[Lldb-commits] [lldb] [lldb] Use LLVM's helper for Unicode conversion (NFC) (PR #112582)
Jonas Devlieghere via lldb-commits
lldb-commits at lists.llvm.org
Wed Oct 30 09:09:50 PDT 2024
https://github.com/JDevlieghere updated https://github.com/llvm/llvm-project/pull/112582
>From de19f43fbee9aff2a99106a23b7f264012ab5485 Mon Sep 17 00:00:00 2001
From: Jonas Devlieghere <jonas at devlieghere.com>
Date: Mon, 21 Oct 2024 13:40:02 -0700
Subject: [PATCH 1/2] [lldb] Use LLVM's helper for Unicode conversion (NFC)
The codecvt header has been deprecated in C++17. Use LLVM's unicode
helpers to convert between UTF-8 and UTF-16.
---
lldb/include/lldb/Host/Editline.h | 25 ---------------
lldb/source/Host/common/Editline.cpp | 48 ++++++++++++++++------------
2 files changed, 28 insertions(+), 45 deletions(-)
diff --git a/lldb/include/lldb/Host/Editline.h b/lldb/include/lldb/Host/Editline.h
index a02f90891599ad..57e2c831e3499d 100644
--- a/lldb/include/lldb/Host/Editline.h
+++ b/lldb/include/lldb/Host/Editline.h
@@ -30,9 +30,6 @@
#include "lldb/Host/Config.h"
-#if LLDB_EDITLINE_USE_WCHAR
-#include <codecvt>
-#endif
#include <locale>
#include <sstream>
#include <vector>
@@ -57,23 +54,6 @@
#include "llvm/ADT/FunctionExtras.h"
-#if defined(__clang__) && defined(__has_warning)
-#if __has_warning("-Wdeprecated-declarations")
-#define LLDB_DEPRECATED_WARNING_DISABLE \
- _Pragma("clang diagnostic push") \
- _Pragma("clang diagnostic ignored \"-Wdeprecated-declarations\"")
-#define LLDB_DEPRECATED_WARNING_RESTORE _Pragma("clang diagnostic pop")
-#endif
-#elif defined(__GNUC__) && __GNUC__ > 6
-#define LLDB_DEPRECATED_WARNING_DISABLE \
- _Pragma("GCC diagnostic push") \
- _Pragma("GCC diagnostic ignored \"-Wdeprecated-declarations\"")
-#define LLDB_DEPRECATED_WARNING_RESTORE _Pragma("GCC diagnostic pop")
-#else
-#define LLDB_DEPRECATED_WARNING_DISABLE
-#define LLDB_DEPRECATED_WARNING_RESTORE
-#endif
-
namespace lldb_private {
namespace line_editor {
@@ -383,11 +363,6 @@ class Editline {
void SetEditLinePromptCallback(EditlinePromptCallbackType callbackFn);
void SetGetCharacterFunction(EditlineGetCharCallbackType callbackFn);
-#if LLDB_EDITLINE_USE_WCHAR
- LLDB_DEPRECATED_WARNING_DISABLE
- std::wstring_convert<std::codecvt_utf8<wchar_t>> m_utf8conv;
- LLDB_DEPRECATED_WARNING_RESTORE
-#endif
::EditLine *m_editline = nullptr;
EditlineHistorySP m_history_sp;
bool m_in_history = false;
diff --git a/lldb/source/Host/common/Editline.cpp b/lldb/source/Host/common/Editline.cpp
index 60117cb5f0e615..b568ede4fa0322 100644
--- a/lldb/source/Host/common/Editline.cpp
+++ b/lldb/source/Host/common/Editline.cpp
@@ -12,6 +12,8 @@
#include "lldb/Host/Editline.h"
+#include <codecvt>
+
#include "lldb/Host/ConnectionFileDescriptor.h"
#include "lldb/Host/FileSystem.h"
#include "lldb/Host/Host.h"
@@ -23,6 +25,7 @@
#include "lldb/Utility/StreamString.h"
#include "lldb/Utility/StringList.h"
#include "lldb/Utility/Timeout.h"
+#include "llvm/Support/ConvertUTF.h"
#include "llvm/Support/FileSystem.h"
#include "llvm/Support/Locale.h"
@@ -444,7 +447,9 @@ StringList Editline::GetInputAsStringList(int line_count) {
if (line_count == 0)
break;
#if LLDB_EDITLINE_USE_WCHAR
- lines.AppendString(m_utf8conv.to_bytes(line));
+ std::string buffer;
+ llvm::convertWideToUTF8(line, buffer);
+ lines.AppendString(buffer);
#else
lines.AppendString(line);
#endif
@@ -636,7 +641,9 @@ unsigned char Editline::BreakLineCommand(int ch) {
if (m_fix_indentation_callback) {
StringList lines = GetInputAsStringList(m_current_line_index + 1);
#if LLDB_EDITLINE_USE_WCHAR
- lines.AppendString(m_utf8conv.to_bytes(new_line_fragment));
+ std::string buffer;
+ llvm::convertWideToUTF8(new_line_fragment, buffer);
+ lines.AppendString(buffer);
#else
lines.AppendString(new_line_fragment);
#endif
@@ -684,8 +691,9 @@ unsigned char Editline::EndOrAddLineCommand(int ch) {
m_input_lines.clear();
