[clang] [clang][Diagnostics] Highlight code snippets (PR #66514)
Timm Baeder via cfe-commits
cfe-commits at lists.llvm.org
Thu Sep 21 05:57:57 PDT 2023
Timm =?utf-8?q?Bäder?= <tbaeder at redhat.com>,
Timm =?utf-8?q?Bäder?= <tbaeder at redhat.com>,
Timm =?utf-8?q?Bäder?= <tbaeder at redhat.com>,
Timm =?utf-8?q?Bäder?= <tbaeder at redhat.com>
Message-ID:
In-Reply-To: <llvm/llvm-project/pull/66514/clang at github.com>
https://github.com/tbaederr updated https://github.com/llvm/llvm-project/pull/66514
>From 4323e6952a577a81a5fdc51ce9571f28d7ccb6f9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timm=20B=C3=A4der?= <tbaeder at redhat.com>
Date: Fri, 15 Sep 2023 15:51:39 +0200
Subject: [PATCH 1/5] [clang][Diagnostics] Highlight code snippets
Add some primitive syntax highlighting to our code snippet output.
---
.../clang/Frontend/CodeSnippetHighlighter.h | 46 +++++++
clang/include/clang/Frontend/TextDiagnostic.h | 2 +
clang/lib/Frontend/CMakeLists.txt | 1 +
clang/lib/Frontend/CodeSnippetHighlighter.cpp | 120 ++++++++++++++++++
clang/lib/Frontend/TextDiagnostic.cpp | 26 ++++
5 files changed, 195 insertions(+)
create mode 100644 clang/include/clang/Frontend/CodeSnippetHighlighter.h
create mode 100644 clang/lib/Frontend/CodeSnippetHighlighter.cpp
diff --git a/clang/include/clang/Frontend/CodeSnippetHighlighter.h b/clang/include/clang/Frontend/CodeSnippetHighlighter.h
new file mode 100644
index 000000000000000..776954b59e2e1a8
--- /dev/null
+++ b/clang/include/clang/Frontend/CodeSnippetHighlighter.h
@@ -0,0 +1,46 @@
+//===--- CodeSnippetHighlighter.h - Code snippet highlighting ---*- 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
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_CLANG_FRONTEND_CODESNIPPETHIGHLIGHTER_H
+#define LLVM_CLANG_FRONTEND_CODESNIPPETHIGHLIGHTER_H
+
+#include "clang/Basic/LangOptions.h"
+#include "llvm/ADT/SmallSet.h"
+#include "llvm/Support/raw_ostream.h"
+#include <vector>
+
+namespace clang {
+
+struct StyleRange {
+ unsigned Start;
+ unsigned End;
+ const enum llvm::raw_ostream::Colors c;
+};
+
+class CodeSnippetHighlighter final {
+public:
+ CodeSnippetHighlighter() = default;
+
+ /// Produce StyleRanges for the given line.
+ /// The returned vector contains non-overlapping style ranges. They are sorted
+ /// from beginning of the line to the end.
+ std::vector<StyleRange> highlightLine(llvm::StringRef SourceLine,
+ const LangOptions &LangOpts);
+
+private:
+ bool Initialized = false;
+ /// Fills Keywords and Literals.
+ void ensureTokenData();
+
+ llvm::SmallSet<StringRef, 12> Keywords;
+ llvm::SmallSet<StringRef, 12> Literals;
+};
+
+} // namespace clang
+
+#endif
diff --git a/clang/include/clang/Frontend/TextDiagnostic.h b/clang/include/clang/Frontend/TextDiagnostic.h
index 7eb0ab0cdc9bca8..59fd4d4f9408d48 100644
--- a/clang/include/clang/Frontend/TextDiagnostic.h
+++ b/clang/include/clang/Frontend/TextDiagnostic.h
@@ -15,6 +15,7 @@
#ifndef LLVM_CLANG_FRONTEND_TEXTDIAGNOSTIC_H
#define LLVM_CLANG_FRONTEND_TEXTDIAGNOSTIC_H
+#include "clang/Frontend/CodeSnippetHighlighter.h"
#include "clang/Frontend/DiagnosticRenderer.h"
namespace clang {
@@ -33,6 +34,7 @@ namespace clang {
/// printing coming out of libclang.
class TextDiagnostic : public DiagnosticRenderer {
raw_ostream &OS;
+ CodeSnippetHighlighter SnippetHighlighter;
public:
TextDiagnostic(raw_ostream &OS,
diff --git a/clang/lib/Frontend/CMakeLists.txt b/clang/lib/Frontend/CMakeLists.txt
index 1e5f0a859dfd568..f3547f771593093 100644
--- a/clang/lib/Frontend/CMakeLists.txt
+++ b/clang/lib/Frontend/CMakeLists.txt
@@ -42,6 +42,7 @@ add_clang_library(clangFrontend
TextDiagnosticPrinter.cpp
VerifyDiagnosticConsumer.cpp
InterfaceStubFunctionsConsumer.cpp
+ CodeSnippetHighlighter.cpp
DEPENDS
ClangDriverOptions
diff --git a/clang/lib/Frontend/CodeSnippetHighlighter.cpp b/clang/lib/Frontend/CodeSnippetHighlighter.cpp
new file mode 100644
index 000000000000000..829a533ad2692e5
--- /dev/null
+++ b/clang/lib/Frontend/CodeSnippetHighlighter.cpp
@@ -0,0 +1,120 @@
+
+#include "clang/Frontend/CodeSnippetHighlighter.h"
+#include "clang/Basic/DiagnosticOptions.h"
+#include "clang/Basic/SourceManager.h"
+#include "clang/Lex/Lexer.h"
+#include "llvm/Support/raw_ostream.h"
+
+using namespace clang;
+
+void CodeSnippetHighlighter::ensureTokenData() {
+ if (Initialized)
+ return;
+
+ // List of keywords, literals and types we want to highlight.
