From 8f9cf0d4856bb53009bb58b53a42e21e2cd1e947 Mon Sep 17 00:00:00 2001 From: Kyle Gunger Date: Sun, 28 Jun 2020 14:30:45 -0400 Subject: [Initial parser] Upload existing --- src/main.go | 24 +++++ src/tparse/parse.go | 203 +++++++++++++++++++++++++++++++++++++++ src/tparse/token.go | 13 +++ src/tparse/type.go | 267 ++++++++++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 507 insertions(+) create mode 100644 src/main.go create mode 100644 src/tparse/parse.go create mode 100644 src/tparse/token.go create mode 100644 src/tparse/type.go (limited to 'src') diff --git a/src/main.go b/src/main.go new file mode 100644 index 0000000..ec3b14b --- /dev/null +++ b/src/main.go @@ -0,0 +1,24 @@ +package main + +import "fmt" +import "tparse" +import "flag" +import "os" + +func main() { + inputFile := flag.String("in", "", "The file to parse") + outputFile := flag.String("out", "out.tnp", "The file to store the parse in") + + flag.Parse() + + fd, err := os.Create(*outputFile) + + if err != nil { + fmt.Println(err.Error()) + return + } + + fd.WriteString(fmt.Sprint(tparse.ParseFile(*inputFile))) + + fd.Close() +} diff --git a/src/tparse/parse.go b/src/tparse/parse.go new file mode 100644 index 0000000..e9e1ee6 --- /dev/null +++ b/src/tparse/parse.go @@ -0,0 +1,203 @@ +package tparse + +import ( + "bufio" + "io" + "os" + "strings" + "unicode" + "unicode/utf8" +) + +// Read in a number (may be a float) +func numericLiteral(r *bufio.Reader) Token { + decimal := false + run, _, err := r.ReadRune() + + b := strings.Builder{} + + for ; err == nil; run, _, err = r.ReadRune() { + if (run == '.' || run == ',') && !decimal { + decimal = true + } else if !unicode.IsNumber(run) { + break + } + b.WriteRune(run) + } + + r.UnreadRune() + + return Token{Type: LITERAL, Data: b.String()} +} + +// Parse a string (will escape \" only in this stage) +func stringLiteral(r *bufio.Reader) Token { + escape := false + run, _, err := r.ReadRune() + + if run != '"' { + return Token{Type: LITERAL} + } + + b := strings.Builder{} + + for ; err == nil; run, _, err = r.ReadRune() { + b.WriteRune(run) + if run == '\\' && !escape { + escape = true + } else if run == '"' && !escape { + break + } + + } + + return Token{Type: LITERAL, Data: b.String()} +} + +// Parse a character in (escape \\ or \') +func charLiteral(r *bufio.Reader) Token { + escape := false + run, _, err := r.ReadRune() + + if run != '\'' { + return Token{Type: LITERAL} + } + + b := strings.Builder{} + + for ; err == nil; run, _, err = r.ReadRune() { + b.WriteRune(run) + if run == '\\' && !escape { + escape = true + } else if run == '\'' && !escape { + break + } + + } + + return Token{Type: LITERAL, Data: b.String()} +} + +// Split reserved runes into rune groups +func splitResRunes(str string, max int) []Token { + out := []Token{} + + rs := StringAsRunes(str) + s, e := 0, max + + if max > len(rs) { + e = len(rs) + } + + for e <= len(rs) && s < len(rs) { + if checkRuneGroup(RunesAsString(rs[s:e])) != -1 || e == s+1 { + tmp := RunesAsString(rs[s:e]) + out = append(out, Token{Type: checkRuneGroup(tmp), Data: tmp}) + s = e + if s+max < len(rs) { + e = s + max + } else { + e = len(rs) + } + } else if e != s+1 { + e-- + } + } + + return out +} + +// ParseFile tries to read