完善jsonl文件打印

.idea/Gencoding.iml

@@ -1,8 +1,10 @@
 <?xml version="1.0" encoding="UTF-8"?>
 <module type="PYTHON_MODULE" version="4">
   <component name="NewModuleRootManager">
-    <content url="file://$MODULE_DIR$" />
-    <orderEntry type="inheritedJdk" />
+    <content url="file://$MODULE_DIR$">
+      <sourceFolder url="file://$MODULE_DIR$/python_lib" isTestSource="false" />
+    </content>
+    <orderEntry type="jdk" jdkName="Python 2.7" jdkType="Python SDK" />
     <orderEntry type="sourceFolder" forTests="false" />
   </component>
 </module>

.idea/misc.xml

@@ -1,4 +1,4 @@
 <?xml version="1.0" encoding="UTF-8"?>
 <project version="4">
-  <component name="ProjectRootManager" version="2" project-jdk-name="D:\Program\miniconda3\envs\pyqt" project-jdk-type="Python SDK" />
+  <component name="ProjectRootManager" version="2" project-jdk-name="Python 2.7" project-jdk-type="Python SDK" />
 </project>

README.md

@@ -16,3 +16,9 @@ This project provides two components of Genius, a graph-based bug search framewo
 	
 	The online search is based on nearpy (https://github.com/pixelogik/NearPy). 
 
+```
+D:\IDA_Pro_v6.8\idaq.exe -c -S"raw-feature-extractor/preprocessing_ida.py" file_name
+D:\IDA_Pro_v6.8\idaq.exe -c -S"raw-feature-extractor/preprocessing_ida.py --path XXX" file_name
+
+```
+

python_lib/PySide/__init__.py

@@ -0,0 +1,3 @@
+__all__ = ['QtCore', 'QtGui', 'QtNetwork', 'QtOpenGL', 'QtSql', 'QtSvg', 'QtTest', 'QtWebKit', 'QtScript']
+__version__         = "1.1.2"
+__version_info__    = (1, 1, 2, "final", 1)

