Merge pull request #81 from ErwannLesech/dev

Step2.2 - Add tests and last fixes

src/lexer/Makefile.am

@@ -1,5 +1,5 @@
 lib_LIBRARIES = liblexer.a
 
-liblexer_a_SOURCES = lexer.c lexer_utils.c lexer.h token.h
+liblexer_a_SOURCES = lexer.c lexer_utils.c lexer_utils2.c lexer.h token.h
 liblexer_a_CFLAGS = -Wall -Wextra -Werror -Wvla -pedantic -std=c99
 liblexer_a_CPPFLAGS = -I$(top_srcdir)/src

src/lexer/lexer.c

@@ -149,7 +149,8 @@ char *get_word(struct lexer *lexer, bool *is_diactivated)
             // Handle the variable
             if (lexer->data[lexer->index] == '$')
             {
-                if (word_index != 0)
+                if (word_index != 0
+                    && check_variable_name_simulated(lexer->data, lexer->index))
                 {
                     break;
                 }
@@ -181,7 +182,6 @@ char *get_word(struct lexer *lexer, bool *is_diactivated)
             {
                 lexer->index += 1;
             }
-
             // Take next char and put it in the word
             word[word_index] = lexer->data[lexer->index];
             ++word_index;
@@ -198,6 +198,16 @@ char *get_word(struct lexer *lexer, bool *is_diactivated)
                     word_index -= 1;
                     lexer->curr_tok.type = TOKEN_DOUBLE_QUOTE;
                 }
+                else if (word_index > 0
+                         && lexer->data[lexer->index - 1] == '\\')
+                {
+                    word_index -= 1;
+                    handle_back_slash_in_double_quote(lexer, word, &word_index);
+                    if (lexer->data[lexer->index] == '$')
+                    {
+                        break;
+                    }
+                }
 
                 // Handle the double quote
                 word = handle_double_quote(lexer, is_diactivated, word,
@@ -207,8 +217,10 @@ char *get_word(struct lexer *lexer, bool *is_diactivated)
                 {
                     return NULL;
                 }
-                word[word_index] = '\0';
-                return word;
+                if (lexer->curr_tok.type == TOKEN_VARIABLE_AND_DOUBLE_QUOTE)
+                {
+                    return word;
+                }
             }
 
             // Handle backslash

src/lexer/lexer.h

@@ -55,6 +55,12 @@ void lexer_free(struct lexer *lexer);
  */
 void token_free(struct token token);
 
+/**
+ * \brief Print the given token.
+ * \param token The token to print.
+ */
+void print_token(struct token token);
+
 /**
  * \brief Get the next token from the lexer.
  * \param lexer The lexer.
@@ -121,6 +127,25 @@ bool check_variable_assignement(char *word);
 bool check_variable_name(struct lexer *lexer, char **word, unsigned *word_index,
                          bool *is_in_braces);
 
+/**
+ * \brief Check if the current word is a valid variable name (simulated mode)
+ * meaning that pointers to the lexer and the word won't be modified.
+ * \param data The word to check
+ * \param index The index of the data to check
+ * \return true if it's a valid variable name, false otherwise
+ */
+bool check_variable_name_simulated(const char *data, int index);
+
+/**
+ * \brief Handle the backslash character in a double quote.
+ * \param lexer The lexer.
+ * \param word The word.
+ * \param word_index The index of the word.
+ */
+
+void handle_back_slash_in_double_quote(struct lexer *lexer, char *word,
+                                       unsigned *word_index);
+
 /**
  * \brief Handle the dollar character.
  * \param lexer The lexer.

src/lexer/lexer_utils.c

@@ -98,7 +98,8 @@ bool check_variable_name(struct lexer *lexer, char **word, unsigned *word_index,
     char *curr_word = *word;
     *is_in_braces = false;
     // Handle variable in double quote
-    if (lexer->curr_tok.type == TOKEN_DOUBLE_QUOTE)
+    if (lexer->curr_tok.type == TOKEN_DOUBLE_QUOTE
+        || lexer->curr_tok.type == TOKEN_VARIABLE_AND_DOUBLE_QUOTE)
     {
         lexer->curr_tok.type = TOKEN_VARIABLE_AND_DOUBLE_QUOTE;
     }
@@ -149,7 +150,7 @@ bool check_variable_name(struct lexer *lexer, char **word, unsigned *word_index,
     // Not a valid variable name
     else
     {
-        if (lexer->curr_tok.type != TOKEN_DOUBLE_QUOTE)
+        if (lexer->curr_tok.type != TOKEN_VARIABLE_AND_DOUBLE_QUOTE)
         {
             lexer->curr_tok.type = TOKEN_WORD;
         }
@@ -204,6 +205,7 @@ bool handle_dollar(struct lexer *lexer, char **word, unsigned *word_index,
     // Check if the name of the variable is correct
     return check_variable_name(lexer, word, word_index, is_in_braces);
 }
+
 char *handle_double_quote(struct lexer *lexer, bool *is_diactivated, char *word,
                           unsigned *word_index)
 {
@@ -241,27 +243,7 @@ char *handle_double_quote(struct lexer *lexer, bool *is_diactivated, char *word,
         {
             lexer->index += 1;
             word = realloc(word, sizeof(char) * (*word_index + 1));
-            if (lexer->data[lexer->index] == '\"'
-                || lexer->data[lexer->index] == '$'
-                || lexer->data[lexer->index] == '\\'
-                || lexer->data[lexer->index] == '\n'
-                || lexer->data[lexer->index] == '`')
-            {
-                if (lexer->data[lexer->index] != '\n')
-                {
-                    word[*word_index] = lexer->data[lexer->index];
-                }
-                else
-                {
-                    *word_index -= 1;
-                }
-                lexer->index += 1;
-            }
-            else
-            {
-                word[*word_index] = '\\';
-            }
-            *word_index += 1;
+            handle_back_slash_in_double_quote(lexer, word, word_index);
         }
         else
         {

