutTerminator) { + // FIXME: That disables source size. + string_opts.SetSourceSize(0); + string_opts.SetNeedsZeroTermination(false); + buffer_opts.SetSourceSize(0); + buffer_opts.SetNeedsZeroTermination(false); + RUN_STRING_AND_BUFFER_TEST("abcd", "\"abcd\""); +} + +TEST_F(StringPrinterTest, SourceSizeZeroWithTerminator) { + // FIXME: That disables source size. + string_opts.SetSourceSize(0); + string_opts.SetNeedsZeroTermination(true); + buffer_opts.SetSourceSize(0); + buffer_opts.SetNeedsZeroTermination(true); + RUN_STRING_AND_BUFFER_TEST("abcd", "\"abcd\""); +} + +TEST_F(StringPrinterTest, MatchingSourceSizeWithoutTerminator) { + string_opts.SetSourceSize(3); + string_opts.SetNeedsZeroTermination(false); + buffer_opts.SetSourceSize(3); + buffer_opts.SetNeedsZeroTermination(false); + RUN_STRING_AND_BUFFER_TEST("abcd", "\"abc\""); +} + +TEST_F(StringPrinterTest, TooShortSourceSizeWithoutTerminator) { + string_opts.SetSourceSize(3); + string_opts.SetNeedsZeroTermination(false); + buffer_opts.SetSourceSize(3); + buffer_opts.SetNeedsZeroTermination(false); + RUN_STRING_AND_BUFFER_TEST("abcdef", "\"abc\""); +} + +TEST_F(StringPrinterTest, TooLongSourceSizeWithoutTerminator) { + string_opts.SetSourceSize(8); + string_opts.SetNeedsZeroTermination(false); + buffer_opts.SetSourceSize(8); + buffer_opts.SetNeedsZeroTermination(false); + RUN_STRING_AND_BUFFER_TEST("abc\0cdefadsf", "\"abc\""); +} + +TEST_F(StringPrinterTest, MatchingSourceSizeWithTerminator) { + string_opts.SetSourceSize(3); + string_opts.SetNeedsZeroTermination(true); + buffer_opts.SetSourceSize(3); + buffer_opts.SetNeedsZeroTermination(true); + // FIXME: Seems like we forgot to implement zero terminator support for + // ASCII. + RUN_USTRING_TEST("abcd", "\"ab\""); + RUN_ASCII_STRING_TEST("abcd", "\"abc\""); + RUN_BUFFER_TEST("abcd", "\"abc\""); +} + +TEST_F(StringPrinterTest, TooShortSourceSizeWithTerminator) { + string_opts.SetSourceSize(3); + string_opts.SetNeedsZeroTermination(true); + buffer_opts.SetSourceSize(3); + buffer_opts.SetNeedsZeroTermination(true); + // FIXME: Seems like we forgot to implement zero terminator support for + // ASCII. + RUN_USTRING_TEST("abcdef", "\"ab\""); + RUN_ASCII_STRING_TEST("abcd", "\"abc\""); + RUN_BUFFER_TEST("abcdef", "\"abc\""); +} + +TEST_F(StringPrinterTest, TooLongSourceSizeWithTerminator) { + string_opts.SetSourceSize(8); + string_opts.SetNeedsZeroTermination(true); + buffer_opts.SetSourceSize(8); + buffer_opts.SetNeedsZeroTermination(true); + RUN_STRING_AND_BUFFER_TEST("abc", "\"abc\""); +} + +//------------------------------------------------------------------------------ +// Ignore zero terminator tests. + +TEST_F(StringPrinterTest, IgnoredZeroAfterString) { + string_opts.SetSourceSize(2); + string_opts.SetBinaryZeroIsTerminator(false); + string_opts.SetNeedsZeroTermination(true); + buffer_opts.SetSourceSize(2); + buffer_opts.SetBinaryZeroIsTerminator(false); + buffer_opts.SetNeedsZeroTermination(true); + // FIXME: It seems we terminate unicode string in this mode with an actual + // '0' *character* and not a '\0' *byte*... + RUN_USTRING_TEST("abc\0adsfasdf", "\"a\\0\""); + RUN_ASCII_STRING_TEST("abc\0adsfasdf", "\"ab\""); + RUN_BUFFER_TEST("abc\0adsfasdf", "\"ab\""); +} + +TEST_F(StringPrinterTest, DontIgnoreEmbeddedZero) { + string_opts.SetSourceSize(8); + string_opts.SetBinaryZeroIsTerminator(true); + buffer_opts.SetSourceSize(8); + buffer_opts.SetBinaryZeroIsTerminator(true); + RUN_STRING_AND_BUFFER_TEST("abc\0adsfasdf", "\"abc\""); +} + +TEST_F(StringPrinterTest, IgnoreEmbeddedZero) { + string_opts.SetSourceSize(8); + string_opts.SetBinaryZeroIsTerminator(false); + buffer_opts.SetSourceSize(8); + buffer_opts.SetBinaryZeroIsTerminator(false); + // FIXME: Unicode ignores embedded zero, but just keeps adding escaped zeroes? + RUN_USTRING_TEST("abc\0adsfasdf", "\"abc\\0\\0\\0\\0\\0\""); + // FIXME: ASCII doesn't even support ignoring the embedded zero... + RUN_ASCII_STRING_TEST("abc\0adsfasdf", "\"abc\""); + RUN_BUFFER_TEST("abc\0adsfasdf", "\"abc\\0adsf\""); +} + +//------------------------------------------------------------------------------ +// ASCII escaping tests + +TEST_F(StringPrinterTest, Escape) { + RUN_STRING_AND_BUFFER_TEST("\n\"\t\r\b\a\f\v", // + "\"\\n\\\"\\t\\r\\b\\a\\f\\v\""); +} + +//------------------------------------------------------------------------------ +// String prefix tests + +TEST_F(StringPrinterTest, Prefix) { + string_opts.SetPrefixToken("PREFIX"); + buffer_opts.SetPrefixToken("PREFIX"); + RUN_STRING_AND_BUFFER_TEST("", "PREFIX\"\""); +} + +TEST_F(StringPrinterTest, PrefixEmptyString) { + string_opts.SetPrefixToken("PREFIX"); + buffer_opts.SetPrefixToken("PREFIX"); + RUN_STRING_AND_BUFFER_TEST("", "PREFIX\"\""); +} + +TEST_F(StringPrinterTest, EmptyPrefix) { + string_opts.SetPrefixToken(""); + buffer_opts.SetPrefixToken(""); + RUN_STRING_AND_BUFFER_TEST("", "\"\""); +} + +TEST_F(StringPrinterTest, EmptyPrefixEmptyString) { + string_opts.SetPrefixToken(""); + buffer_opts.SetPrefixToken(""); + RUN_STRING_AND_BUFFER_TEST("", "\"\""); +} + +//------------------------------------------------------------------------------ +// String suffix tests + +TEST_F(StringPrinterTest, Suffix) { + SetCharLocation(1); + string_opts.SetSuffixToken("SUFFIX"); + buffer_opts.SetSuffixToken("SUFFIX"); + RUN_STRING_AND_BUFFER_TEST("", "\"\"SUFFIX"); +} + +TEST_F(StringPrinterTest, SuffixEmptyString) { + SetCharLocation(1); + string_opts.SetSuffixToken("SUFFIX"); + buffer_opts.SetSuffixToken("SUFFIX"); + RUN_STRING_AND_BUFFER_TEST("", "\"\"SUFFIX"); +} + +TEST_F(StringPrinterTest, SuffixPrefix) { + string_opts.SetSuffixToken(""); + buffer_opts.SetSuffixToken(""); + RUN_STRING_AND_BUFFER_TEST("", "\"\""); +} + +TEST_F(StringPrinterTest, SuffixPrefixEmptyString) { + string_opts.SetSuffixToken(""); + buffer_opts.SetSuffixToken(""); + RUN_STRING_AND_BUFFER_TEST("", "\"\""); +} + +//------------------------------------------------------------------------------ +// Max size setting tests. + +TEST_F(StringPrinterTest, MaxSizeWithoutTerminator) { + max_read_size = 3; + string_opts.SetIgnoreMaxLength(false); + string_opts.SetNeedsZeroTermination(false); + // FIXME: ASCII and Unicode don't agree on the output string length. + RUN_USTRING_TEST("1234", "\"12\""); + RUN_ASCII_STRING_TEST("1234", "\"123\""); +} + +TEST_F(StringPrinterTest, MaxSizeWithTerminator) { + max_read_size = 3; + string_opts.SetIgnoreMaxLength(false); + string_opts.SetNeedsZeroTermination(true); + // FIXME: ASCII and Unicode don't agree on the output string length. + RUN_USTRING_TEST("1234", "\"12\""); + RUN_ASCII_STRING_TEST("1234", "\"123\""); +} + +TEST_F(StringPrinterTest, MaxSizeWithoutTerminatorAndShorterSourceLength) { + max_read_size = 3; + string_opts.SetIgnoreMaxLength(false); + string_opts.SetNeedsZeroTermination(false); + string_opts.SetSourceSize(1); + RUN_STRING_TEST("1234", "\"1\""); +} + +TEST_F(StringPrinterTest, MaxSizeWithTerminatorAndShorterSourceLength) { + max_read_size = 3; + string_opts.SetIgnoreMaxLength(false); + string_opts.SetNeedsZeroTermination(true); + string_opts.SetSourceSize(1); + // FIXME: ASCII and Unicode don't agree on the output string length. + RUN_USTRING_TEST("1234", "\"\""); + RUN_ASCII_STRING_TEST("1234", "\"1\""); +} + +TEST_F(StringPrinterTest, MaxSizeWithoutTerminatorAndSameSourceLength) { + max_read_size = 3; + string_opts.SetIgnoreMaxLength(false); + string_opts.SetNeedsZeroTermination(false); + string_opts.SetSourceSize(3); + RUN_STRING_TEST("1234", "\"123\""); +} + +TEST_F(StringPrinterTest, MaxSizeWithTerminatorAndSameSourceLength) { + max_read_size = 3; + string_opts.SetIgnoreMaxLength(false); + string_opts.SetNeedsZeroTermination(true); + string_opts.SetSourceSize(3); + // FIXME: ASCII and Unicode don't agree on the output string length. + RUN_USTRING_TEST("1234", "\"12\""); + RUN_ASCII_STRING_TEST("1234", "\"123\""); +} + +TEST_F(StringPrinterTest, MaxSizeWithoutTerminatorAndLongerSourceLength) { + max_read_size = 3; + string_opts.SetIgnoreMaxLength(false); + string_opts.SetNeedsZeroTermination(false); + string_opts.SetSourceSize(4); + RUN_STRING_TEST("1234", "\"123\"..."); +} + +TEST_F(StringPrinterTest, MaxSizeWithTerminatorAndLongerSourceLength) { + max_read_size = 3; + string_opts.SetIgnoreMaxLength(false); + string_opts.SetNeedsZeroTermination(true); + string_opts.SetSourceSize(4); + // FIXME: ASCII and Unicode don't agree on the output string length. + RUN_USTRING_TEST("1234", "\"12\"..."); + RUN_ASCII_STRING_TEST("1234", "\"123\"..."); +} + +//------------------------------------------------------------------------------ +// Truncated buffer. + +TEST_F(StringPrinterTest, IsTruncated) { + buffer_opts.SetIsTruncated(true); + RUN_BUFFER_TEST("1234", "\"1234\"..."); +} + +TEST_F(StringPrinterTest, IsNotTruncated) { + buffer_opts.SetIsTruncated(false); + RUN_BUFFER_TEST("1234", "\"1234\""); +} + +//------------------------------------------------------------------------------ +// Language tests + +TEST_F(StringPrinterTest, LanguageC) { + string_opts.SetLanguage(lldb::eLanguageTypeC); + buffer_opts.SetLanguage(lldb::eLanguageTypeC); + RUN_STRING_AND_BUFFER_TEST(" \t@\"'\\\0", "\" \\t@\\\"'\\\\\""); +} + +TEST_F(StringPrinterTest, LanguageC11) { + string_opts.SetLanguage(lldb::eLanguageTypeC11); + buffer_opts.SetLanguage(lldb::eLanguageTypeC11); + RUN_STRING_AND_BUFFER_TEST(" \t@\"'\\\0", "\" \\t@\\\"'\\\\\""); +} + +TEST_F(StringPrinterTest, LanguageObjCxx) { + string_opts.SetLanguage(lldb::eLanguageTypeObjC_plus_plus); + buffer_opts.SetLanguage(lldb::eLanguageTypeObjC_plus_plus); + RUN_STRING_AND_BUFFER_TEST(" \t@\"'\\\0", "\" \\t@\\\"'\\\\\""); +} + +TEST_F(StringPrinterTest, LanguageObjC) { + string_opts.SetLanguage(lldb::eLanguageTypeObjC); + buffer_opts.SetLanguage(lldb::eLanguageTypeObjC); + RUN_STRING_AND_BUFFER_TEST(" \t@\"'\\\0", "\" \\t@\\\"'\\\\\""); +} + +TEST_F(StringPrinterTest, LanguageSwift) { + string_opts.SetLanguage(lldb::eLanguageTypeSwift); + buffer_opts.SetLanguage(lldb::eLanguageTypeSwift); + RUN_STRING_AND_BUFFER_TEST(" \t@\"'\\\0", "\" \\t@\\\"'\\\\\""); +} + +TEST_F(StringPrinterTest, LanguageUnknown) { + string_opts.SetLanguage(lldb::eLanguageTypeUnknown); + buffer_opts.SetLanguage(lldb::eLanguageTypeUnknown); + RUN_STRING_AND_BUFFER_TEST(" \t@\"'\\\0", "\" \\t@\\\"'\\\\\""); +} + +TEST_F(StringPrinterTest, LanguageUnimplemented) { + // Not implemented, but we should at least do something reasonable. + string_opts.SetLanguage(lldb::eLanguageTypeModula2); + buffer_opts.SetLanguage(lldb::eLanguageTypeModula2); + RUN_STRING_AND_BUFFER_TEST(" \t@\"'\\\0", "\" \\t@\\\"'\\\\\""); +} + +//------------------------------------------------------------------------------ +// Quote character tests + +TEST_F(StringPrinterTest, SingleQuoteAsQuote) { + SetCharLocation(1); + string_opts.SetQuote('\''); + buffer_opts.SetQuote('\''); + // FIXME: We should probably escape ' when we use it as a