python_lib/idautils.py

@@ -0,0 +1,830 @@
+#---------------------------------------------------------------------
+# IDAPython - Python plugin for Interactive Disassembler
+#
+# Copyright (c) 2004-2010 Gergely Erdelyi <gergely.erdelyi@d-dome.net>
+#
+# All rights reserved.
+#
+# For detailed copyright information see the file COPYING in
+# the root of the distribution archive.
+#---------------------------------------------------------------------
+"""
+idautils.py - High level utility functions for IDA
+"""
+import idaapi
+import idc
+import types
+import os
+
+
+def refs(ea, funcfirst, funcnext):
+    """
+    Generic reference collector - INTERNAL USE ONLY.
+    """
+    ref = funcfirst(ea)
+    while ref != idaapi.BADADDR:
+        yield ref
+        ref = funcnext(ea, ref)
+
+
+def CodeRefsTo(ea, flow):
+    """
+    Get a list of code references to 'ea'
+
+    @param ea:   Target address
+    @param flow: Follow normal code flow or not
+    @type  flow: Boolean (0/1, False/True)
+
+    @return: list of references (may be empty list)
+
+    Example::
+
+        for ref in CodeRefsTo(ScreenEA(), 1):
+            print ref
+    """
+    if flow == 1:
+        return refs(ea, idaapi.get_first_cref_to, idaapi.get_next_cref_to)
+    else:
+        return refs(ea, idaapi.get_first_fcref_to, idaapi.get_next_fcref_to)
+
+
+def CodeRefsFrom(ea, flow):
+    """
+    Get a list of code references from 'ea'
+
+    @param ea:   Target address
+    @param flow: Follow normal code flow or not
+    @type  flow: Boolean (0/1, False/True)
+
+    @return: list of references (may be empty list)
+
+    Example::
+
+        for ref in CodeRefsFrom(ScreenEA(), 1):
+            print ref
+    """
+    if flow == 1:
+        return refs(ea, idaapi.get_first_cref_from, idaapi.get_next_cref_from)
+    else:
+        return refs(ea, idaapi.get_first_fcref_from, idaapi.get_next_fcref_from)
+
+
+def DataRefsTo(ea):
+    """
+    Get a list of data references to 'ea'
+
+    @param ea:   Target address
+
+    @return: list of references (may be empty list)
+
+    Example::
+
+        for ref in DataRefsTo(ScreenEA()):
+            print ref
+    """
+    return refs(ea, idaapi.get_first_dref_to, idaapi.get_next_dref_to)
+
+
+def DataRefsFrom(ea):
+    """
+    Get a list of data references from 'ea'
+
+    @param ea:   Target address
+
+    @return: list of references (may be empty list)
+
+    Example::
+
+        for ref in DataRefsFrom(ScreenEA()):
+            print ref
+    """
+    return refs(ea, idaapi.get_first_dref_from, idaapi.get_next_dref_from)
+
+
+def XrefTypeName(typecode):
+    """
+    Convert cross-reference type codes to readable names
+
+    @param typecode: cross-reference type code
+    """
+    ref_types = {
+        0  : 'Data_Unknown',
+        1  : 'Data_Offset',
+        2  : 'Data_Write',
+        3  : 'Data_Read',
+        4  : 'Data_Text',
+        5  : 'Data_Informational',
+        16 : 'Code_Far_Call',
+        17 : 'Code_Near_Call',
+        18 : 'Code_Far_Jump',
+        19 : 'Code_Near_Jump',
+        20 : 'Code_User',
+        21 : 'Ordinary_Flow'
+        }
+    assert typecode in ref_types, "unknown reference type %d" % typecode
+    return ref_types[typecode]
+
+
+def _copy_xref(xref):
+    """ Make a private copy of the xref class to preserve its contents """
+    class _xref(object):
+        pass
+
+    xr = _xref()
+    for attr in [ 'frm', 'to', 'iscode', 'type', 'user' ]:
+        setattr(xr, attr, getattr(xref, attr))
+    return xr
+
+
+def XrefsFrom(ea, flags=0):
+    """
+    Return all references from address 'ea'
+
+    @param ea: Reference address
+    @param flags: any of idaapi.XREF_* flags
+
+    Example::
+           for xref in XrefsFrom(here(), 0):
+               print xref.type, XrefTypeName(xref.type), \
+                         'from', hex(xref.frm), 'to', hex(xref.to)
+    """
+    xref = idaapi.xrefblk_t()
+    if xref.first_from(ea, flags):
+        yield _copy_xref(xref)
+        while xref.next_from():
+            yield _copy_xref(xref)
+
+
+def XrefsTo(ea, flags=0):
+    """
+    Return all references to address 'ea'
+
+    @param ea: Reference address
+    @param flags: any of idaapi.XREF_* flags
+
+    Example::
+           for xref in XrefsTo(here(), 0):
+               print xref.type, XrefTypeName(xref.type), \
+                         'from', hex(xref.frm), 'to', hex(xref.to)
+    """
+    xref = idaapi.xrefblk_t()
+    if xref.first_to(ea, flags):
+        yield _copy_xref(xref)
+        while xref.next_to():
+            yield _copy_xref(xref)
+
+
+def Threads():
+    """Returns all thread IDs"""
+    for i in xrange(0, idc.GetThreadQty()):
+        yield idc.GetThreadId(i)
+
+
+def Heads(start=None, end=None):
+    """
+    Get a list of heads (instructions or data)
+
+    @param start: start address (default: inf.minEA)
+    @param end:   end address (default: inf.maxEA)
+
+    @return: list of heads between start and end
+    """
+    if not start: start = idaapi.cvar.inf.minEA
+    if not end:   end = idaapi.cvar.inf.maxEA
+
+    ea = start
+    if not idc.isHead(idc.GetFlags(ea)):
+        ea = idaapi.next_head(ea, end)
+    while ea != idaapi.BADADDR:
+        yield ea
+        ea = idaapi.next_head(ea, end)
+
+
+def Functions(start=None, end=None):
+    """
+    Get a list of functions
+
+    @param start: start address (default: inf.minEA)
+    @param end:   end address (default: inf.maxEA)
+
+    @return: list of heads between start and end
+
+    @note: The last function that starts before 'end' is included even
+    if it extends beyond 'end'. Any function that has its chunks scattered
+    in multiple segments will be reported multiple times, once in each segment
+    as they are listed.
+    """
+    if not start: start = idaapi.cvar.inf.minEA
+    if not end:   end = idaapi.cvar.inf.maxEA
+
+    # find first function head chunk in the range
+    chunk = idaapi.get_fchunk(start)
+    if not chunk:
+        chunk = idaapi.get_next_fchunk(start)
+    while chunk and chunk.startEA < end and (chunk.flags & idaapi.FUNC_TAIL) != 0:
+        chunk = idaapi.get_next_fchunk(chunk.startEA)
+    func = chunk
+
+    while func and func.startEA < end:
+        startea = func.startEA
+        yield startea
+        func = idaapi.get_next_func(startea)
+
+
+def Chunks(start):
+    """
+    Get a list of function chunks
+
+    @param start: address of the function
+
+    @return: list of funcion chunks (tuples of the form (start_ea, end_ea))
+             belonging to the function
+    """
+    func_iter = idaapi.func_tail_iterator_t( idaapi.get_func( start ) )
+    status = func_iter.main()
+    while status:
+        chunk = func_iter.chunk()
+        yield (chunk.startEA, chunk.endEA)
+        status = func_iter.next()
+
+
+def Modules():
+    """
+    Returns a list of module objects with name,size,base and the rebase_to attributes
+    """
+    mod = idaapi.module_info_t()
+    result = idaapi.get_first_module(mod)
+    while result:
+        yield idaapi.object_t(name=mod.name, size=mod.size, base=mod.base, rebase_to=mod.rebase_to)
+        result = idaapi.get_next_module(mod)
+
+
+def Names():
+    """
+    Returns a list of names
+
+    @return: List of tuples (ea, name)
+    """
+    for i in xrange(idaapi.get_nlist_size()):
+        ea   = idaapi.get_nlist_ea(i)
+        name = idaapi.get_nlist_name(i)
+        yield (ea, name)
+
+
+def Segments():
+    """
+    Get list of segments (sections) in the binary image
+
+    @return: List of segment start addresses.
+    """
+    for n in xrange(idaapi.get_segm_qty()):
+        seg = idaapi.getnseg(n)
+        if seg:
+            yield seg.startEA
+
+
+def Entries():
+    """
+    Returns a list of entry points
+
+    @return: List of tuples (index, ordinal, ea, name)
+    """
+    n = idaapi.get_entry_qty()
+    for i in xrange(0, n):
+        ordinal = idaapi.get_entry_ordinal(i)
+        ea      = idaapi.get_entry(ordinal)
+        name    = idaapi.get_entry_name(ordinal)
+        yield (i, ordinal, ea, name)
+
+
+def FuncItems(start):
+    """
+    Get a list of function items
+
+    @param start: address of the function
+
+    @return: ea of each item in the function
+    """
+    func = idaapi.get_func(start)
+    if not func:
+        return
+    fii = idaapi.func_item_iterator_t()
+    ok = fii.set(func)
+    while ok:
+        yield fii.current()
+        ok = fii.next_code()
+
+
+def Structs():
+    """
+    Get a list of structures
+
+    @return: List of tuples (idx, sid, name)
+    """
+    idx  = idc.GetFirstStrucIdx()
+    while idx != idaapi.BADADDR:
+        sid = idc.GetStrucId(idx)
+        yield (idx, sid, idc.GetStrucName(sid))
+        idx = idc.GetNextStrucIdx(idx)
+
+
+def StructMembers(sid):
+    """
+    Get a list of structure members information (or stack vars if given a frame).
+
+    @param sid: ID of the structure.
+
+    @return: List of tuples (offset, name, size)
+
+    @note: If 'sid' does not refer to a valid structure,
+           an exception will be raised.
+    @note: This will not return 'holes' in structures/stack frames;
+           it only returns defined structure members.
+    """
+    m = idc.GetFirstMember(sid)
+    if m == -1:
+        raise Exception("No structure with ID: 0x%x" % sid)
+    while (m != idaapi.BADADDR):
+        name = idc.GetMemberName(sid, m)
+        if name:
+            yield (m, name, idc.GetMemberSize(sid, m))
+        m = idc.GetStrucNextOff(sid, m)
+
+
+def DecodePrecedingInstruction(ea):
+    """
+    Decode preceding instruction in the execution flow.
+
+    @param ea: address to decode
+    @return: (None or the decode instruction, farref)
+             farref will contain 'true' if followed an xref, false otherwise
+    """
+    prev_addr, farref  = idaapi.decode_preceding_insn(ea)
+    if prev_addr == idaapi.BADADDR:
+        return (None, False)
+    else:
+        return (idaapi.cmd.copy(), farref)
+
+
+
+def DecodePreviousInstruction(ea):
+    """
+    Decodes the previous instruction and returns an insn_t like class
+
+    @param ea: address to decode
+    @return: None or a new insn_t instance
+    """
+    prev_addr = idaapi.decode_prev_insn(ea)
+    if prev_addr == idaapi.BADADDR:
+        return None
+
+    return idaapi.cmd.copy()
+
+
+def DecodeInstruction(ea):
+    """
+    Decodes an instruction and returns an insn_t like class
+
+    @param ea: address to decode
+    @return: None or a new insn_t instance
+    """
+    inslen = idaapi.decode_insn(ea)
+    if inslen == 0:
+        return None
+
+    return idaapi.cmd.copy()
+
+
+def GetDataList(ea, count, itemsize=1):
+    """
+    Get data list - INTERNAL USE ONLY
+    """
+    if itemsize == 1:
+        getdata = idaapi.get_byte
+    elif itemsize == 2:
+        getdata = idaapi.get_word
+    elif itemsize == 4:
+        getdata = idaapi.get_long
+    elif itemsize == 8:
+        getdata = idaapi.get_qword
+    else:
+        raise ValueError, "Invalid data size! Must be 1, 2, 4 or 8"
+
+    endea = ea + itemsize * count
+    curea = ea
+    while curea < endea:
+        yield getdata(curea)
+        curea += itemsize
+
+
+def PutDataList(ea, datalist, itemsize=1):
+    """
+    Put data list - INTERNAL USE ONLY
+    """
+    putdata = None
+
+    if itemsize == 1:
+        putdata = idaapi.patch_byte
+    if itemsize == 2:
+        putdata = idaapi.patch_word
+    if itemsize == 4:
+        putdata = idaapi.patch_long
+
+    assert putdata, "Invalid data size! Must be 1, 2 or 4"
+
+    for val in datalist:
+        putdata(ea, val)
+        ea = ea + itemsize
+
+
+def MapDataList(ea, length, func, wordsize=1):
+    """
+    Map through a list of data words in the database
+
+    @param ea:       start address
+    @param length:   number of words to map
+    @param func:     mapping function
+    @param wordsize: size of words to map [default: 1 byte]
+
+    @return: None
+    """
+    PutDataList(ea, map(func, GetDataList(ea, length, wordsize)), wordsize)
+
+
+def GetInputFileMD5():
+    """
+    Return the MD5 hash of the input binary file
+
+    @return: MD5 string or None on error
+    """
+    return idc.GetInputMD5()
+
+
+class Strings(object):
+    """
+    Allows iterating over the string list. The set of strings will not be modified.
+    , unless asked explicitly at setup()-time..
+
+    Example:
+        s = Strings()
+
+        for i in s:
+            print "%x: len=%d type=%d -> '%s'" % (i.ea, i.length, i.type, str(i))
+
+    """
+    class StringItem(object):
+        """
+        Class representing each string item.
+        """
+        def __init__(self, si):
+            self.ea     = si.ea
+            """String ea"""
+            self.type   = si.type
+            """string type (ASCSTR_xxxxx)"""
+            self.length = si.length
+            """string length"""
+
+        def is_1_byte_encoding(self):
+            return not self.is_2_bytes_encoding() and not self.is_4_bytes_encoding()
+
+        def is_2_bytes_encoding(self):
+            return (self.type & 7) in [idaapi.ASCSTR_UTF16, idaapi.ASCSTR_ULEN2, idaapi.ASCSTR_ULEN4]
+
+        def is_4_bytes_encoding(self):
+            return (self.type & 7) == idaapi.ASCSTR_UTF32
+
+        def _toseq(self, as_unicode):
+            if self.is_2_bytes_encoding():
+                conv = idaapi.ACFOPT_UTF16
+                pyenc = "utf-16"
+            elif self.is_4_bytes_encoding():
+                conv = idaapi.ACFOPT_UTF8
+                pyenc = "utf-8"
+            else:
+                conv = idaapi.ACFOPT_ASCII
+                pyenc = 'ascii'
+            strbytes = idaapi.get_ascii_contents2(self.ea, self.length, self.type, conv)
+            return unicode(strbytes, pyenc, 'replace') if as_unicode else strbytes
+
+        def __str__(self):
+            return self._toseq(False)
+
+        def __unicode__(self):
+            return self._toseq(True)
+
+
+    STR_C       = 0x0001
+    """C-style ASCII string"""
+    STR_PASCAL  = 0x0002
+    """Pascal-style ASCII string (length byte)"""
+    STR_LEN2    = 0x0004
+    """Pascal-style, length is 2 bytes"""
+    STR_UNICODE = 0x0008
+    """Unicode string"""
+    STR_LEN4    = 0x0010
+    """Pascal-style, length is 4 bytes"""
+    STR_ULEN2   = 0x0020
+    """Pascal-style Unicode, length is 2 bytes"""
+    STR_ULEN4   = 0x0040
+    """Pascal-style Unicode, length is 4 bytes"""
+
+    def clear_cache(self):
+        """Clears the strings list cache"""
+        self.refresh(0, 0) # when ea1=ea2 the kernel will clear the cache
+
+    def __init__(self, default_setup = False):
+        """
+        Initializes the Strings enumeration helper class
+
+        @param default_setup: Set to True to use default setup (C strings, min len 5, ...)
+        """
+        self.size = 0
+        if default_setup:
+            self.setup()
+        else:
+            self.refresh()
+
+        self._si  = idaapi.string_info_t()
+
+    def refresh(self, ea1=None, ea2=None):
+        """Refreshes the strings list"""
+        if ea1 is None:
+            ea1 = idaapi.cvar.inf.minEA
+        if ea2 is None:
+            ea2 = idaapi.cvar.inf.maxEA
+
+        idaapi.refresh_strlist(ea1, ea2)
+        self.size = idaapi.get_strlist_qty()
+
+
+    def setup(self,
+              strtypes = STR_C,
+              minlen = 5,
+              only_7bit = True,
+              ignore_instructions = False,
+              ea1 = None,
+              ea2 = None,
+              display_only_existing_strings = False):
+
+        if ea1 is None:
+            ea1 = idaapi.cvar.inf.minEA
+
+        if ea2 is None:
+            ea2 = idaapi.cvar.inf.maxEA
+
+        t = idaapi.strwinsetup_t()
+        t.strtypes = strtypes
+        t.minlen = minlen
+        t.only_7bit = only_7bit
+        t.ea1 = ea1
+        t.ea2 = ea2
+        t.display_only_existing_strings = display_only_existing_strings
+        idaapi.set_strlist_options(t)
+
+        # Automatically refreshes
+        self.refresh()
+
+
+    def _get_item(self, index):
+        if not idaapi.get_strlist_item(index, self._si):
+            return None
+        else:
+            return Strings.StringItem(self._si)
+
+
+    def __iter__(self):
+        return (self._get_item(index) for index in xrange(0, self.size))
+
+
+    def __getitem__(self, index):
+        """Returns a string item or None"""
+        if index >= self.size:
+            raise KeyError
+        else:
+            return self._get_item(index)
+
+# -----------------------------------------------------------------------
+def GetIdbDir():
+    """
+    Get IDB directory
+
+    This function returns directory path of the current IDB database
+    """
+    return os.path.dirname(idaapi.cvar.database_idb) + os.sep
+
+# -----------------------------------------------------------------------
+def GetRegisterList():
+    """Returns the register list"""
+    return idaapi.ph_get_regnames()
+
+# -----------------------------------------------------------------------
+def GetInstructionList():
+    """Returns the instruction list of the current processor module"""
+    return [i[0] for i in idaapi.ph_get_instruc() if i[0]]
+
+# -----------------------------------------------------------------------
+def _Assemble(ea, line):
+    """
+    Please refer to Assemble() - INTERNAL USE ONLY
+    """
+    if type(line) == types.StringType:
+        lines = [line]
+    else:
+        lines = line
+    ret = []
+    for line in lines:
+        seg = idaapi.getseg(ea)
+        if not seg:
+            return (False, "No segment at ea")
+        ip  = ea - (idaapi.ask_selector(seg.sel) << 4)
+        buf = idaapi.AssembleLine(ea, seg.sel, ip, seg.bitness, line)
+        if not buf:
+            return (False, "Assembler failed: " + line)
+        ea += len(buf)
+        ret.append(buf)
+
+    if len(ret) == 1:
+        ret = ret[0]
+    return (True, ret)
+
+
+def Assemble(ea, line):
+    """
+    Assembles one or more lines (does not display an message dialogs)
+    If line is a list then this function will attempt to assemble all the lines
+    This function will turn on batch mode temporarily so that no messages are displayed on the screen
+
+    @param ea:       start address
+    @return: (False, "Error message") or (True, asm_buf) or (True, [asm_buf1, asm_buf2, asm_buf3])
+    """
+    old_batch = idc.Batch(1)
+    ret = _Assemble(ea, line)
+    idc.Batch(old_batch)
+    return ret
+
+def _copy_obj(src, dest, skip_list = None):
+    """
+    Copy non private/non callable attributes from a class instance to another
+    @param src: Source class to copy from
+    @param dest: If it is a string then it designates the new class type that will be created and copied to.
+                 Otherwise dest should be an instance of another class
+    @return: A new instance or "dest"
+    """
+    if type(dest) == types.StringType:
+        # instantiate a new destination class of the specified type name?
+        dest = new.classobj(dest, (), {})
+    for x in dir(src):
+        # skip special and private fields
+        if x.startswith("__") and x.endswith("__"):
+            continue
+        # skip items in the skip list
+        if skip_list and x in skip_list:
+            continue
+        t = getattr(src, x)
+        # skip callable
+        if callable(t):
+            continue
+        setattr(dest, x, t)
+    return dest
+
+# -----------------------------------------------------------------------
+class _reg_dtyp_t(object):
+    """
+    INTERNAL
+    This class describes a register's number and dtyp.
+    The equal operator is overloaded so that two instances can be tested for equality
+    """
+    def __init__(self, reg, dtyp):
+        self.reg  = reg
+        self.dtyp = dtyp
+
+    def __eq__(self, other):
+        return (self.reg == other.reg) and (self.dtyp == other.dtyp)
+
+# -----------------------------------------------------------------------
+class _procregs(object):
+    """Utility class allowing the users to identify registers in a decoded instruction"""
+    def __getattr__(self, attr):
+        ri = idaapi.reg_info_t()
+        if not idaapi.parse_reg_name(attr, ri):
+            raise AttributeError()
+        r = _reg_dtyp_t(ri.reg, ord(idaapi.get_dtyp_by_size(ri.size)))
+        self.__dict__[attr] = r
+        return r
+
+    def __setattr__(self, attr, value):
+        raise AttributeError(attr)
+
+
+# -----------------------------------------------------------------------
+class _cpu(object):
+    "Simple wrapper around GetRegValue/SetRegValue"
+    def __getattr__(self, name):
+        #print "cpu.get(%s)" % name
+        return idc.GetRegValue(name)
+
+    def __setattr__(self, name, value):
+        #print "cpu.set(%s)" % name
+        return idc.SetRegValue(value, name)
+
+
+# --------------------------------------------------------------------------
+class __process_ui_actions_helper(object):
+    def __init__(self, actions, flags = 0):
+        """Expect a list or a string with a list of actions"""
+        if isinstance(actions, str):
+            lst = actions.split(";")
+        elif isinstance(actions, (list, tuple)):
+            lst = actions
+        else:
+            raise ValueError, "Must pass a string, list or a tuple"
+
+        # Remember the action list and the flags
+        self.__action_list = lst
+        self.__flags = flags
+
+        # Reset action index
+        self.__idx = 0
+
+    def __len__(self):
+        return len(self.__action_list)
+
+    def __call__(self):
+        if self.__idx >= len(self.__action_list):
+            return False
+
+        # Execute one action
+        idaapi.process_ui_action(
+                self.__action_list[self.__idx],
+                self.__flags)
+
+        # Move to next action
+        self.__idx += 1
+
+        # Reschedule
+        return True
+
+
+# --------------------------------------------------------------------------
+def ProcessUiActions(actions, flags=0):
+    """
+    @param actions: A string containing a list of actions separated by semicolon, a list or a tuple
+    @param flags: flags to be passed to process_ui_action()
+    @return: Boolean. Returns False if the action list was empty or execute_ui_requests() failed.
+    """
+
+    # Instantiate a helper
+    helper = __process_ui_actions_helper(actions, flags)
+    return False if len(helper) < 1 else idaapi.execute_ui_requests((helper,))
+
+
+# -----------------------------------------------------------------------
+class peutils_t(object):
+    """
+    PE utility class. Retrieves PE information from the database.
+
+    Constants from pe.h
+    """
+    PE_NODE = "$ PE header" # netnode name for PE header
+    PE_ALT_DBG_FPOS   = idaapi.BADADDR & -1 #  altval() -> translated fpos of debuginfo
+    PE_ALT_IMAGEBASE  = idaapi.BADADDR & -2 #  altval() -> loading address (usually pe.imagebase)
+    PE_ALT_PEHDR_OFF  = idaapi.BADADDR & -3 #  altval() -> offset of PE header
+    PE_ALT_NEFLAGS    = idaapi.BADADDR & -4 #  altval() -> neflags
+    PE_ALT_TDS_LOADED = idaapi.BADADDR & -5 #  altval() -> tds already loaded(1) or invalid(-1)
+    PE_ALT_PSXDLL     = idaapi.BADADDR & -6 #  altval() -> if POSIX(x86) imports from PSXDLL netnode
+
+    def __init__(self):
+        self.__penode = idaapi.netnode()
+        self.__penode.create(peutils_t.PE_NODE)
+
+    imagebase = property(
+        lambda self: self.__penode.altval(peutils_t.PE_ALT_IMAGEBASE)
+      )
+
+    header = property(
+        lambda self: self.__penode.altval(peutils_t.PE_ALT_PEHDR_OFF)
+      )
+
+    def __str__(self):
+        return "peutils_t(imagebase=%s, header=%s)" % (hex(self.imagebase), hex(self.header))
+
+    def header(self):
+        """
+        Returns the complete PE header as an instance of peheader_t (defined in the SDK).
+        """
+        return self.__penode.valobj()
+
+# -----------------------------------------------------------------------
+cpu = _cpu()
+"""This is a special class instance used to access the registers as if they were attributes of this object.
+For example to access the EAX register:
+    print "%x" % cpu.Eax
+"""
+
+procregs = _procregs()
+"""This object is used to access the processor registers. It is useful when decoding instructions and you want to see which instruction is which.
+For example:
+    x = idautils.DecodeInstruction(here())
+    if x[0] == procregs.Esp:
+        print "This operand is the register ESP
+"""

