Age | Commit message (Collapse) | Author | |
---|---|---|---|
2010-07-11 | Allow static upb_strings. | Joshua Haberman | |
This can allow strings to reference static data, and reduced the memory footprint of test_def by about 10% (3k). | |||
2010-07-07 | Defined the function for getting a upb_symtab for descriptor.proto. | Joshua Haberman | |
2010-07-07 | Change descriptor.c/h to be just a byte array of serialized descriptor.proto. | Joshua Haberman | |
2010-06-27 | upb_def compiles again, though with lots of #if 0. | Joshua Haberman | |
2010-01-04 | Modified descriptor files. | Joshua Haberman | |
2010-01-02 | Move string representations back upb.h -> upb_data.h. | Joshua Haberman | |
2010-01-02 | upb_string* -> upb_strptr, to follow aliasing rules. | Joshua Haberman | |
2009-12-31 | Correct descriptor: initialize base. | Joshua Haberman | |
2009-12-31 | Correct descriptor.h: for now messages always have a upb_data. | Joshua Haberman | |
2009-12-28 | upbc compiles, but doesn't link due to unimplemented stuff in upb_data. | Joshua Haberman | |
2009-12-23 | Core compiles, though with warnings, and is still incomplete. | Joshua Haberman | |
2009-12-22 | More progress, upb_data and upb_def compile (but are incomplete). | Joshua Haberman | |
2009-12-21 | In the midst of a major refactoring. | Joshua Haberman | |
2009-12-05 | Make defs refcounted, rename upb_context->upbsymtab. | Joshua Haberman | |
There is currently a memory leak when type definitions form cycles. This will need to be dealt with. | |||
2009-11-28 | Overhaul defs to derive from a common base. | Joshua Haberman | |
2009-11-26 | Make upb_msgdef own all its data. | Joshua Haberman | |
This is in anticipation of making upb_msgdef's easy to dup. This involved removing all traces of any descriptors from the defs. | |||
2009-08-30 | Small bugfix: enums are int32, not uint32. | Joshua Haberman | |
2009-08-24 | Significant memory-management refactoring any Python extension. | Joshua Haberman | |
2009-08-12 | Refactoring: unify upb_msg. | Joshua Haberman | |
The cost is that a upb_msg will now always have an overhead of 2*sizeof(void*). This is comparable to proto2 overhead. The benefit is that upb_msg is now self-describing, and read-only algorithms can now operate on a upb_msg regardless of the memory-management scheme. Also, upb_array and upb_string now know inherently if they own their associated memory, and upb_array has a generic pointer for memory management purposes like upb_msg does. | |||
2009-08-02 | Add a friendly comment at the top of auto-generated C files. | Joshua Haberman | |
2009-08-01 | Fixed a bug in set_flags order in the compiler. | Joshua Haberman | |
2009-07-23 | New descriptor.c and descriptor.h (generated by upbc!) | Joshua Haberman | |
2009-07-20 | Move upb_array into its own header, upb_context parses fds instead of fd. | Joshua Haberman | |
2009-07-08 | Add text version of descriptor.proto. | Joshua Haberman | |
2009-07-08 | Directory restructuring. | Joshua Haberman | |