yin parser CHANGE validate argument value according to its type
helpful functions and macros were moved from yang_parser and are used also in yin parser.
diff --git a/src/parser_yang.c b/src/parser_yang.c
index 42eb69d..5489051 100644
--- a/src/parser_yang.c
+++ b/src/parser_yang.c
@@ -31,18 +31,6 @@
#include "tree_schema.h"
#include "tree_schema_internal.h"
-/* Macro to check YANG's yang-char grammar rule */
-#define is_yangutf8char(c) ((c >= 0x20 && c <= 0xd77) || c == 0x09 || c == 0x0a || c == 0x0d || \
- (c >= 0xe000 && c <= 0xfdcf) || (c >= 0xfdf0 && c <= 0xfffd) || \
- (c >= 0x10000 && c <= 0x1fffd) || (c >= 0x20000 && c <= 0x2fffd) || \
- (c >= 0x30000 && c <= 0x3fffd) || (c >= 0x40000 && c <= 0x2fffd) || \
- (c >= 0x50000 && c <= 0x5fffd) || (c >= 0x60000 && c <= 0x6fffd) || \
- (c >= 0x70000 && c <= 0x7fffd) || (c >= 0x80000 && c <= 0x8fffd) || \
- (c >= 0x90000 && c <= 0x9fffd) || (c >= 0xa0000 && c <= 0xafffd) || \
- (c >= 0xb0000 && c <= 0xbfffd) || (c >= 0xc0000 && c <= 0xcfffd) || \
- (c >= 0xd0000 && c <= 0xdfffd) || (c >= 0xe0000 && c <= 0xefffd) || \
- (c >= 0xf0000 && c <= 0xffffd) || (c >= 0x100000 && c <= 0x10fffd))
-
/**
* @brief Try to find object with MEMBER string matching the IDENT in the given ARRAY.
* Macro logs an error message and returns LY_EVALID in case of existence of a matching object.
@@ -169,62 +157,6 @@
}
/**
- * @brief Check that \p c is valid UTF8 code point for YANG string.
- *
- * @param[in] ctx yang parser context for logging.
- * @param[in] c UTF8 code point of a character to check.
- * @return LY_ERR values.
- */
-static LY_ERR
-check_stringchar(struct lys_parser_ctx *ctx, unsigned int c)
-{
- if (!is_yangutf8char(c)) {
- LOGVAL_PARSER(ctx, LY_VCODE_INCHAR, c);
- return LY_EVALID;
- }
- return LY_SUCCESS;
-}
-
-/**
- * @brief Check that \p c is valid UTF8 code point for YANG identifier.
- *
- * @param[in] ctx yang parser context for logging.
- * @param[in] c UTF8 code point of a character to check.
- * @param[in] first Flag to check the first character of an identifier, which is more restricted.
- * @param[in,out] prefix Storage for internally used flag in case of possible prefixed identifiers:
- * 0 - colon not yet found (no prefix)
- * 1 - \p c is the colon character
- * 2 - prefix already processed, now processing the identifier
- *
- * If the identifier cannot be prefixed, NULL is expected.
- * @return LY_ERR values.
- */
-LY_ERR
-check_identifierchar(struct lys_parser_ctx *ctx, unsigned int c, int first, int *prefix)
-{
- if (first || (prefix && (*prefix) == 1)) {
- if (!is_yangidentstartchar(c)) {
- LOGVAL_PARSER(ctx, LYVE_SYNTAX_YANG, "Invalid identifier first character '%c'.", c);
- return LY_EVALID;
- }
- if (prefix) {
- if (first) {
- (*prefix) = 0;
- } else {
- (*prefix) = 2;
- }
- }
- } else if (c == ':' && prefix && (*prefix) == 0) {
- (*prefix) = 1;
- } else if (!is_yangidentchar(c)) {
- LOGVAL_PARSER(ctx, LYVE_SYNTAX_YANG, "Invalid identifier character '%c'.", c);
- return LY_EVALID;
- }
-
- return LY_SUCCESS;
-}
-
-/**
* @brief Store a single UTF8 character. It depends whether in a dynamically-allocated buffer or just as a pointer to the data.
*
* @param[in] ctx yang parser context for logging.