python_lib/init.py

@@ -0,0 +1,111 @@
+#!/usr/bin/env python
+# -----------------------------------------------------------------------
+# IDAPython - Python plugin for Interactive Disassembler
+#
+# Copyright (c) The IDAPython Team <idapython@googlegroups.com>
+#
+# All rights reserved.
+#
+# For detailed copyright information see the file COPYING in
+# the root of the distribution archive.
+# -----------------------------------------------------------------------
+# init.py - Essential init routines
+# -----------------------------------------------------------------------
+import os
+import sys
+import time
+import warnings
+import _idaapi
+
+# __EA64__ is set if IDA is running in 64-bit mode
+__EA64__ = _idaapi.BADADDR == 0xFFFFFFFFFFFFFFFFL
+
+# -----------------------------------------------------------------------
+# Take over the standard text outputs
+# -----------------------------------------------------------------------
+class IDAPythonStdOut:
+    """
+    Dummy file-like class that receives stout and stderr
+    """
+    def write(self, text):
+        # NB: in case 'text' is Unicode, msg() will decode it
+        # and call umsg() to print it
+        _idaapi.msg(text)
+
+    def flush(self):
+        pass
+
+    def isatty(self):
+        return False
+
+# -----------------------------------------------------------------------
+def runscript(script):
+    """
+    Executes a script.
+    This function is present for backward compatiblity. Please use idaapi.IDAPython_ExecScript() instead
+
+    @param script: script path
+
+    @return: Error string or None on success
+    """
+
+    import idaapi
+    return idaapi.IDAPython_ExecScript(script, globals())
+
+# -----------------------------------------------------------------------
+def print_banner():
+    banner = [
+      "Python %s " % sys.version,
+      "IDAPython" + (" 64-bit" if __EA64__ else "") + " v%d.%d.%d %s (serial %d) (c) The IDAPython Team <idapython@googlegroups.com>" % IDAPYTHON_VERSION
+    ]
+    sepline = '-' * (max([len(s) for s in banner])+1)
+
+    print(sepline)
+    print("\n".join(banner))
+    print(sepline)
+
+# -----------------------------------------------------------------------
+
+# Redirect stderr and stdout to the IDA message window
+_orig_stdout = sys.stdout;
+_orig_stderr = sys.stderr;
+sys.stdout = sys.stderr = IDAPythonStdOut()
+
+# -----------------------------------------------------------------------
+# Initialize the help, with our own stdin wrapper, that'll query the user
+# -----------------------------------------------------------------------
+import pydoc
+class IDAPythonHelpPrompter:
+    def readline(self):
+        return idaapi.askstr(0, '', 'Help topic?')
+help = pydoc.Helper(input = IDAPythonHelpPrompter(), output = sys.stdout)
+
+# Assign a default sys.argv
+sys.argv = [""]
+
+# Have to make sure Python finds our modules
+sys.path.append(_idaapi.idadir("python"))
+
+# Remove current directory from the top of the patch search
+if '' in sys.path: # On non Windows, the empty path is added
+    sys.path.remove('')
+
+if os.getcwd() in sys.path:
+    sys.path.remove(os.getcwd())
+
+# ...and add it to the end if needed
+if not IDAPYTHON_REMOVE_CWD_SYS_PATH:
+    sys.path.append(os.getcwd())
+
+# Import all the required modules
+from idaapi import Choose, get_user_idadir, cvar, Choose2, Appcall, Form
+from idc      import *
+from idautils import *
+import idaapi
+
+# Load the users personal init file
+userrc = os.path.join(get_user_idadir(), "idapythonrc.py")
+if os.path.exists(userrc):
+    idaapi.IDAPython_ExecScript(userrc, globals())
+
+# All done, ready to rock.

raw-feature-extractor/cfg_constructor.py

@@ -2,8 +2,9 @@ import copy
 import networkx as nx
 from idautils import *
 from idaapi import *
+from idaapi import *
 from idc import *
-
+#from idc_bc695 import *
 import copy
 import networkx as nx
 from idautils import *
@@ -12,6 +13,8 @@ from idc import *
 from graph_analysis_ida import *
 
 
+
+
 def getCfg(func, externs_eas, ea_externs):
 	func_start = func.startEA
 	func_end = func.endEA
@@ -155,6 +158,15 @@ def attributingRe(cfg, externs_eas, ea_externs):
 		cfg.node[node_id]['externs'] = externs
 		numTIs = calTransferIns(bl)
 		cfg.node[node_id]['numTIs'] = numTIs
+		numCompare = calCompareIns(bl)
+		cfg.node[node_id]['numCom'] = numCompare
+		numMove = calMoveIns(bl)
+		cfg.node[node_id]['numMov'] = numMove
+		numTerm = calTerminationIns(bl)
+		cfg.node[node_id]['numTerm'] = numTerm
+		numDD = calDateDecIns(bl)
+		cfg.node[node_id]['numDD'] = numDD
+		
 
 
 def attributing(cfg):

raw-feature-extractor/discovRe.py

@@ -10,92 +10,95 @@ import cPickle as pickle
 import pdb
 from graph_analysis_ida import *
 from graph_property import *
+from cfg_constructor import cfg_construct
+
 #import wingdbstub
 #wingdbstub.Ensure()
 
 def get_funcs(ea):
-        funcs = {}
-        # Get current ea
-        # Loop from start to end in the current segment
+	funcs = {}
+		# Get current ea
+		# Loop from start to end in the current segment
+
 	for funcea in Functions(SegStart(ea)):
 		funcname = GetFunctionName(funcea)
 		func = get_func(funcea)
 		blocks = FlowChart(func)
 		funcs[funcname] = []
 		for bl in blocks:
-		        start = bl.startEA
-		        end = bl.endEA
-		        funcs[funcname].append((start, end))
-        return funcs
+				start = bl.startEA
+				end = bl.endEA
+				funcs[funcname].append((start, end))
+		return funcs
 
 def get_funcs_for_discoverRe(ea):
-    features = {}
-    for funcea in Functions(SegStart(ea)):
-        funcname = GetFunctionName(funcea)
-        print funcname
-        func = get_func(funcea)
-        feature = get_discoverRe_feature(func)
-        features[funcname] = feature
-    return features
+	features = {}
+	for funcea in Functions(SegStart(ea)):
+		funcname = GetFunctionName(funcea)
+		print funcname
+		func = get_func(funcea)
+		feature = get_discoverRe_feature(func)
+		features[funcname] = feature
+	return features
 
