diff options
Diffstat (limited to 'src/tokenizer.c')
-rw-r--r-- | src/tokenizer.c | 58 |
1 files changed, 56 insertions, 2 deletions
diff --git a/src/tokenizer.c b/src/tokenizer.c index 635d507df..c1b64f959 100644 --- a/src/tokenizer.c +++ b/src/tokenizer.c @@ -90,6 +90,60 @@ tokenizer_next_word(char **input_p, GError **error_r) return word; } +static inline bool +valid_unquoted_char(char ch) +{ + return (unsigned char)ch > 0x20 && ch != '"' && ch != '\''; +} + +char * +tokenizer_next_unquoted(char **input_p, GError **error_r) +{ + char *word, *input; + + assert(input_p != NULL); + assert(*input_p != NULL); + + word = input = *input_p; + + if (*input == 0) + return NULL; + + /* check the first character */ + + if (!valid_unquoted_char(*input)) { + g_set_error(error_r, tokenizer_quark(), 0, + "Invalid unquoted character"); + return NULL; + } + + /* now iterate over the other characters until we find a + whitespace or end-of-string */ + + while (*++input != 0) { + if (g_ascii_isspace(*input)) { + /* a whitespace: the word ends here */ + *input = 0; + /* skip all following spaces, too */ + input = g_strchug(input + 1); + break; + } + + if (!valid_unquoted_char(*input)) { + *input_p = input; + g_set_error(error_r, tokenizer_quark(), 0, + "Invalid unquoted character"); + return NULL; + } + } + + /* end of string: the string is already null-terminated + here */ + + *input_p = input; + return word; +} + char * tokenizer_next_string(char **input_p, GError **error_r) { @@ -155,7 +209,7 @@ tokenizer_next_string(char **input_p, GError **error_r) } char * -tokenizer_next_word_or_string(char **input_p, GError **error_r) +tokenizer_next_param(char **input_p, GError **error_r) { assert(input_p != NULL); assert(*input_p != NULL); @@ -163,5 +217,5 @@ tokenizer_next_word_or_string(char **input_p, GError **error_r) if (**input_p == '"') return tokenizer_next_string(input_p, error_r); else - return tokenizer_next_word(input_p, error_r); + return tokenizer_next_unquoted(input_p, error_r); } |