@@ -265,14 +197,14 @@
/* check character validity */
switch (arg) {
case Y_IDENTIF_ARG:
- LY_CHECK_RET(check_identifierchar(ctx, c, !(*word_len), NULL));
+ LY_CHECK_RET(lysp_check_identifierchar(ctx, c, !(*word_len), NULL));
break;
case Y_PREF_IDENTIF_ARG:
- LY_CHECK_RET(check_identifierchar(ctx, c, !(*word_len), &prefix));
+ LY_CHECK_RET(lysp_check_identifierchar(ctx, c, !(*word_len), &prefix));
break;
case Y_STR_ARG:
case Y_MAYBE_STR_ARG:
- LY_CHECK_RET(check_stringchar(ctx, c));
+ LY_CHECK_RET(lysp_check_stringchar(ctx, c));
break;
}
@@ -809,7 +741,7 @@
LOGVAL_PARSER(ctx, LY_VCODE_INCHAR, (*data)[-len]), LY_EVALID);
++ctx->indent;
/* check character validity */
- LY_CHECK_RET(check_identifierchar(ctx, c, *data - len == word_start ? 1 : 0, &prefix));
+ LY_CHECK_RET(lysp_check_identifierchar(ctx, c, *data - len == word_start ? 1 : 0, &prefix));
}
if (!**data) {
LOGVAL_PARSER(ctx, LY_VCODE_EOF);
diff --git a/src/parser_yin.c b/src/parser_yin.c
index b1de36f..dfbba27 100644
--- a/src/parser_yin.c
+++ b/src/parser_yin.c
@@ -193,13 +193,44 @@
return ret;
}
+LY_ERR
+yin_validate_value(struct yin_parser_ctx *ctx, enum yang_arg val_type, char *val, size_t len)
+{
+ int prefix = 0;
+ unsigned int c;
+ size_t utf8_char_len;
+ size_t already_read = 0;
+ while (already_read < len) {
+ LY_CHECK_ERR_RET(ly_getutf8((const char **)&val, &c, &utf8_char_len),
+ LOGVAL_PARSER((struct lys_parser_ctx *)ctx, LY_VCODE_INCHAR, (val)[-utf8_char_len]), LY_EVALID);
+ already_read += utf8_char_len;
+ LY_CHECK_ERR_RET(already_read > len, LOGINT(ctx->xml_ctx.ctx), LY_EINT);
+
+ switch (val_type) {
+ case Y_IDENTIF_ARG:
+ LY_CHECK_RET(lysp_check_identifierchar((struct lys_parser_ctx *)ctx, c, !already_read, NULL));
+ break;
+ case Y_PREF_IDENTIF_ARG:
+ LY_CHECK_RET(lysp_check_identifierchar((struct lys_parser_ctx *)ctx, c, !already_read, &prefix));
+ break;
+ case Y_STR_ARG:
+ case Y_MAYBE_STR_ARG:
+ LY_CHECK_RET(lysp_check_stringchar((struct lys_parser_ctx *)ctx, c));
+ break;
+ }
+ }
+
+ return LY_SUCCESS;
+}
+
/**
* @brief Parse yin argument.
*
* @param[in,out] ctx Yin parser context for logging and to store current state.
* @param[in] attrs ([Sized array](@ref sizedarrays)) of attributes.
* @param[in,out] data Data to read from.
- * @param[in] arg_type Type of argument that is expected in parsed element (use YIN_ARG_NONE for elements without special argument).
+ * @param[in] arg_type Type of argument that is expected in parsed element (use YIN_ARG_NONE for elements without
+ * special argument).
* @param[out] arg_val Where value of argument should be stored. Can be NULL if arg_type is specified as YIN_ARG_NONE.
* @param[in] val_type Type of expected value of attribute.
* @param[in] current_element Identification of current element, used for logging.
@@ -225,6 +256,7 @@
LY_CHECK_ERR_RET(found, LOGVAL_PARSER((struct lys_parser_ctx *)ctx, LYVE_SYNTAX_YIN, "Duplicit definition of %s attribute in %s element",
yin_attr2str(arg), ly_stmt2str(current_element)), LY_EVALID);
found = true;
+ LY_CHECK_RET(yin_validate_value(ctx, val_type, iter->content, iter->content_len));
if (iter->dynamic_content) {
*arg_val = lydict_insert_zc(ctx->xml_ctx.ctx, iter->content);
LY_CHECK_RET(!(*arg_val), LY_EMEM);
@@ -247,8 +279,6 @@
return LY_EVALID;
}
- /* TODO check validity according to val_type */
-
return LY_SUCCESS;
}
diff --git a/src/parser_yin.h b/src/parser_yin.h
index b4ef46b..d4e22bf 100644
--- a/src/parser_yin.h
+++ b/src/parser_yin.h
@@ -71,7 +71,7 @@
struct yin_subelement {
enum yang_keyword type; /**< type of keyword */
- void *dest; /**< meta infromation passed to responsible function (information about where parsed subelement should be stored) */
+ void *dest; /**< meta infromation passed to responsible function (mostly information about where parsed subelement should be stored) */
uint8_t flags; /**< describes cardianlity of subelement can be set to YIN_SUBELEM_MANDATORY and YIN_SUBELEM_UNIQUE and YIN_SUBELEM_FIRST */
};
@@ -131,6 +131,19 @@
struct lysp_ext_instance **exts);
/**
+ * @brief Check that val is valid UTF8 character sequence of val_type.
+ * Doesn't check empty string, only character validity.
+ *
+ * @param[in] ctx Yin parser context for logging.
+ * @param[in] val_type Type of the input string to select method of checking character validity.
+ * @param[in] val Input to validate.
+ * @param[in] len Length of input.
+ *
+ * @return LY_ERR values.
+ */
+LY_ERR yin_validate_value(struct yin_parser_ctx *ctx, enum yang_arg val_type, char *val, size_t len);
+
+/**
* @brief Parse import element.
