From: Kevin Day <kevin@kevux.org>
Date: Mon, 27 Mar 2023 12:35:06 +0000 (-0500)
Subject: Bugfix: The f_parse parsing functions do not check buffer.used.
X-Git-Url: https://git.kevux.org/?a=commitdiff_plain;h=7edc624484027af9392e8a3ed06c0fd263bc7aaf;p=fll

Bugfix: The f_parse parsing functions do not check buffer.used.

The range length is being performed but not the buffer used length.
Add the appropriate return codes to represent end of string (buffer used length is reached).
This is applicable for the dynamic strings but is not applicable for the regular strings.

Do some code clean up while I am at it.
---

diff --git a/level_0/f_parse/c/parse.c b/level_0/f_parse/c/parse.c
index 237efb1..dd6b81c 100644
--- a/level_0/f_parse/c/parse.c
+++ b/level_0/f_parse/c/parse.c
@@ -12,6 +12,7 @@ extern "C" {
 
     if (!buffer.used) return F_data_not;
     if (range->start > range->stop) return F_data_not_stop;
+    if (range->start >= buffer.used) return F_data_not_eos;
 
     const unsigned short seek_width = macro_f_utf_char_t_width(seek_to_this);
 
@@ -20,7 +21,7 @@ extern "C" {
     unsigned short width = 0;
     f_array_length_t width_max = 0;
 
-    while (range->start <= range->stop) {
+    while (range->start <= range->stop && range->start < buffer.used) {
 
       width_max = (range->stop - range->start) + 1;
       width = macro_f_utf_byte_width_is(buffer.string[range->start]);
@@ -37,6 +38,7 @@ extern "C" {
       }
       else {
         if (range->start + width > range->stop) return F_status_set_error(F_complete_not_utf_stop);
+        if (range->start + width >= buffer.used) return F_status_set_error(F_complete_not_utf_eos);
 
         if (width == seek_width) {
           f_utf_char_t character = 0;
@@ -50,6 +52,7 @@ extern "C" {
       range->start += width;
 
       if (range->start >= range->stop) return F_none_stop;
+      if (range->start > buffer.used) return F_none_eos;
     } // while
 
     return F_none_eos;
@@ -64,6 +67,7 @@ extern "C" {
 
     if (!buffer.used) return F_data_not;
     if (range->start > range->stop) return F_data_not_stop;
+    if (range->start >= buffer.used) return F_data_not_eos;
 
     f_status_t status = F_none;
     unsigned short width = 0;
@@ -87,11 +91,13 @@ extern "C" {
       }
       else {
         if (range->start + width > range->stop) return F_status_set_error(F_complete_not_utf_stop);
+        if (range->start + width >= buffer.used) return F_status_set_error(F_complete_not_utf_eos);
       }
 
       range->start += width;
 
       if (range->start > range->stop) return F_none_stop;
+      if (range->start >= buffer.used) return F_none_eos;
 
       width_max = (range->stop - range->start) + 1;
     } // while
@@ -110,6 +116,7 @@ extern "C" {
 
     if (!buffer.used) return F_data_not;
     if (range->start > range->stop) return F_data_not_stop;
+    if (range->start >= buffer.used) return F_data_not_eos;
 
     f_status_t status = F_none;
     unsigned short width = 0;
@@ -133,11 +140,13 @@ extern "C" {
       }
       else {
         if (range->start + width > range->stop) return F_status_set_error(F_complete_not_utf_stop);
+        if (range->start + width >= buffer.used) return F_status_set_error(F_complete_not_utf_eos);
       }
 
       range->start += width;
 
       if (range->start > range->stop) return F_none_stop;
+      if (range->start >= buffer.used) return F_none_eos;
 
       width_max = (range->stop - range->start) + 1;
     } // while
@@ -157,6 +166,7 @@ extern "C" {
     if (!buffer.used) return F_data_not;
 
     if (range->start > range->stop) return F_data_not_stop;
+    if (range->start > buffer.used) return F_data_not_eos;
 
     const unsigned short seek_width = macro_f_utf_char_t_width(seek_to_this);
 
@@ -166,7 +176,7 @@ extern "C" {
 
     f_array_length_t width_max = 0;
 
-    while (range->start <= range->stop) {
+    while (range->start <= range->stop && range->start < buffer.used) {
 
       width_max = (range->stop - range->start) + 1;
       width = macro_f_utf_byte_width_is(buffer.string[range->start]);
@@ -183,9 +193,8 @@ extern "C" {
         return F_status_set_error(F_complete_not_utf);
       }
       else {
-        if (range->start + width > range->stop) {
-          return F_status_set_error(F_complete_not_utf_stop);
-        }
+        if (range->start + width > range->stop) return F_status_set_error(F_complete_not_utf_stop);
+        if (range->start + width >= buffer.used) return F_status_set_error(F_complete_not_utf_eos);
 
         if (width == seek_width) {
           f_utf_char_t character = 0;
@@ -197,10 +206,10 @@ extern "C" {
       }
 
       range->start += width;
-
-      if (range->start >= range->stop) return F_none_stop;
     } // while
 
