Add parsing function and test

.gitignore

@@ -1,3 +1,4 @@
 /target
 Cargo.lock
-.idea
+.idea
+test-files

src/lib.rs

@@ -1,13 +1,34 @@
-#[macro_use] extern crate maplit;
+#[macro_use]
-#[macro_use] extern crate lazy_static;
+extern crate maplit;
+#[macro_use]
+extern crate lazy_static;
 pub mod elements;
 pub mod tokenizer;
 mod tokens;
 
 #[cfg(test)]
 mod tests {
+    use crate::tokenizer::Tokenizer;
+    use crate::tokens::{Grouping, Misc, Operation, Relation, Text, Token};
+
     #[test]
     fn it_works() {
-        assert_eq!(2 + 2, 4);
+        let expression = "sum_(i=1)^n";
+        let mut tokenizer = Tokenizer::new(expression.to_string());
+        let tokens = tokenizer.parse();
+        assert_eq!(
+            tokens,
+            vec![
+                Token::Operation(Operation::Sum),
+                Token::Misc(Misc::Sub),
+                Token::Grouping(Grouping::RParen),
+                Token::Text(Text::Plain("i".to_string())),
+                Token::Relation(Relation::Eq),
+                Token::Text(Text::Plain("1".to_string())),
+                Token::Grouping(Grouping::LParen),
+                Token::Misc(Misc::Pow),
+                Token::Text(Text::Plain("n".to_string()))
+            ]
+        );
     }
 }

src/tokenizer/mod.rs

@@ -5,25 +5,74 @@ use crate::tokens::mappings::{
     get_relation_mapping,
 };
 use crate::tokens::{
-    Accent, Arrow, FontCommand, Greek, Grouping, Logical, Misc, Operation, Relation, Token,
+    Accent, Arrow, FontCommand, Greek, Grouping, Logical, Misc, Operation, Relation, Text, Token,
 };
 use charred::tapemachine::CharTapeMachine;
 use std::collections::HashMap;
-use std::fmt::Debug;
 
 pub struct Tokenizer {
     ctm: CharTapeMachine,
-    tokens: Vec<Token>,
 }
 
 impl Tokenizer {
     pub fn new(text: String) -> Self {
+        let mut chars = text.chars().collect::<Vec<char>>();
+        chars.push('\n');
         Self {
-            ctm: CharTapeMachine::new(text.chars().collect()),
+            ctm: CharTapeMachine::new(chars),
-            tokens: Vec::new(),
         }
     }
 
+    pub fn parse(&mut self) -> Vec<Token> {
+        let mut tokens = Vec::<Token>::new();
+        let mut tmp_string = String::new();
+        self.ctm.seek_whitespace();
+
+        while !self.ctm.check_eof() {
+            if let Some(grouping) = self.parse_grouping() {
+                tokens.push(Token::Grouping(grouping))
+            } else if let Some(arrow) = self.parse_arrows() {
+                tokens.push(Token::Arrow(arrow))
+            } else if let Some(relation) = self.parse_relation() {
+                tokens.push(Token::Relation(relation))
+            } else if let Some(operation) = self.parse_operation() {
+                tokens.push(Token::Operation(operation))
+            } else if let Some(misc) = self.parse_misc() {
+                tokens.push(Token::Misc(misc))
+            } else if let Some(logical) = self.parse_logical() {
+                tokens.push(Token::Logical(logical))
+            } else if let Some(accent) = self.parse_accent() {
+                tokens.push(Token::Accent(accent))
+            } else if let Some(greek) = self.parse_greek() {
+                tokens.push(Token::Greek(greek))
+            } else if let Some(font) = self.parse_font_command() {
+                tokens.push(Token::Font(font))
+            } else if let Some(whitespace) = self.parse_whitespace() {
+                tokens.push(Token::Text(whitespace))
+            } else {
+                tmp_string.push(self.ctm.get_current());
+                let _ = self.ctm.seek_one();
+                continue;
+            }
+            if !tmp_string.is_empty() {
+                let last = tokens.pop().unwrap();
+                tokens.push(Token::Text(Text::Plain(tmp_string.clone())));
+                tmp_string.clear();
+                tokens.push(last);
+            }
+            let _ = self.ctm.seek_one();
+        }
+        if !tmp_string.is_empty() {
+            tokens.push(Token::Text(Text::Plain(tmp_string)));
+        }
+        // stripping the whitespace at the end
+        if let Some(Token::Text(Text::Whitespace)) = tokens.last() {
+            tokens.pop().unwrap();
+        }
+
+        tokens
+    }
+
     fn parse_misc(&mut self) -> Option<Misc> {
         lazy_static! {
             static ref MISC_MAPPINGS: Vec<HashMap<TokenPattern, Misc>> = get_misc_mappings();
@@ -153,4 +202,14 @@ impl Tokenizer {
         }
         None
     }
+
+    fn parse_whitespace(&mut self) -> Option<Text> {
+        if self.ctm.get_current().is_whitespace() {
+            self.ctm.seek_whitespace();
+            self.ctm.rewind(self.ctm.get_index() - 1);
+            Some(Text::Whitespace)
+        } else {
+            None
+        }
+    }
 }

src/tokens/constants/misc.rs

@@ -1,6 +1,7 @@
 pub const G_A_FRAC: &'static [&str] = &["/"];
 pub const G_T_FRAC: &'static [&str] = &["frac"];
 
