fix<double_quotes>: fix and add double quote and variable assignment

This commit should add the double quote handling and variable assignment handling in lexer

README.md

@@ -13,9 +13,10 @@
 ## Installation
 
 ```sh
-git clone <path>
-cd <path>
-make
+git clone [email protected]:ErwannLesech/42-Sh.git
+cd 42-Sh
+./42-install.sh
+cd src/
 ```
 
 ## Usage

src/Makefile.am

@@ -11,7 +11,7 @@ bin_PROGRAMS = 42sh
 
 42sh_CPPFLAGS = -I%D%
 
-42sh_CFLAGS = -std=c99 -Werror -Wall -Wextra -Wvla -pedantic -fsanitize=address -g
+42sh_CFLAGS = -std=c99 -Werror -Wall -Wextra -Wvla -pedantic
 
 42sh_LDADD =  lexer/liblexer.a \
        ast/libast.a \

src/lexer/lexer.c

@@ -23,9 +23,7 @@ struct lex_match lex_match[] = {
     { "&&", TOKEN_AND },     { "||", TOKEN_OR },     { "|", TOKEN_PIPE },
     { "!", TOKEN_NEGATE },   { "<", TOKEN_REDIR },   { ">", TOKEN_REDIR },
     { ">>", TOKEN_REDIR },   { "<&", TOKEN_REDIR },  { ">&", TOKEN_REDIR },
-    { ">|", TOKEN_REDIR },   { "<>", TOKEN_REDIR },
-
-    { "$*", TOKEN_VARIABLE }
+    { ">|", TOKEN_REDIR },   { "<>", TOKEN_REDIR }
 };
 
 struct lexer *lexer_new(const char *input)
@@ -105,16 +103,37 @@ char *get_word(struct lexer *lexer, bool *is_diactivated)
                && lexer->data[lexer->index] != '|'
                && lexer->data[lexer->index] != '&')
         {
+            if (lexer->data[lexer->index] == '=' && word_index > 0 && lexer->curr_tok.type != TOKEN_DOUBLE_QUOTE)
+            {
+                printf("word1: %s\n", word);
+                lexer->curr_tok.type = TOKEN_WORD_ASSIGNMENT;
+                lexer->index += 1;
+                break;
+            }
             word = realloc(word, sizeof(char) * (word_index + 1));
             word[word_index] = lexer->data[lexer->index];
             ++word_index;
             ++lexer->index;
-            if (lexer->data[lexer->index - 1] == '\\')
+            if (lexer->data[lexer->index - 1] == '\"'
+                || lexer->curr_tok.type == TOKEN_DOUBLE_QUOTE)
             {
-                if (!handle_backslash(lexer, is_diactivated, word, word_index))
+                if (lexer->data[lexer->index - 1] == '\"')
+                {
+                    word_index -= 1;
+                    lexer->curr_tok.type = TOKEN_DOUBLE_QUOTE;
+                }
+                word = handle_double_quote(lexer, is_diactivated, word,
+                                           &word_index);
+                if (!word)
                 {
-                    return word;
+                    return NULL;
                 }
+                word[word_index] = '\0';
+                return word;
+            }
+            else if (lexer->data[lexer->index - 1] == '\\')
+            {
+                handle_backslash(lexer, is_diactivated, word, word_index);
             }
             else if (lexer->data[lexer->index - 1] == '\'')
             {
@@ -161,6 +180,15 @@ struct token parse_input_for_tok(struct lexer *lexer)
         return token;
     }
 
+    if (lexer->curr_tok.type == TOKEN_WORD_ASSIGNMENT)
+    {
+        token.type = TOKEN_WORD_ASSIGNMENT;
+        token.data = word;
+        lexer->curr_tok.type = TOKEN_EOL;
+        printf("word: %s\n", word);
+        return token;
+    }
+
     for (unsigned i = 0; i < sizeof(lex_match) / sizeof(*lex_match); ++i)
     {
         if (fnmatch(lex_match[i].str, word, 0) == 0 && !is_diactivated)
@@ -170,7 +198,7 @@ struct token parse_input_for_tok(struct lexer *lexer)
             return token;
         }
     }
-
+    
     token.type = TOKEN_WORD;
     token.data = word;
     return token;
@@ -201,7 +229,7 @@ struct token lexer_pop(struct lexer *lexer)
         return token;
     }
     struct token token = parse_input_for_tok(lexer);
-    if (token.type != TOKEN_EOF)
+    if (token.type == TOKEN_EOF)
     {
         lexer->curr_tok = token;
     }

src/lexer/lexer.h

@@ -80,7 +80,7 @@ struct token lexer_pop(struct lexer *lexer);
  *
  * \return false if it's the end of the string, true otherwise.
  */
-bool handle_backslash(struct lexer *lexer, bool *is_diactivated, char *word,
+void handle_backslash(struct lexer *lexer, bool *is_diactivated, char *word,
                       unsigned word_index);
 
 /**
@@ -91,6 +91,14 @@ bool handle_backslash(struct lexer *lexer, bool *is_diactivated, char *word,
 char *handle_simple_quote(struct lexer *lexer, bool *is_diactivated, char *word,
                           unsigned *word_index);
 
+/**
+ * \brief Handle the double quote character.
+ * \param lexer The lexer.
+ * \return The next word.
+ */
+char *handle_double_quote(struct lexer *lexer, bool *is_diactivated, char *word,
+                          unsigned *word_index);
+
 /**
  * \brief Handle the comment character.
  *

src/lexer/lexer_utils.c

@@ -1,8 +1,10 @@
+#include <fnmatch.h>
 #include <stdlib.h>
+#include <stdio.h>
 
 #include "lexer.h"
 