+ // These are best-effort, as is everything we do wrt. highlighting.
+ Keywords.insert("_Static_assert");
+ Keywords.insert("auto");
+ Keywords.insert("concept");
+ Keywords.insert("const");
+ Keywords.insert("consteval");
+ Keywords.insert("constexpr");
+ Keywords.insert("delete");
+ Keywords.insert("do");
+ Keywords.insert("else");
+ Keywords.insert("final");
+ Keywords.insert("for");
+ Keywords.insert("if");
+ Keywords.insert("mutable");
+ Keywords.insert("namespace");
+ Keywords.insert("new");
+ Keywords.insert("private");
+ Keywords.insert("public");
+ Keywords.insert("requires");
+ Keywords.insert("return");
+ Keywords.insert("static");
+ Keywords.insert("static_assert");
+ Keywords.insert("using");
+ Keywords.insert("void");
+ Keywords.insert("volatile");
+ Keywords.insert("while");
+
+ // Builtin types we highlight
+ Keywords.insert("void");
+ Keywords.insert("char");
+ Keywords.insert("short");
+ Keywords.insert("int");
+ Keywords.insert("unsigned");
+ Keywords.insert("long");
+ Keywords.insert("float");
+ Keywords.insert("double");
+
+ Literals.insert("true");
+ Literals.insert("false");
+ Literals.insert("nullptr");
+
+ Initialized = true;
+}
+
+static SourceManager createTempSourceManager() {
+ FileSystemOptions FileOpts;
+ FileManager FileMgr(FileOpts);
+ llvm::IntrusiveRefCntPtr<DiagnosticIDs> DiagIDs(new DiagnosticIDs());
+ llvm::IntrusiveRefCntPtr<DiagnosticOptions> DiagOpts(new DiagnosticOptions());
+ DiagnosticsEngine diags(DiagIDs, DiagOpts);
+ return SourceManager(diags, FileMgr);
+}
+
+static Lexer createTempLexer(llvm::MemoryBufferRef B, SourceManager &FakeSM,
+ const LangOptions &LangOpts) {
+ return Lexer(FakeSM.createFileID(B), B, FakeSM, LangOpts);
+}
+
+std::vector<StyleRange>
+CodeSnippetHighlighter::highlightLine(StringRef SourceLine,
+ const LangOptions &LangOpts) {
+ ensureTokenData();
+
+ constexpr raw_ostream::Colors CommentColor = raw_ostream::BLACK;
+ constexpr raw_ostream::Colors LiteralColor = raw_ostream::GREEN;
+ constexpr raw_ostream::Colors KeywordColor = raw_ostream::YELLOW;
+
+ const auto MemBuf = llvm::MemoryBuffer::getMemBuffer(SourceLine);
+ SourceManager FakeSM = createTempSourceManager();
+ Lexer L = createTempLexer(MemBuf->getMemBufferRef(), FakeSM, LangOpts);
+ L.SetKeepWhitespaceMode(true);
+
+ std::vector<StyleRange> Styles;
+ bool Stop = false;
+ while (!Stop) {
+ Token tok;
+ Stop = L.LexFromRawLexer(tok);
+ if (tok.is(tok::unknown))
+ continue;
+
+ bool Invalid;
+ unsigned Start =
+ FakeSM.getSpellingColumnNumber(tok.getLocation(), &Invalid) - 1;
+ if (Invalid)
+ continue;
+
+ if (tok.is(tok::raw_identifier)) {
+ // Almost everything we lex is an identifier, since we use a raw lexer.
+ // Some should be highlightes as literals, others as keywords.
+ if (Keywords.contains(tok.getRawIdentifier()))
+ Styles.push_back(
+ StyleRange{Start, Start + tok.getLength(), KeywordColor});
+ else if (Literals.contains(tok.getRawIdentifier()))
+ Styles.push_back(
+ StyleRange{Start, Start + tok.getLength(), LiteralColor});
+ } else if (tok::isLiteral(tok.getKind())) {
+ Styles.push_back(
+ StyleRange{Start, Start + tok.getLength(), LiteralColor});
+ } else if (tok.is(tok::comment)) {
+ Styles.push_back(
+ StyleRange{Start, Start + tok.getLength(), CommentColor});
+ }
+ }
+
+ return Styles;
+}
diff --git a/clang/lib/Frontend/TextDiagnostic.cpp b/clang/lib/Frontend/TextDiagnostic.cpp
index eaa6e8d29a1dece..8bd13aa72b13235 100644
--- a/clang/lib/Frontend/TextDiagnostic.cpp
+++ b/clang/lib/Frontend/TextDiagnostic.cpp
@@ -11,6 +11,7 @@
#include "clang/Basic/DiagnosticOptions.h"
#include "clang/Basic/FileManager.h"
#include "clang/Basic/SourceManager.h"
+#include "clang/Frontend/CodeSnippetHighlighter.h"
#include "clang/Lex/Lexer.h"
#include "llvm/ADT/SmallString.h"
#include "llvm/ADT/StringExtras.h"
@@ -1279,6 +1280,9 @@ void TextDiagnostic::emitSnippetAndCaret(
void TextDiagnostic::emitSnippet(StringRef SourceLine,
unsigned MaxLineNoDisplayWidth,
unsigned LineNo) {
+ std::vector<StyleRange> Styles =
+ SnippetHighlighter.highlightLine(SourceLine, LangOpts);
+
// Emit line number.
