640 lines
25 KiB
C#

using System;
using System.Collections.Generic;
using System.IO;
using System.Text;
namespace Assignment_1
{
class Program
{
[Flags]
enum VariableFlags
{
Empty = 0,
Reserved = 1,
NoPrint = 2
}
/// <summary>
/// This captures the end-point of each part of an expression (in the stream), to validate the syntax
/// Optionally also captures the parsed string for each expression.
///
/// For this program, a word is considered to be any non-whitespace value bounded by whitespace or the array boundary.
///
/// </summary>
static Dictionary<string, Tuple<string, VariableFlags>> Symbols = new Dictionary<string, Tuple<string, VariableFlags>>
{
{ "SPACE", new Tuple<string, VariableFlags>(" ", VariableFlags.Reserved | VariableFlags.NoPrint) },
{ "TAB", new Tuple<string, VariableFlags>("\t", VariableFlags.Reserved | VariableFlags.NoPrint) },
{ "NEWLINE", new Tuple<string, VariableFlags>("\n", VariableFlags.Reserved | VariableFlags.NoPrint) },
{ "CARRIAGE_RETURN", new Tuple<string, VariableFlags>("\r", VariableFlags.Reserved | VariableFlags.NoPrint) }
};
static void Main(string[] args)
{
Console.WriteLine("┌──────────────────────────────────────────┐");
Console.WriteLine("│ 159.341 2021 Semester 1, Assignment 1 │");
Console.WriteLine("│ Submitted by Brychan Dempsey, 14299890 │");
Console.WriteLine("└──────────────────────────────────────────┘");
MemoryStream sourceStream = new MemoryStream(1024); // Creates a memory stream to retain source while being interpreted.
Parser parser = new Parser();
bool dynamicInput = false;
// From https://stackoverflow.com/questions/3453220/how-to-detect-if-console-in-stdin-has-been-redirected
// Reading from pipes is equivalent to reading user input, though the input is redirected
if (Console.IsInputRedirected)
{
sourceStream.Write(Encoding.UTF8.GetBytes(Console.In.ReadToEnd()));
sourceStream.Position = 0;
}
else
{
sourceStream.Write(Encoding.UTF8.GetBytes("{ \r\n"));
sourceStream.Position = 0;
dynamicInput = true;
}
parser.StartParsing(sourceStream, dynamicInput);
Console.ReadLine();
}
public class Parser
{
public enum statements
{
exit,
append,
list,
print,
printlength,
printwords,
printwordcount,
set,
reverse,
h
}
public void StartParsing(Stream source, bool dynamicInput = false)
{
if ((byte)source.ReadByte() == '{')
{
while (true)
{
if (dynamicInput)
{
Console.WriteLine("Enter a command: ");
string s = Console.ReadLine();
long pos = source.Position;
source.Write(Encoding.UTF8.GetBytes(s));
source.Position = pos;
int g = 0;
}
// parse the statement or list of statements;
// This is done by reading the next word
SkipWhitespace(source);
long position = FindNextWord(source, out string word);
object statementType;
if (Enum.TryParse(typeof(statements), word, out statementType))
{
source.Position = position;
switch ((statements)statementType)
{
case statements.exit:
Exit();
break;
case statements.append:
Append(source);
break;
case statements.list:
List();
break;
case statements.print:
Print(source, 0);
break;
case statements.printlength:
Print(source, 1);
break;
case statements.printwordcount:
Print(source, 2);
break;
case statements.printwords:
Print(source, 3);
break;
case statements.set:
Set(source);
break;
case statements.reverse:
Reverse(source);
break;
case statements.h:
Console.WriteLine("Commands are: ");
foreach (var item in Enum.GetValues(typeof(statements)))
{
Console.WriteLine("\t{0}", ((statements)item).ToString());
}
break;
}
}
else Console.WriteLine("Failed parsing statement");
}
}
}
#region Function Handling
/// <summary>
/// Handles the append x y case.
