Added various macros to check return values and log any errors and abort
if the call fails.
Updated logging to handle early errors before logging initialized.
Added configure checks needed to detect missing functions on OSX
Added compat clock functions using clock_get_time on OSX
Added compat timer functions using clock_gettime and setitimer on OSX
Added byteswap functions on OSX
encoded/converted. Characters above x7F were replaced by '?' character
although the rfc defines a ISO−8859−1 encoding for descriptive
field-content.
According to rfc2616 the field-content is defined as follows:
<the OCTETs making up the field-value and consisting of either *TEXT or
combinations of token, separators, and quoted-string>
The TEXT rule is only used for descriptive field contents and values
that are not intended to be interpreted by the message parser. Words of
*TEXT MAY contain characters from character sets other than ISO- 8859-1
only when encoded according to the rules of RFC 2047.
In the previous implementation the icy metadata was converted based on
fromcode "ascii".
Following incoming icy header field-values should be encoded as
"ISO−8859−1" before adding them to the metadata structure.
- misc.c unicode_fixup_string enhanced by an additional parameter to
define the fromcode
- misc.h unicode_fixup_string prototype updated
- filescanner.c function fixup_tags updated to stay compatible to the
previous implementation using fromcode "ascii"
- db.c function unicode_fixup_mfi updated to stay compatible to the
previous implementation using fromcode "ascii"
- http.c function metadata_header_get enhanced to encode the header
field-content as "ISO−8859−1" to comply with rfc2616
The byte-order marks are useless for UTF-8, but that doesn't mean we don't
find them in the wild. Get rid of them, they confuse the hell out of the
collation functions.
Reported by Kai Elwert.
Properly handle the tail of the data buffer, accessing the tail as an
unsigned char array instead of a uint32_t array.
Issue spotted by Peter Carmichael.