if (MaxLineNoDisplayWidth > 0) {
unsigned LineNoDisplayWidth = getNumDisplayWidth(LineNo);
@@ -1288,11 +1292,33 @@ void TextDiagnostic::emitSnippet(StringRef SourceLine,
// Print the source line one character at a time.
bool PrintReversed = false;
+ bool HighlightingEnabled = DiagOpts->ShowColors;
size_t I = 0;
while (I < SourceLine.size()) {
auto [Str, WasPrintable] =
printableTextForNextCharacter(SourceLine, &I, DiagOpts->TabStop);
+ // Just stop highlighting anything for this line if we found a non-printable
+ // character.
+ if (!WasPrintable)
+ HighlightingEnabled = false;
+
+ // FIXME: I hope we can do this in some nicer way.
+ if (HighlightingEnabled) {
+ std::optional<enum raw_ostream::Colors> H;
+ for (auto &P : Styles) {
+ if (P.Start < I && P.End >= I) {
+ H = P.c;
+ break;
+ }
+ }
+
+ if (H) {
+ OS.changeColor(*H, false);
+ } else
+ OS.resetColor();
+ }
+
// Toggle inverted colors on or off for this character.
if (DiagOpts->ShowColors) {
if (WasPrintable == PrintReversed) {
>From d473e6c037a09199df3a12bae56209c8e4835ec7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timm=20B=C3=A4der?= <tbaeder at redhat.com>
Date: Wed, 20 Sep 2023 15:28:10 +0200
Subject: [PATCH 2/5] Get identifier table from Preprocessor
---
.../clang/Frontend/CodeSnippetHighlighter.h | 11 +-
clang/include/clang/Frontend/TextDiagnostic.h | 7 +-
clang/lib/Frontend/CodeSnippetHighlighter.cpp | 101 +++++-------------
clang/lib/Frontend/TextDiagnostic.cpp | 8 +-
clang/lib/Frontend/TextDiagnosticPrinter.cpp | 2 +-
5 files changed, 39 insertions(+), 90 deletions(-)
diff --git a/clang/include/clang/Frontend/CodeSnippetHighlighter.h b/clang/include/clang/Frontend/CodeSnippetHighlighter.h
index 776954b59e2e1a8..ec03375221f9ffc 100644
--- a/clang/include/clang/Frontend/CodeSnippetHighlighter.h
+++ b/clang/include/clang/Frontend/CodeSnippetHighlighter.h
@@ -22,6 +22,8 @@ struct StyleRange {
const enum llvm::raw_ostream::Colors c;
};
+class Preprocessor;
+
class CodeSnippetHighlighter final {
public:
CodeSnippetHighlighter() = default;
@@ -30,15 +32,8 @@ class CodeSnippetHighlighter final {
/// The returned vector contains non-overlapping style ranges. They are sorted
/// from beginning of the line to the end.
std::vector<StyleRange> highlightLine(llvm::StringRef SourceLine,
+ const Preprocessor *PP,
const LangOptions &LangOpts);
-
-private:
- bool Initialized = false;
- /// Fills Keywords and Literals.
- void ensureTokenData();
-
- llvm::SmallSet<StringRef, 12> Keywords;
- llvm::SmallSet<StringRef, 12> Literals;
};
} // namespace clang
diff --git a/clang/include/clang/Frontend/TextDiagnostic.h b/clang/include/clang/Frontend/TextDiagnostic.h
index 59fd4d4f9408d48..8cdb9b141a8a4af 100644
--- a/clang/include/clang/Frontend/TextDiagnostic.h
+++ b/clang/include/clang/Frontend/TextDiagnostic.h
@@ -19,7 +19,6 @@
#include "clang/Frontend/DiagnosticRenderer.h"
namespace clang {
-
/// Class to encapsulate the logic for formatting and printing a textual
/// diagnostic message.
///
@@ -34,12 +33,12 @@ namespace clang {
/// printing coming out of libclang.
class TextDiagnostic : public DiagnosticRenderer {
raw_ostream &OS;
+ const Preprocessor *PP;
CodeSnippetHighlighter SnippetHighlighter;
public:
- TextDiagnostic(raw_ostream &OS,
- const LangOptions &LangOpts,
- DiagnosticOptions *DiagOpts);
+ TextDiagnostic(raw_ostream &OS, const LangOptions &LangOpts,
+ const Preprocessor *PP, DiagnosticOptions *DiagOpts);
~TextDiagnostic() override;
diff --git a/clang/lib/Frontend/CodeSnippetHighlighter.cpp b/clang/lib/Frontend/CodeSnippetHighlighter.cpp
index 829a533ad2692e5..63b3707fbb7ef83 100644
--- a/clang/lib/Frontend/CodeSnippetHighlighter.cpp
+++ b/clang/lib/Frontend/CodeSnippetHighlighter.cpp
@@ -3,59 +3,12 @@
#include "clang/Basic/DiagnosticOptions.h"
#include "clang/Basic/SourceManager.h"
#include "clang/Lex/Lexer.h"
+#include "clang/Lex/Preprocessor.h"
+#include "clang/Lex/PreprocessorOptions.h"
#include "llvm/Support/raw_ostream.h"
using namespace clang;
-void CodeSnippetHighlighter::ensureTokenData() {
- if (Initialized)
- return;
-
- // List of keywords, literals and types we want to highlight.