quote character + // (otherwise printing a `char c = '\''` produces ''' as value. + RUN_STRING_AND_BUFFER_TEST(" \t@\"'\\\0", "' \\t@\\\"'\\\\'"); +} + +TEST_F(StringPrinterTest, LetterAsQuote) { + string_opts.SetQuote('a'); + buffer_opts.SetQuote('a'); + RUN_STRING_AND_BUFFER_TEST(" \t@\"'\\\0", "a \\t@\\\"'\\\\a"); +} + +TEST_F(StringPrinterTest, NoQuoteCharacter) { + SetCharLocation(1); + // Disables quoting. + string_opts.SetQuote('\0'); + buffer_opts.SetQuote('\0'); + // FIXME: Disabling quoting when printing ASCII instead actually quotes the + // string with NULL bytes. For now only run this with unicode strings where + // the quoting isn't doing that. We can't really test this as a \0-quoted + // string is just an empty string for the Stream we use... + RUN_USTRING_TEST(" \t@\"'\\\0", " \\t@\\\"'\\\\"); + RUN_BUFFER_TEST(" \t@\"'\\\0", " \\t@\\\"'\\\\"); +} Index: lldb/unittests/DataFormatters/CMakeLists.txt =================================================================== --- /dev/null +++ lldb/unittests/DataFormatters/CMakeLists.txt @@ -0,0 +1,10 @@ +add_lldb_unittest(DataFormattersTests + StringPrinterTest.cpp + + LINK_LIBS + lldbCore + lldbDataFormatters + lldbUtility + lldbUtilityHelpers + LLVMTestingSupport + ) Index: lldb/unittests/CMakeLists.txt =================================================================== --- lldb/unittests/CMakeLists.txt +++ lldb/unittests/CMakeLists.txt @@ -61,6 +61,7 @@ add_subdirectory(TestingSupport) add_subdirectory(Breakpoint) add_subdirectory(Core) +add_subdirectory(DataFormatters) add_subdirectory(Disassembler) add_subdirectory(Editline) add_subdirectory(Expression) Index: lldb/source/DataFormatters/StringPrinter.cpp =================================================================== --- lldb/source/DataFormatters/StringPrinter.cpp +++ lldb/source/DataFormatters/StringPrinter.cpp @@ -387,6 +387,13 @@ valobj.GetTargetSP()->GetDebugger().GetEscapeNonPrintables()); } +void StringPrinter::ReadStringAndDumpToStreamOptions::SetProcessSP( + ProcessSP p) { + m_memory_source = p.get(); + if (m_memory_source) + m_max_size = p->GetTarget().GetMaximumSizeOfStringSummary(); +} + lldb_private::formatters::StringPrinter::ReadBufferAndDumpToStreamOptions:: ReadBufferAndDumpToStreamOptions(ValueObject &valobj) : ReadBufferAndDumpToStreamOptions() { @@ -418,13 +425,14 @@ assert(options.GetStream() && "need a Stream to print the string to"); Status my_error; - ProcessSP process_sp(options.GetProcessSP()); + MemorySource *m_source = options.GetMemorySource(); - if (process_sp.get() == nullptr || options.GetLocation() == 0) + if (m_source == nullptr || options.GetLocation() == 0 || + options.GetLocation() == LLDB_INVALID_ADDRESS) return false; size_t size; - const auto max_size = process_sp->GetTarget().GetMaximumSizeOfStringSummary(); + const auto max_size = options.GetMaxSize(); bool is_truncated = false; if (options.GetSourceSize() == 0) @@ -440,7 +448,7 @@ lldb::DataBufferSP buffer_sp(new DataBufferHeap(size, 0)); - process_sp->ReadCStringFromMemory( + m_source->ReadCStringFromMemory( options.GetLocation(), (char *)buffer_sp->GetBytes(), size, my_error); if (my_error.Fail()) @@ -521,9 +529,9 @@ options.GetLocation() == LLDB_INVALID_ADDRESS) return false; - lldb::ProcessSP process_sp(options.GetProcessSP()); + MemorySource *m_source = options.GetMemorySource(); - if (!process_sp) + if (!m_source) return false; const int type_width = sizeof(SourceDataType); @@ -541,7 +549,7 @@ bool needs_zero_terminator = options.GetNeedsZeroTermination(); bool is_truncated = false; - const auto max_size = process_sp->GetTarget().GetMaximumSizeOfStringSummary(); + const