+    if (range->start >= range->stop) return F_none_stop;
+
     return F_none_eos;
   }
 #endif // _di_f_parse_dynamic_seek_to_utf_character_
@@ -217,7 +226,7 @@ extern "C" {
     }
 
     // Skip past all leading NULLs.
-    for (; range->start <= range->stop; ++range->start) {
+    for (; range->start <= range->stop && range->start < buffer.used; ++range->start) {
       if (buffer.string[range->start]) break;
     } // for
 
@@ -229,7 +238,7 @@ extern "C" {
     if (range->stop - range->start < 5) {
 
       // Increment until stop, while taking into consideration UTF-8 character widths.
-      for (; range->start <= range->stop; ) {
+      for (; range->start <= range->stop && range->start < buffer.used; ) {
 
         if (buffer.string[range->start] == f_string_eol_s.string[0]) {
           ++range->start;
@@ -245,7 +254,7 @@ extern "C" {
 
     f_status_t status = F_none;
 
-    for (; range->start <= range->stop; ) {
+    for (; range->start <= range->stop && range->start < buffer.used; ) {
 
       status = f_utf_is_whitespace(buffer.string + range->start, (range->stop - range->start) + 1, F_false);
 
@@ -274,14 +283,12 @@ extern "C" {
       range->start += macro_f_utf_byte_width(buffer.string[range->start]);
     } // for
 
-    if (range->start > range->stop) {
-      return F_found_not;
-    }
+    if (range->start > range->stop || range->start > buffer.used) return F_found_not;
 
     if (range->stop - range->start < 5) {
 
       // Increment until stop, while taking into consideration UTF-8 character widths.
-      for (; range->start <= range->stop; ) {
+      for (; range->start <= range->stop && range->start < buffer.used; ) {
 
         if (buffer.string[range->start] == f_string_eol_s.string[0]) {
           ++range->start;
@@ -297,7 +304,7 @@ extern "C" {
 
     f_array_length_t i = range->start;
 
-    for (; range->start <= range->stop; ) {
+    for (; range->start <= range->stop && range->start < buffer.used; ) {
 
       status = f_utf_is_word(buffer.string + range->start, (range->stop - range->start) + 1, F_true);
       if (F_status_is_error(status)) return status;
@@ -315,10 +322,10 @@ extern "C" {
       range->start += macro_f_utf_byte_width(buffer.string[range->start]);
     } // for
 
-    if (range->start > range->stop || buffer.string[range->start] != f_string_ascii_minus_s.string[0]) {
+    if (range->start > range->stop || range->start >= buffer.used || buffer.string[range->start] != f_string_ascii_minus_s.string[0]) {
 
       // Increment until stop, while taking into consideration UTF-8 character widths.
-      for (; range->start <= range->stop; ) {
+      for (; range->start <= range->stop && range->start < buffer.used; ) {
 
         if (buffer.string[range->start] == f_string_eol_s.string[0]) {
           ++range->start;
@@ -336,13 +343,13 @@ extern "C" {
       f_array_length_t j = 0;
       f_char_t number[5] = { 0, 0, 0, 0, 0 };
 
-      for (++range->start; range->start <= range->stop && j < 4; ++range->start, ++j) {
+      for (++range->start; range->start <= range->stop && range->start < buffer.used && j < 4; ++range->start, ++j) {
 
         // The hexidecimal representing the number may only be ASCII.
         if (macro_f_utf_byte_width_is(buffer.string[range->start])) {
 
           // Increment until stop, while taking into consideration UTF-8 character widths.
-          for (; range->start <= range->stop; ) {
+          for (; range->start <= range->stop && range->start < buffer.used; ) {
 
             if (buffer.string[range->start] == f_string_eol_s.string[0]) {
               ++range->start;
@@ -378,7 +385,7 @@ extern "C" {
       else {
 
         // Increment until stop, while taking into consideration UTF-8 character widths.
-        for (; range->start <= range->stop; ) {
+        for (; range->start <= range->stop && range->start < buffer.used; ) {
 
           if (buffer.string[range->start] == f_string_eol_s.string[0]) {
             ++range->start;
@@ -398,7 +405,7 @@ extern "C" {
     }
 
     // Skip past all NULLs.
-    for (; range->start <= range->stop; ++range->start) {
+    for (; range->start <= range->stop && range->start < buffer.used; ++range->start) {
       if (buffer.string[range->start]) break;
     } // for
 
@@ -417,7 +424,7 @@ extern "C" {
       if (status == F_false) {
 
         // Increment until stop, while taking into consideration UTF-8 character widths.
-        for (; range->start <= range->stop; ) {
+        for (; range->start <= range->stop && range->start < buffer.used; ) {
 
           if (buffer.string[range->start] == f_string_eol_s.string[0]) {
             ++range->start;
@@ -441,7 +448,7 @@ extern "C" {
     }
 
     if (id) {
-      for (f_array_length_t j = i, i = 0; j <= range->stop; ++j) {
+      for (f_array_length_t j = i, i = 0; j <= range->stop && j < buffer.used; ++j) {
 
         if (!buffer.string[j]) continue;
         if (buffer.string[j] == f_string_ascii_minus_s.string[0]) break;
@@ -466,9 +473,7 @@ extern "C" {
       if (!range) return F_status_set_error(F_parameter);
     #endif // _di_level_0_parameter_checking_
 