- // These are best-effort, as is everything we do wrt. highlighting.
- Keywords.insert("_Static_assert");
- Keywords.insert("auto");
- Keywords.insert("concept");
- Keywords.insert("const");
- Keywords.insert("consteval");
- Keywords.insert("constexpr");
- Keywords.insert("delete");
- Keywords.insert("do");
- Keywords.insert("else");
- Keywords.insert("final");
- Keywords.insert("for");
- Keywords.insert("if");
- Keywords.insert("mutable");
- Keywords.insert("namespace");
- Keywords.insert("new");
- Keywords.insert("private");
- Keywords.insert("public");
- Keywords.insert("requires");
- Keywords.insert("return");
- Keywords.insert("static");
- Keywords.insert("static_assert");
- Keywords.insert("using");
- Keywords.insert("void");
- Keywords.insert("volatile");
- Keywords.insert("while");
-
- // Builtin types we highlight
- Keywords.insert("void");
- Keywords.insert("char");
- Keywords.insert("short");
- Keywords.insert("int");
- Keywords.insert("unsigned");
- Keywords.insert("long");
- Keywords.insert("float");
- Keywords.insert("double");
-
- Literals.insert("true");
- Literals.insert("false");
- Literals.insert("nullptr");
-
- Initialized = true;
-}
-
static SourceManager createTempSourceManager() {
FileSystemOptions FileOpts;
FileManager FileMgr(FileOpts);
@@ -70,49 +23,51 @@ static Lexer createTempLexer(llvm::MemoryBufferRef B, SourceManager &FakeSM,
return Lexer(FakeSM.createFileID(B), B, FakeSM, LangOpts);
}
-std::vector<StyleRange>
-CodeSnippetHighlighter::highlightLine(StringRef SourceLine,
- const LangOptions &LangOpts) {
- ensureTokenData();
-
+std::vector<StyleRange> CodeSnippetHighlighter::highlightLine(
+ StringRef SourceLine, const Preprocessor *PP, const LangOptions &LangOpts) {
constexpr raw_ostream::Colors CommentColor = raw_ostream::BLACK;
constexpr raw_ostream::Colors LiteralColor = raw_ostream::GREEN;
constexpr raw_ostream::Colors KeywordColor = raw_ostream::YELLOW;
- const auto MemBuf = llvm::MemoryBuffer::getMemBuffer(SourceLine);
SourceManager FakeSM = createTempSourceManager();
+ const auto MemBuf = llvm::MemoryBuffer::getMemBuffer(SourceLine);
Lexer L = createTempLexer(MemBuf->getMemBufferRef(), FakeSM, LangOpts);
L.SetKeepWhitespaceMode(true);
std::vector<StyleRange> Styles;
bool Stop = false;
while (!Stop) {
- Token tok;
- Stop = L.LexFromRawLexer(tok);
- if (tok.is(tok::unknown))
+ Token T;
+ Stop = L.LexFromRawLexer(T);
+ if (T.is(tok::unknown))
continue;
bool Invalid;
unsigned Start =
- FakeSM.getSpellingColumnNumber(tok.getLocation(), &Invalid) - 1;
+ FakeSM.getSpellingColumnNumber(T.getLocation(), &Invalid) - 1;
if (Invalid)
continue;
- if (tok.is(tok::raw_identifier)) {
- // Almost everything we lex is an identifier, since we use a raw lexer.
- // Some should be highlightes as literals, others as keywords.
- if (Keywords.contains(tok.getRawIdentifier()))
- Styles.push_back(
- StyleRange{Start, Start + tok.getLength(), KeywordColor});
- else if (Literals.contains(tok.getRawIdentifier()))
+ if (T.is(tok::raw_identifier)) {
+ StringRef RawIdent = T.getRawIdentifier();
+ // Special case true/false/nullptr literals, since they will otherwise be
+ // treated as keywords.
+ if (RawIdent == "true" || RawIdent == "false" || RawIdent == "nullptr") {
Styles.push_back(
- StyleRange{Start, Start + tok.getLength(), LiteralColor});
- } else if (tok::isLiteral(tok.getKind())) {
- Styles.push_back(
- StyleRange{Start, Start + tok.getLength(), LiteralColor});
- } else if (tok.is(tok::comment)) {
- Styles.push_back(
- StyleRange{Start, Start + tok.getLength(), CommentColor});
+ StyleRange{Start, Start + T.getLength(), LiteralColor});
+ } else {
+ const IdentifierInfo *II = PP->getIdentifierInfo(RawIdent);
+ assert(II);
+
+ if (II->isKeyword(LangOpts)) {
+ Styles.push_back(
+ StyleRange{Start, Start + T.getLength(), KeywordColor});
+ }
+ }
+ } else if (tok::isLiteral(T.getKind())) {
+ Styles.push_back(StyleRange{Start, Start + T.getLength(), LiteralColor});
+ } else if (T.is(tok::comment)) {
+ Styles.push_back(StyleRange{Start, Start + T.getLength(), CommentColor});
}
}
diff --git a/clang/lib/Frontend/TextDiagnostic.cpp b/clang/lib/Frontend/TextDiagnostic.cpp
index 8bd13aa72b13235..e5de17794323383 100644
--- a/clang/lib/Frontend/TextDiagnostic.cpp
+++ b/clang/lib/Frontend/TextDiagnostic.cpp
@@ -646,10 +646,10 @@ static bool printWordWrapped(raw_ostream &OS, StringRef Str, unsigned Columns,
return Wrapped;
}
-TextDiagnostic::TextDiagnostic(raw_ostream &OS,
- const LangOptions &LangOpts,
+TextDiagnostic::TextDiagnostic(raw_ostream &OS, const LangOptions &LangOpts,
+ const Preprocessor *PP,
DiagnosticOptions *DiagOpts)
- : DiagnosticRenderer(LangOpts, DiagOpts), OS(OS) {}
+ : DiagnosticRenderer(LangOpts, DiagOpts), OS(OS), PP(PP) {}
TextDiagnostic::~TextDiagnostic() {}
@@ -1281,7 +1281,7 @@ void TextDiagnostic::emitSnippet(StringRef SourceLine,
unsigned MaxLineNoDisplayWidth,
unsigned LineNo) {
std::vector<StyleRange> Styles =
- SnippetHighlighter.highlightLine(SourceLine, LangOpts);
+ SnippetHighlighter.highlightLine(SourceLine, PP, LangOpts);
// Emit line number.