auto max_size = options.GetMaxSize(); if (!sourceSize) { sourceSize = max_size; @@ -564,20 +572,20 @@ char *buffer = reinterpret_cast<char *>(buffer_sp->GetBytes()); if (needs_zero_terminator) - process_sp->ReadStringFromMemory(options.GetLocation(), buffer, - bufferSPSize, error, type_width); + m_source->ReadStringFromMemory(options.GetLocation(), buffer, bufferSPSize, + error, type_width); else - process_sp->ReadMemoryFromInferior(options.GetLocation(), - (char *)buffer_sp->GetBytes(), - bufferSPSize, error); + m_source->ReadMemoryFromInferior(options.GetLocation(), + (char *)buffer_sp->GetBytes(), + bufferSPSize, error); if (error.Fail()) { options.GetStream()->Printf("unable to read data"); return true; } - DataExtractor data(buffer_sp, process_sp->GetByteOrder(), - process_sp->GetAddressByteSize()); + DataExtractor data(buffer_sp, m_source->GetByteOrder(), + m_source->GetAddressByteSize()); StringPrinter::ReadBufferAndDumpToStreamOptions dump_options(options); dump_options.SetData(data); Index: lldb/include/lldb/Target/Process.h =================================================================== --- lldb/include/lldb/Target/Process.h +++ lldb/include/lldb/Target/Process.h @@ -349,6 +349,86 @@ return (!lhs.StopIDEqual(rhs) || !lhs.MemoryIDEqual(rhs)); } +struct MemorySource { + virtual ~MemorySource() = default; + + /// Read a NULL terminated string from memory + /// + /// This function will read a cache page at a time until a NULL string + /// terminator is found. It will stop reading if an aligned sequence of NULL + /// termination \a type_width bytes is not found before reading \a + /// cstr_max_len bytes. The results are always guaranteed to be NULL + /// terminated, and that no more than (max_bytes - type_width) bytes will be + /// read. + /// + /// \param[in] vm_addr + /// The virtual load address to start the memory read. + /// + /// \param[in] str + /// A character buffer containing at least max_bytes. + /// + /// \param[in] max_bytes + /// The maximum number of bytes to read. + /// + /// \param[in] error + /// The error status of the read operation. + /// + /// \param[in] type_width + /// The size of the null terminator (1 to 4 bytes per + /// character). Defaults to 1. + /// + /// \return + /// The error status or the number of bytes prior to the null terminator. + virtual size_t ReadStringFromMemory(lldb::addr_t vm_addr, char *str, + size_t max_bytes, Status &error, + size_t type_width = 1) = 0; + + /// Read a NULL terminated C string from memory + /// + /// This function will read a cache page at a time until the NULL + /// C string terminator is found. It will stop reading if the NULL + /// termination byte isn't found before reading \a cstr_max_len bytes, and + /// the results are always guaranteed to be NULL terminated (at most + /// cstr_max_len - 1 bytes will be read). + virtual size_t ReadCStringFromMemory(lldb::addr_t vm_addr, char *cstr, + size_t cstr_max_len, Status &error) = 0; + + /// Read of memory from a process. + /// + /// This function has the same semantics of ReadMemory except that it + /// bypasses caching. + /// + /// \param[in] vm_addr + /// A virtual load address that indicates where to start reading + /// memory from. + /// + /// \param[out] buf + /// A byte buffer that is at least \a size bytes long that + /// will receive the memory bytes. + /// + /// \param[in] size + /// The number of bytes to read. + /// + /// \param[out] error + /// An error that indicates the success or failure of this + /// operation. If error indicates success (error.Success()), + /// then the value returned can be trusted, otherwise zero + /// will be