- Jun 05, 2012
-
-
Ryan Kelly authored
-
- Jun 04, 2012
-
-
Ryan Kelly authored
-
Ryan Kelly authored
-
- Feb 13, 2012
-
-
Ryan Kelly authored
-
- Jan 08, 2012
-
-
Ryan Kelly authored
-
- May 06, 2011
-
-
Ryan Kelly authored
-
Ryan Kelly authored
-
Ryan Kelly authored
-
Ryan Kelly authored
-
- Apr 13, 2011
-
-
Ryan Kelly authored
-
- Apr 11, 2011
-
-
Ryan Kelly authored
-
- Apr 10, 2011
-
-
Ryan Kelly authored
-
Ryan Kelly authored
This costs a little in performance, since the compiler can't inline trivial functions like tns_get_null(). But, it will make it much easier to provide slightly different parsing/rending strategies e.g. automatically turning unicode strings into utf8 bytes.
-
- Apr 06, 2011
-
-
Ryan Kelly authored
-
- Apr 04, 2011
-
-
Ryan Kelly authored
-
Ryan Kelly authored
-
Ryan Kelly authored
-
Ryan Kelly authored
-
Ryan Kelly authored
-
Ryan Kelly authored
-
- Apr 03, 2011
-
-
Ryan Kelly authored
-
Ryan Kelly authored
-
Ryan Kelly authored
-
Ryan Kelly authored
-
Ryan Kelly authored
-
Ryan Kelly authored
-
Ryan Kelly authored
-
Ryan Kelly authored
-
Ryan Kelly authored
-
Ryan Kelly authored
This is what the stdlib json module uses, and it seems appropriate. Custom error classes don't really buy us anything.
-
Ryan Kelly authored
dump() is just a simpler wrapper around dumps(), while load() does some extra work to ensure it only reads as much data as it needs.
-
- Apr 02, 2011
-
-
Ryan Kelly authored
-
Ryan Kelly authored
-
Ryan Kelly authored
-
Ryan Kelly authored
-
- Mar 31, 2011
-
-
Ryan Kelly authored
tns_outbuf_rev.c now contains the original write-in-reverse routines, while tns_outbuf_push.c contains routines that write from the back of the buffer but keep the data right-way-round.
-
- Mar 27, 2011
-
-
Ryan Kelly authored
-
Ryan Kelly authored
-
Ryan Kelly authored
This attempt uses a deque to collect fragments of the output, generated last-chunk-first. We save the overhead of reversing lots of little strings.
-
Ryan Kelly authored
-