Mixing python3 strings with bytes requires .encode and .decode

modified:   stub/scripts/bin2h.py
	modified:   stub/scripts/brandelf.py
	modified:   stub/scripts/gpp_inc.py
	modified:   stub/scripts/xstrip.py
This commit is contained in:
John Reiser 2025-10-10 08:38:59 -07:00
parent a3d5a0a7b8
commit 4ec866594b
4 changed files with 44 additions and 38 deletions

View File

@ -45,7 +45,7 @@ class opts:
# ************************************************************************/ # ************************************************************************/
def w_header_c(w, ifile, ofile, n): def w_header_c(w, ifile, ofile, n):
w("/* %s\n created from %s, %d (0x%x) bytes\n" % (os.path.basename(ofile), os.path.basename(ifile), n, n)) w(("/* %s\n created from %s, %d (0x%x) bytes\n" % (os.path.basename(ofile), os.path.basename(ifile), n, n)).encode())
w("""\n\ w("""\n\
This file is part of the UPX executable compressor. This file is part of the UPX executable compressor.
@ -74,7 +74,7 @@ def w_header_c(w, ifile, ofile, n):
John F. Reiser John F. Reiser
<jreiser@users.sourceforge.net> <jreiser@users.sourceforge.net>
*/\n\n""") */\n\n""".encode())
# /*********************************************************************** # /***********************************************************************
@ -87,11 +87,11 @@ class DataWriter:
self.pos = None self.pos = None
def w_bol(self, pos): def w_bol(self, pos):
self.w("/* 0x%04x */ " % (pos)) self.w(("/* 0x%04x */ " % (pos)).encode())
self.pos = pos self.pos = pos
def w_eol(self, fill=""): def w_eol(self, fill=""):
if self.pos is not None: if self.pos is not None:
self.w(fill.rstrip() + "\n") self.w((fill.rstrip() + "\n").encode())
class DataWriter_c(DataWriter): class DataWriter_c(DataWriter):
@ -101,8 +101,8 @@ class DataWriter_c(DataWriter):
if i & 15 == 0: if i & 15 == 0:
self.w_eol() self.w_eol()
self.w_bol(i) self.w_bol(i)
w("%3d" % ord(data[i])) w(("%3d" % data[i]).encode())
if i != n - 1: w(",") if i != n - 1: w(",".encode())
self.w_eol() self.w_eol()
@ -169,10 +169,10 @@ class DataWriter_nasm(DataWriter):
# ************************************************************************/ # ************************************************************************/
def w_checksum_c(w, s, data): def w_checksum_c(w, s, data):
w("#define %s_SIZE %d\n" % (s, len(data))) w(("#define %s_SIZE %d\n" % (s, len(data))).encode())
w("#define %s_ADLER32 0x%08x\n" % (s, 0xffffffff & zlib.adler32(data))) w(("#define %s_ADLER32 0x%08x\n" % (s, 0xffffffff & zlib.adler32(data))).encode())
w("#define %s_CRC32 0x%08x\n" % (s, 0xffffffff & zlib.crc32(data))) w(("#define %s_CRC32 0x%08x\n" % (s, 0xffffffff & zlib.crc32(data))).encode())
w("\n") w("\n".encode())
def write_stub(w, odata, method_index, methods): def write_stub(w, odata, method_index, methods):
@ -194,7 +194,7 @@ def write_stub(w, odata, method_index, methods):
#w('__attribute__((__section__("upx_stubs")))\n') #w('__attribute__((__section__("upx_stubs")))\n')
#w("#endif\n") #w("#endif\n")
w("ATTRIBUTE_FOR_STUB(%s)\n" % (opts.ident)) w("ATTRIBUTE_FOR_STUB(%s)\n" % (opts.ident))
w("unsigned char %s[%d] = {\n" % (opts.ident, len(odata))) w(("unsigned char %s[%d] = {\n" % (opts.ident, len(odata))).encode())
if opts.mode == "c": if opts.mode == "c":
DataWriter_c(w).w_data(odata) DataWriter_c(w).w_data(odata)
elif opts.mode == "gas": elif opts.mode == "gas":
@ -209,11 +209,11 @@ def write_stub(w, odata, method_index, methods):
assert 0, ("invalid mode", opts.mode) assert 0, ("invalid mode", opts.mode)
if opts.ident: if opts.ident:
if opts.mode == "c": if opts.mode == "c":
w("};\n") w("};\n".encode())
if len(methods) > 1: if len(methods) > 1:
if method_index == len(methods) - 1: if method_index == len(methods) - 1:
w("\n#endif\n") w("\n#endif\n".encode())
# /*********************************************************************** # /***********************************************************************
@ -321,9 +321,12 @@ def main(argv):
# compress stubs # compress stubs
# (process in reverse order so that incompressible do not get sorted first) # (process in reverse order so that incompressible do not get sorted first)
mdata, mdata_odata = [], {} mdata, mdata_odata = [], {}
assert len(opts.methods) >= 1 ### FIXME assert len(opts.methods) >= 1
r_methods = opts.methods[:] ### r_methods = opts.methods[:]
r_methods.reverse() ### r_methods.reverse()
r_methods = [];
for m in opts.methods:
r_methods.insert(0, m)
for method in r_methods: for method in r_methods:
method, odata = compress_stub(method, idata) method, odata = compress_stub(method, idata)
if method in mdata_odata: if method in mdata_odata:
@ -349,7 +352,7 @@ def main(argv):
if opts.mode == "c": if opts.mode == "c":
if opts.verbose >= 0: if opts.verbose >= 0:
w_header_c(w, ifile, ofile, len(idata)) w_header_c(w, ifile, ofile, len(idata))
w("/* clang" + "-format" + " off */\n\n") w(("/* clang" + "-format" + " off */\n\n").encode())
for i in range(len(mdata)): for i in range(len(mdata)):
write_stub(w, mdata_odata[mdata[i]], i, mdata) write_stub(w, mdata_odata[mdata[i]], i, mdata)
if ofp: if ofp:

