diff options
author | Tim-Philipp Müller <tim@centricular.net> | 2006-04-28 11:37:22 +0000 |
---|---|---|
committer | Tim-Philipp Müller <tim@centricular.net> | 2006-04-28 11:37:22 +0000 |
commit | d5296f28959c0cd0a6960ef36a8db840ba93aba2 (patch) | |
tree | 1ca82101f9bff6be983917c75446379a62501d40 /gst/id3demux | |
parent | b88deddeca5ed94c19baa5321d04447897a4840d (diff) |
gst/id3demux/id3v2frames.c: Recognise and skip any byte order marker (BOM) in
Original commit message from CVS:
* gst/id3demux/id3v2frames.c: (has_utf16_bom),
(parse_split_strings):
Recognise and skip any byte order marker (BOM) in
UTF-16 strings.
Diffstat (limited to 'gst/id3demux')
-rw-r--r-- | gst/id3demux/id3v2frames.c | 24 |
1 files changed, 24 insertions, 0 deletions
diff --git a/gst/id3demux/id3v2frames.c b/gst/id3demux/id3v2frames.c index cf56f8bd..b97e4d01 100644 --- a/gst/id3demux/id3v2frames.c +++ b/gst/id3demux/id3v2frames.c @@ -668,6 +668,24 @@ parse_insert_string_field (const gchar * encoding, gchar * data, gint data_size, g_array_append_val (fields, field); } +static gboolean +has_utf16_bom (gchar * data, const gchar ** p_in_encoding) +{ + guint16 marker = (GST_READ_UINT8 (data) << 8) | GST_READ_UINT8 (data + 1); + + switch (marker) { + case 0xFFFE: + *p_in_encoding = "UTF16LE"; + return TRUE; + case 0xFEFF: + *p_in_encoding = "UTF16BE"; + return TRUE; + default: + break; + } + return FALSE; +} + static void parse_split_strings (guint8 encoding, gchar * data, gint data_size, GArray ** out_fields) @@ -719,6 +737,9 @@ parse_split_strings (guint8 encoding, gchar * data, gint data_size, /* Find '\0\0' terminator */ for (text_pos = 0; text_pos < data_size - 1; text_pos += 2) { if (data[text_pos] == '\0' && data[text_pos + 1] == '\0') { + if (has_utf16_bom (data + prev, &in_encode)) { + prev += 2; /* skip BOM */ + } /* found a delimiter */ parse_insert_string_field (in_encode, data + prev, text_pos - prev + 2, fields); @@ -729,6 +750,9 @@ parse_split_strings (guint8 encoding, gchar * data, gint data_size, } if (data_size - prev > 1 && (data[prev] != 0x00 || data[prev + 1] != 0x00)) { + if (has_utf16_bom (data + prev, &in_encode)) { + prev += 2; /* skip BOM */ + } /* There were 2 or more non-null chars left, convert those too */ parse_insert_string_field (in_encode, data + prev, data_size - prev, fields); |