Introduce cpj_path_join_and_normalize

Author: lygstate

.clang-format

@@ -1,6 +1,6 @@
 ---
 BasedOnStyle: LLVM
-AlignAfterOpenBracket: DontAlign
+AlignAfterOpenBracket: BlockIndent
 BinPackArguments: true
 BinPackParameters: true
 BreakBeforeBraces: Mozilla
@@ -16,3 +16,4 @@ AllowShortLoopsOnASingleLine: false
 PenaltyBreakAssignment: 10000
 PenaltyBreakBeforeFirstCallParameter: 100000
 IndentExternBlock: NoIndent
+ColumnLimit: 120

CMakeLists.txt

@@ -175,6 +175,7 @@ if(ENABLE_TESTS)
   create_test(DEFAULT normalize mixed)
   create_test(DEFAULT normalize overlap)
   create_test(DEFAULT normalize empty)
+  create_test(DEFAULT normalize zero_length)
   create_test(DEFAULT normalize only_separators)
   create_test(DEFAULT normalize back_after_root)
   create_test(DEFAULT normalize forward_slashes)

include/cpj.h

@@ -5,6 +5,7 @@
 
 #include <stdbool.h>
 #include <stddef.h>
+#include <stdint.h>
 
 #if defined(_WIN32) || defined(__CYGWIN__)
 #define CPJ_EXPORT __declspec(dllexport)
@@ -32,7 +33,17 @@ extern "C"
 {
 #endif
 
-typedef size_t cpj_size_t;
+typedef uint8_t cpj_char_t;
+typedef uint32_t cpj_size_t;
+
+/**
+ * Description of a JerryScript string for arguments passing
+ */
+typedef struct
+{
+  const cpj_char_t *ptr; /**< pointer to the zero-terminated ASCII/UTF-8/CESU-8 string */
+  cpj_size_t size;       /**< size of the string, excluding '\0' terminator */
+} cpj_string_t;
 
 /**
  * A segment represents a single component of a path. For instance, on linux a
@@ -41,10 +52,10 @@ typedef size_t cpj_size_t;
  */
 struct cpj_segment
 {
-  const char *path;
-  const char *segments;
-  const char *begin;
-  const char *end;
+  const cpj_char_t *path;
+  const cpj_char_t *segments;
+  const cpj_char_t *begin;
+  const cpj_char_t *end;
   cpj_size_t size;
 };
 
@@ -67,11 +78,11 @@ enum cpj_segment_type
  * @brief Determines the style which is used for the path parsing and
  * generation.
  */
-enum cpj_path_style
+typedef enum
 {
   CPJ_STYLE_WINDOWS,
-  CPJ_STYLE_UNIX
-};
+  CPJ_STYLE_UNIX,
+} cpj_path_style_t;
 
 /**
  * @brief Generates an absolute path based on a base.
@@ -93,8 +104,10 @@ enum cpj_path_style
  * @param buffer_size The size of the result buffer.
  * @return Returns the total amount of characters of the new absolute path.
  */
-CPJ_PUBLIC cpj_size_t cpj_path_get_absolute(enum cpj_path_style path_style,
-  const char *base, const char *path, char *buffer, cpj_size_t buffer_size);
+CPJ_PUBLIC cpj_size_t cpj_path_get_absolute(
+  cpj_path_style_t path_style, const cpj_char_t *base, const cpj_char_t *path, cpj_char_t *buffer,
+  cpj_size_t buffer_size
+);
 
 /**
  * @brief Generates a relative path based on a base.
@@ -116,9 +129,35 @@ CPJ_PUBLIC cpj_size_t cpj_path_get_absolute(enum cpj_path_style path_style,
  * @param buffer_size The size of the result buffer.
  * @return Returns the total amount of characters of the full path.
  */
-CPJ_PUBLIC cpj_size_t cpj_path_get_relative(enum cpj_path_style path_style,
-  const char *base_directory, const char *path, char *buffer,
-  cpj_size_t buffer_size);
+CPJ_PUBLIC cpj_size_t cpj_path_get_relative(
+  cpj_path_style_t path_style, const cpj_char_t *base_directory, const cpj_char_t *path, cpj_char_t *buffer,
+  cpj_size_t buffer_size
+);
+
+/**
+ * @brief Joins two paths together.
+ *
+ * This function generates a new path by combining the two submitted paths. It
+ * will remove double separators, and unlike cpj_path_get_absolute it permits
+ * the use of two relative paths to combine. The result will be written to a
+ * buffer, which might be truncated if the buffer is not large enough to hold
+ * the full path. However, the truncated result will always be
+ * null-terminated. The returned value is the amount of characters which the
+ * resulting path would take if it was not truncated (excluding the
+ * null-terminating character).
+ *
+ * @param path_style Style depending on the operating system. So this should
+ * detect whether we should use windows or unix paths.
+ * @param path_a The first path which comes first.
+ * @param path_b The second path which comes after the first.
+ * @param buffer The buffer where the result will be written to.
+ * @param buffer_size The size of the result buffer.
+ * @return Returns the total amount of characters of the full, combined path.
+ */
+CPJ_PUBLIC cpj_size_t cpj_path_join(
+  cpj_path_style_t path_style, const cpj_char_t *path_a, const cpj_char_t *path_b, cpj_char_t *buffer,
+  cpj_size_t buffer_size
+);
 