if (MaxLineNoDisplayWidth > 0) {
diff --git a/clang/lib/Frontend/TextDiagnosticPrinter.cpp b/clang/lib/Frontend/TextDiagnosticPrinter.cpp
index 0ff5376098ffe8d..3bc3935078baada 100644
--- a/clang/lib/Frontend/TextDiagnosticPrinter.cpp
+++ b/clang/lib/Frontend/TextDiagnosticPrinter.cpp
@@ -36,7 +36,7 @@ TextDiagnosticPrinter::~TextDiagnosticPrinter() {
void TextDiagnosticPrinter::BeginSourceFile(const LangOptions &LO,
const Preprocessor *PP) {
// Build the TextDiagnostic utility.
- TextDiag.reset(new TextDiagnostic(OS, LO, &*DiagOpts));
+ TextDiag.reset(new TextDiagnostic(OS, LO, PP, &*DiagOpts));
}
void TextDiagnosticPrinter::EndSourceFile() {
>From c7cc185eb2309b69da58a5697e8f80ef724c5c82 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timm=20B=C3=A4der?= <tbaeder at redhat.com>
Date: Wed, 20 Sep 2023 17:24:42 +0200
Subject: [PATCH 3/5] Move the PP parameter to the end of the TextDiagnostic
ctor
---
clang/include/clang/Frontend/TextDiagnostic.h | 2 +-
clang/lib/Frontend/CodeSnippetHighlighter.cpp | 2 ++
clang/lib/Frontend/TextDiagnostic.cpp | 3 +--
clang/lib/Frontend/TextDiagnosticPrinter.cpp | 2 +-
4 files changed, 5 insertions(+), 4 deletions(-)
diff --git a/clang/include/clang/Frontend/TextDiagnostic.h b/clang/include/clang/Frontend/TextDiagnostic.h
index 8cdb9b141a8a4af..43c39ff96a2d1ce 100644
--- a/clang/include/clang/Frontend/TextDiagnostic.h
+++ b/clang/include/clang/Frontend/TextDiagnostic.h
@@ -38,7 +38,7 @@ class TextDiagnostic : public DiagnosticRenderer {
public:
TextDiagnostic(raw_ostream &OS, const LangOptions &LangOpts,
- const Preprocessor *PP, DiagnosticOptions *DiagOpts);
+ DiagnosticOptions *DiagOpts, const Preprocessor *PP = nullptr);
~TextDiagnostic() override;
diff --git a/clang/lib/Frontend/CodeSnippetHighlighter.cpp b/clang/lib/Frontend/CodeSnippetHighlighter.cpp
index 63b3707fbb7ef83..32bd61f3746023c 100644
--- a/clang/lib/Frontend/CodeSnippetHighlighter.cpp
+++ b/clang/lib/Frontend/CodeSnippetHighlighter.cpp
@@ -25,6 +25,8 @@ static Lexer createTempLexer(llvm::MemoryBufferRef B, SourceManager &FakeSM,
std::vector<StyleRange> CodeSnippetHighlighter::highlightLine(
StringRef SourceLine, const Preprocessor *PP, const LangOptions &LangOpts) {
+ if (!PP)
+ return {};
constexpr raw_ostream::Colors CommentColor = raw_ostream::BLACK;
constexpr raw_ostream::Colors LiteralColor = raw_ostream::GREEN;
constexpr raw_ostream::Colors KeywordColor = raw_ostream::YELLOW;
diff --git a/clang/lib/Frontend/TextDiagnostic.cpp b/clang/lib/Frontend/TextDiagnostic.cpp
index e5de17794323383..8675d3b08e86b71 100644
--- a/clang/lib/Frontend/TextDiagnostic.cpp
+++ b/clang/lib/Frontend/TextDiagnostic.cpp
@@ -647,8 +647,7 @@ static bool printWordWrapped(raw_ostream &OS, StringRef Str, unsigned Columns,
}
TextDiagnostic::TextDiagnostic(raw_ostream &OS, const LangOptions &LangOpts,
- const Preprocessor *PP,
- DiagnosticOptions *DiagOpts)
+ DiagnosticOptions *DiagOpts, const Preprocessor *PP)
: DiagnosticRenderer(LangOpts, DiagOpts), OS(OS), PP(PP) {}
TextDiagnostic::~TextDiagnostic() {}
diff --git a/clang/lib/Frontend/TextDiagnosticPrinter.cpp b/clang/lib/Frontend/TextDiagnosticPrinter.cpp
index 3bc3935078baada..b2fb762537573ef 100644
--- a/clang/lib/Frontend/TextDiagnosticPrinter.cpp
+++ b/clang/lib/Frontend/TextDiagnosticPrinter.cpp
@@ -36,7 +36,7 @@ TextDiagnosticPrinter::~TextDiagnosticPrinter() {
void TextDiagnosticPrinter::BeginSourceFile(const LangOptions &LO,
const Preprocessor *PP) {
// Build the TextDiagnostic utility.