src/lexer/lexer_utils2.c

@@ -0,0 +1,123 @@
+#include "lexer.h"
+
+void print_token(struct token token)
+{
+    char *tokens[] = { // Step1
+                       [TOKEN_IF] = "TOKEN_IF",
+                       [TOKEN_THEN] = "TOKEN_THEN",
+                       [TOKEN_ELIF] = "TOKEN_ELIF",
+                       [TOKEN_ELSE] = "TOKEN_ELSE",
+                       [TOKEN_FI] = "TOKEN_FI",
+                       [TOKEN_SEMICOLON] = "TOKEN_SEMICOLON",
+                       [TOKEN_WORD] = "TOKEN_WORD",
+                       [TOKEN_EOL] = "TOKEN_EOL",
+                       [TOKEN_EOF] = "TOKEN_EOF",
+                       [TOKEN_ERROR] = "TOKEN_ERROR",
+                       // Step 2
+                       [TOKEN_DONE] = "TOKEN_DONE",
+                       [TOKEN_AND] = "TOKEN_AND",
+                       [TOKEN_OR] = "TOKEN_OR",
+                       [TOKEN_PIPE] = "TOKEN_PIPE",
+                       [TOKEN_NEGATE] = "TOKEN_NEGATE",
+                       [TOKEN_IONUMBER] = "TOKEN_IONUMBER",
+                       [TOKEN_REDIR] = "TOKEN_REDIR",
+                       [TOKEN_DOUBLE_QUOTE] = "TOKEN_DOUBLE_QUOTE",
+                       [TOKEN_WORD_DOUBLE_QUOTE] = "TOKEN_WORD_DOUBLE_QUOTE",
+                       [TOKEN_WORD_ASSIGNMENT] = "TOKEN_WORD_ASSIGNMENT",
+                       [TOKEN_VARIABLE] = "TOKEN_VARIABLE",
+
+                       // Internal values for lexer
+                       [TOKEN_VARIABLE_VALUE] = "TOKEN_VARIABLE_VALUE",
+                       [TOKEN_VARIABLE_AND_DOUBLE_QUOTE] =
+                           "TOKEN_VARIABLE_AND_DOUBLE_QUOTE"
+    };
+    printf("Token: %s\n", tokens[token.type]);
+}
+
+bool check_variable_name_simulated(const char *data, int index)
+{
+    bool is_in_braces = false;
+
+    index += 1;
+
+    // Check if it's a special variable (like $?, $*, $@, $# or $$)
+    if (data[index] == '?' || data[index] == '*' || data[index] == '@'
+        || data[index] == '#' || data[index] == '$')
+    {
+        return true;
+    }
+
+    // Chech if it's a special variable (like $n)
+    else if (data[index] >= '0' && data[index] <= '9')
+    {
+        return true;
+    }
+
+    else if (data[index] == '{')
+    {
+        index += 1;
+        is_in_braces = true;
+    }
+
+    // Classic variable name
+    else if (data[index] == '_' || data[index] == '-'
+             || (data[index] >= 'a' && data[index] <= 'z')
+             || (data[index] >= 'A' && data[index] <= 'Z'))
+    {
+        index += 1;
+    }
+    // Not a valid variable name
+    else
+    {
+        return false;
+    }
+
+    // Check the rest of the variable name break
+    while (data[index] == '_' || data[index] == '-'
+           || (data[index] >= 'a' && data[index] <= 'z')
+           || (data[index] >= 'A' && data[index] <= 'Z')
+           || (data[index] >= '0' && data[index] <= '9'))
+    {
+        index += 1;
+    }
+
+    if (is_in_braces)
+    {
+        if (data[index] == '}')
+        {
+            index += 1;
+            return true;
+        }
+        else
+        {
+            return false;
+        }
+    }
+
+    return true;
+}
+
+void handle_back_slash_in_double_quote(struct lexer *lexer, char *word,
+                                       unsigned *word_index)
+{
+    if (lexer->data[lexer->index] == '\"' || lexer->data[lexer->index] == '$'
+        || lexer->data[lexer->index] == '\\'
+        || lexer->data[lexer->index] == '\n'
+        || lexer->data[lexer->index] == '`')
+    {
+        if (lexer->data[lexer->index] != '\n')
+        {
+            word[*word_index] = lexer->data[lexer->index];
+        }
+        else
+        {
+            *word_index -= 1;
+        }
+        lexer->index += 1;
+    }
+    else
+    {
+        word[*word_index] = '\\';
+    }
+    *word_index += 1;
+}