 /**
  * @brief Joins two paths together.
@@ -140,8 +179,10 @@ CPJ_PUBLIC cpj_size_t cpj_path_get_relative(enum cpj_path_style path_style,
  * @param buffer_size The size of the result buffer.
  * @return Returns the total amount of characters of the full, combined path.
  */
-CPJ_PUBLIC cpj_size_t cpj_path_join(enum cpj_path_style path_style,
-  const char *path_a, const char *path_b, char *buffer, cpj_size_t buffer_size);
+CPJ_PUBLIC cpj_size_t cpj_path_join_module(
+  cpj_path_style_t path_style, const cpj_char_t *path_a, const cpj_char_t *path_b, cpj_char_t *buffer,
+  cpj_size_t buffer_size
+);
 
 /**
  * @brief Joins multiple paths together.
@@ -163,8 +204,9 @@ CPJ_PUBLIC cpj_size_t cpj_path_join(enum cpj_path_style path_style,
  * @param buffer_size The size of the result buffer.
  * @return Returns the total amount of characters of the full, combined path.
  */
-CPJ_PUBLIC cpj_size_t cpj_path_join_multiple(enum cpj_path_style path_style,
-  const char **paths, char *buffer, cpj_size_t buffer_size);
+CPJ_PUBLIC cpj_size_t cpj_path_join_multiple(
+  cpj_path_style_t path_style, const cpj_char_t **paths, cpj_char_t *buffer, cpj_size_t buffer_size
+);
 
 /**
  * @brief Determines the root of a path.
@@ -178,8 +220,7 @@ CPJ_PUBLIC cpj_size_t cpj_path_join_multiple(enum cpj_path_style path_style,
  * @param path The path which will be inspected.
  * @param length The output of the root length.
  */
-CPJ_PUBLIC void cpj_path_get_root(enum cpj_path_style path_style,
-  const char *path, cpj_size_t *length);
+CPJ_PUBLIC void cpj_path_get_root(cpj_path_style_t path_style, const cpj_char_t *path, cpj_size_t *length);
 
 /**
  * @brief Changes the root of a path.
@@ -200,8 +241,10 @@ CPJ_PUBLIC void cpj_path_get_root(enum cpj_path_style path_style,
  * written to.
  * @return Returns the total amount of characters of the new path.
  */
-CPJ_PUBLIC cpj_size_t cpj_path_change_root(enum cpj_path_style path_style,
-  const char *path, const char *new_root, char *buffer, cpj_size_t buffer_size);
+CPJ_PUBLIC cpj_size_t cpj_path_change_root(
+  cpj_path_style_t path_style, const cpj_char_t *path, const cpj_char_t *new_root, cpj_char_t *buffer,
+  cpj_size_t buffer_size
+);
 
 /**
  * @brief Determine whether the path is absolute or not.
@@ -214,8 +257,7 @@ CPJ_PUBLIC cpj_size_t cpj_path_change_root(enum cpj_path_style path_style,
  * @param path The path which will be checked.
  * @return Returns true if the path is absolute or false otherwise.
  */
-CPJ_PUBLIC bool cpj_path_is_absolute(enum cpj_path_style path_style,
-  const char *path);
+CPJ_PUBLIC bool cpj_path_is_absolute(cpj_path_style_t path_style, const cpj_char_t *path);
 
 /**
  * @brief Determine whether the path is relative or not.
@@ -228,8 +270,7 @@ CPJ_PUBLIC bool cpj_path_is_absolute(enum cpj_path_style path_style,
  * @param path The path which will be checked.
  * @return Returns true if the path is relative or false otherwise.
  */
-CPJ_PUBLIC bool cpj_path_is_relative(enum cpj_path_style path_style,
-  const char *path);
+CPJ_PUBLIC bool cpj_path_is_relative(cpj_path_style_t path_style, const cpj_char_t *path);
 
 /**
  * @brief Gets the basename of a file path.
@@ -248,8 +289,9 @@ CPJ_PUBLIC bool cpj_path_is_relative(enum cpj_path_style path_style,
  * @param length The output of the length of the basename. This may be
  * null if not required.
  */
-CPJ_PUBLIC void cpj_path_get_basename(enum cpj_path_style path_style,
-  const char *path, const char **basename, cpj_size_t *length);
+CPJ_PUBLIC void cpj_path_get_basename(
+  cpj_path_style_t path_style, const cpj_char_t *path, const cpj_char_t **basename, cpj_size_t *length
+);
 