for (unsigned index = 0; index < lines.GetSize(); index++) {
#if LLDB_EDITLINE_USE_WCHAR
- m_input_lines.insert(m_input_lines.end(),
- m_utf8conv.from_bytes(lines[index]));
+ std::wstring wbuffer;
+ llvm::ConvertUTF8toWide(lines[index], wbuffer);
+ m_input_lines.insert(m_input_lines.end(), wbuffer);
#else
m_input_lines.insert(m_input_lines.end(), lines[index]);
#endif
@@ -869,7 +877,9 @@ unsigned char Editline::FixIndentationCommand(int ch) {
currentLine = currentLine.erase(0, -indent_correction);
}
#if LLDB_EDITLINE_USE_WCHAR
- m_input_lines[m_current_line_index] = m_utf8conv.from_bytes(currentLine);
+ std::wstring wbuffer;
+ llvm::ConvertUTF8toWide(currentLine, wbuffer);
+ m_input_lines[m_current_line_index] = wbuffer;
#else
m_input_lines[m_current_line_index] = currentLine;
#endif
@@ -1502,7 +1512,7 @@ bool Editline::GetLine(std::string &line, bool &interrupted) {
} else {
m_history_sp->Enter(input);
#if LLDB_EDITLINE_USE_WCHAR
- line = m_utf8conv.to_bytes(SplitLines(input)[0]);
+ llvm::convertWideToUTF8(SplitLines(input)[0], line);
#else
line = SplitLines(input)[0];
#endif
@@ -1574,25 +1584,23 @@ bool Editline::CompleteCharacter(char ch, EditLineGetCharType &out) {
out = (unsigned char)ch;
return true;
#else
- LLDB_DEPRECATED_WARNING_DISABLE
- std::codecvt_utf8<wchar_t> cvt;
- LLDB_DEPRECATED_WARNING_RESTORE
llvm::SmallString<4> input;
for (;;) {
- const char *from_next;
- wchar_t *to_next;
- std::mbstate_t state = std::mbstate_t();
input.push_back(ch);
- switch (cvt.in(state, input.begin(), input.end(), from_next, &out, &out + 1,
- to_next)) {
- case std::codecvt_base::ok:
+ const char *cur_ptr = input.begin();
+ const char *end_ptr = input.end();
+ llvm::UTF32 code_point = 0;
+ llvm::ConversionResult cr = llvm::convertUTF8Sequence(
+ (const llvm::UTF8 **)&cur_ptr, (const llvm::UTF8 *)end_ptr, &code_point,
+ llvm::lenientConversion);
+ switch (cr) {
+ case llvm::conversionOK:
+ out = code_point;
return out != (EditLineGetCharType)WEOF;
-
- case std::codecvt_base::error:
- case std::codecvt_base::noconv:
+ case llvm::targetExhausted:
+ case llvm::sourceIllegal:
return false;
-
- case std::codecvt_base::partial:
+ case llvm::sourceExhausted:
lldb::ConnectionStatus status;
size_t read_count = m_input_connection.Read(
&ch, 1, std::chrono::seconds(0), status, nullptr);
>From dac7b1cb9f3418354f6d5efb3fbd1034a5e3d607 Mon Sep 17 00:00:00 2001
From: Jonas Devlieghere <jonas at devlieghere.com>
Date: Wed, 30 Oct 2024 09:09:34 -0700
Subject: [PATCH 2/2] Address Pavel's feedback
---
lldb/source/Host/common/Editline.cpp | 12 ++++--------
1 file changed, 4 insertions(+), 8 deletions(-)
diff --git a/lldb/source/Host/common/Editline.cpp b/lldb/source/Host/common/Editline.cpp
index b568ede4fa0322..f95f854c5f220c 100644
--- a/lldb/source/Host/common/Editline.cpp
+++ b/lldb/source/Host/common/Editline.cpp
@@ -10,11 +10,8 @@
#include <iomanip>
#include <optional>
-#include "lldb/Host/Editline.h"
-
-#include <codecvt>
-
#include "lldb/Host/ConnectionFileDescriptor.h"
+#include "lldb/Host/Editline.h"
#include "lldb/Host/FileSystem.h"
#include "lldb/Host/Host.h"
#include "lldb/Utility/CompletionRequest.h"
@@ -1587,12 +1584,11 @@ bool Editline::CompleteCharacter(char ch, EditLineGetCharType &out) {
llvm::SmallString<4> input;
for (;;) {
input.push_back(ch);
- const char *cur_ptr = input.begin();
- const char *end_ptr = input.end();
+ auto *cur_ptr = reinterpret_cast<const llvm::UTF8 *>(input.begin());
+ auto *end_ptr = reinterpret_cast<const llvm::UTF8 *>(input.end());
llvm::UTF32 code_point = 0;
llvm::ConversionResult cr = llvm::convertUTF8Sequence(
- (const llvm::UTF8 **)&cur_ptr, (const llvm::UTF8 *)end_ptr, &code_point,
- llvm::lenientConversion);
+ &cur_ptr, end_ptr, &code_point, llvm::lenientConversion);
switch (cr) {
case llvm::conversionOK:
out = code_point;
More information about the lldb-commits
mailing list