 def get_discoverRe_feature(func, icfg):
-    start = func.startEA
-    end = func.endEA
-    features = []
-    FunctionCalls = getFuncCalls(func)
-    #1
-    features.append(FunctionCalls)
-    LogicInstr = getLogicInsts(func)
-    #2
-    features.append(LogicInstr)
-    Transfer = getTransferInsts(func)
-    #3
-    features.append(Transfer)
-    Locals = getLocalVariables(func)
-    #4
-    features.append(Locals)
-    BB = getBasicBlocks(func)
-    #5
-    features.append(BB)
-    Edges = len(icfg.edges())
-    #6
-    features.append(Edges)
-    Incoming = getIncommingCalls(func)
-    #7
-    features.append(Incoming)
-    #8
-    Instrs = getIntrs(func)
-    features.append(Instrs)
-    between = retrieveGP(icfg)
-    #9
-    features.append(between)
+	start = func.startEA
+	end = func.endEA
+	features = []
+	FunctionCalls = getFuncCalls(func)
+	#1
+	features.append(FunctionCalls)
+	LogicInstr = getLogicInsts(func)
+	#2
+	features.append(LogicInstr)
+	Transfer = getTransferInsts(func)
+	#3
+	features.append(Transfer)
+	Locals = getLocalVariables(func)
+	#4
+	features.append(Locals)
+	BB = getBasicBlocks(func)
+	#5
+	features.append(BB)
+	Edges = len(icfg.edges())
+	#6
+	features.append(Edges)
+	Incoming = getIncommingCalls(func)
+	#7
+	features.append(Incoming)
+	#8
+	Instrs = getIntrs(func)
+	features.append(Instrs)
+	between = retrieveGP(icfg)
+	#9
+	features.append(between)
 
-    strings, consts = getfunc_consts(func)
-    features.append(strings)
-    features.append(consts)
-    return features
+	strings, consts = getfunc_consts(func)
+	features.append(strings)
+	features.append(consts)
+	return features
 
 def get_func_names(ea):
-    funcs = {}
-    for funcea in Functions(SegStart(ea)):
-            funcname = GetFunctionName(funcea)
-            funcs[funcname] = funcea
-    return funcs
+	funcs = {}
+	for funcea in Functions(SegStart(ea)):
+			funcname = GetFunctionName(funcea)
+			funcs[funcname] = funcea
+	return funcs
 
 def get_func_bases(ea):
-        funcs = {}
-        for funcea in Functions(SegStart(ea)):
-                funcname = GetFunctionName(funcea)
-                funcs[funcea] = funcname
-        return funcs
+		funcs = {}
+		for funcea in Functions(SegStart(ea)):
+				funcname = GetFunctionName(funcea)
+				funcs[funcea] = funcname
+		return funcs
 
 def get_func_range(ea):
-        funcs = {}
-        for funcea in Functions(SegStart(ea)):
-                funcname = GetFunctionName(funcea)
+	funcs = {}
+	for funcea in Functions(SegStart(ea)):
+		funcname = GetFunctionName(funcea)
 		func = get_func(funcea)
-                funcs[funcname] = (func.startEA, func.endEA)
-        return funcs
+		funcs[funcname] = (func.startEA, func.endEA)
+	return funcs
 
 def get_func_sequences(ea):
 	funcs_bodylist = {}
@@ -109,120 +112,120 @@ def get_func_sequences(ea):
 				opcode = GetMnem(inst_addr)
 				funcs_bodylist[funcname].append(opcode)
 				inst_addr = NextHead(inst_addr)
-        return funcs_bodylist
+		return funcs_bodylist
 
 def get_func_cfgs(ea):
-    func_cfglist = {}
-    i = 0
-    start, end = get_section('LOAD')
-    #print start, end
-    for funcea in Functions(SegStart(ea)):
-        if start <= funcea <= end:
-            funcname = GetFunctionName(funcea)
-            func = get_func(funcea)
-            print i
-            i += 1
-            try:
-                icfg = cfg.cfg_construct(func)
-                func_cfglist[funcname] = icfg
-            except:
-                pass
+	func_cfglist = {}
+	i = 0
+	start, end = get_section('LOAD')
+	#print start, end
+	for funcea in Functions(SegStart(ea)):
+		if start <= funcea <= end:
+			funcname = GetFunctionName(funcea)
+			func = get_func(funcea)
+			print i
+			i += 1
+			try:
+				icfg = cfg_construct(func)
+				func_cfglist[funcname] = icfg
+			except:
+				pass
 
-    return func_cfglist
+	return func_cfglist
 
 def get_section(t):
-    base = SegByName(t)
-    start = SegByBase(base)
-    end = SegEnd(start)
-    return start, end
+	base = SegByName(t)
+	start = SegByBase(base)
+	end = SegEnd(start)
+	return start, end
 
 
 def get_func_cfg_sequences(func_cfglist):
-    func_cfg_seqlist = {}
-    for funcname in func_cfglist:
-        func_cfg_seqlist[funcname] = {}
-        cfg = func_cfglist[funcname][0]
-        for start, end in cfg:
-            codesq = get_sequences(start, end)
-            func_cfg_seqlist[funcname][(start,end)] = codesq
+	func_cfg_seqlist = {}
+	for funcname in func_cfglist:
+		func_cfg_seqlist[funcname] = {}
+		cfg = func_cfglist[funcname][0]
+		for start, end in cfg:
+			codesq = get_sequences(start, end)
+			func_cfg_seqlist[funcname][(start,end)] = codesq
 
-    return func_cfg_seqlist
+	return func_cfg_seqlist
 
 
 def get_sequences(start, end):
-    seq = []
-    inst_addr = start
-    while inst_addr <= end:
-        opcode = GetMnem(inst_addr)
-        seq.append(opcode)
-        inst_addr = NextHead(inst_addr)
-    return seq
+	seq = []
+	inst_addr = start
+	while inst_addr <= end:
+		opcode = GetMnem(inst_addr)
+		seq.append(opcode)
+		inst_addr = NextHead(inst_addr)
+	return seq
 
 def get_stack_arg(func_addr):
-    print func_addr
-    args = []
-    stack = GetFrame(func_addr)
-    if not stack:
-            return []
-    firstM = GetFirstMember(stack)
-    lastM = GetLastMember(stack)
-    i = firstM
-    while i <=lastM:
-        mName = GetMemberName(stack,i)
-        mSize = GetMemberSize(stack,i)
-        if mSize:
-                i = i + mSize
-        else:
-                i = i+4
-        if mName not in args and mName and ' s' not in mName and ' r' not in mName:
-            args.append(mName)
-    return args
+	print func_addr
+	args = []
+	stack = GetFrame(func_addr)
+	if not stack:
+			return []
+	firstM = GetFirstMember(stack)
+	lastM = GetLastMember(stack)
+	i = firstM
+	while i <=lastM:
+		mName = GetMemberName(stack,i)
+		mSize = GetMemberSize(stack,i)
+		if mSize:
+				i = i + mSize
+		else:
+				i = i+4
+		if mName not in args and mName and ' s' not in mName and ' r' not in mName:
+			args.append(mName)
+	return args
 
-        #pickle.dump(funcs, open('C:/Documents and Settings/Administrator/Desktop/funcs','w'))
+		#pickle.dump(funcs, open('C:/Documents and Settings/Administrator/Desktop/funcs','w'))
 
 def processDataSegs():
-    funcdata = {}
-    datafunc = {}
-    for n in xrange(idaapi.get_segm_qty()):
-        seg = idaapi.getnseg(n)
-        ea = seg.startEA
-        segtype = idc.GetSegmentAttr(ea, idc.SEGATTR_TYPE)
-        if segtype in [idc.SEG_DATA, idc.SEG_BSS]:
-            start = idc.SegStart(ea)
-            end = idc.SegEnd(ea)
-            cur = start
-            while cur <= end:
-                refs = [v for v in DataRefsTo(cur)]
-                for fea in refs:
-                    name = GetFunctionName(fea)
-                    if len(name)== 0:
-                        continue
-                    if name not in funcdata:
-                        funcdata[name] = [cur]
-                    else:
-                        funcdata[name].append(cur)
-                    if cur not in datafunc:
-                        datafunc[cur] = [name]
-                    else:
-                        datafunc[cur].append(name)
-                cur = NextHead(cur)
-    return funcdata, datafunc
+	funcdata = {}
+	datafunc = {}
+	for n in xrange(idaapi.get_segm_qty()):
+		seg = idaapi.getnseg(n)
+		ea = seg.startEA
+		segtype = idc.GetSegmentAttr(ea, idc.SEGATTR_TYPE)
+		if segtype in [idc.SEG_DATA, idc.SEG_BSS]:
+			start = idc.SegStart(ea)
+			end = idc.SegEnd(ea)
+			cur = start
+			while cur <= end:
+				refs = [v for v in DataRefsTo(cur)]
+				for fea in refs:
+					name = GetFunctionName(fea)
+					if len(name)== 0:
+						continue
+					if name not in funcdata:
+						funcdata[name] = [cur]
+					else:
+						funcdata[name].append(cur)
+					if cur not in datafunc:
+						datafunc[cur] = [name]
+					else:
+						datafunc[cur].append(name)
+				cur = NextHead(cur)
+	return funcdata, datafunc
 
 def obtainDataRefs(callgraph):
-    datarefs = {}
-    funcdata, datafunc = processDataSegs()
-    for node in callgraph:
-        if node in funcdata:
-            datas = funcdata[node]
-            for dd in datas:
-                refs = datafunc[dd]
-                refs = list(set(refs))
-                if node in datarefs:
-                    print refs
-                    datarefs[node] += refs
-                    datarefs[node] = list(set(datarefs[node]))
-                else:
-                    datarefs[node] = refs
-    return datarefs
+	datarefs = {}
+	funcdata, datafunc = processDataSegs()
+	for node in callgraph:
+		if node in funcdata:
+			datas = funcdata[node]
+			for dd in datas:
+				refs = datafunc[dd]
+				refs = list(set(refs))
+				if node in datarefs:
+					print refs
+					datarefs[node] += refs
+					datarefs[node] = list(set(datarefs[node]))
+				else:
+					datarefs[node] = refs
+	return datarefs
 
 

raw-feature-extractor/func.py

@@ -204,8 +204,8 @@ def get_stack_arg(func_addr):
 def processExternalSegs():
 	funcdata = {}
 	datafunc = {}
-	for n in xrange(idaapi.get_segm_qty()):
-		seg = idaapi.getnseg(n)
+	for n in xrange(get_segm_qty()):
+		seg = getnseg(n)
 		ea = seg.startEA
 		segtype = idc.GetSegmentAttr(ea, idc.SEGATTR_TYPE)
 		if segtype in [idc.SEG_XTRN]:
@@ -221,8 +221,8 @@ def processExternalSegs():
 def processpltSegs():
 	funcdata = {}
 	datafunc = {}
-	for n in xrange(idaapi.get_segm_qty()):
-		seg = idaapi.getnseg(n)
+	for n in xrange(get_segm_qty()):
+		seg = getnseg(n)
 		ea = seg.startEA
 		segname = SegName(ea)
 		if segname in ['.plt', 'extern', '.MIPS.stubs']:
@@ -240,8 +240,8 @@ def processpltSegs():
 def processDataSegs():
 	funcdata = {}
 	datafunc = {}
-	for n in xrange(idaapi.get_segm_qty()):
-		seg = idaapi.getnseg(n)
+	for n in xrange(get_segm_qty()):
+		seg = getnseg(n)
 		ea = seg.startEA
 		segtype = idc.GetSegmentAttr(ea, idc.SEGATTR_TYPE)
 		if segtype in [idc.SEG_DATA, idc.SEG_BSS]:

raw-feature-extractor/graph_analysis_ida.py

@@ -1,257 +1,386 @@
+# coding=utf-8
 from idautils import *
 from idaapi import *
+
 from idc import *
 