- TextDiag.reset(new TextDiagnostic(OS, LO, PP, &*DiagOpts));
+ TextDiag.reset(new TextDiagnostic(OS, LO, &*DiagOpts, PP));
}
void TextDiagnosticPrinter::EndSourceFile() {
>From fec4eeee662aca00833f0b60fa42be03bbe487de Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timm=20B=C3=A4der?= <tbaeder at redhat.com>
Date: Thu, 21 Sep 2023 06:38:24 +0200
Subject: [PATCH 4/5] Tune colors
---
clang/include/clang/Frontend/CodeSnippetHighlighter.h | 2 +-
clang/lib/Frontend/CodeSnippetHighlighter.cpp | 6 +++---
clang/lib/Frontend/TextDiagnostic.cpp | 7 ++++---
3 files changed, 8 insertions(+), 7 deletions(-)
diff --git a/clang/include/clang/Frontend/CodeSnippetHighlighter.h b/clang/include/clang/Frontend/CodeSnippetHighlighter.h
index ec03375221f9ffc..c2a0184085d5da4 100644
--- a/clang/include/clang/Frontend/CodeSnippetHighlighter.h
+++ b/clang/include/clang/Frontend/CodeSnippetHighlighter.h
@@ -19,7 +19,7 @@ namespace clang {
struct StyleRange {
unsigned Start;
unsigned End;
- const enum llvm::raw_ostream::Colors c;
+ const enum llvm::raw_ostream::Colors color;
};
class Preprocessor;
diff --git a/clang/lib/Frontend/CodeSnippetHighlighter.cpp b/clang/lib/Frontend/CodeSnippetHighlighter.cpp
index 32bd61f3746023c..dba7f5d2848505a 100644
--- a/clang/lib/Frontend/CodeSnippetHighlighter.cpp
+++ b/clang/lib/Frontend/CodeSnippetHighlighter.cpp
@@ -27,9 +27,9 @@ std::vector<StyleRange> CodeSnippetHighlighter::highlightLine(
StringRef SourceLine, const Preprocessor *PP, const LangOptions &LangOpts) {
if (!PP)
return {};
- constexpr raw_ostream::Colors CommentColor = raw_ostream::BLACK;
- constexpr raw_ostream::Colors LiteralColor = raw_ostream::GREEN;
- constexpr raw_ostream::Colors KeywordColor = raw_ostream::YELLOW;
+ constexpr raw_ostream::Colors CommentColor = raw_ostream::GREEN;
+ constexpr raw_ostream::Colors LiteralColor = raw_ostream::CYAN;
+ constexpr raw_ostream::Colors KeywordColor = raw_ostream::BLUE;
SourceManager FakeSM = createTempSourceManager();
const auto MemBuf = llvm::MemoryBuffer::getMemBuffer(SourceLine);
diff --git a/clang/lib/Frontend/TextDiagnostic.cpp b/clang/lib/Frontend/TextDiagnostic.cpp
index 8675d3b08e86b71..5ba78f9e51e1259 100644
--- a/clang/lib/Frontend/TextDiagnostic.cpp
+++ b/clang/lib/Frontend/TextDiagnostic.cpp
@@ -19,6 +19,7 @@
#include "llvm/Support/ErrorHandling.h"
#include "llvm/Support/Locale.h"
#include "llvm/Support/Path.h"
+#include "llvm/Support/Process.h"
#include "llvm/Support/raw_ostream.h"
#include <algorithm>
#include <optional>
@@ -1307,14 +1308,14 @@ void TextDiagnostic::emitSnippet(StringRef SourceLine,
std::optional<enum raw_ostream::Colors> H;
for (auto &P : Styles) {
if (P.Start < I && P.End >= I) {
- H = P.c;
+ H = P.color;
break;
}
}
- if (H) {
+ if (H)
OS.changeColor(*H, false);
- } else
+ else
OS.resetColor();
}
>From 6b733700f611fcbcbe774e7398f9ac84ab18d82e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timm=20B=C3=A4der?= <tbaeder at redhat.com>
Date: Thu, 21 Sep 2023 11:01:43 +0200
Subject: [PATCH 5/5] Lex the entire file
---
.../clang/Frontend/CodeSnippetHighlighter.h | 7 +-
clang/include/clang/Frontend/TextDiagnostic.h | 3 +-
clang/lib/Frontend/CodeSnippetHighlighter.cpp | 135 ++++++++++++------
clang/lib/Frontend/TextDiagnostic.cpp | 14 +-
4 files changed, 109 insertions(+), 50 deletions(-)
diff --git a/clang/include/clang/Frontend/CodeSnippetHighlighter.h b/clang/include/clang/Frontend/CodeSnippetHighlighter.h
index c2a0184085d5da4..51c14880fb95485 100644
--- a/clang/include/clang/Frontend/CodeSnippetHighlighter.h
+++ b/clang/include/clang/Frontend/CodeSnippetHighlighter.h
@@ -23,6 +23,8 @@ struct StyleRange {
};
class Preprocessor;
+class FileID;
+class SourceManager;
class CodeSnippetHighlighter final {
public:
@@ -31,9 +33,10 @@ class CodeSnippetHighlighter final {
/// Produce StyleRanges for the given line.