a file and turn it into a series of tokens +func ParseFile(path string) []Token { + out := []Token{} + + fd, err := os.Open(path) + + if err != nil { + return out + } + + read := bufio.NewReader(fd) + + b := strings.Builder{} + + max := maxResRunes() + + for r := ' '; ; r, _, err = read.ReadRune() { + // If error in stream or EOF, break + if err != nil { + if err != io.EOF { + out = append(out, Token{Type: -1}) + } + break + } + + // Checking for a space + if unicode.IsSpace(r) { + if b.String() != "" { + out = append(out, Token{Type: checkToken(b.String()), Data: b.String()}) + b.Reset() + } + continue + } + + // Checking for a rune group + if checkResRune(r) != -1 { + if b.String() != "" { + out = append(out, Token{Type: checkToken(b.String()), Data: b.String()}) + b.Reset() + } + + for ; err == nil; r, _, err = read.ReadRune() { + if checkResRune(r) == -1 { + break + } + b.WriteRune(r) + } + + read.UnreadRune() + + out = append(out, splitResRunes(b.String(), max)...) + + b.Reset() + + continue + } + + // Accumulate + b.WriteRune(r) + } + + return out +} + +// StringAsRunes returns a string as a rune slice +func StringAsRunes(s string) []rune { + out := []rune{} + for i, j := 0, 0; i < len(s); i += j { + r, w := utf8.DecodeRuneInString(s[i:]) + out = append(out, r) + j = w + } + return out +} + +// BytesAsRunes returns a byte slice as a rune slice +func BytesAsRunes(b []byte) []rune { + out := []rune{} + for i, j := 0, 0; i < len(b); i += j { + r, w := utf8.DecodeRune(b[i:]) + out = append(out, r) + j = w + } + return out +} + +// RunesAsString returns a string from a slice of runes +func RunesAsString(rs []rune) string { + b := strings.Builder{} + for _, r := range rs { + b.WriteRune(r) + } + return b.String() +} diff --git a/src/tparse/token.go b/src/tparse/token.go new file mode 100644 index 0000000..712b746 --- /dev/null +++ b/src/tparse/token.go @@ -0,0 +1,13 @@ +package tparse + +// Token represents a token in a program +type Token struct { + Type int + Data string +} + +// Container represents a container of data +type Container struct { + Data []interface{} + Holds bool +} diff --git a/src/tparse/type.go b/src/tparse/type.go new file mode 100644 index 0000000..ee8b5e7 --- /dev/null +++ b/src/tparse/type.go @@ -0,0 +1,267 @@ +package tparse + +import () + +// LINESEP represents a line seperator +const LINESEP = 0 + +// ARGNSEP represents an inline seperator +const ARGNSEP = 1 + +// DELIMIT represents an opening or closing delimiter +const DELIMIT = 2 + +// AUGMENT represents an augmentation +const AUGMENT = 3 + +// LITERAL represents a literal value +const LITERAL = 4 + +// KEYTYPE represents a built in type +const KEYTYPE = 5 + +// PREWORD represents a reserved pre-processor directive +const PREWORD = 6 + +// KEYWORD represents a reserved word +const KEYWORD = 7 + +// DEFWORD represents a user-defined word such as a variable, method, or struct +const DEFWORD = 8 + +// RESWORD represents all the reserved words and what type of tokens they are. +var RESWORD = map[string]int{ + "import": PREWORD, + + "int": KEYTYPE, + "bool": KEYTYPE, + "float": KEYTYPE, + "char": KEYTYPE, + + "struct": KEYWORD, + "type": KEYWORD, + + "loop": KEYWORD, + "continue": KEYWORD, + "break": KEYWORD, + + "switch": KEYWORD, + "case": KEYWORD, + "default": KEYWORD, + + "label": KEYWORD, + "goto": KEYWORD, + + "if": KEYWORD, + "else": KEYWORD, + + "const": KEYWORD, + "static": KEYWORD, + "volatile": KEYWORD, + + "true": LITERAL, + "false": LITERAL, + + "null": LITERAL, +} + +func checkResWord(s string) int { + out, prs := RESWORD[s] + if !prs { + return -1 + } + return out +} + +// RESRUNE represents all the reserved runes +var RESRUNE = map[rune]int{ + // Starting condition open + '(': DELIMIT, + // Starting condition close + ')': DELIMIT, + // Ending condition open + '[': DELIMIT, + // Ending condition close + ']': DELIMIT, + // Array mark open + '{': DELIMIT, + // Array mark close + '}': DELIMIT, + // String literal + '\'': DELIMIT, + // String + '"': DELIMIT, + + // Start of pre-proc directive + ':': LINESEP, + // Start of line + ';': LINESEP, + // Start of comment + '#': LINESEP, + + // Seperate arguments + ',': ARGNSEP, + + // Assignment + '=': AUGMENT, + + // Get + '.': AUGMENT, + + // Bitwise and + '&': AUGMENT, + // Bitwise or + '|': AUGMENT, + // Bitwise xor + '^': AUGMENT, + + // Greater than + '>': AUGMENT, + // Less than + '<': AUGMENT, + + // Not (prefix any bool or bitwise) + '!': AUGMENT, + + // Addition + '+': AUGMENT, + // Subtraction + '-': AUGMENT, + // Multiplication + '*': AUGMENT, + // Division + '/': AUGMENT, + // Mod + '%': AUGMENT, + + // Address of + '~': AUGMENT, + // De-ref + '_': AUGMENT, +} + +func checkResRune(r rune) int { + out, prs := RESRUNE[r] + if !prs { + return -1 + } + return out +} + +// RESRUNES Reserved sets of reserved runes which mean something +var RESRUNES = map[string]int{ + // Pre-processor block + "/:": DELIMIT, + ":/": DELIMIT, + // Code block + "/;": DELIMIT, + ";/": DELIMIT, + // Comment block + "/#": DELIMIT, + "#/": DELIMIT, + + // Quick chain + "::": DELIMIT, + ":;": DELIMIT, + ":#": DELIMIT, + + ";;": DELIMIT, + ";:": DELIMIT, + ";#": DELIMIT, + + "##": DELIMIT, + "#:": DELIMIT, + "#;": DELIMIT, + + // Boolean equ + "==": AUGMENT, + // Boolean and + "&&": AUGMENT, + // Boolean or + "||": AUGMENT, + + // Bitwise l-shift + "<<": AUGMENT, + // Bitwise r-shift + ">>": AUGMENT, + + // PREaugmented augmentors + "&=": AUGMENT, + "|=": AUGMENT, + "^=": AUGMENT, + "!=": AUGMENT, + "+=": AUGMENT, + "-=": AUGMENT, + "*=": AUGMENT, + "/=": AUGMENT, + "%=": AUGMENT, + "~=": AUGMENT, + "_=": AUGMENT, + + // POSTaugmented augmentors + "!&": AUGMENT, + "!|": AUGMENT, + "!^": AUGMENT, + "!==": AUGMENT, + "!&&": AUGMENT, + "!||": AUGMENT, + "!>": AUGMENT, + "!<": AUGMENT, + ">==": AUGMENT, + "<==": AUGMENT, +} + +func maxResRunes() int { + max := 0 + + for k := range RESRUNES { + if len(k) > max { + max = len(k) + } + } + + return max +} + +func checkRuneGroup(s string) int { + rs := StringAsRunes(s) + + if len(rs) == 1 { + return checkResRune(rs[0]) + } + + out, prs := RESRUNES[s] + if !prs { + return -1 + } + return out +} + +func checkToken(s string) int { + rs := StringAsRunes(s) + + if len(rs) == 0 { + return -1 + } + + if len(rs) == 1 { + o := checkResRune(rs[0]) + if o > -1 { + return o + } + } + + o := checkResWord(s) + + if o > -1 { + return o + } + + o = checkRuneGroup(s) + + if o > -1 { + return o + } + + return DEFWORD +} -- cgit v1.2.3