+
 def getfunc_consts(func):
-	strings = []
-	consts = []
-	blocks = [(v.startEA, v.endEA) for v in FlowChart(func)]
-	for bl in blocks:
-		strs, conts = getBBconsts(bl)
-		strings += strs
-		consts += conts
-	return strings, consts
+    strings = []
+    consts = []
+    blocks = [(v.startEA, v.endEA) for v in FlowChart(func)]
+    for bl in blocks:
+        strs, conts = getBBconsts(bl)
+        strings += strs
+        consts += conts
+    return strings, consts
+
 
 def getConst(ea, offset):
-	strings = []
-	consts = []
-	optype1 = GetOpType(ea, offset)
-	if optype1 == idaapi.o_imm:
-		imm_value = GetOperandValue(ea, offset)
-		if 0<= imm_value <= 10:
-			consts.append(imm_value)
-		else:
-			if idaapi.isLoaded(imm_value) and idaapi.getseg(imm_value):
-				str_value = GetString(imm_value)
-				if str_value is None:
-					str_value = GetString(imm_value+0x40000)
-					if str_value is None:
-						consts.append(imm_value)
-					else:
-						re = all(40 <= ord(c) < 128 for c in str_value)
-						if re:
-							strings.append(str_value)
-						else:
-							consts.append(imm_value)
-				else:
-					re = all(40 <= ord(c) < 128 for c in str_value)
-					if re:
-						strings.append(str_value)
-					else:
-						consts.append(imm_value)
-			else:
-				consts.append(imm_value)
-	return strings, consts
+    strings = []
+    consts = []
+    optype1 = GetOpType(ea, offset)
+    if optype1 == o_imm:
+        imm_value = GetOperandValue(ea, offset)
+        if 0 <= imm_value <= 10:
+            consts.append(imm_value)
+        else:
+            if isLoaded(imm_value) and getseg(imm_value):
+                str_value = GetString(imm_value)
+                if str_value is None:
+                    str_value = GetString(imm_value + 0x40000)
+                    if str_value is None:
+                        consts.append(imm_value)
+                    else:
+                        re = all(40 <= ord(c) < 128 for c in str_value)
+                        if re:
+                            strings.append(str_value)
+                        else:
+                            consts.append(imm_value)
+                else:
+                    re = all(40 <= ord(c) < 128 for c in str_value)
+                    if re:
+                        strings.append(str_value)
+                    else:
+                        consts.append(imm_value)
+            else:
+                consts.append(imm_value)
+    return strings, consts
+
 
 def getBBconsts(bl):
-	strings = []
-	consts = []
-	start = bl[0]
-	end = bl[1]
-	invoke_num = 0
-	inst_addr = start
-	while inst_addr < end:
-		opcode = GetMnem(inst_addr)
-		if opcode in ['la','jalr','call', 'jal']:
-			inst_addr = NextHead(inst_addr)
-			continue
-		strings_src, consts_src = getConst(inst_addr, 0)
-		strings_dst, consts_dst = getConst(inst_addr, 1)
-		strings += strings_src
-		strings += strings_dst
-		consts += consts_src
-		consts += consts_dst
-		try:
-			strings_dst, consts_dst = getConst(inst_addr, 2)
-			consts += consts_dst
-			strings += strings_dst
-		except:
-			pass
+    strings = []
+    consts = []
+    start = bl[0]
+    end = bl[1]
+    invoke_num = 0
+    inst_addr = start
+    while inst_addr < end:
+        opcode = GetMnem(inst_addr)
+        if opcode in ['la', 'jalr', 'call', 'jal']:
+            inst_addr = NextHead(inst_addr)
+            continue
+        strings_src, consts_src = getConst(inst_addr, 0)
+        strings_dst, consts_dst = getConst(inst_addr, 1)
+        strings += strings_src
+        strings += strings_dst
+        consts += consts_src
+        consts += consts_dst
+        try:
+            strings_dst, consts_dst = getConst(inst_addr, 2)
+            consts += consts_dst
+            strings += strings_dst
+        except:
+            pass
+
+        inst_addr = NextHead(inst_addr)
+    return strings, consts
 
-		inst_addr = NextHead(inst_addr)
-	return strings, consts
 
 def getFuncCalls(func):
-	blocks = [(v.startEA, v.endEA) for v in FlowChart(func)]
-	sumcalls = 0
-	for bl in blocks:
-		callnum = calCalls(bl)
-		sumcalls += callnum
-	return sumcalls
+    blocks = [(v.startEA, v.endEA) for v in FlowChart(func)]
+    sumcalls = 0
+    for bl in blocks:
+        callnum = calCalls(bl)
+        sumcalls += callnum
+    return sumcalls
+
 
 def getLogicInsts(func):
-	blocks = [(v.startEA, v.endEA) for v in FlowChart(func)]
-	sumcalls = 0
-	for bl in blocks:
-		callnum = calLogicInstructions(bl)
-		sumcalls += callnum
-	return sumcalls
+    blocks = [(v.startEA, v.endEA) for v in FlowChart(func)]
+    sumcalls = 0
+    for bl in blocks:
+        callnum = calLogicInstructions(bl)
+        sumcalls += callnum
+    return sumcalls
+
 
 def getTransferInsts(func):
-	blocks = [(v.startEA, v.endEA) for v in FlowChart(func)]
-	sumcalls = 0
-	for bl in blocks:
-		callnum = calTransferIns(bl)
-		sumcalls += callnum
-	return sumcalls
+    blocks = [(v.startEA, v.endEA) for v in FlowChart(func)]
+    sumcalls = 0
+    for bl in blocks:
+        callnum = calTransferIns(bl)
+        sumcalls += callnum
+    return sumcalls
+
 
 def getIntrs(func):
-	blocks = [(v.startEA, v.endEA) for v in FlowChart(func)]
-	sumcalls = 0
-	for bl in blocks:
-		callnum = calInsts(bl)
-		sumcalls += callnum
-	return sumcalls	
+    blocks = [(v.startEA, v.endEA) for v in FlowChart(func)]
+    sumcalls = 0
+    for bl in blocks:
+        callnum = calInsts(bl)
+        sumcalls += callnum
+    return sumcalls
+
 
 def getLocalVariables(func):
-	args_num = get_stackVariables(func.startEA)
-	return args_num
+    args_num = get_stackVariables(func.startEA)
+    return args_num
+
 
 def getBasicBlocks(func):
-	blocks = [(v.startEA, v.endEA) for v in FlowChart(func)]
-	return len(blocks)
+    blocks = [(v.startEA, v.endEA) for v in FlowChart(func)]
+    return len(blocks)
+
 
 def getIncommingCalls(func):
-	refs = CodeRefsTo(func.startEA, 0)
-	re = len([v for v in refs])
-	return re
+    refs = CodeRefsTo(func.startEA, 0)
+    re = len([v for v in refs])
+    return re
 
 
 def get_stackVariables(func_addr):
-    #print func_addr
+    # print func_addr
     args = []
     stack = GetFrame(func_addr)
     if not stack:
-            return 0
+        return 0
     firstM = GetFirstMember(stack)
     lastM = GetLastMember(stack)
     i = firstM
-    while i <=lastM:
-        mName = GetMemberName(stack,i)
-        mSize = GetMemberSize(stack,i)
+    while i <= lastM:
+        mName = GetMemberName(stack, i)
+        mSize = GetMemberSize(stack, i)
         if mSize:
-                i = i + mSize
+            i = i + mSize
         else:
-                i = i+4
+            i = i + 4
         if mName not in args and mName and 'var_' in mName:
             args.append(mName)
     return len(args)
 
 
-
 def calArithmeticIns(bl):
-	x86_AI = {'add':1, 'sub':1, 'div':1, 'imul':1, 'idiv':1, 'mul':1, 'shl':1, 'dec':1, 'inc':1}
-	mips_AI = {'add':1, 'addu':1, 'addi':1, 'addiu':1, 'mult':1, 'multu':1, 'div':1, 'divu':1}
-	calls = {}
-	calls.update(x86_AI)
-	calls.update(mips_AI)
-	start = bl[0]
-	end = bl[1]
-	invoke_num = 0
-	inst_addr = start
-	while inst_addr < end:
-		opcode = GetMnem(inst_addr)
-		if opcode in calls:
-			invoke_num += 1
-		inst_addr = NextHead(inst_addr)
-	return invoke_num
+    """
+    基本块算术指令统计
+    """
+    x86_AI = {'add': 1, 'sub': 1, 'div': 1, 'imul': 1, 'idiv': 1, 'mul': 1, 'shl': 1, 'dec': 1, 'inc': 1}
+    mips_AI = {'add': 1, 'addu': 1, 'addi': 1, 'addiu': 1, 'mult': 1, 'multu': 1, 'div': 1, 'divu': 1}
+    calls = {}
+    calls.update(x86_AI)
+    calls.update(mips_AI)
+    start = bl[0]
+    end = bl[1]
+    invoke_num = 0
+    inst_addr = start
+    while inst_addr < end:
+        opcode = GetMnem(inst_addr)
+        if opcode in calls:
+            invoke_num += 1
+        inst_addr = NextHead(inst_addr)
+    return invoke_num
+
 
 def calCalls(bl):
-	calls = {'call':1, 'jal':1, 'jalr':1}
-	start = bl[0]
-	end = bl[1]
-	invoke_num = 0
-	inst_addr = start
-	while inst_addr < end:
-		opcode = GetMnem(inst_addr)
-		if opcode in calls:
-			invoke_num += 1
-		inst_addr = NextHead(inst_addr)
-	return invoke_num
+    """
+    基本快转移指令
+    :param bl:
+    :return:
+    """
+    calls = {'call': 1, 'jal': 1, 'jalr': 1}
+    start = bl[0]
+    end = bl[1]
+    invoke_num = 0
+    inst_addr = start
+    while inst_addr < end:
+        opcode = GetMnem(inst_addr)
+        if opcode in calls:
+            invoke_num += 1
+        inst_addr = NextHead(inst_addr)
+    return invoke_num
+
 
 def calInsts(bl):
-	start = bl[0]
-	end = bl[1]
-	ea = start
-	num = 0
-	while ea < end:
-		num += 1
-		ea = NextHead(ea)
-	return num
+    """
+    基本快指令数
+    :param bl:
+    :return:
+    """
+    start = bl[0]
+    end = bl[1]
+    ea = start
+    num = 0
+    while ea < end:
+        num += 1
+        ea = NextHead(ea)
+    return num
+
 