///
/// </summary>
/// <param name="source"></param>
/// <param name="lineStart"></param>
/// <returns></returns>
bool Append(Stream source, long lineStart = -1)
{
if (lineStart == -1)
{
lineStart = GetLineStart(source, "append");
}
string key;
long keyEndPos = FindIdentifier(source, out key);
if (keyEndPos < 0 || !Symbols.ContainsKey(key))
{
WriteDebugLine(lineStart, lineStart + "append ".Length, "could not identify object", source);
return false;
}
else
{
source.Position = keyEndPos;
}
string value;
long valuePos = FindValue(source, out value);
if (valuePos < 0)
{
// Error on finding object
WriteDebugLine(lineStart, keyEndPos, "could not evaluate expression", source);
return false;
}
else
{
source.Position = valuePos;
}
string eol;
FindNextWord(source, out eol);
if (eol.Length == 0 || eol[0] != ';')
{
WriteDebugLine(lineStart, valuePos, "expected a semicolon", source);
return false;
}
if (Symbols[key].Item2.HasFlag(VariableFlags.Reserved))
{
WriteDebugLine(lineStart, keyEndPos - (key.Length + 1), "cannot assign a value to a reserved constant", source);
return false;
}
Symbols[key] = new Tuple<string, VariableFlags>(Symbols[key].Item1 + value, Symbols[key].Item2);
return true;
}
void List()
{
Console.WriteLine("┌" + new string('─', 15) + "┬" + new string('─', 25) + "┬" + new string('─', 9) + "┐");
Console.WriteLine("│{0,-15}│{1,-25}│{2,9}│", "Symbol", "Value", "Flags");
Console.WriteLine("├" + new string('─', 15) + "┼" + new string('─', 25) + "┼" + new string('─', 9) + "┤");
int keyPos = 0;
foreach (var item in Symbols)
{
Console.WriteLine("│{0,-15}│{1,-25}│{2,9}│", item.Key, item.Value.Item1.Replace("\r","\\r").Replace("\n","\\n").Replace("\t", "\\t"), Convert.ToString((byte)item.Value.Item2,2).PadLeft(8,'0'));
if (keyPos == Symbols.Count-1)
{
Console.WriteLine("└" + new string('─', 15) + "┴" + new string('─', 25) + "┴" + new string('─', 9) + "┘");
}
else
{
Console.WriteLine("├" + new string('─', 15) + "┼" + new string('─', 25) + "┼" + new string('─', 9) + "┤");
}
keyPos++;
}
}
void Exit()
{
Environment.Exit(0);
}
bool Print(Stream source, int mode=0)
{
string expression;
long result = FindExpression(source, out expression);
if (result < 0)
{
// Could not print
return false;
}
if (mode == 0)
{
Console.WriteLine(expression);
}
else if (mode == 1)
{
Console.Write("Length of the expression is: ");
Console.WriteLine(expression.Length);
}
else if (mode >= 2)
{
string[] words = expression.Split(' ');
if (mode == 3)
{
Console.Write("Wordcount is: ");
Console.WriteLine(words.Length);
}
else
{
Console.WriteLine("Words are:");
foreach (string word in words)
{
Console.WriteLine(word);
}
}
}
source.Position = result;
return true;
}
bool Set(Stream source, long lineStart=-1)
{
if(lineStart == -1)
{
lineStart = GetLineStart(source, "set");
}
string identifier;
long identifierEndPos = FindIdentifier(source, out identifier);
if (identifierEndPos <= source.Position + 1 )
{
WriteDebugLine(lineStart, "set ".Length, "???", source);
return false;
}
source.Position = identifierEndPos;
string expression;
long expressionEndPos = FindExpression(source, out expression);
if (expressionEndPos < 0)
{
WriteDebugLine(lineStart, identifierEndPos, "failed parsing expression", source);
// Couldn't match expression
return false;
}
if (Symbols.ContainsKey(identifier))
{
if (Symbols[identifier].Item2.HasFlag(VariableFlags.Reserved))
{
WriteDebugLine(lineStart, identifierEndPos - identifier.Length, "cannot assign to a reserved constant", source);
return false;
}
Symbols[identifier] = new Tuple<string, VariableFlags>(expression, Symbols[identifier].Item2);
}
else
{
Symbols.Add(identifier, new Tuple<string, VariableFlags>(expression, VariableFlags.Empty));
}
source.Position = expressionEndPos;
return true;
}
bool Reverse(Stream source)
{
string identifier;
long resultPos = FindIdentifier(source, out identifier);
if (resultPos < 0)
{
// Couldn't match an identifier
// If ID Doesn't exist, we should make it
return false;
}
string ToReverse = Symbols[identifier].Item1;
string[] words = ToReverse.Split(' ');
StringBuilder reversed = new StringBuilder();
for (int i = words.Length-1; i < 0; i--)
{
reversed.Append(words[i]);
reversed.Append(' ');
}
Symbols[identifier] = new Tuple<string, VariableFlags>(reversed.ToString(), Symbols[identifier].Item2);
return true;
}
/// <summary>
/// Writes the debug info to the screen in the form:<br/>
/// line read from stream (lineStart) to line end<br/>
/// &lt;whitespace@caratPos&gt; ^ &lt;errorMessage&gt;
/// </summary>
/// <param name="lineStart"></param>
/// <param name="caratPos"></param>
/// <param name="errorMessage"></param>
/// <param name="source"></param>
void WriteDebugLine(long lineStart, long caratPos, string errorMessage, Stream source)
{
source.Position = lineStart;
string fullLine = GetNextLine(source);
string errorMSG = new string (' ', (caratPos - lineStart) >= 0 ? (int)(caratPos - lineStart):0) + "^ " + errorMessage;
Console.WriteLine(fullLine);
Console.WriteLine(errorMSG);
source.SetLength(source.Position);
}
/// <summary>
/// Gets the starting point of the expression at expected line.