View File

@ -54,39 +54,42 @@ def do_file(fn):
def write(s): def write(s):
if not opts.dry_run: if not opts.dry_run:
fp.write(s) fp.write(s.encode())
def brand_arm(s): def brand_arm(s):
if e_ident[4:7] != s: if e_ident[4:7] != s.encode():
raise Exception("%s is not %s" % (fn, opts.bfdname)) raise Exception("%s is not %s" % (fn, opts.bfdname))
write("\x61") # ELFOSABI_ARM write("\x61") # ELFOSABI_ARM
def brand_freebsd(s): def brand_freebsd(s):
if e_ident[4:7] != s: if e_ident[4:7] != s.encode():
raise Exception("%s is not %s" % (fn, opts.bfdname)) raise Exception("%s is not %s" % (fn, opts.bfdname))
write("\x09") write("\x09")
def brand_linux(s): def brand_linux(s):
if e_ident[4:7] != s: print ("brand_linux ", s.encode(), "e_ident ", e_ident[4:7])
if e_ident[4:7] != s.encode():
raise Exception("%s is not %s" % (fn, opts.bfdname)) raise Exception("%s is not %s" % (fn, opts.bfdname))
##write("\x00Linux\x00\x00\x00") ##write("\x00Linux\x00\x00\x00")
write("\x00" * 9) write("\x00" * 9)
def brand_netbsd(s): def brand_netbsd(s):
if e_ident[4:7] != s: if e_ident[4:7] != s.encode():
raise Exception("%s is not %s" % (fn, opts.bfdname)) raise Exception("%s is not %s" % (fn, opts.bfdname))
write("\x02") write("\x02")
def brand_openbsd(s): def brand_openbsd(s):
if e_ident[4:7] != s: if e_ident[4:7] != s.encode():
raise Exception("%s is not %s" % (fn, opts.bfdname)) raise Exception("%s is not %s" % (fn, opts.bfdname))
write("\x0c") write("\x0c")
if opts.bfdname[:3] == "elf": if opts.bfdname[:3] == "elf":
if e_ident[:4] != "\x7f\x45\x4c\x46": if e_ident[:4] != "\x7f\x45\x4c\x46".encode():
raise Exception("%s is not %s" % (fn, "ELF")) raise Exception("%s is not %s" % (fn, "ELF"))
fp.seek(7, 0) fp.seek(7, 0)
if opts.bfdname == "elf32-bigarm" and opts.elfosabi == "arm": if opts.bfdname == "elf32-bigarm" and opts.elfosabi == "arm":
brand_arm("\x01\x02\x01") brand_arm("\x01\x02\x01")
elif opts.bfdname == "elf32-i386" and opts.elfosabi == "freebsd": elif opts.bfdname == "elf32-i386" and opts.elfosabi == "freebsd":
print ("case 4")
brand_freebsd("\x01\x01\x01") brand_freebsd("\x01\x01\x01")
elif opts.bfdname == "elf32-i386" and opts.elfosabi == "linux": elif opts.bfdname == "elf32-i386" and opts.elfosabi == "linux":
print ("case 5")
brand_linux("\x01\x01\x01") brand_linux("\x01\x01\x01")
elif opts.bfdname == "elf32-i386" and opts.elfosabi == "netbsd": elif opts.bfdname == "elf32-i386" and opts.elfosabi == "netbsd":
brand_netbsd("\x01\x01\x01") brand_netbsd("\x01\x01\x01")