 def calLogicInstructions(bl):
-	x86_LI = {'and':1, 'andn':1, 'andnpd':1, 'andpd':1, 'andps':1, 'andnps':1, 'test':1, 'xor':1, 'xorpd':1, 'pslld':1}
-	mips_LI = {'and':1, 'andi':1, 'or':1, 'ori':1, 'xor':1, 'nor':1, 'slt':1, 'slti':1, 'sltu':1}
-	calls = {}
-	calls.update(x86_LI)
-	calls.update(mips_LI)
-	start = bl[0]
-	end = bl[1]
-	invoke_num = 0
-	inst_addr = start
-	while inst_addr < end:
-		opcode = GetMnem(inst_addr)
-		if opcode in calls:
-			invoke_num += 1
-		inst_addr = NextHead(inst_addr)
-	return invoke_num
+    """
+    基本快逻辑运算
+    :param bl:
+    :return:
+    """
+    x86_LI = {'and': 1, 'andn': 1, 'andnpd': 1, 'andpd': 1, 'andps': 1, 'andnps': 1, 'test': 1, 'xor': 1, 'xorpd': 1,
+              'pslld': 1}
+    mips_LI = {'and': 1, 'andi': 1, 'or': 1, 'ori': 1, 'xor': 1, 'nor': 1, 'slt': 1, 'slti': 1, 'sltu': 1}
+    calls = {}
+    calls.update(x86_LI)
+    calls.update(mips_LI)
+    start = bl[0]
+    end = bl[1]
+    invoke_num = 0
+    inst_addr = start
+    while inst_addr < end:
+        opcode = GetMnem(inst_addr)
+        if opcode in calls:
+            invoke_num += 1
+        inst_addr = NextHead(inst_addr)
+    return invoke_num
+
 
 def calSconstants(bl):
-	start = bl[0]
-	end = bl[1]
-	invoke_num = 0
-	inst_addr = start
-	while inst_addr < end:
-		opcode = GetMnem(inst_addr)
-		if opcode in calls:
-			invoke_num += 1
-		inst_addr = NextHead(inst_addr)
-	return invoke_num
+    """
+    基本快字符串常量
+    :param bl:
+    :return:
+    """
+    calls = {}
+    start = bl[0]
+    end = bl[1]
+    invoke_num = 0
+    inst_addr = start
+    while inst_addr < end:
+        opcode = GetMnem(inst_addr)
+        if opcode in calls:
+            invoke_num += 1
+        inst_addr = NextHead(inst_addr)
+    return invoke_num
 
 
 def calNconstants(bl):
-	start = bl[0]
-	end = bl[1]
-	invoke_num = 0
-	inst_addr = start
-	while inst_addr < end:
-		optype1 = GetOpType(inst_addr, 0)
-		optype2 = GetOpType(inst_addr, 1)
-		if optype1 == 5 or optype2 == 5:
-			invoke_num += 1
-		inst_addr = NextHead(inst_addr)
-	return invoke_num
+    """
+    基本快整数常量
+    :param bl:
+    :return:
+    """
+    start = bl[0]
+    end = bl[1]
+    invoke_num = 0
+    inst_addr = start
+    while inst_addr < end:
+        optype1 = GetOpType(inst_addr, 0)
+        optype2 = GetOpType(inst_addr, 1)
+        if optype1 == 5 or optype2 == 5:
+            invoke_num += 1
+        inst_addr = NextHead(inst_addr)
+    return invoke_num
+
 
 def retrieveExterns(bl, ea_externs):
-	externs = []
-	start = bl[0]
-	end = bl[1]
-	inst_addr = start
-	while inst_addr < end:
-		refs = CodeRefsFrom(inst_addr, 1)
-		try:
-			ea = [v for v in refs if v in ea_externs][0]
-			externs.append(ea_externs[ea])
-		except:
-			pass
-		inst_addr = NextHead(inst_addr)
-	return externs
+    externs = []
+    start = bl[0]
+    end = bl[1]
+    inst_addr = start
+    while inst_addr < end:
+        refs = CodeRefsFrom(inst_addr, 1)
+        try:
+            ea = [v for v in refs if v in ea_externs][0]
+            externs.append(ea_externs[ea])
+        except:
+            pass
+        inst_addr = NextHead(inst_addr)
+    return externs
+
 
 def calTransferIns(bl):
-	x86_TI = {'jmp':1, 'jz':1, 'jnz':1, 'js':1, 'je':1, 'jne':1, 'jg':1, 'jle':1, 'jge':1, 'ja':1, 'jnc':1, 'call':1}
-	mips_TI = {'beq':1, 'bne':1, 'bgtz':1, "bltz":1, "bgez":1, "blez":1, 'j':1, 'jal':1, 'jr':1, 'jalr':1}
-	arm_TI = {'MVN':1, "MOV":1}
-	calls = {}
-	calls.update(x86_TI)
-	calls.update(mips_TI)
-	start = bl[0]
-	end = bl[1]
-	invoke_num = 0
-	inst_addr = start
-	while inst_addr < end:
-		opcode = GetMnem(inst_addr)
-		re = [v for v in calls if opcode in v]
-		if len(re) > 0:
-			invoke_num += 1
-		inst_addr = NextHead(inst_addr)
-	return invoke_num
+    """
+    基本快转移指令
+    :param bl:
+    :return:
+    """
+    x86_TI = {'jmp': 1, 'jz': 1, 'jnz': 1, 'js': 1, 'je': 1, 'jne': 1, 'jg': 1, 'jle': 1, 'jge': 1, 'ja': 1, 'jnc': 1,
+              'call': 1}
+    mips_TI = {'beq': 1, 'bne': 1, 'bgtz': 1, "bltz": 1, "bgez": 1, "blez": 1, 'j': 1, 'jal': 1, 'jr': 1, 'jalr': 1}
+    arm_TI = {'MVN': 1, "MOV": 1}
+    calls = {}
+    calls.update(x86_TI)
+    calls.update(mips_TI)
+    start = bl[0]
+    end = bl[1]
+    invoke_num = 0
+    inst_addr = start
+    while inst_addr < end:
+        opcode = GetMnem(inst_addr)
+        re = [v for v in calls if opcode in v]
+        if len(re) > 0:
+            invoke_num += 1
+        inst_addr = NextHead(inst_addr)
+    return invoke_num
+
+
+def calCompareIns(bl):
+    """
+    基本快比较指令
+    :param bl:
+    :return:
+    """
+    calls = {'cmp': 1}
+    start = bl[0]
+    end = bl[1]
+    invoke_num = 0
+    inst_addr = start
+    while inst_addr < end:
+        opcode = GetMnem(inst_addr)
+        if opcode in calls:
+            invoke_num += 1
+        inst_addr = NextHead(inst_addr)
+    return invoke_num
+
+
+def calMoveIns(bl):
+    """
+    基本快移动指令
+    :param bl:
+    :return:
+    """
+    calls = {'mov': 1, 'lea': 1, 'xchg': 1}
+    start = bl[0]
+    end = bl[1]
+    invoke_num = 0
+    inst_addr = start
+    while inst_addr < end:
+        opcode = GetMnem(inst_addr)
+        if opcode in calls:
+            invoke_num += 1
+        inst_addr = NextHead(inst_addr)
+    return invoke_num
+
+
+def calTerminationIns(bl):
+    """
+    基本块终止指令
+    :param bl:
+    :return:
+    """
+    calls = {'ret': 1, 'retn': 1, 'hlt': 1, 'sys_exit': 1}
+    start = bl[0]
+    end = bl[1]
+    invoke_num = 0
+    inst_addr = start
+    while inst_addr < end:
+        opcode = GetMnem(inst_addr)
+        if opcode in calls:
+            invoke_num += 1
+        inst_addr = NextHead(inst_addr)
+    return invoke_num
+
+
+def calDateDecIns(bl):
+    """
+    基本快数据声明
+    :param bl:
+    :return:
+    """
+    calls = {'db': 1, 'dw': 1, 'dd': 1, 'dq': 1}
+    start = bl[0]
+    end = bl[1]
+    invoke_num = 0
+    inst_addr = start
+    while inst_addr < end:
+        opcode = GetMnem(inst_addr)
+        if opcode in calls:
+            invoke_num += 1
+        inst_addr = NextHead(inst_addr)
+    return invoke_num

raw-feature-extractor/preprocessing_ida.py

@@ -4,24 +4,24 @@ from idc import *
 import os
 import argparse
 
+
 def parse_command():
-	parser = argparse.ArgumentParser(description='Process some integers.')
-	parser.add_argument("--path", type=str, help="The directory where to store the generated .ida file")
-	args = parser.parse_args()
-	return args
+    parser = argparse.ArgumentParser(description='Process some integers.')
+    parser.add_argument("--path", type=str, help="The directory where to store the generated .ida file")
+    args = parser.parse_args()
+    return args
+
 
 if __name__ == '__main__':
-
-	args = parse_command()
-	path = args.path
-	analysis_flags = idc.GetShortPrm(idc.INF_START_AF)
-	analysis_flags &= ~idc.AF_IMMOFF
-	# turn off "automatically make offset" heuristic
-	idc.SetShortPrm(idc.INF_START_AF, analysis_flags)
-	idaapi.autoWait()
-	cfgs = get_func_cfgs_c(FirstSeg())
-	binary_name = idc.GetInputFile() + '.ida'
-	fullpath = os.path.join(path, binary_name)
-	pickle.dump(cfgs, open(fullpath,'w'))
-	print binary_name
-	idc.Exit(0)
+    args = parse_command()
+    path = idc.ARGV[2]
+    analysis_flags = idc.GetShortPrm(idc.INF_START_AF)
+    analysis_flags &= ~idc.AF_IMMOFF
+    # turn off "automatically make offset" heuristic
+    idc.SetShortPrm(idc.INF_START_AF, analysis_flags)
+    idaapi.autoWait()
+    cfgs = get_func_cfgs_c(FirstSeg())
+    binary_name = idc.GetInputFile() + '.ida'
+    fullpath = os.path.join(path, binary_name)
+    pickle.dump(cfgs, open(fullpath, 'w'))
+    idc.Exit(0)

