Merge pull request #47 from ErwannLesech/lexer_part2

Lexer part2

.github/workflows/clang_tidy.sh

@@ -24,10 +24,12 @@ for file in $(find "$root_dir/src" -type f -name '*.c'); do
 
         if [[ "$parameters" -gt 4 ]]; then
             echo "Too many parameters in function: $function_name"
+            echo "Parameters: $parameters"
         fi
 
         if [[ "$lines_in_function" -gt 40 ]]; then
             echo "Too many lines in function: $function_name"
+            echo "Lines in function: $lines_in_function"
         fi
 
         # echo "Function: $function_name"
@@ -40,5 +42,6 @@ for file in $(find "$root_dir/src" -type f -name '*.c'); do
     # echo "Total functions: $function_count"
     if [[ "$function_count" -gt 10 ]]; then
         echo "Too many functions in file: $file"
+        echo "Total functions: $function_count"
     fi
 done

.gitignore

@@ -25,6 +25,9 @@ config.status
 
 *.diff
 
+*.gv
+*.png
+
 # Doxygen
 Doxyfile.bak
 html

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/ast/ast.h

@@ -9,11 +9,10 @@
 #ifndef AST_H
 #define AST_H
 
+#include <stdbool.h>
 #include <stdio.h>
 #include <stdlib.h>
 
-#include "../options/options.h"
-
 /**
  * \enum ast_type
  * \brief Enumerate the different types of AST nodes.
@@ -67,6 +66,13 @@ void ast_free(struct ast_node *node);
  */
 struct ast_node *ast_node_word(char *value);
 
+/**
+ * \brief Convert an AST type to a string.
+ * \param type The type to convert.
+ * \return The AST type string.
+ */
+char *ast_type_to_string(enum ast_type type);
+
 /**
  * \brief Create a new AST node of type AST_SIMPLE_COMMAND.
  * \param value The value of the node.

src/lexer/Makefile.am

@@ -1,5 +1,5 @@
 lib_LIBRARIES = liblexer.a
 
-liblexer_a_SOURCES = lexer.c lexer.h token.h
+liblexer_a_SOURCES = lexer.c lexer_utils.c lexer.h token.h
 liblexer_a_CFLAGS = -Wall -Wextra -Werror -std=c99 -pedantic
 liblexer_a_CPPFLAGS = -I$(top_srcdir)

src/lexer/lexer.c

@@ -9,15 +9,21 @@
 #include "lexer.h"
 
 #include <err.h>
+#include <fnmatch.h>
 #include <stdbool.h>
 #include <stdio.h>
 #include <stdlib.h>
 #include <string.h>
 
-struct lex_match lex_match[8] = {
-    { "if", TOKEN_IF },     { "then", TOKEN_THEN }, { "elif", TOKEN_ELIF },
-    { "else", TOKEN_ELSE }, { "fi", TOKEN_FI },     { ";", TOKEN_SEMICOLON },
-    { "\n", TOKEN_EOL },    { "\0", TOKEN_EOF }
+struct lex_match lex_match[] = {
+    { "if", TOKEN_IF },      { "then", TOKEN_THEN }, { "elif", TOKEN_ELIF },
+    { "else", TOKEN_ELSE },  { "fi", TOKEN_FI },     { ";", TOKEN_SEMICOLON },
+    { "\n", TOKEN_EOL },     { "\0", TOKEN_EOF },
+
+    { "&&", TOKEN_AND },     { "||", TOKEN_OR },     { "|", TOKEN_PIPE },
+    { "!", TOKEN_NEGATE },   { "<", TOKEN_REDIR },   { ">", TOKEN_REDIR },
+    { ">>", TOKEN_REDIR },   { "<&", TOKEN_REDIR },  { ">&", TOKEN_REDIR },
+    { ">|", TOKEN_REDIR },   { "<>", TOKEN_REDIR }
 };
 
 struct lexer *lexer_new(const char *input)
@@ -45,140 +51,107 @@ void token_free(struct token token)
     free(token.data);
 }
 
-/**
- * \brief Handle the backslash character.
- *
- * \return false if it's the end of the string, true otherwise.
- */
-bool handle_backslash(struct lexer *lexer, bool *is_diactivated, char *word,
-                      unsigned word_index)
+char *get_word(struct lexer *lexer, bool *is_diactivated)
 {
-    *is_diactivated = true;
-    if (lexer->data[lexer->index] != '\0')
-    {
-        word[word_index - 1] = lexer->data[lexer->index];
-        ++lexer->index;
-    }
-    else
-    {
-        word[word_index - 1] = '\0';
-        return false;
-    }
-
-    return true;
-}
+    char *word = malloc(sizeof(char) * 2);
+    unsigned word_index = 0;
 