View File

@ -83,7 +83,7 @@ def parse_comment(state, l, comment):
def handle_inc_c(state, l, ofp): def handle_inc_c(state, l, ofp):
m = re.search(r"^\s*\#\s*include\s+([\"\<])(.+?)([\"\>])(.*)$", l) m = re.search(r"^\s*\#\s*include\s+([\"\<])(.+?)([\"\>])(.*)$", l.decode())
if not m: if not m:
return l return l
q1, inc, q2, comment = m.groups() q1, inc, q2, comment = m.groups()
@ -130,13 +130,13 @@ def handle_file(ifn, ofp, parent_state=None):
ifp = open(ifn, "rb") ifp = open(ifn, "rb")
for l in ifp.readlines(): for l in ifp.readlines():
state[2] += 1 # line counter state[2] += 1 # line counter
l = l.rstrip("\n") l = l.rstrip("\n".encode())
if opts.mode == "c": if opts.mode == "c":
l = handle_inc_c(state, l, ofp) l = handle_inc_c(state, l, ofp)
elif opts.mode == "nasm": elif opts.mode == "nasm":
l = handle_inc_nasm(state, l, ofp) l = handle_inc_nasm(state, l, ofp)
if l is not None: if l is not None:
ofp.write(l + "\n") ofp.write(l + "\n".encode())
def main(argv): def main(argv):
@ -179,12 +179,12 @@ def main(argv):
os.unlink(fn) os.unlink(fn)
if files_mmd: if files_mmd:
fp = open(fn, "wb") fp = open(fn, "wb")
fp.write("%s : \\\n" % opts.target_mmd) fp.write(("%s : \\\n" % opts.target_mmd).encode())
for i, f in enumerate(files_mmd): for i, f in enumerate(files_mmd):
if i < len(files_mmd) - 1: if i < len(files_mmd) - 1:
fp.write(" %s \\\n" % f) fp.write((" %s \\\n" % f).encode())
else: else:
fp.write(" %s\n" % f) fp.write((" %s\n" % f).encode())
fp.close() fp.close()

