The type `ida_hexrays.DecompilationFailure` is still present (but is a dummy) because otherwise scripts might break. It is never thrown anymore.
Note that `vds_xrefs.py` still need the `str(cfunc))` kludge, because of a caveat in when/how the decompiler actually updates the global xrefs cache. This will be fixed in the next IDA minor release
In addition, this commit introduces the notion of a typemap to convert form an `ea_t` to a `func_t *` for the `decompile()` call, but doesn't generalize its use.
Since this is a relatively far-ranging change, this will only be introduced in the next IDA minor release as well.
Specifically:
BUGFIX: IDAPython's choose.choose() was broken.
BUGFIX: calling idaapi.del_hotkey() twice with the same argument could crash IDA.
BUGFIX: IDAPython's IDC emulation of idc.GetTevRegMem() was not working.
BUGFIX: IDAPython: execute_ui_requests() could crash IDA.
BUGFIX: IDAPython: IDP_Hooks instances could prevent the decompiler from working properly
What's new:
- added the decompiler bindings
- Expose simpleline_t type to IDAPython. That lets the user to set the bgcolor & text for each line in the decompilation.
- Wrapped new functions from the IDA SDK
Various fixes:
for non-code locations, idc.GetOpnd() would create instructions instead of returning empty result
- idb_event::area_cmt_changed was never received in IDB_Hooks (and descendants)
- idb_event::ti_changed, and idb_event::op_ti_changed notifications were not accessible in IDAPython
- op_t.value was truncated to 32 bits under IDA64.
- print_tinfo() wouldn't return a valid string.
- readsel2() was not usable.
- read_selection() was buggy for 64-bit programs.
- StructMembers() considered holes in structures, and didn't properly iterate through the whole structure definition.
- There was no way to call calc_switch_cases() from IDAPython.
- when using multi-select/multi-edit choosers, erroneous event codes could be sent at beginning & end of batch deletion of lines.
- When, in a PluginForm#OnCreate, the layout of IDA was requested to change (for example by starting a debugging session), that PluginForm could be deleted and create an access violation.
- tinfo_t objects created from IDAPython could cause an assertion failure at exit time.
- Usage of IDAPython's DropdownListControl was broken.
What's new:
- Proper multi-threaded support
- Better PyObject reference counting with ref_t and newref_t helper classes
- Improved the pywraps/deployment script
- Added IDAViewWrapper class and example
- Added idc.GetDisasmEx()
- Added idc.AddSegEx()
- Added idc.GetLocalTinfo()
- Added idc.ApplyType()
- Updated type information implementation
- Introduced the idaapi.require() - see http://www.hexblog.com/?p=749
- set REMOVE_CWD_SYS_PATH=1 by default in python.cfg (remove current directory from the import search path).
Various bugfixes:
- fixed various memory leaks
- asklong/askaddr/asksel (and corresponding idc.py functions) were returning results truncated to 32 bits in IDA64
- fix wrong documentation for idc.SizeOf
- GetFloat/GetDouble functions did not take into account endianness of the processor
- idaapi.NO_PROCESS was not defined, and was causing GetProcessPid() to fail
- idc.py: insert escape characters to string parameter when call Eval()
- idc.SaveFile/savefile were always overwriting an existing file instead of writing only the new data
- PluginForm.Close() wasn't passing its arguments to the delegate function, resulting in an error.
- added support for 'long' addresses
- updated copyright notice in hexrays.i
- added methods to remove or insert elements in the qlist<cinsn_t> list
- added support for ctree_visitor_t / ctree_parentee_t / cfunc_parentee_t / user_lvar_visitor_t
- updated vds3 sample
- added vds4 and vds7 python samples (ported from their C++ counter parts)
https://github.com/EiNSTeiN-/hexrays-python
- Added '--with-hexrays' switch to the build script so it wrap Hex-Rays Decompiler API
- Added one Hex-Rays decompiler sample: vds1.py
Arnaud Diederen