1 /*
2 * Copyright (c) 2023 Huawei Device Co., Ltd.
3 * Licensed under the Apache License, Version 2.0 (the "License");
4 * you may not use this file except in compliance with the License.
5 * You may obtain a copy of the License at
6 *
7 * http://www.apache.org/licenses/LICENSE-2.0
8 *
9 * Unless required by applicable law or agreed to in writing, software
10 * distributed under the License is distributed on an "AS IS" BASIS,
11 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
12 * See the License for the specific language governing permissions and
13 * limitations under the License.
14 */
15
16 #ifndef MAPLE_IR_INCLUDE_LEXER_H
17 #define MAPLE_IR_INCLUDE_LEXER_H
18 #include "cstdio"
19 #include <fstream>
20 #include "types_def.h"
21 #include "tokens.h"
22 #include "mempool_allocator.h"
23 #include "mir_module.h"
24
25 namespace maple {
26 class MIRParser; // circular dependency exists, no other choice
27 class MIRLexer {
28 friend MIRParser;
29
30 public:
31 explicit MIRLexer(MIRModule &mod);
~MIRLexer()32 ~MIRLexer()
33 {
34 airFile = nullptr;
35 if (airFileInternal.is_open()) {
36 airFileInternal.close();
37 }
38 }
39
40 void PrepareForFile(const std::string &filename);
41 void PrepareForString(const std::string &src);
42 TokenKind NextToken();
43 TokenKind LexToken();
GetTokenKind()44 TokenKind GetTokenKind() const
45 {
46 return kind;
47 }
48
GetLineNum()49 uint32 GetLineNum() const
50 {
51 return lineNum;
52 }
53
GetCurIdx()54 uint32 GetCurIdx() const
55 {
56 return curIdx;
57 }
58
59 // get the identifier name after the % or $ prefix
GetName()60 const std::string &GetName() const
61 {
62 return name;
63 }
64
GetTheIntVal()65 uint64 GetTheIntVal() const
66 {
67 return theIntVal;
68 }
69
GetTheFloatVal()70 float GetTheFloatVal() const
71 {
72 return theFloatVal;
73 }
74
GetTheDoubleVal()75 double GetTheDoubleVal() const
76 {
77 return theDoubleVal;
78 }
79
80 std::string GetTokenString() const; // for error reporting purpose
81
82 private:
83 MIRModule &module;
84 // for storing the different types of constant values
85 uint64 theIntVal = 0; // also indicates preg number under TK_preg
86 float theFloatVal = 0.0;
87 double theDoubleVal = 0.0;
88 MapleVector<std::string> seenComments;
89 std::ifstream *airFile = nullptr;
90 std::ifstream airFileInternal;
91 std::string line;
92 size_t lineBufSize = 0; // the allocated size of line(buffer).
93 uint32 currentLineSize = 0;
94 uint32 curIdx = 0;
95 uint32 lineNum = 0;
96 TokenKind kind = TK_invalid;
97 std::string name = ""; // store the name token without the % or $ prefix
98 MapleUnorderedMap<std::string, TokenKind> keywordMap;
99 std::queue<std::string> mirQueue;
100 bool needFile = true;
RemoveReturnInline(std::string & line)101 void RemoveReturnInline(std::string &line)
102 {
103 if (line.empty()) {
104 return;
105 }
106 if (line.back() == '\n') {
107 line.pop_back();
108 }
109 if (line.back() == '\r') {
110 line.pop_back();
111 }
112 }
113
114 int ReadALine(); // read a line from MIR (text) file.
115 int ReadALineByMirQueue(); // read a line from MIR Queue.
116 void GenName();
117 TokenKind GetConstVal();
118 TokenKind GetSpecialFloatConst();
119 TokenKind GetHexConst(uint32 valStart, bool negative);
120 TokenKind GetIntConst(uint32 valStart, bool negative);
121 TokenKind GetFloatConst(uint32 valStart, uint32 startIdx, bool negative);
122 TokenKind GetSpecialTokenUsingOneCharacter(char c);
123 TokenKind GetTokenWithPrefixDollar();
124 TokenKind GetTokenWithPrefixPercent();
125 TokenKind GetTokenWithPrefixAmpersand();
126 TokenKind GetTokenWithPrefixAtOrCircumflex(char prefix);
127 TokenKind GetTokenWithPrefixExclamation();
128 TokenKind GetTokenWithPrefixQuotation();
129 TokenKind GetTokenWithPrefixDoubleQuotation();
130 TokenKind GetTokenSpecial();
131
GetCharAt(uint32 idx)132 char GetCharAt(uint32 idx) const
133 {
134 return line[idx];
135 }
136
GetCharAtWithUpperCheck(uint32 idx)137 char GetCharAtWithUpperCheck(uint32 idx) const
138 {
139 return idx < currentLineSize ? line[idx] : 0;
140 }
141
GetCharAtWithLowerCheck(uint32 idx)142 char GetCharAtWithLowerCheck(uint32 idx) const
143 {
144 return idx >= 0 ? line[idx] : 0;
145 }
146
GetCurrentCharWithUpperCheck()147 char GetCurrentCharWithUpperCheck()
148 {
149 return curIdx < currentLineSize ? line[curIdx] : 0;
150 }
151
GetNextCurrentCharWithUpperCheck()152 char GetNextCurrentCharWithUpperCheck()
153 {
154 ++curIdx;
155 return curIdx < currentLineSize ? line[curIdx] : 0;
156 }
157
SetFile(std::ifstream & file)158 void SetFile(std::ifstream &file)
159 {
160 airFile = &file;
161 }
162
GetFile()163 std::ifstream *GetFile() const
164 {
165 return airFile;
166 }
167
SetMirQueue(const std::string & fileText)168 void SetMirQueue(const std::string &fileText)
169 {
170 StringUtils::Split(fileText, mirQueue, '\n');
171 needFile = false;
172 }
173 };
174
IsPrimitiveType(TokenKind tk)175 inline bool IsPrimitiveType(TokenKind tk)
176 {
177 return (tk >= TK_void) && (tk < TK_unknown);
178 }
179
IsVarName(TokenKind tk)180 inline bool IsVarName(TokenKind tk)
181 {
182 return (tk == TK_lname) || (tk == TK_gname);
183 }
184
IsExprBinary(TokenKind tk)185 inline bool IsExprBinary(TokenKind tk)
186 {
187 return (tk >= TK_add) && (tk <= TK_sub);
188 }
189
IsConstValue(TokenKind tk)190 inline bool IsConstValue(TokenKind tk)
191 {
192 return (tk >= TK_intconst) && (tk <= TK_doubleconst);
193 }
194
IsConstAddrExpr(TokenKind tk)195 inline bool IsConstAddrExpr(TokenKind tk)
196 {
197 return (tk == TK_addrof) || (tk == TK_addroffunc) || (tk == TK_addroflabel) || (tk == TK_conststr) ||
198 (tk == TK_conststr16);
199 }
200 } // namespace maple
201 #endif // MAPLE_IR_INCLUDE_LEXER_H
202