Fix scc2srt

examples/scc2srt.c

@@ -37,15 +37,13 @@ int main(int argc, char** argv)
     utf8_char_t* scc_data_ptr = utf8_load_text_file(argv[1], &scc_size);
     utf8_char_t* scc_data = scc_data_ptr;
 
+    srt = srt_new();
     caption_frame_init(&frame);
     scc_data += scc_to_608(&scc, scc_data);
 
-    srt = srt_new();
-
     while (scc->cc_size) {
         for (i = 0; i < scc->cc_size; ++i) {
             // eia608_dump (scc->cc_data[i]);
-
             if (LIBCAPTION_READY == caption_frame_decode(&frame, scc->cc_data[i], scc->timestamp)) {
                 srt_cue_from_caption_frame(&frame, srt);
             }

src/scc.c

@@ -81,15 +81,8 @@ size_t scc_to_608(scc_t** scc, const utf8_char_t* data)
     }
 
     // Skip blank lines
-    for (;;) {
-        llen = utf8_line_length(data);
-
-        if (0 == llen || 0 != utf8_trimmed_length(data, llen)) {
-            break;
-        }
-
-        data += llen;
-        size += llen;
+    while(utf8_char_whitespace(data)) {
+        data += 1, size += 1;
     }
 
     if (4 == sscanf(data, "%2d:%2d:%2d%*1[:;]%2d", &hh, &mm, &ss, &ff)) {
@@ -102,6 +95,7 @@ size_t scc_to_608(scc_t** scc, const utf8_char_t* data)
         (*scc)->timestamp = scc_time_to_timestamp(hh, mm, ss, ff);
         (*scc)->cc_size = 0;
 
+
         while ((*scc)->cc_size < max_cc_count && 1 == sscanf(data, "%04x", &cc_data)) {
             (*scc)->cc_data[(*scc)->cc_size] = (uint16_t)cc_data;
             (*scc)->cc_size += 1;

src/utf8.c

@@ -117,16 +117,16 @@ utf8_size_t utf8_char_count(const char* data, size_t size)
     return count;
 }
 
-// returnes the length of the line in bytes triming not printable charcters at the end
+// returns the length of the line in bytes triming not printable charcters at the end
 size_t utf8_trimmed_length(const utf8_char_t* data, utf8_size_t charcters)
 {
     size_t l, t = 0, split_at = 0;
     for (size_t c = 0; (*data) && c < charcters; ++c) {
         l = utf8_char_length(data);
-        t += l, data += l;
         if (!utf8_char_whitespace(data)) {
-            split_at = t;
+            split_at = t + l;
         }
+         t += l, data += l;
     }
 
     return split_at;
@@ -152,6 +152,8 @@ size_t utf8_line_length(const utf8_char_t* data)
         if (0 < (n = _utf8_newline(data))) {
             return len + n;
         }
+
+        data += utf8_char_length(data);
     }
 
     return len;