/// The returned vector contains non-overlapping style ranges. They are sorted
/// from beginning of the line to the end.
- std::vector<StyleRange> highlightLine(llvm::StringRef SourceLine,
+ std::vector<StyleRange> highlightLine(unsigned LineNumber,
const Preprocessor *PP,
- const LangOptions &LangOpts);
+ const LangOptions &LangOpts, FileID FID,
+ const SourceManager &SM);
};
} // namespace clang
diff --git a/clang/include/clang/Frontend/TextDiagnostic.h b/clang/include/clang/Frontend/TextDiagnostic.h
index 43c39ff96a2d1ce..102b33aedd5ef98 100644
--- a/clang/include/clang/Frontend/TextDiagnostic.h
+++ b/clang/include/clang/Frontend/TextDiagnostic.h
@@ -105,7 +105,8 @@ class TextDiagnostic : public DiagnosticRenderer {
ArrayRef<FixItHint> Hints);
void emitSnippet(StringRef SourceLine, unsigned MaxLineNoDisplayWidth,
- unsigned LineNo);
+ FileID FID, const SourceManager &SM, unsigned LineNo,
+ unsigned DisplayLineNo);
void emitParseableFixits(ArrayRef<FixItHint> Hints, const SourceManager &SM);
};
diff --git a/clang/lib/Frontend/CodeSnippetHighlighter.cpp b/clang/lib/Frontend/CodeSnippetHighlighter.cpp
index dba7f5d2848505a..3c3f86b7b9c3fae 100644
--- a/clang/lib/Frontend/CodeSnippetHighlighter.cpp
+++ b/clang/lib/Frontend/CodeSnippetHighlighter.cpp
@@ -9,34 +9,47 @@
using namespace clang;
-static SourceManager createTempSourceManager() {
- FileSystemOptions FileOpts;
- FileManager FileMgr(FileOpts);
- llvm::IntrusiveRefCntPtr<DiagnosticIDs> DiagIDs(new DiagnosticIDs());
- llvm::IntrusiveRefCntPtr<DiagnosticOptions> DiagOpts(new DiagnosticOptions());
- DiagnosticsEngine diags(DiagIDs, DiagOpts);
- return SourceManager(diags, FileMgr);
-}
-
-static Lexer createTempLexer(llvm::MemoryBufferRef B, SourceManager &FakeSM,
- const LangOptions &LangOpts) {
- return Lexer(FakeSM.createFileID(B), B, FakeSM, LangOpts);
-}
+static constexpr raw_ostream::Colors CommentColor = raw_ostream::GREEN;
+static constexpr raw_ostream::Colors LiteralColor = raw_ostream::CYAN;
+static constexpr raw_ostream::Colors KeywordColor = raw_ostream::BLUE;
std::vector<StyleRange> CodeSnippetHighlighter::highlightLine(
- StringRef SourceLine, const Preprocessor *PP, const LangOptions &LangOpts) {
+ unsigned LineNumber, const Preprocessor *PP, const LangOptions &LangOpts,
+ FileID FID, const SourceManager &SM) {
if (!PP)
return {};
- constexpr raw_ostream::Colors CommentColor = raw_ostream::GREEN;
- constexpr raw_ostream::Colors LiteralColor = raw_ostream::CYAN;
- constexpr raw_ostream::Colors KeywordColor = raw_ostream::BLUE;
- SourceManager FakeSM = createTempSourceManager();
- const auto MemBuf = llvm::MemoryBuffer::getMemBuffer(SourceLine);
- Lexer L = createTempLexer(MemBuf->getMemBufferRef(), FakeSM, LangOpts);
+ // Classify the given token and append it to the given vector.
+ auto appendStyle = [PP, &LangOpts](std::vector<StyleRange> &Vec,
+ const Token &T, unsigned Start,
+ unsigned Length) -> void {
+ if (T.is(tok::raw_identifier)) {
+ StringRef RawIdent = T.getRawIdentifier();
+ // Special case true/false/nullptr literals, since they will otherwise be
+ // treated as keywords.
+ if (RawIdent == "true" || RawIdent == "false" || RawIdent == "nullptr") {
+ Vec.push_back(StyleRange{Start, Start + Length, LiteralColor});
+ } else {
+ const IdentifierInfo *II = PP->getIdentifierInfo(RawIdent);
+ assert(II);
+
+ if (II->isKeyword(LangOpts)) {
+ Vec.push_back(StyleRange{Start, Start + Length, KeywordColor});
+ }
+ }
+ } else if (tok::isLiteral(T.getKind())) {
+ Vec.push_back(StyleRange{Start, Start + Length, LiteralColor});
+ } else if (T.is(tok::comment)) {
+ Vec.push_back(StyleRange{Start, Start + Length, CommentColor});
+ }
+ };
+
+ auto Buff = SM.getBufferOrNone(FID);
+ assert(Buff);
+ Lexer L = Lexer(FID, *Buff, SM, LangOpts);
L.SetKeepWhitespaceMode(true);
+ std::vector<std::vector<StyleRange>> Lines;
- std::vector<StyleRange> Styles;
bool Stop = false;
while (!Stop) {
Token T;
@@ -45,33 +58,73 @@ std::vector<StyleRange> CodeSnippetHighlighter::highlightLine(
continue;
bool Invalid;
- unsigned Start =
- FakeSM.getSpellingColumnNumber(T.getLocation(), &Invalid) - 1;
+ unsigned StartCol =
+ SM.getSpellingColumnNumber(T.getLocation(), &Invalid) - 1;
+ if (Invalid)
+ continue;
+ unsigned StartLine =
+ SM.getSpellingLineNumber(T.getLocation(), &Invalid) - 1;
if (Invalid)
continue;
- if (T.is(tok::raw_identifier)) {
- StringRef RawIdent = T.getRawIdentifier();
- // Special case true/false/nullptr literals, since they will otherwise be
- // treated as keywords.