-bool handle_backslash(struct lexer *lexer, bool *is_diactivated, char *word,
+void handle_backslash(struct lexer *lexer, bool *is_diactivated, char *word,
                       unsigned word_index)
 {
     *is_diactivated = true;
@@ -14,10 +16,7 @@ bool handle_backslash(struct lexer *lexer, bool *is_diactivated, char *word,
     else
     {
         word[word_index - 1] = '\0';
-        return false;
     }
-
-    return true;
 }
 
 char *handle_simple_quote(struct lexer *lexer, bool *is_diactivated, char *word,
@@ -42,6 +41,134 @@ char *handle_simple_quote(struct lexer *lexer, bool *is_diactivated, char *word,
     return word;
 }
 
+/**
+ * \brief Check if the given word is a variable name.
+ * \param lexer The lexer.
+ * \param word The word to check.
+ */
+char *check_variable_name(struct lexer *lexer, char *word, unsigned *word_index)
+{
+    if (lexer->data[lexer->index] == '?' || lexer->data[lexer->index] == '*'
+        || lexer->data[lexer->index] == '@' || lexer->data[lexer->index] == '#'
+        || lexer->data[lexer->index] == '$')
+    {
+        word = realloc(word, sizeof(char) * (*word_index + 1));
+        word[*word_index] = lexer->data[lexer->index];
+        *word_index += 1;
+        lexer->index += 1;
+        return word;
+    }
+    else if (lexer->data[lexer->index] >= '0'
+             && lexer->data[lexer->index] <= '9')
+    {
+        while (lexer->data[lexer->index] >= '0'
+               && lexer->data[lexer->index] <= '9')
+        {
+            word = realloc(word, sizeof(char) * (*word_index + 1));
+            word[*word_index] = lexer->data[lexer->index];
+            *word_index += 1;
+            lexer->index += 1;
+        }
+        return word;
+    }
+    else if (lexer->data[lexer->index] == '_'
+             || lexer->data[lexer->index] == '-'
+             || (lexer->data[lexer->index] >= 'a'
+                 && lexer->data[lexer->index] <= 'z')
+             || (lexer->data[lexer->index] >= 'A'
+                 && lexer->data[lexer->index] <= 'Z'))
+    {
+        word = realloc(word, sizeof(char) * (*word_index + 1));
+        word[*word_index] = lexer->data[lexer->index];
+        *word_index += 1;
+        lexer->index += 1;
+    }
+    else
+    {
+        return NULL;
+    }
+
+    while (lexer->data[lexer->index] != ' ' && lexer->data[lexer->index] != '\t'
+           && lexer->data[lexer->index] != '\n'
+           && lexer->data[lexer->index] != '\0')
+    {
+        if (lexer->data[lexer->index] == '_' || lexer->data[lexer->index] == '-'
+            || (lexer->data[lexer->index] >= 'a'
+                && lexer->data[lexer->index] <= 'z')
+            || (lexer->data[lexer->index] >= 'A'
+                && lexer->data[lexer->index] <= 'Z')
+            || (lexer->data[lexer->index] >= '0'
+                && lexer->data[lexer->index] <= '9'))
+        {
+            word = realloc(word, sizeof(char) * (*word_index + 1));
+            word[*word_index] = lexer->data[lexer->index];
+            *word_index += 1;
+            lexer->index += 1;
+        }
+        else
+        {
+            return word;
+        }
+    }
+
+    return word;
+}
+
+char *handle_double_quote(struct lexer *lexer, bool *is_diactivated, char *word,
+                          unsigned *word_index)
+{
+    *is_diactivated = true;
+    if (lexer->data[lexer->index] == '$')
+    {
+        word = realloc(word, sizeof(char) * (*word_index + 1));
+        word[*word_index] = lexer->data[lexer->index];
+        *word_index += 1;
+        lexer->index += 1;
+        char *word_tmp = check_variable_name(lexer, word, word_index);
+        if (word_tmp != NULL)
+        {
+            word = word_tmp;
+        }
+        else
+        {
+            return word;
+        }
+    }
+    while (lexer->data[lexer->index] != '\"' && lexer->data[lexer->index] != '$')
+    {
+        if (lexer->data[lexer->index] == '\0')
+        {
+            free(word);
+            word = NULL;
+            return NULL;
+        }
+        else if (lexer->data[lexer->index] == '\\')
+        {
+            lexer->index += 1;
+            if (lexer->data[lexer->index] == '\"'
+                || lexer->data[lexer->index] == '$'
+                || lexer->data[lexer->index] == '\\'
+                || lexer->data[lexer->index] == '\n')
+            {
+                word = realloc(word, sizeof(char) * (*word_index + 1));
+                word[*word_index] = lexer->data[lexer->index];
+                *word_index += 1;
+                lexer->index += 1;
+            }
+        }
+        word = realloc(word, sizeof(char) * (*word_index + 1));
+        word[*word_index] = lexer->data[lexer->index];
+        *word_index += 1;
+        lexer->index += 1;
+    }
+    if (lexer->data[lexer->index] == '\"')
+    {
+        lexer->curr_tok.type = TOKEN_EOL;
+        ++lexer->index;
+    }
+    return word;
+}
+
 char *handle_comment(struct lexer *lexer, char *word, unsigned *word_index)
 {
     // Skip the comment
@@ -90,4 +217,4 @@ char *handle_redir(struct lexer *lexer, unsigned *word_index)
         ++lexer->index;
     }
     return redir;
-}
+}