raw-feature-extractor/print_test.py

@@ -0,0 +1,97 @@
+# -*- coding: UTF-8 -*-
+import sys
+from matplotlib import pyplot as plt
+import networkx as nx
+
+import hashlib
+import json
+
+
+def print_obj(obj):
+    "打印对象的所有属性"
+    print(obj.__dict__)
+
+
+def calc_sha256(file_path):
+    with open(file_path, 'rb') as f:
+        bytes = f.read()
+        sha256obj = hashlib.sha256(bytes)
+        sha256 = sha256obj.hexdigest()
+        return sha256
+
+
+import pickle
+
+# sub_10F20 308  反编译代码有字符串，但是这个特征提取里没有字符串 constant，可能是间接引用的，不识别。看了下所有函数的特征，几乎都没有字符串常量，可能都是写在别的地方然后引用的。
+# sub_166C4 393
+if __name__ == '__main__':
+    file_path = '../3c580f5beca53b6599e5f04d3aa68a34bd50521d7ec5d7163849eb69f53a4150.exe'
+    testpath = '../store/3c580f5beca53b6599e5f04d3aa68a34bd50521d7ec5d7163849eb69f53a4150.exe.ida'
+    fr = open(testpath, 'r')
+    data1 = pickle.load(fr)  # 一个二进制文件的acfgs
+    # function_edges
+    function_edge_start = []
+    function_edge_end = []
+    for item in data1.raw_graph_list[0].old_g.edges:
+        function_edge_start.append(item[0])
+        function_edge_end.append(item[1])
+    function_edges = [function_edge_start, function_edge_end]
+    fun_name_temp = []
+    # function hsah
+    file_hash = calc_sha256(file_path)
+    # funtion num
+    function_number = len(data1.raw_graph_list)
+    acfg_list = []
+    # 函数级特征
+    for i in range(len(data1.raw_graph_list)):
+
+        # function name
+        fun_name_temp.append(data1.raw_graph_list[i].funcname)
+        # block features
+        temp_G = data1.raw_graph_list[i].old_g
+        # block_number
+        block_number = len(temp_G.node)
+        # block_features
+        acfg_list_item_feature = []
+        for temp in range(len(temp_G.node)):
+            block_features = []
+            # call
+            block_features.append(temp_G.node[temp]['numCalls'])
+            # transfer
+            block_features.append(temp_G.node[temp]['numTIs'])
+            # arithmetic
+            block_features.append(temp_G.node[temp]['numAs'])
+            # logic
+            block_features.append(temp_G.node[temp]['numLIs'])
+            # compare
+            block_features.append(temp_G.node[temp]['numCom'])
+            # move
+            block_features.append(temp_G.node[temp]['numMov'])
+            # termination
+            block_features.append(temp_G.node[temp]['numTerm'])
+            # date declaration
+            block_features.append(temp_G.node[temp]['numDD'])
+            # total instructions
+            block_features.append(temp_G.node[temp]['numIns'])
+            # string or integer constants
+            block_features.append(len(temp_G.node[temp]['strings']) if len(temp_G.node[temp]['strings']) != 0 else len(
+                temp_G.node[temp]['consts']))
+            # offspring
+            block_features.append(temp_G.node[temp]['offs'])
+            acfg_list_item_feature.append(block_features)
+        edge_list_start = []
+        edge_list_end = []
+        for item in temp_G.edges:
+            edge_list_start.append(item[0])
+            edge_list_end.append(item[1])
+        block_edges = [edge_list_start, edge_list_end]
+        acfg_list_item = {"block_number": block_number, "block_edges": block_edges, "block_features": acfg_list_item_feature}
+        acfg_list.append(acfg_list_item)
+
+    json_temp = {"function_edges": function_edges, "acfg_list": acfg_list, "function_names": fun_name_temp, "hash": file_hash, "function_number": function_number}
+    json_str = json.dumps(json_temp)
+    print json_str
+
+
+
+    

raw-feature-extractor/raw_graphs.py

@@ -1,286 +1,298 @@
 import itertools
 import sys
-sys.path.insert(0, '/usr/local/lib/python2.7/dist-packages/')
+
+sys.path.insert(1, 'C:/Python27/Lib/site-packages')
 import networkx as nx
-#import numpy as np
+# import numpy as np
 from subprocess import Popen, PIPE
 import pdb
 import os
-import re,mmap
-#from graph_edit_new import *
+import re, mmap
+
+
+# from graph_edit_new import *
 
 class raw_graph:
-	def __init__(self, funcname, g, func_f):
-		self.funcname = funcname
-		self.old_g = g[0]
-		self.g = nx.DiGraph()
-		self.entry = g[1]
-		self.fun_features = func_f
-		self.attributing()
+    def __init__(self, funcname, g, func_f):
+        self.funcname = funcname
+        self.old_g = g[0]
+        self.g = nx.DiGraph()
+        self.entry = g[1]
+        self.fun_features = func_f
+        self.attributing()
 
-	def __len__(self):
-		return len(self.g)
+    def __len__(self):
+        return len(self.g)
 
-	def attributing(self):
-		self.obtainOffsprings(self.old_g)
-		for node in self.old_g:
-			fvector = self.retrieveVec(node, self.old_g)
-			self.g.add_node(node)
-			self.g.node[node]['v'] = fvector
+    def attributing(self):
+        self.obtainOffsprings(self.old_g)
+        for node in self.old_g:
+            fvector = self.retrieveVec(node, self.old_g)
+            self.g.add_node(node)
+            self.g.node[node]['v'] = fvector
 
-		for edge in self.old_g.edges():
-			node1 = edge[0]
-			node2 = edge[1]
-			self.g.add_edge(node1, node2)
+        for edge in self.old_g.edges():
+            node1 = edge[0]
+            node2 = edge[1]
+            self.g.add_edge(node1, node2)
 
-	def obtainOffsprings(self,g):
-		nodes = g.nodes()
-		for node in nodes:
-			offsprings = {}
-			self.getOffsprings(g, node, offsprings)
-			g.node[node]['offs'] = len(offsprings)
-		return g
+    def obtainOffsprings(self, g):
+        nodes = g.nodes()
+        for node in nodes:
+            offsprings = {}
+            self.getOffsprings(g, node, offsprings)
+            g.node[node]['offs'] = len(offsprings)
+        return g
 
-	def getOffsprings(self, g, node, offsprings):
-		node_offs = 0
-		sucs = g.successors(node)
-		for suc in sucs:
-			if suc not in offsprings:
-				offsprings[suc] = 1
-				self.getOffsprings(g, suc, offsprings)
+    def getOffsprings(self, g, node, offsprings):
+        node_offs = 0
+        sucs = g.successors(node)
+        for suc in sucs:
+            if suc not in offsprings:
+                offsprings[suc] = 1
+                self.getOffsprings(g, suc, offsprings)
 
-	def retrieveVec(self, id_, g):
-		feature_vec = []
-		#numC0
-		numc = g.node[id_]['consts']
-		feature_vec.append(numc)
-		#nums1
-		nums = g.node[id_]['strings']
-		feature_vec.append(nums)
-		#offsprings2
-		offs = g.node[id_]['offs']
-		feature_vec.append(offs)
-		#numAs3
-		numAs = g.node[id_]['numAs']
-		feature_vec.append(numAs)
-		# of calls4
-		calls = g.node[id_]['numCalls']
-		feature_vec.append(calls)
-		# of insts5
-		insts = g.node[id_]['numIns']
-		feature_vec.append(insts)
-		# of LIs6
-		insts = g.node[id_]['numLIs']
-		feature_vec.append(insts)
-		# of TIs7
-		insts = g.node[id_]['numTIs']
-		feature_vec.append(insts)	
-		return feature_vec
+    def retrieveVec(self, id_, g):
+        feature_vec = []
+        # numC0
+        numc = g.node[id_]['consts']
+        feature_vec.append(numc)
+        # nums1
+        nums = g.node[id_]['strings']
+        feature_vec.append(nums)
+        # offsprings2
+        offs = g.node[id_]['offs']
+        feature_vec.append(offs)
+        # numAs3
+        numAs = g.node[id_]['numAs']
+        feature_vec.append(numAs)
+        # of calls4
+        calls = g.node[id_]['numCalls']
+        feature_vec.append(calls)
+        # of insts5
+        insts = g.node[id_]['numIns']
+        feature_vec.append(insts)
+        # of LIs6
+        insts = g.node[id_]['numLIs']
+        feature_vec.append(insts)
+        # of TIs7
+        insts = g.node[id_]['numTIs']
+        feature_vec.append(insts)
 
+        feature_vec.append(g.node[id_]['numCom'])
+        feature_vec.append(g.node[id_]['numMov'])
+        feature_vec.append(g.node[id_]['numTerm'])
+        feature_vec.append(g.node[id_]['numDD'])
+        return feature_vec
 
-	def enumerating(self, n):
-		subgs = []
-		#pdb.set_trace()
-		for sub_nodes in itertools.combinations(self.g.nodes(), n):
-		    subg = self.g.subgraph(sub_nodes)
-		    u_subg = subg.to_undirected()
-		    if nx.is_connected(u_subg):
-		        subgs.append(subg)
-		return subgs
+    def enumerating(self, n):
+        subgs = []
+        # pdb.set_trace()
+        for sub_nodes in itertools.combinations(self.g.nodes(), n):
+            subg = self.g.subgraph(sub_nodes)
+            u_subg = subg.to_undirected()
+            if nx.is_connected(u_subg):
+                subgs.append(subg)
+        return subgs
 
+    def genMotifs(self, n):
+        motifs = {}
+        subgs = self.enumerating(n)
+        for subg in subgs:
+            if len(motifs) == 0:
+                motifs[subg] = [subg]
+            else:
+                nomatch = True
+                for mt in motifs:
+                    if nx.is_isomorphic(mt, subg):
+                        motifs[mt].append(subg)
+                        nomatch = False
+                if nomatch:
+                    motifs[subg] = [subg]
+        return motifs
 
-	def genMotifs(self, n):
-		motifs = {}
-		subgs = enumerating(n)
-		for subg in subgs:
-			if len(motifs) == 0:
-				motifs[subg] = [subg]
-			else:
-				nomatch = True
-				for mt in motifs:
-					if nx.is_isomorphic(mt, subg):
-						motifs[mt].append(subg)
-						nomatch = False
-				if nomatch:
-					motifs[subg] = [subg]
-		return motifs
+    def enumerating_efficient(self, n):
+        # pdb.set_trace()
+        if len(self.g) >= 200:
+            return []
+        with open('/home/qian/workspace/gEnding/gencoding/encoding/labeled/data/preprocessing/OUTPUT.txt', 'wb') as f:
+            nx.write_edgelist(self.g, f, data=False)
+        # pdb.set_trace()
+        process = Popen(
+            ["/home/qian/workspace/FANMOD-command_line-source/executables/./fanmod_command_line_linux", str(n),
+             "100000", "1", "/home/qian/workspace/gEnding/gencoding/encoding/labeled/data/preprocessing/OUTPUT.txt",
+             "1", "0", "0", "2", "0", "0", "0", "1000", "3", "3",
+             "/home/qian/workspace/gEnding/gencoding/encoding/labeled/data/preprocessing/MotifCount.txt", "0", "1"],
+            stdout=PIPE, stderr=PIPE)
+        stdout, stderr = process.communicate()
+        if process.returncode >= 0:
+            # os.system("/home/qian/software/FANMOD-command_line-source/executables/./fanmod_command_line_linux " +str(n) + " 100000 1 /home/qian/workspace/gEnding/gencoding/encoding/labeled/data/preprocessing/OUTPUT.txt 1 0 0 2 0 0 0 1000 3 3 /home/qian/workspace/gEnding/gencoding/encoding/labeled/data/preprocessing/MotifCount.txt 0 1")
+            # pdb.set_trace()
+            # pdb.set_trace()
+            subgs = self.parseOutput(
+                "/home/qian/workspace/gEnding/gencoding/encoding/labeled/data/preprocessing/MotifCount.txt.dump", n)
+            # pdb.set_trace()
+            os.remove("/home/qian/workspace/gEnding/gencoding/encoding/labeled/data/preprocessing/MotifCount.txt.dump")
+            return subgs
+        return []
 