-/**
- * \brief Handle the simple quote character.
- *
- * \return false if a closing quote was not found, true otherwise.
- */
-char *handle_simple_quote(struct lexer *lexer, bool *is_diactivated, char *word,
-                          unsigned *word_index)
-{
-    *is_diactivated = true;
-    *word_index -= 1;
-    while (lexer->data[lexer->index] != '\'')
+    if (lexer->data[lexer->index] == ';' || lexer->data[lexer->index] == '\n')
     {
-        if (lexer->data[lexer->index] == '\0')
+        word[0] = lexer->data[lexer->index];
+        word_index = 1;
+        ++lexer->index;
+        if (lexer->data[lexer->index] == ' ')
         {
-            free(word);
-            word = NULL;
-            return NULL;
+            ++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;
-}
-
-char *handle_comment(struct lexer *lexer, char *word, unsigned word_index)
-{
-    // Skip the comment
-    ++lexer->index;
-
-    // Find the end of the comment
-    while (lexer->data[lexer->index] != '\n'
-           && lexer->data[lexer->index] != '\0')
-    {
-        ++lexer->index;
-    }
-    word[word_index] = lexer->data[lexer->index];
-    ++lexer->index;
-    // If the comment isn't the last thing in the string, we need to add a '\0'
-    // at the end of the word.
-    if (word[word_index] != '\0')
-    {
-        word[word_index + 1] = '\0';
     }
-
-    // Skip the spaces after the comment
-    while (lexer->data[lexer->index] == ' ')
+    else if (lexer->data[lexer->index] == '#')
     {
-        ++lexer->index;
+        word = handle_comment(lexer, word, &word_index);
     }
-    return word;
-}
-
-char *get_word(struct lexer *lexer, bool *is_diactivated)
-{
-    char *word = malloc(sizeof(char) * 2);
-    unsigned word_index = 0;
-    if (lexer->data[lexer->index] == '\0')
+    else if (lexer->data[lexer->index] == '>'
+             || lexer->data[lexer->index] == '<')
     {
-        ++lexer->index;
-        word[0] = '\0';
-        return word;
+        word = handle_redir(lexer, &word_index);
     }
-    if (lexer->data[lexer->index] == ';' || lexer->data[lexer->index] == '\n')
+    else if (lexer->data[lexer->index] == '|'
+             || lexer->data[lexer->index] == '&')
     {
         word[0] = lexer->data[lexer->index];
-        word[1] = '\0';
+        word_index = 1;
         ++lexer->index;
-        if (lexer->data[lexer->index] == ' ')
+
+        if (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 = 2;
             ++lexer->index;
         }
-        return word;
     }
-    if (lexer->data[lexer->index] == '#')
-    {
-        return handle_comment(lexer, word, 0);
-    }
-    while (lexer->data[lexer->index] != ' ' && lexer->data[lexer->index] != '\0'
-           && lexer->data[lexer->index] != ';'
-           && lexer->data[lexer->index] != '\n'
-           && lexer->data[lexer->index] != '\t')
+    else
     {
-        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] == '\\')
+        while (lexer->data[lexer->index] != ' '
+               && lexer->data[lexer->index] != '\0'
+               && lexer->data[lexer->index] != ';'
+               && lexer->data[lexer->index] != '\n'
+               && lexer->data[lexer->index] != '\t'
+               && lexer->data[lexer->index] != '>'
+               && lexer->data[lexer->index] != '<'
+               && lexer->data[lexer->index] != '|'
+               && lexer->data[lexer->index] != '&')
         {
-            if (!handle_backslash(lexer, is_diactivated, word, word_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] == '\"'
+                || lexer->curr_tok.type == TOKEN_DOUBLE_QUOTE)
+            {
+                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 NULL;
+                }
+                word[word_index] = '\0';
                 return word;
             }
-        }
-        else if (lexer->data[lexer->index - 1] == '\'')
-        {
-            word =
-                handle_simple_quote(lexer, is_diactivated, word, &word_index);
-            if (!word)
+            else if (lexer->data[lexer->index - 1] == '\\')
+            {
+                handle_backslash(lexer, is_diactivated, word, word_index);
+            }
+            else if (lexer->data[lexer->index - 1] == '\'')
             {
-                return NULL;
+                word = handle_simple_quote(lexer, is_diactivated, word,
+                                           &word_index);
+                if (!word)
+                {
+                    return NULL;
+                }
+                lexer->index += 1;
             }
-            lexer->index += 1;
         }
     }
     word = realloc(word, sizeof(char) * (word_index + 1));
     word[word_index] = '\0';
 
-    while (lexer->data[lexer->index] == ' ')
+    while (lexer->data[lexer->index] == ' '
+           || lexer->data[lexer->index] == '\t')
     {
         ++lexer->index;
     }
@@ -207,16 +180,25 @@ 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 (!strcmp(word, lex_match[i].str) && !is_diactivated)
+        if (fnmatch(lex_match[i].str, word, 0) == 0 && !is_diactivated)
         {
             token.type = lex_match[i].type;
             token.data = word;
             return token;
         }
     }
-
+    
     token.type = TOKEN_WORD;
     token.data = word;
     return token;
@@ -247,6 +229,9 @@ struct token lexer_pop(struct lexer *lexer)
         return token;
     }
     struct token token = parse_input_for_tok(lexer);
-    lexer->curr_tok = token;
+    if (token.type == TOKEN_EOF)
+    {
+        lexer->curr_tok = token;
+    }
     return token;
 }