 /**
  * @brief Changes the basename of a file path.
@@ -271,8 +313,10 @@ CPJ_PUBLIC void cpj_path_get_basename(enum cpj_path_style path_style,
  * @return Returns the size which the complete new path would have if it was
  * not truncated.
  */
-CPJ_PUBLIC cpj_size_t cpj_path_change_basename(enum cpj_path_style path_style,
-  const char *path, const char *new_basename, char *buffer, cpj_size_t buffer_size);
+CPJ_PUBLIC cpj_size_t cpj_path_change_basename(
+  cpj_path_style_t path_style, const cpj_char_t *path, const cpj_char_t *new_basename, cpj_char_t *buffer,
+  cpj_size_t buffer_size
+);
 
 /**
  * @brief Gets the dirname of a file path.
@@ -287,8 +331,7 @@ CPJ_PUBLIC cpj_size_t cpj_path_change_basename(enum cpj_path_style path_style,
  * @param path The path which will be inspected.
  * @param length The length of the dirname.
  */
-CPJ_PUBLIC void cpj_path_get_dirname(enum cpj_path_style path_style,
-  const char *path, cpj_size_t *length);
+CPJ_PUBLIC void cpj_path_get_dirname(cpj_path_style_t path_style, const cpj_char_t *path, cpj_size_t *length);
 
 /**
  * @brief Gets the extension of a file path.
@@ -307,8 +350,9 @@ CPJ_PUBLIC void cpj_path_get_dirname(enum cpj_path_style path_style,
  * @param length The output of the length of the extension.
  * @return Returns true if an extension is found or false otherwise.
  */
-CPJ_PUBLIC bool cpj_path_get_extension(enum cpj_path_style path_style,
-  const char *path, const char **extension, cpj_size_t *length);
+CPJ_PUBLIC bool cpj_path_get_extension(
+  cpj_path_style_t path_style, const cpj_char_t *path, const cpj_char_t **extension, cpj_size_t *length
+);
 
 /**
  * @brief Determines whether the file path has an extension.
@@ -321,8 +365,7 @@ CPJ_PUBLIC bool cpj_path_get_extension(enum cpj_path_style path_style,
  * @param path The path which will be inspected.
  * @return Returns true if the path has an extension or false otherwise.
  */
-CPJ_PUBLIC bool cpj_path_has_extension(enum cpj_path_style path_style,
-  const char *path);
+CPJ_PUBLIC bool cpj_path_has_extension(cpj_path_style_t path_style, const cpj_char_t *path);
 
 /**
  * @brief Changes the extension of a file path.
@@ -347,9 +390,10 @@ CPJ_PUBLIC bool cpj_path_has_extension(enum cpj_path_style path_style,
  * @return Returns the total size which the output would have if it was not
  * truncated.
  */
-CPJ_PUBLIC cpj_size_t cpj_path_change_extension(enum cpj_path_style path_style,
-  const char *path, const char *new_extension, char *buffer,
-  cpj_size_t buffer_size);
+CPJ_PUBLIC cpj_size_t cpj_path_change_extension(
+  cpj_path_style_t path_style, const cpj_char_t *path, const cpj_char_t *new_extension, cpj_char_t *buffer,
+  cpj_size_t buffer_size
+);
 
 /**
  * @brief Creates a normalized version of the path.
@@ -376,8 +420,8 @@ CPJ_PUBLIC cpj_size_t cpj_path_change_extension(enum cpj_path_style path_style,
  * @return The size which the complete normalized path has if it was not
  * truncated.
  */
-CPJ_PUBLIC cpj_size_t cpj_path_normalize(enum cpj_path_style path_style,
-  const char *path, char *buffer, cpj_size_t buffer_size);
+CPJ_PUBLIC cpj_size_t
+cpj_path_normalize(cpj_path_style_t path_style, const cpj_char_t *path, cpj_char_t *buffer, cpj_size_t buffer_size);
 
 /**
  * @brief Finds common portions in two paths.
@@ -392,8 +436,8 @@ CPJ_PUBLIC cpj_size_t cpj_path_normalize(enum cpj_path_style path_style,
  * @param path_other The other path which will compared with the base path.
  * @return Returns the number of characters which are common in the base path.
  */
-CPJ_PUBLIC cpj_size_t cpj_path_get_intersection(enum cpj_path_style path_style,
-  const char *path_base, const char *path_other);
+CPJ_PUBLIC cpj_size_t
+cpj_path_get_intersection(cpj_path_style_t path_style, const cpj_char_t *path_base, const cpj_char_t *path_other);
 