*
* @param[in,out] ctx Yin parser context for logging and to store current state.
diff --git a/src/tree_schema_helpers.c b/src/tree_schema_helpers.c
index 8cfdbc7..c367f96 100644
--- a/src/tree_schema_helpers.c
+++ b/src/tree_schema_helpers.c
@@ -796,6 +796,41 @@
}
LY_ERR
+lysp_check_stringchar(struct lys_parser_ctx *ctx, unsigned int c)
+{
+ if (!is_yangutf8char(c)) {
+ LOGVAL_PARSER(ctx, LY_VCODE_INCHAR, c);
+ return LY_EVALID;
+ }
+ return LY_SUCCESS;
+}
+
+LY_ERR
+lysp_check_identifierchar(struct lys_parser_ctx *ctx, unsigned int c, int first, int *prefix)
+{
+ if (first || (prefix && (*prefix) == 1)) {
+ if (!is_yangidentstartchar(c)) {
+ LOGVAL_PARSER(ctx, LYVE_SYNTAX_YANG, "Invalid identifier first character '%c'.", c);
+ return LY_EVALID;
+ }
+ if (prefix) {
+ if (first) {
+ (*prefix) = 0;
+ } else {
+ (*prefix) = 2;
+ }
+ }
+ } else if (c == ':' && prefix && (*prefix) == 0) {
+ (*prefix) = 1;
+ } else if (!is_yangidentchar(c)) {
+ LOGVAL_PARSER(ctx, LYVE_SYNTAX_YANG, "Invalid identifier character '%c'.", c);
+ return LY_EVALID;
+ }
+
+ return LY_SUCCESS;
+}
+
+LY_ERR
lysp_load_submodule(struct lys_parser_ctx *ctx, struct lysp_module *mod, struct lysp_include *inc)
{
struct lysp_submodule *submod = NULL;
diff --git a/src/tree_schema_internal.h b/src/tree_schema_internal.h
index 5720e65..813dede 100644
--- a/src/tree_schema_internal.h
+++ b/src/tree_schema_internal.h
@@ -27,6 +27,18 @@
#define is_yangidentchar(c) ((c >= 'a' && c <= 'z') || (c >= 'A' && c <= 'Z') || (c >= '0' && c <= '9') || \
c == '_' || c == '-' || c == '.')
+/* Macro to check YANG's yang-char grammar rule */
+#define is_yangutf8char(c) ((c >= 0x20 && c <= 0xd77) || c == 0x09 || c == 0x0a || c == 0x0d || \
+ (c >= 0xe000 && c <= 0xfdcf) || (c >= 0xfdf0 && c <= 0xfffd) || \
+ (c >= 0x10000 && c <= 0x1fffd) || (c >= 0x20000 && c <= 0x2fffd) || \
+ (c >= 0x30000 && c <= 0x3fffd) || (c >= 0x40000 && c <= 0x2fffd) || \
+ (c >= 0x50000 && c <= 0x5fffd) || (c >= 0x60000 && c <= 0x6fffd) || \
+ (c >= 0x70000 && c <= 0x7fffd) || (c >= 0x80000 && c <= 0x8fffd) || \
+ (c >= 0x90000 && c <= 0x9fffd) || (c >= 0xa0000 && c <= 0xafffd) || \
+ (c >= 0xb0000 && c <= 0xbfffd) || (c >= 0xc0000 && c <= 0xcfffd) || \
+ (c >= 0xd0000 && c <= 0xdfffd) || (c >= 0xe0000 && c <= 0xefffd) || \
+ (c >= 0xf0000 && c <= 0xffffd) || (c >= 0x100000 && c <= 0x10fffd))
+
/**
* @brief List of YANG statement groups - the (sub)module's substatements
*/
@@ -80,6 +92,31 @@
};
/**
+ * @brief Check that \p c is valid UTF8 code point for YANG string.
+ *
+ * @param[in] ctx yang parser context for logging.
+ * @param[in] c UTF8 code point of a character to check.
+ * @return LY_ERR values.
+ */
+LY_ERR lysp_check_stringchar(struct lys_parser_ctx *ctx, unsigned int c);
+
+/**
+ * @brief Check that \p c is valid UTF8 code point for YANG identifier.
+ *
+ * @param[in] ctx yang parser context for logging.
+ * @param[in] c UTF8 code point of a character to check.
+ * @param[in] first Flag to check the first character of an identifier, which is more restricted.
+ * @param[in,out] prefix Storage for internally used flag in case of possible prefixed identifiers:
+ * 0 - colon not yet found (no prefix)
+ * 1 - \p c is the colon character
+ * 2 - prefix already processed, now processing the identifier
+ *
+ * If the identifier cannot be prefixed, NULL is expected.
+ * @return LY_ERR values.
+ */
+LY_ERR lysp_check_identifierchar(struct lys_parser_ctx *ctx, unsigned int c, int first, int *prefix);
+
+/**
* @brief Check the currently present prefixes in the module for collision with the new one.
*
* @param[in] ctx Context for logging.