-	def enumerating_efficient(self, n):
-		#pdb.set_trace()
-		if len(self.g) >= 200:
-			return []
-		with open('/home/qian/workspace/gEnding/gencoding/encoding/labeled/data/preprocessing/OUTPUT.txt','wb') as f:
-			nx.write_edgelist(self.g,f,data=False)
-		#pdb.set_trace()
-		process = Popen(["/home/qian/workspace/FANMOD-command_line-source/executables/./fanmod_command_line_linux", str(n), "100000", "1", "/home/qian/workspace/gEnding/gencoding/encoding/labeled/data/preprocessing/OUTPUT.txt", "1", "0", "0", "2", "0", "0", "0", "1000", "3", "3", "/home/qian/workspace/gEnding/gencoding/encoding/labeled/data/preprocessing/MotifCount.txt", "0", "1"], stdout=PIPE, stderr=PIPE)
-		stdout, stderr = process.communicate()
-		if process.returncode >= 0:
-		#os.system("/home/qian/software/FANMOD-command_line-source/executables/./fanmod_command_line_linux " +str(n) + " 100000 1 /home/qian/workspace/gEnding/gencoding/encoding/labeled/data/preprocessing/OUTPUT.txt 1 0 0 2 0 0 0 1000 3 3 /home/qian/workspace/gEnding/gencoding/encoding/labeled/data/preprocessing/MotifCount.txt 0 1")
-		#pdb.set_trace()
-			#pdb.set_trace()
-			subgs = self.parseOutput("/home/qian/workspace/gEnding/gencoding/encoding/labeled/data/preprocessing/MotifCount.txt.dump", n)
-			#pdb.set_trace()
-			os.remove("/home/qian/workspace/gEnding/gencoding/encoding/labeled/data/preprocessing/MotifCount.txt.dump")
-			return subgs
-		return []
+    def parseOutput(self, path, n):
+        pattern = re.compile('[0-9]+\,[0-9]+\,[0-9]+\,[0-9]+')
+        subgraphs = []
+        with open(path, 'r') as f:
+            data = mmap.mmap(f.fileno(), 0, prot=mmap.PROT_READ)
+            mo = re.findall(pattern, data)
+            if mo:
+                results = [map(int, v.split(',')[1:]) for v in mo]
+                subgraphs = self.createGraphDirectly(results)
+        return subgraphs
 
-	def parseOutput(self, path, n):
-		pattern = re.compile('[0-9]+\,[0-9]+\,[0-9]+\,[0-9]+')
-		subgraphs = []
-		with open(path,'r') as f:
-			data = mmap.mmap(f.fileno(), 0, prot=mmap.PROT_READ)
-			mo = re.findall(pattern, data)
-			if mo:
-				results = [map(int, v.split(',')[1:]) for v in mo]
-				subgraphs = self.createGraphDirectly(results)
-		return subgraphs
+    def parseOutputByconditions(self, path, n):
+        pattern = re.compile('[0-9]+\,[0-9]+\,[0-9]+\,[0-9]+')
+        subgraphs = []
+        with open(path, 'r') as f:
+            data = mmap.mmap(f.fileno(), 0, prot=mmap.PROT_READ)
+            mo = re.findall(pattern, data)
+            if mo:
+                results = [map(int, v.split(',')[1:]) for v in mo]
+                subgraphs = self.create_Graphbycondition_Directly(results)
+        return subgraphs
 
-	def parseOutputByconditions(self, path, n):
-		pattern = re.compile('[0-9]+\,[0-9]+\,[0-9]+\,[0-9]+')
-		subgraphs = []
-		with open(path,'r') as f:
-			data = mmap.mmap(f.fileno(), 0, prot=mmap.PROT_READ)
-			mo = re.findall(pattern, data)
-			if mo:
-				results = [map(int, v.split(',')[1:]) for v in mo]
-				subgraphs = self.create_Graphbycondition_Directly(results)
-		return subgraphs
+    def create_Graphbycondition_Directly(self, results):
+        subgs = []
+        for indexes in results:
+            tg = template_graph()
+            subg = self.g.subgraph(indexes)
+            tg.updateG(subg)
+            subgs.append(tg)
+            del tg
+        return subgs
 
-	def create_Graphbycondition_Directly(self, results):
-		subgs = []
-		for indexes in results:
-			tg = template_graph()
-			subg = self.g.subgraph(indexes)
-			tg.updateG(subg)
-			subgs.append(tg)
-			del tg
-		return subgs
+    def createGraphDirectly(self, results):
+        # pdb.set_trace()
+        # subgs = [self.g.subgraph(indexes) for indexes in results]
+        subgs = []
+        for indexes in results:
+            tg = template_graph()
+            subg = self.g.subgraph(indexes)
+            tg.updateG(subg)
+            subgs.append(tg)
+            del tg
+        return subgs
 
-	def createGraphDirectly(self, results):
-		#pdb.set_trace()
-		#subgs = [self.g.subgraph(indexes) for indexes in results]
-		subgs = []
-		for indexes in results:
-			tg = template_graph()
-			subg = self.g.subgraph(indexes)
-			tg.updateG(subg)
-			subgs.append(tg)
-			del tg
-		return subgs
-
-	def createGraph(self, results, n):
-		binary_value = int(results[0],2)
-		indexes = [int(v) for v in results[1:]]
-		fang = self.createG(results[0], n)
-		if fang:
-			tg = template_graph(binary_value)
-			tg.updateG(fang, indexes, self.g)
-			return tg
-		pdb.set_trace()
-		print "there is g which is none"
-
-	def createG(self, binary_str, n):
-		g = nx.DiGraph()
-		l = [int(v) for v in binary_str]
-		#pdb.set_trace()
-		shape = (n, n)
-		data = np.array(l)
-		ad_matrix = data.reshape(shape)
-		for i in xrange(n):
-			for j in xrange(n):
-				if ad_matrix[i][j] == 1:
-					g.add_edge(i, j)
-		return g
+    def createGraph(self, results, n):
+        binary_value = int(results[0], 2)
+        indexes = [int(v) for v in results[1:]]
+        fang = self.createG(results[0], n)
+        if fang:
+            tg = template_graph(binary_value)
+            tg.updateG(fang, indexes, self.g)
+            return tg
+        pdb.set_trace()
+        print "there is g which is none"
 
+    def createG(self, binary_str, n):
+        g = nx.DiGraph()
+        l = [int(v) for v in binary_str]
+        # pdb.set_trace()
+        shape = (n, n)
+        data = np.array(l)
+        ad_matrix = data.reshape(shape)
+        for i in xrange(n):
+            for j in xrange(n):
+                if ad_matrix[i][j] == 1:
+                    g.add_edge(i, j)
+        return g
 
 
 class raw_graphs:
-	def __init__(self, binary_name):
-		self.binary_name = binary_name
-		self.raw_graph_list = []
+    def __init__(self, binary_name):
+        self.binary_name = binary_name
+        self.raw_graph_list = []
 
-	def append(self, raw_g):
-		self.raw_graph_list.append(raw_g)
+    def append(self, raw_g):
+        self.raw_graph_list.append(raw_g)
 
-	def __len__(self):
-		return len(self.raw_graph_list)
+    def __len__(self):
+        return len(self.raw_graph_list)
 
 
 class graphlets:
-	def __init__(self, funcname):
-		self.funcname = funcname
-		self.graphlets_list = []
-		self.binary_name = None
+    def __init__(self, funcname):
+        self.funcname = funcname
+        self.graphlets_list = []
+        self.binary_name = None
 
-	def updateBN(self, binary_name):
-		self.binary_name = binary_name
+    def updateBN(self, binary_name):
+        self.binary_name = binary_name
 
-	def append(self, subg):
-		self.graphlets_list.append(subg)
+    def append(self, subg):
+        self.graphlets_list.append(subg)
 
-	def appendSet(self, subgs):
-		self.graphlets_list += subgs
+    def appendSet(self, subgs):
+        self.graphlets_list += subgs
+
+    def __len__(self):
+        return len(self.graphlets_list)
 
-	def __len__(self):
-		return len(self.graphlets_list)
 
 class template_graph:
-	def __init__(self, value=None):
-		self.value = value
-		self.g = None
+    def __init__(self, value=None):
+        self.value = value
+        self.g = None
 
-	def updateG(self,g):
-		self.g = g
-	#def updateIndexes(self, indexes):
-	#	self.indexes = indexes
+    def updateG(self, g):
+        self.g = g
+# def updateIndexes(self, indexes):
+#	self.indexes = indexes
 
-	#def updateAttributes(self, pg, indexes, maing):
-	#	for id_ in xrange(len(indexes)):
-	#		index = indexes[id_]
-	#		gnode = self.findNode(index, maing)
-	#		self.g.node[gnode] = pg.node[index]
+# def updateAttributes(self, pg, indexes, maing):
+#	for id_ in xrange(len(indexes)):
+#		index = indexes[id_]
+#		gnode = self.findNode(index, maing)
+#		self.g.node[gnode] = pg.node[index]
 
 
 class template_graphs:
-	def __init__(self, size):
-		self.size = size
-		self.gs = []
-		self.bit_len = None
+    def __init__(self, size):
+        self.size = size
+        self.gs = []
+        self.bit_len = None
 
-	def enumeratingAll(self):
-		subgs = []
-		binary_value = self.genBinValue()
-		for i in xrange(binary_value):
-			if i == 0 :
-				continue
-			g = self.createG(i)
-			if g:
-				tg = template_graph(i)
-				tg.updateG(g)
-				self.gs.append(tg)
+    def enumeratingAll(self):
+        subgs = []
+        binary_value = self.genBinValue()
+        for i in xrange(binary_value):
+            if i == 0:
+                continue
+            g = self.createG(i)
+            if g:
+                tg = template_graph(i)
+                tg.updateG(g)
+                self.gs.append(tg)
 
-	def genBinValue(self):
-		n = self.size
-		self.bit_len = n*n
-		return 2**(self.bit_len)
+    def genBinValue(self):
+        n = self.size
+        self.bit_len = n * n
+        return 2 ** (self.bit_len)
 
-	def createG(self, i):
-		g = nx.DiGraph()
-		l = self.genArray(i)
-		#pdb.set_trace()
-		shape = (self.size, self.size)
-		data = np.array(l)
-		ad_matrix = data.reshape(shape)
-		for i in xrange(self.size):
-			for j in xrange(self.size):
-				if ad_matrix[i][j] == 1:
-					g.add_edge(i, j)
-		u_g = g.to_undirected()
-		if len(g) == self.size and nx.is_connected(u_g):
-			return g
-		return False
+    def createG(self, i):
+        g = nx.DiGraph()
+        l = self.genArray(i)
+        # pdb.set_trace()
+        shape = (self.size, self.size)
+        data = np.array(l)
+        ad_matrix = data.reshape(shape)
+        for i in xrange(self.size):
+            for j in xrange(self.size):
+                if ad_matrix[i][j] == 1:
+                    g.add_edge(i, j)
+        u_g = g.to_undirected()
+        if len(g) == self.size and nx.is_connected(u_g):
+            return g
+        return False
 
-	def genArray(self, i):
-		l = [int(x) for x in bin(i)[2:]]
-		x = [0 for v in xrange(self.bit_len - len(l))]
-		return x + l
+    def genArray(self, i):
+        l = [int(x) for x in bin(i)[2:]]
+        x = [0 for v in xrange(self.bit_len - len(l))]
+        return x + l