returned. + /// + /// \return + /// The number of bytes that were actually read into \a buf. If + /// the returned number is greater than zero, yet less than \a + /// size, then this function will get called again with \a + /// vm_addr, \a buf, and \a size updated appropriately. Zero is + /// returned in the case of an error. + virtual size_t ReadMemoryFromInferior(lldb::addr_t vm_addr, void *buf, + size_t size, Status &error) = 0; + + virtual lldb::ByteOrder GetByteOrder() const = 0; + + virtual uint32_t GetAddressByteSize() const = 0; +}; + /// \class Process Process.h "lldb/Target/Process.h" /// A plug-in interface definition class for debugging a process. class Process : public std::enable_shared_from_this<Process>, @@ -356,7 +436,8 @@ public UserID, public Broadcaster, public ExecutionContextScope, - public PluginInterface { + public PluginInterface, + public MemorySource { friend class FunctionCaller; // For WaitForStateChangeEventsPrivate friend class Debugger; // For PopProcessIOHandler and ProcessIOHandlerIsActive friend class DynamicLoader; // For LoadOperatingSystemPlugin @@ -550,9 +631,9 @@ int signo, // Zero for no signal int status); // Exit value of process if signal is zero - lldb::ByteOrder GetByteOrder() const; + lldb::ByteOrder GetByteOrder() const override; - uint32_t GetAddressByteSize() const; + uint32_t GetAddressByteSize() const override; uint32_t GetUniqueID() const { return m_process_unique_id; } @@ -1488,7 +1569,7 @@ /// vm_addr, \a buf, and \a size updated appropriately. Zero is /// returned in the case of an error. size_t ReadMemoryFromInferior(lldb::addr_t vm_addr, void *buf, size_t size, - Status &error); + Status &error) override; /// Read a NULL terminated string from memory /// @@ -1518,7 +1599,7 @@ /// \return /// The error status or the number of bytes prior to the null terminator. size_t ReadStringFromMemory(lldb::addr_t vm_addr, char *str, size_t max_bytes, - Status &error, size_t type_width = 1); + Status &error, size_t type_width = 1) override; /// Read a NULL terminated C string from memory /// @@ -1528,7 +1609,7 @@ /// the results are always guaranteed to be NULL terminated (at most /// cstr_max_len - 1 bytes will be read). size_t ReadCStringFromMemory(lldb::addr_t vm_addr, char *cstr, - size_t cstr_max_len, Status &error); + size_t cstr_max_len, Status &error) override; size_t ReadCStringFromMemory(lldb::addr_t vm_addr, std::string &out_str, Status &error); Index: lldb/include/lldb/DataFormatters/StringPrinter.h =================================================================== --- lldb/include/lldb/DataFormatters/StringPrinter.h +++ lldb/include/lldb/DataFormatters/StringPrinter.h @@ -17,6 +17,7 @@ #include "lldb/Utility/DataExtractor.h" namespace lldb_private { +struct MemorySource; namespace formatters { class StringPrinter { public: @@ -111,13 +112,22 @@ uint64_t GetLocation() const { return m_location; } - void SetProcessSP(lldb::ProcessSP p) { m_process_sp = p; } + void SetProcessSP(lldb::ProcessSP p); - lldb::ProcessSP GetProcessSP() const { return m_process_sp; } + void setMemorySourceAndMaxSize(MemorySource *memory_source, + size_t max_size) { + m_memory_source = memory_source; + m_max_size = max_size; + } + + MemorySource *GetMemorySource() const { return m_memory_source; } + + size_t GetMaxSize() const { return m_max_size; } private: uint64_t m_location = 0; - lldb::ProcessSP m_process_sp; + MemorySource *m_memory_source = nullptr; + size_t m_max_size = 1024; }; class ReadBufferAndDumpToStreamOptions : public DumpToStreamOptions {
_______________________________________________ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