/// </summary>
/// <param name="source"></param>
/// <param name="word"></param>
/// <returns></returns>
long GetLineStart(Stream source, string word)
{
// Decrement and check the previous value is 32 (space), before incrementing back to our current pos
// Don't need to guard against oob - implied it's within bounds by the set command being at least 3 bytes long
source.Position--;
if (source.ReadByte() == 32)
{
source.Position--;
}
//source.Position++;
return source.Position - word.Length;
}
#endregion
#region Data Handling
// Data Handling
/// <summary>
/// Parses the expression from the point in the string
/// </summary>
/// <param name="s"></param>
/// <param name="expression"></param>
/// <returns></returns>
long FindExpression(Stream s, out string expression)
{
// must contain at least one value
string result;
long wordEnd = FindValue(s, out result);
while (true)
{
string nextWord;
wordEnd = FindNextWord(s, out nextWord);
if (wordEnd > 0 && nextWord == "+")
{
s.Position = wordEnd;
}
else
{
break;
}
s.Position = wordEnd;
wordEnd = FindNextWord(s, out nextWord);
result += nextWord;
}
expression = result;
return wordEnd;
}
// Most atomic unit is 'value':
/// <summary>
/// Finds the next value in the stream
/// </summary>
/// <param name="s"></param>
/// <param name="returnedValue"></param>
/// <returns></returns>
long FindValue(Stream s, out string returnedValue)
{
SkipWhitespace(s);
int result = s.ReadByte();
if (result == '\"')
{
return FindLiteral(s, out returnedValue);
}
else
{
s.Position--;
string keyValue;
long t = FindExistingIdentifier(s, out keyValue);
if (!Symbols.ContainsKey(keyValue))
{
returnedValue = "";
return -1;
}
returnedValue = Symbols[keyValue].Item1;
return t;
}
}
long FindIdentifier(Stream s, out string returnedKey)
{
long wordEnd = FindNextWord(s, out returnedKey);
return wordEnd;
}
long FindExistingIdentifier(Stream s, out string returnedKey)
{
string identifier;
long wordEnd = FindNextWord(s, out identifier);
if (identifier.EndsWith(';') && identifier.Length > 1)
{
// Remove the trailing semicolon from the parse & backtrack the identifier length one spot
identifier = identifier.TrimEnd(';');
wordEnd--;
}
// Lookup the value in the symbol table
returnedKey = identifier;
return wordEnd;
}
long FindLiteral(Stream s, out string returnedLiteral)
{
// Is a literal. Now we must parse until we find the end of the literal
string resultLiteral;
long resultPosition = FindNextOccurance(s, (c, s) =>
{
if (c == '\"')
{
long pos = s.Position--;
if (GetChar(s) == '\\')
{
// TODO: handle the \\ escape
return false;
}
else
{
return true;
}
}
return false;
}, out resultLiteral);
if (resultPosition > -1)
{
returnedLiteral = resultLiteral;
}
else
{
returnedLiteral = "";
}
return resultPosition;
}
#endregion
}
/// <summary>
/// Reads the memory stream as a UTF-8 encoded string until the next occurance of '\n' or '\r\n' (consuming, and excluded)
/// </summary>
/// <param name="s"></param>
/// <returns></returns>
static string GetNextLine(Stream s)
{
string nextLine;
FindNextOccurance(s, '\n', out nextLine);
return nextLine;
}
/// <summary>
/// Finds the next word in the string
/// </summary>
/// <param name="s"></param>
/// <param name="nextWord"></param>
/// <returns>A value <0 if an error occurred, else the position of the end of the word</returns>