src/lexer/tests/lexer2_tests.c

@@ -997,12 +997,7 @@ Test(lexer2, variable_distinction_access_deactivate_brace)
 
     tok = lexer_pop(lexer);
     cr_assert_eq(tok.type, TOKEN_WORD, "got %d", tok.type);
-    cr_assert_str_eq(tok.data, "variable:");
-    token_free(tok);
-
-    tok = lexer_pop(lexer);
-    cr_assert_eq(tok.type, TOKEN_WORD);
-    cr_assert_str_eq(tok.data, "${a_123as}");
+    cr_assert_str_eq(tok.data, "variable:${a_123as}");
     token_free(tok);
 
     tok = lexer_pop(lexer);

src/main.c

@@ -39,6 +39,13 @@ int main(int argc, char **argv)
     {
         errx(1, "Error while creating lexer");
     }
+    /*    while (lexer_peek(lexer).type != TOKEN_EOF)
+    {
+        struct token token = lexer_pop(lexer);
+        print_token(token);
+        print_token(lexer->curr_tok);
+        printf("value: %s\n", token.data);
+    }*/
 
     int val = parser_loop(lexer, pretty_print_enabled);
     if (val == 2)

tests/quote/double_quote/double_quote10.test

@@ -0,0 +1 @@
+a=1;echo "$a\"$a"

tests/quote/double_quote/double_quote11.test

@@ -0,0 +1 @@
+echo "Special characters: !@#%$^&*()"

tests/quote/double_quote/double_quote12.test

@@ -0,0 +1 @@
+bb=1; echo "$ddd\""

tests/quote/double_quote/double_quote5.test

@@ -0,0 +1 @@
+a=1;echo "$a\""

tests/quote/double_quote/double_quote6.test

@@ -0,0 +1 @@
+a=1;echo "idiej$a:jodejoj" "$a" "dikefok"'

tests/quote/double_quote/double_quote7.test

@@ -0,0 +1 @@
+a=1;echo "lpikopo"pojo^é$

tests/quote/double_quote/double_quote8.test

@@ -0,0 +1,4 @@
+a=1;echo "lpikopo\
+
+
+"

tests/quote/double_quote/double_quote9.test

@@ -0,0 +1 @@
+a=1;echo "lpikopo\$$\$$\\$\$\*\\*\\\*\*\\***\ "

tests/quote/double_quote/testsuite.sh

@@ -1,4 +1,12 @@
 run_test double_quote1.test
 run_test double_quote2.test
 run_test double_quote3.test
-run_test double_quote4.test
+run_test double_quote4.test
+run_test double_quote5.test
+run_test double_quote6.test
+run_test double_quote7.test
+run_test double_quote8.test
+run_test double_quote9.test
+run_test double_quote10.test
+run_test double_quote11.test
+run_test double_quote12.test

tests/quote/double_quote_error/double_quote_error1.test

@@ -0,0 +1 @@
+bb=1; echo "$bb

tests/quote/double_quote_error/double_quote_error2.test

@@ -0,0 +1 @@
+bb=1; echo "$ddd\"

tests/quote/double_quote_error/double_quote_error3.test

@@ -0,0 +1 @@
+tata=1; echo "${tata}

tests/quote/double_quote_error/double_quote_error4.test

@@ -0,0 +1 @@
+tata=1; echo "${tata}\"

tests/quote/double_quote_error/double_quote_error5.test

@@ -0,0 +1 @@
+tata=1; echo "${tata"

tests/quote/double_quote_error/double_quote_error6.test

@@ -0,0 +1 @@
+tata=1; echo "$

tests/quote/double_quote_error/double_quote_error7.test

@@ -0,0 +1 @@
+tata=1; echo "$lff

tests/quote/double_quote_error/double_quote_error8.test

@@ -0,0 +1 @@
+tata=1; echo \"${tata}"

tests/quote/double_quote_error/testsuite.sh

@@ -0,0 +1,8 @@
+run_test double_quote_error1.test
+run_test double_quote_error2.test
+run_test double_quote_error3.test
+run_test double_quote_error4.test
+run_test double_quote_error5.test
+run_test double_quote_error6.test
+run_test double_quote_error7.test
+run_test double_quote_error8.test

tests/quote/single_quote_error/testsuite.sh

@@ -1,5 +1,5 @@
 run_test single_quote_error1.test
 run_test single_quote_error2.test
 run_test single_quote_error3.test
-run_test single_quote_errorMouli1.test
-run_test single_quote_errorMouli2.test
+run_test single_quoteMouli1.test
+run_test single_quoteMouli2.test

tests/redirections/redir_&input/complex1.test

@@ -0,0 +1,5 @@
+echo 'hello' > input.txt
+ls > output.txt 2<&1 < input.txt
+cat output.txt
+cat input.txt
+rm input.txt output.txt