+pub const G_SUB: &'static [&str] = &["_"];
 pub const G_POW: &'static [&str] = &["^"];
 pub const G_SQRT: &'static [&str] = &["sqrt"];
 pub const G_ROOT: &'static [&str] = &["root"];

src/tokens/mappings.rs

@@ -12,7 +12,6 @@ use crate::tokens::constants::TokenPattern;
 use crate::tokens::{
     Accent, Arrow, FontCommand, Greek, Grouping, Logical, Misc, Operation, Relation,
 };
-use std::cell::RefCell;
 use std::collections::HashMap;
 
 pub fn get_operation_mappings() -> Vec<HashMap<TokenPattern, Operation>> {
@@ -64,6 +63,7 @@ pub fn get_misc_mappings() -> Vec<HashMap<TokenPattern, Misc>> {
             G_A_FRAC => Misc::AsciiFrac,
             G_T_FRAC => Misc::LatexFrac,
             G_POW => Misc::Pow,
+            G_SUB => Misc::Sub,
             G_SQRT => Misc::Sqrt,
             G_ROOT => Misc::Root,
             G_INT => Misc::Int,

src/tokens/mod.rs

@@ -1,7 +1,7 @@
 pub mod constants;
 pub mod mappings;
 
-#[derive(Debug, Clone)]
+#[derive(Debug, Clone, PartialOrd, PartialEq)]
 pub enum Token {
     Operation(Operation),
     Misc(Misc),
@@ -12,9 +12,16 @@ pub enum Token {
     Accent(Accent),
     Greek(Greek),
     Font(FontCommand),
+    Text(Text),
 }
 
-#[derive(Debug, Clone)]
+#[derive(Debug, Clone, PartialOrd, PartialEq)]
+pub enum Text {
+    Plain(String),
+    Whitespace,
+}
+
+#[derive(Debug, Clone, PartialOrd, PartialEq)]
 pub enum Operation {
     Plus,
     Minus,
@@ -44,10 +51,11 @@ pub enum Operation {
     BigCup,
 }
 
-#[derive(Debug, Clone)]
+#[derive(Debug, Clone, PartialOrd, PartialEq)]
 pub enum Misc {
     AsciiFrac,
     LatexFrac,
+    Sub,
     Pow,
     Sqrt,
     Root,
@@ -85,7 +93,7 @@ pub enum Misc {
     LatexText,
 }
 
-#[derive(Debug, Clone)]
+#[derive(Debug, Clone, PartialOrd, PartialEq)]
 pub enum Relation {
     Eq,
     Ne,
@@ -109,7 +117,7 @@ pub enum Relation {
     PropTo,
 }
 
-#[derive(Debug, Clone)]
+#[derive(Debug, Clone, PartialOrd, PartialEq)]
 pub enum Logical {
     And,
     Or,
@@ -125,7 +133,7 @@ pub enum Logical {
     Models,
 }
 
-#[derive(Debug, Clone)]
+#[derive(Debug, Clone, PartialOrd, PartialEq)]
 pub enum Grouping {
     RParen,
     LParen,
@@ -143,7 +151,7 @@ pub enum Grouping {
     Norm,
 }
 
-#[derive(Debug, Clone)]
+#[derive(Debug, Clone, PartialOrd, PartialEq)]
 pub enum Arrow {
     UpArrow,
     DownArrow,
@@ -160,7 +168,7 @@ pub enum Arrow {
     BigLeftRightArrow,
 }
 
-#[derive(Debug, Clone)]
+#[derive(Debug, Clone, PartialOrd, PartialEq)]
 pub enum Accent {
     Hat,
     Overline,
@@ -176,7 +184,7 @@ pub enum Accent {
     Cancel,
 }
 
-#[derive(Debug, Clone)]
+#[derive(Debug, Clone, PartialOrd, PartialEq)]
 pub enum Greek {
     Alpha,
     Beta,
@@ -216,7 +224,7 @@ pub enum Greek {
     BigOmega,
 }
 
-#[derive(Debug, Clone)]
+#[derive(Debug, Clone, PartialOrd, PartialEq)]
 pub enum FontCommand {
     Big,
     BigOutline,