View File

@ -46,7 +46,7 @@ def strip_with_dump(dump_fn, eh, idata):
new_len = 0 new_len = 0
lines = open(dump_fn, "rb").readlines() lines = open(dump_fn, "rb").readlines()
for l in lines: for l in lines:
l = re.sub(r"\s+", " ", l.strip()) l = re.sub(r"\s+", " ", l.strip().decode())
f = l.split(" ") f = l.split(" ")
if len(f) >= 8: if len(f) >= 8:
if f[7].startswith("CONTENTS"): if f[7].startswith("CONTENTS"):
@ -67,7 +67,7 @@ def strip_with_dump(dump_fn, eh, idata):
def check_dump(dump_fn): def check_dump(dump_fn):
lines = open(dump_fn, "rb").readlines() lines = open(dump_fn, "rb").readlines()
lines = map(lambda l: re.sub(r"\s+", " ", l.strip()).strip(), lines) lines = map(lambda l: re.sub(r"\s+", " ", l.strip().decode()).strip(), lines)
lines = filter(None, lines) lines = filter(None, lines)
d = "\n".join(lines) d = "\n".join(lines)
psections = d.find("Sections:\n") psections = d.find("Sections:\n")
@ -140,25 +140,25 @@ def do_file(fn):
fp = open(fn, "r+b") fp = open(fn, "r+b")
fp.seek(0, 0) fp.seek(0, 0)
idata = fp.read() idata = fp.read()
if idata[:4] != "\x7f\x45\x4c\x46": if idata[:4] != "\x7f\x45\x4c\x46".encode():
raise Exception("%s is not %s" % (fn, "ELF")) raise Exception("%s is not %s" % (fn, "ELF"))
if idata[4:7] == "\x01\x01\x01": if idata[4:7] == "\x01\x01\x01".encode():
# ELF32 LE # ELF32 LE
eh, idata = idata[:52], idata[52:] eh, idata = idata[:52], idata[52:]
e_shnum, e_shstrndx = struct.unpack("<HH", eh[48:52]) e_shnum, e_shstrndx = struct.unpack("<HH", eh[48:52])
assert e_shstrndx + 3 == e_shnum assert e_shstrndx + 3 == e_shnum
##eh = eh[:48] + struct.pack("<HH", e_shnum - 3, e_shstrndx) ##eh = eh[:48] + struct.pack("<HH", e_shnum - 3, e_shstrndx)
elif idata[4:7] == "\x01\x02\x01": elif idata[4:7] == "\x01\x02\x01".encode():
# ELF32 BE # ELF32 BE
eh, idata = idata[:52], idata[52:] eh, idata = idata[:52], idata[52:]
e_shnum, e_shstrndx = struct.unpack(">HH", eh[48:52]) e_shnum, e_shstrndx = struct.unpack(">HH", eh[48:52])
assert e_shstrndx + 3 == e_shnum assert e_shstrndx + 3 == e_shnum
elif idata[4:7] == "\x02\x01\x01": elif idata[4:7] == "\x02\x01\x01".encode():
# ELF64 LE # ELF64 LE
eh, idata = idata[:64], idata[64:] eh, idata = idata[:64], idata[64:]
e_shnum, e_shstrndx = struct.unpack("<HH", eh[60:64]) e_shnum, e_shstrndx = struct.unpack("<HH", eh[60:64])
assert e_shstrndx + 3 == e_shnum assert e_shstrndx + 3 == e_shnum
elif idata[4:7] == "\x02\x02\x01": elif idata[4:7] == "\x02\x02\x01".encode():
# ELF64 BE # ELF64 BE
eh, idata = idata[:64], idata[64:] eh, idata = idata[:64], idata[64:]
e_shnum, e_shstrndx = struct.unpack(">HH", eh[60:64]) e_shnum, e_shstrndx = struct.unpack(">HH", eh[60:64])
@ -167,7 +167,7 @@ def do_file(fn):
raise Exception("%s is not %s" % (fn, "ELF")) raise Exception("%s is not %s" % (fn, "ELF"))
odata = None odata = None
pos = idata.find("\0.symtab\0.strtab\0.shstrtab\0") pos = idata.find("\0.symtab\0.strtab\0.shstrtab\0".encode())
if opts.with_dump: if opts.with_dump:
eh, odata = strip_with_dump(opts.with_dump, eh, idata) eh, odata = strip_with_dump(opts.with_dump, eh, idata)
# Other compilers can intermix the contents of .rela sections # Other compilers can intermix the contents of .rela sections