- if (RawIdent == "true" || RawIdent == "false" || RawIdent == "nullptr") {
- Styles.push_back(
- StyleRange{Start, Start + T.getLength(), LiteralColor});
- } else {
- const IdentifierInfo *II = PP->getIdentifierInfo(RawIdent);
- assert(II);
+ while (Lines.size() <= StartLine)
+ Lines.push_back({});
- if (II->isKeyword(LangOpts)) {
- Styles.push_back(
- StyleRange{Start, Start + T.getLength(), KeywordColor});
+ unsigned EndLine = SM.getSpellingLineNumber(T.getEndLoc(), &Invalid) - 1;
+ if (Invalid)
+ continue;
+
+ // Simple tokens.
+ if (StartLine == EndLine) {
+ appendStyle(Lines[StartLine], T, StartCol, T.getLength());
+ continue;
+ }
+ unsigned NumLines = EndLine - StartLine;
+
+ // For tokens that span multiple lines (think multiline comments), we
+ // divide them into multiple StyleRanges.
+ unsigned EndCol = SM.getSpellingColumnNumber(T.getEndLoc(), &Invalid) - 1;
+ if (Invalid)
+ continue;
+
+ std::string Spelling = Lexer::getSpelling(T, SM, LangOpts);
+
+ unsigned L = 0;
+ unsigned LineLength = 0;
+ for (unsigned I = 0; I <= Spelling.size(); ++I) {
+ // This line is done.
+ if (Spelling[I] == '\n' || Spelling[I] == '\r' || I == Spelling.size()) {
+ if (Lines.size() <= StartLine + L) {
+ Lines.push_back({});
}
+
+ if (L == 0) // First line
+ appendStyle(Lines[StartLine + L], T, StartCol, LineLength);
+ else if (L == NumLines) // Last line
+ appendStyle(Lines[StartLine + L], T, 0, EndCol);
+ else
+ appendStyle(Lines[StartLine + L], T, 0, LineLength);
+ ++L;
+ LineLength = 0;
+ continue;
}
- } else if (tok::isLiteral(T.getKind())) {
- Styles.push_back(StyleRange{Start, Start + T.getLength(), LiteralColor});
- } else if (T.is(tok::comment)) {
- Styles.push_back(StyleRange{Start, Start + T.getLength(), CommentColor});
+ ++LineLength;
}
}
- return Styles;
+#if 0
+ llvm::errs() << "--\nLine Style info: \n";
+ int I = 0;
+ for (std::vector<StyleRange> &Line : Lines) {
+ llvm::errs() << I << ": ";
+ for (const auto &R : Line) {
+ llvm::errs() << "{" << R.Start << ", " << R.End << "}, ";
+ }
+ llvm::errs() << "\n";
+
+ ++I;
+ }
+#endif
+
+ return Lines[LineNumber];
}
diff --git a/clang/lib/Frontend/TextDiagnostic.cpp b/clang/lib/Frontend/TextDiagnostic.cpp
index 5ba78f9e51e1259..8c08d26657254f2 100644
--- a/clang/lib/Frontend/TextDiagnostic.cpp
+++ b/clang/lib/Frontend/TextDiagnostic.cpp
@@ -1249,7 +1249,8 @@ void TextDiagnostic::emitSnippetAndCaret(
}
// Emit what we have computed.
- emitSnippet(SourceLine, MaxLineNoDisplayWidth, DisplayLineNo);
+ emitSnippet(SourceLine, MaxLineNoDisplayWidth, FID, SM, LineNo,
+ DisplayLineNo);
if (!CaretLine.empty()) {
indentForLineNumbers();
@@ -1278,16 +1279,17 @@ void TextDiagnostic::emitSnippetAndCaret(
}
void TextDiagnostic::emitSnippet(StringRef SourceLine,
- unsigned MaxLineNoDisplayWidth,
- unsigned LineNo) {
+ unsigned MaxLineNoDisplayWidth, FileID FID,
+ const SourceManager &SM, unsigned LineNo,
+ unsigned DisplayLineNo) {
std::vector<StyleRange> Styles =
- SnippetHighlighter.highlightLine(SourceLine, PP, LangOpts);
+ SnippetHighlighter.highlightLine(LineNo - 1, PP, LangOpts, FID, SM);
// Emit line number.
if (MaxLineNoDisplayWidth > 0) {
- unsigned LineNoDisplayWidth = getNumDisplayWidth(LineNo);
+ unsigned LineNoDisplayWidth = getNumDisplayWidth(DisplayLineNo);
OS.indent(MaxLineNoDisplayWidth - LineNoDisplayWidth + 1)
- << LineNo << " | ";
+ << DisplayLineNo << " | ";
}
// Print the source line one character at a time.
More information about the cfe-commits
mailing list