 /**
  * @brief Gets the first segment of a path.
@@ -408,8 +452,8 @@ CPJ_PUBLIC cpj_size_t cpj_path_get_intersection(enum cpj_path_style path_style,
  * @param segment The segment which will be extracted.
  * @return Returns true if there is a segment or false if there is none.
  */
-CPJ_PUBLIC bool cpj_path_get_first_segment(enum cpj_path_style path_style,
-  const char *path, struct cpj_segment *segment);
+CPJ_PUBLIC bool
+cpj_path_get_first_segment(cpj_path_style_t path_style, const cpj_char_t *path, struct cpj_segment *segment);
 
 /**
  * @brief Gets the last segment of the path.
@@ -426,8 +470,8 @@ CPJ_PUBLIC bool cpj_path_get_first_segment(enum cpj_path_style path_style,
  * @param segment The segment which will be extracted.
  * @return Returns true if there is a segment or false if there is none.
  */
-CPJ_PUBLIC bool cpj_path_get_last_segment(enum cpj_path_style path_style,
-  const char *path, struct cpj_segment *segment);
+CPJ_PUBLIC bool
+cpj_path_get_last_segment(cpj_path_style_t path_style, const cpj_char_t *path, struct cpj_segment *segment);
 
 /**
  * @brief Advances to the next segment.
@@ -441,8 +485,7 @@ CPJ_PUBLIC bool cpj_path_get_last_segment(enum cpj_path_style path_style,
  * @param segment The current segment which will be advanced to the next one.
  * @return Returns true if another segment was found or false otherwise.
  */
-CPJ_PUBLIC bool cpj_path_get_next_segment(enum cpj_path_style path_style,
-  struct cpj_segment *segment);
+CPJ_PUBLIC bool cpj_path_get_next_segment(cpj_path_style_t path_style, struct cpj_segment *segment);
 
 /**
  * @brief Moves to the previous segment.
@@ -457,8 +500,7 @@ CPJ_PUBLIC bool cpj_path_get_next_segment(enum cpj_path_style path_style,
  * @return Returns true if there is a segment before this one or false
  * otherwise.
  */
-CPJ_PUBLIC bool cpj_path_get_previous_segment(enum cpj_path_style path_style,
-  struct cpj_segment *segment);
+CPJ_PUBLIC bool cpj_path_get_previous_segment(cpj_path_style_t path_style, struct cpj_segment *segment);
 
 /**
  * @brief Gets the type of the submitted path segment.
@@ -471,8 +513,7 @@ CPJ_PUBLIC bool cpj_path_get_previous_segment(enum cpj_path_style path_style,
  * @param segment The segment which will be inspected.
  * @return Returns the type of the segment.
  */
-CPJ_PUBLIC enum cpj_segment_type cpj_path_get_segment_type(
-  const struct cpj_segment *segment);
+CPJ_PUBLIC enum cpj_segment_type cpj_path_get_segment_type(const struct cpj_segment *segment);
 
 /**
  * @brief Changes the content of a segment.
@@ -493,9 +534,10 @@ CPJ_PUBLIC enum cpj_segment_type cpj_path_get_segment_type(
  * @return Returns the total size which would have been written if the output
  * was not truncated.
  */
-CPJ_PUBLIC cpj_size_t cpj_path_change_segment(enum cpj_path_style path_style,
-  struct cpj_segment *segment, const char *value, char *buffer,
-  cpj_size_t buffer_size);
+CPJ_PUBLIC cpj_size_t cpj_path_change_segment(
+  cpj_path_style_t path_style, struct cpj_segment *segment, const cpj_char_t *value, cpj_char_t *buffer,
+  cpj_size_t buffer_size
+);
 
 /**
  * @brief Checks whether the submitted pointer points to a separator.
@@ -510,8 +552,7 @@ CPJ_PUBLIC cpj_size_t cpj_path_change_segment(enum cpj_path_style path_style,
  * @param str A pointer to a string.
  * @return Returns true if it is a separator, or false otherwise.
  */
-CPJ_PUBLIC bool cpj_path_is_separator(enum cpj_path_style path_style,
-  const char *str);
+CPJ_PUBLIC bool cpj_path_is_separator(cpj_path_style_t path_style, const cpj_char_t str);
 
 /**
  * @brief Guesses the path style.
@@ -523,7 +564,7 @@ CPJ_PUBLIC bool cpj_path_is_separator(enum cpj_path_style path_style,
  * @param path The path which will be inspected.
  * @return Returns the style which is most likely used for the path.
  */
-CPJ_PUBLIC enum cpj_path_style cpj_path_guess_style(const char *path);
+CPJ_PUBLIC cpj_path_style_t cpj_path_guess_style(const cpj_char_t *path);
 
 #ifdef __cplusplus
 } // extern "C"