-    if (range->start > range->stop) {
-      return F_data_not_stop;
-    }
+    if (range->start > range->stop) return F_data_not_stop;
 
     const unsigned short seek_width = macro_f_utf_char_t_width(seek_to);
 
diff --git a/level_0/f_parse/c/parse.h b/level_0/f_parse/c/parse.h
index 2ecd78a..41fff43 100644
--- a/level_0/f_parse/c/parse.h
+++ b/level_0/f_parse/c/parse.h
@@ -40,12 +40,15 @@ extern "C" {
  * @return
  *   F_none on success.
  *   F_none_eol on success, but stopped at EOL.
+ *   F_none_eos on success, but stopped at end of buffer.
  *   F_none_stop on success, but stopped at end of range.
  *   F_data_not on success, but there was no string data to seek.
+ *   F_data_not_eos on success, but the range.start >= buffer.used.
  *   F_data_not_stop on success, but the range.start > range.stop.
  *
  *   F_complete_not_utf (with error bit) if character is an incomplete UTF-8 fragment.
  *   F_complete_not_utf_stop (with error bit) if the stop location is reached before the complete UTF-8 character can be processed.
+ *   F_complete_not_utf_eos (with error bit) if the end of buffer is reached before the complete UTF-8 character can be processed.
  *   F_parameter (with error bit) if a parameter is invalid.
  *   F_utf_not (with error bit) if character is an invalid UTF-8 character.
  *
@@ -70,12 +73,15 @@ extern "C" {
  * @return
  *   F_none on success.
  *   F_none_eol on success, but stopped at EOL.
+ *   F_none_eos on success, but stopped at end of buffer.
  *   F_none_stop on success, but stopped at end of range.
  *   F_data_not on success, but there was no string data to seek.
+ *   F_data_not_eos on success, but the range.start >= buffer.used.
  *   F_data_not_stop on success, but the range.start > range.stop.
  *
  *   F_complete_not_utf (with error bit) if character is an incomplete UTF-8 fragment.
  *   F_complete_not_utf_stop (with error bit) if the stop location is reached before the complete UTF-8 character can be processed.
+ *   F_complete_not_utf_eos (with error bit) if the end of buffer is reached before the complete UTF-8 character can be processed.
  *   F_memory_not (with error bit) on out of memory.
  *   F_parameter (with error bit) if a parameter is invalid.
  *
@@ -100,12 +106,15 @@ extern "C" {
  * @return
  *   F_none on success.
  *   F_none_eol on success, but stopped at EOL.
+ *   F_none_eos on success, but stopped at end of buffer.
  *   F_none_stop on success, but stopped at end of range.
  *   F_data_not on success, but there was no string data to seek.
+ *   F_data_not_eos on success, but the range.start >= buffer.used.
  *   F_data_not_stop on success, but the range.start > range.stop.
  *
  *   F_complete_not_utf (with error bit) if character is an incomplete UTF-8 fragment.
  *   F_complete_not_utf_stop (with error bit) if the stop location is reached before the complete UTF-8 character can be processed.
+ *   F_complete_not_utf_eos (with error bit) if the end of buffer is reached before the complete UTF-8 character can be processed.
  *   F_memory_not (with error bit) on out of memory.
  *   F_parameter (with error bit) if a parameter is invalid.
  *
@@ -130,12 +139,16 @@ extern "C" {
  *
  * @return
  *   F_none on success.
+ *   F_none_eol on success, but stopped at EOL.
+ *   F_none_eos on success, but stopped at end of buffer.
  *   F_none_stop on success, but stopped at end of range.
  *   F_data_not on success, but there was no string data to seek.
+ *   F_data_not_eos on success, but the range.start >= buffer.used.
  *   F_data_not_stop on success, but the range.start > range.stop.
  *
  *   F_complete_not_utf (with error bit) if character is an incomplete UTF-8 fragment.
  *   F_complete_not_utf_stop (with error bit) if the stop location is reached before the complete UTF-8 character can be processed.
+ *   F_complete_not_utf_eos (with error bit) if the end of buffer is reached before the complete UTF-8 character can be processed.
  *   F_parameter (with error bit) if a parameter is invalid.
  *   F_utf_not (with error bit) if character is an invalid UTF-8 character.
  *
@@ -179,6 +192,8 @@ extern "C" {
  *
  * @return
  *   F_data_not if length is 0.
+ *   F_data_not_eos on success, but the range.start >= buffer.used.
+ *   F_data_not_stop on success, but the range.start > range.stop.
  *   F_found if the buffer does represent a valid FLL Identifier.
  *   F_found_not if the buffer does not represent a valid FLL Identifier.
  *