static long FindNextWord(string s, out string nextWord)
{
// remove whitespace from the start
int wordStart = 0;
if (char.IsWhiteSpace(s[0]))
{
for (int i = 0; i < s.Length; i++)
{
if (char.IsWhiteSpace(s[i])) break;
wordStart = i;
}
}
int wordEnd = wordStart;
for (int i = wordEnd; i < s.Length; i++)
{
if (char.IsWhiteSpace(s[i])) break;
wordEnd = i;
}
nextWord = s.Substring(wordStart, wordEnd);
return wordEnd;
}
/// <summary>
/// Finds the end-boundary of the next word in the stream
/// </summary>
/// <param name="s"></param>
/// <param name="nextWord"></param>
/// <returns></returns>
static long FindNextWord(Stream s, out string nextWord)
{
// Find the next occurance of a whitespace character
long originalPos = s.Position;
string tempstring;
long nextWSOccurance = FindNextOccurance(s, (c, s) => Char.IsWhiteSpace(c), out tempstring);
while (tempstring.Length == 0)
{
nextWSOccurance = FindNextOccurance(s, (c, s) => Char.IsWhiteSpace(c), out tempstring);
s.Position++;
}
nextWord = tempstring;
s.Position = originalPos;
return nextWSOccurance;
}
/// <summary>
/// Finds and returns the position of the next occurance of the Func returning true.
/// </summary>
/// <param name="s"></param>
/// <param name="p"></param>
/// <param name="result"></param>
/// <returns></returns>
static long FindNextOccurance(Stream s, Func<char, Stream, bool> p, out string result)
{
long start = s.Position;
StringBuilder sb = new StringBuilder();
bool charFound = false;
while (!charFound)
{
char nextChar = GetChar(s);
if (nextChar == 0)
{
charFound = true;
}
else if (p(nextChar, s))
{
charFound = true;
}
else
{
sb.Append(nextChar);
}
}
result = sb.ToString();
long newPosition = s.Position;
s.Position = start;
return newPosition--;
}
/// <summary>
/// Finds the next position of the character
/// </summary>
/// <param name="s"></param>
/// <param name="c"></param>
/// <param name="result">Captures the string read in searching for the character</param>
/// <returns></returns>
static long FindNextOccurance(Stream s, char c, out string result)
{
return FindNextOccurance(s, (streamChar, s) => streamChar == c, out result);
}
static char GetChar(Stream s)
{
// As UTF-8 allows codepoints to span multiple bytes, reading a single byte as a character will not always give the expected
// value.
// Fortunately, the standard ASCII table is 7-bits long. The 8th bit is used to determine the character size
int readAmount = 0;
int firstChar = s.ReadByte();
if (firstChar == -1)
{
return (char)0;
}
if ((firstChar >> 3) == 0x1E) // 11110xxx implies a 4-byte length character
{
readAmount = 3;
}
else if ((firstChar >> 4) == 0xE) // 1110xxxx, 3-byte
{
readAmount = 2;
}
else if ((firstChar >> 5) == 0x6) // 110xxxxx, 2-byte
{
readAmount = 1;
}
byte[] charBytes = new byte[readAmount + 1];
charBytes[0] = (byte)firstChar;
for (int i = 1; i < readAmount; i++)
{
int nextChar = s.ReadByte();
if (nextChar >> 6 != 2) throw new Exception("Character is not a valid UTF-8 code point!");
charBytes[i] = (byte)nextChar;
}
s.Position += readAmount;
string converted = Encoding.UTF8.GetString(charBytes);
return converted[0];
}
static void SkipWhitespace(Stream s)
{
int readByte = s.ReadByte();
while(readByte > -1 && char.IsWhiteSpace((char)readByte))
{
readByte = s.ReadByte();
}
s.Position--;
}
}
}