utf8: skip BOM, if found

3 files changed, 27 insertions, 1 deletions

diff --git a/src/file.c b/src/file.c
index 7171f31..6da94a2 100644
--- a/src/file.c
+++ b/src/file.c
@@ -41,11 +41,23 @@ blogc_file_get_contents(const char *path, size_t *len, blogc_error_t **err)
 
     sb_string_t *str = sb_string_new();
     char buffer[BLOGC_FILE_CHUNK_SIZE];
+    char *tmp;
 
     while (!feof(fp)) {
         size_t read_len = fread(buffer, sizeof(char), BLOGC_FILE_CHUNK_SIZE, fp);
+
+        tmp = buffer;
+
+        if (str->len == 0 && read_len > 0) {
+            // skipping BOM before validation, for performance. should be safe
+            // enough
+            size_t skip = blogc_utf8_skip_bom((uint8_t*) buffer, read_len);
+            read_len -= skip;
+            tmp += skip;
+        }
+
         *len += read_len;
-        sb_string_append_len(str, buffer, read_len);
+        sb_string_append_len(str, tmp, read_len);
     }
     fclose(fp);
 
diff --git a/src/utf8.c b/src/utf8.c
index deea46d..0c04d60 100644
--- a/src/utf8.c
+++ b/src/utf8.c
@@ -86,3 +86,16 @@ blogc_utf8_validate_str(sb_string_t *str)
 {
     return blogc_utf8_validate((uint8_t*) str->str, str->len);
 }
+
+
+size_t
+blogc_utf8_skip_bom(const uint8_t *str, size_t len)
+{
+    if (len < 3)
+        return 0;
+
+    if (str[0] == 0xef && str[1] == 0xbb && str[2] == 0xbf)
+        return 3;
+
+    return 0;
+}
diff --git a/src/utf8.h b/src/utf8.h
index 837d02f..582ae1c 100644
--- a/src/utf8.h
+++ b/src/utf8.h
@@ -15,5 +15,6 @@
 
 bool blogc_utf8_validate(const uint8_t *str, size_t len);
 bool blogc_utf8_validate_str(sb_string_t *str);
+size_t blogc_utf8_skip_bom(const uint8_t *str, size_t len);
 
 #endif /* _UTF_8_H */