def test_find_all_text_pointers_in_script_engine_script(self):
address = 0x197637 # 0x197634
script = parse_script_engine_script_at(address, debug=False)
bank = calculate_bank(address)
r = find_all_text_pointers_in_script_engine_script(script, bank=bank, debug=False)
results = list(r)
self.assertIn(0x197661, results)
def line_has_comment_address(line, returnable={}, bank=None):
"""checks that a given line has a comment
with a valid address, and returns the address in the object.
Note: bank is required if you have a 4-letter-or-less address,
because otherwise there is no way to figure out which bank
is curretly being scanned."""
#first set the bank/offset to nada
returnable["bank"] = None
returnable["offset"] = None
returnable["address"] = None
#only valid characters are 0-9a-fA-F
valid = [str(x) for x in range(10)] + \
[chr(x) for x in range(ord('a'), ord('f')+1)] + \
[chr(x) for x in range(ord('A'), ord('F')+1)]
#check if there is a comment in this line
if ";" not in line:
return False
#first throw away anything in quotes
if (line.count("\"") % 2 == 0 and line.count("\"")!=0) \
or (line.count("\'") % 2 == 0 and line.count("\'")!=0):
line = remove_quoted_text(line)
#check if there is still a comment in this line after quotes removed
if ";" not in line:
return False
#but even if there's a semicolon there must be later text
if line[-1] == ";":
return False
#and just a space doesn't count
if line[-2:] == "; ":
return False
#and multiple whitespace doesn't count either
line = line.rstrip(" ").lstrip(" ")
if line[-1] == ";":
return False
#there must be more content after the semicolon
if len(line)-1 == line.find(";"):
return False
#split it up into the main comment part
comment = line[line.find(";")+1:]
#don't want no leading whitespace
comment = comment.lstrip(" ").rstrip(" ")
#split up multi-token comments into single tokens
token = comment
if " " in comment:
#use the first token in the comment
token = comment.split(" ")[0]
if token in ["0x", "$", "x", ":"]:
return False
offset = None
#process a token with a A:B format
if ":" in token: #3:3F0A, $3:$3F0A, 0x3:0x3F0A, 3:3F0A
#split up the token
bank_piece = token.split(":")[0].lower()
offset_piece = token.split(":")[1].lower()
#filter out blanks/duds
if bank_piece in ["$", "0x", "x"] \
or offset_piece in ["$", "0x", "x"]:
return False
#they can't have both "$" and "x"
if "$" in bank_piece and "x" in bank_piece:
return False
if "$" in offset_piece and "x" in offset_piece:
return False
#process the bank piece
if "$" in bank_piece:
bank_piece = bank_piece.replace("$", "0x")
#check characters for validity?
for c in bank_piece.replace("x", ""):
if c not in valid:
return False
bank = int(bank_piece, 16)
#process the offset piece
if "$" in offset_piece:
offset_piece = offset_piece.replace("$", "0x")
#check characters for validity?
for c in offset_piece.replace("x", ""):
if c not in valid:
return False
if len(offset_piece) == 0:
return None
offset = int(offset_piece, 16)
#filter out blanks/duds
elif token in ["$", "0x", "x"]:
return False
#can't have both "$" and "x" in the number
elif "$" in token and "x" in token:
return False
elif "x" in token and not "0x" in token: #it should be 0x
return False
elif "$" in token and not "x" in token:
token = token.replace("$", "0x")
offset = int(token, 16)
elif "0x" in token and not "$" in token:
offset = int(token, 16)
else: #might just be "1" at this point
token = token.lower()
#check if there are bad characters
for c in token:
if c not in valid:
return False
offset = int(token, 16)
if offset == None and bank == None:
return False
if bank == None:
bank = pointers.calculate_bank(offset)
returnable["bank"] = bank
returnable["offset"] = offset
returnable["address"] = pointers.calculate_pointer(offset, bank=bank)
return True
def line_has_comment_address(line, returnable={}, bank=None):
"""checks that a given line has a comment
with a valid address, and returns the address in the object.
Note: bank is required if you have a 4-letter-or-less address,
because otherwise there is no way to figure out which bank
is curretly being scanned."""
#first set the bank/offset to nada
returnable["bank"] = None
returnable["offset"] = None
returnable["address"] = None
#only valid characters are 0-9A-F
valid = [str(x)
for x in range(0, 10)] + [chr(x) for x in range(97, 102 + 1)]
#check if there is a comment in this line
if ";" not in line:
return False
#first throw away anything in quotes
if (line.count("\"") % 2 == 0 and line.count("\"")!=0) \
or (line.count("\'") % 2 == 0 and line.count("\'")!=0):
line = remove_quoted_text(line)
#check if there is still a comment in this line after quotes removed
if ";" not in line:
return False
#but even if there's a semicolon there must be later text
if line[-1] == ";":
return False
#and just a space doesn't count
if line[-2:] == "; ":
return False
#and multiple whitespace doesn't count either
line = line.rstrip(" ").lstrip(" ")
if line[-1] == ";":
return False
#there must be more content after the semicolon
if len(line) - 1 == line.find(";"):
return False
#split it up into the main comment part
comment = line[line.find(";") + 1:]
#don't want no leading whitespace
comment = comment.lstrip(" ").rstrip(" ")
#split up multi-token comments into single tokens
token = comment
if " " in comment:
#use the first token in the comment
token = comment.split(" ")[0]
if token in ["0x", "$", "x", ":"]:
return False
offset = None
#process a token with a A:B format
if ":" in token: #3:3F0A, $3:$3F0A, 0x3:0x3F0A, 3:3F0A
#split up the token
bank_piece = token.split(":")[0].lower()
offset_piece = token.split(":")[1].lower()
#filter out blanks/duds
if bank_piece in ["$", "0x", "x"] \
or offset_piece in ["$", "0x", "x"]:
return False
#they can't have both "$" and "x"
if "$" in bank_piece and "x" in bank_piece:
return False
if "$" in offset_piece and "x" in offset_piece:
return False
#process the bank piece
if "$" in bank_piece:
bank_piece = bank_piece.replace("$", "0x")
#check characters for validity?
for c in bank_piece.replace("x", ""):
if c not in valid:
return False
bank = int(bank_piece, 16)
#process the offset piece
if "$" in offset_piece:
offset_piece = offset_piece.replace("$", "0x")
#check characters for validity?
for c in offset_piece.replace("x", ""):
if c not in valid:
return False
if len(offset_piece) == 0:
return None
offset = int(offset_piece, 16)
#filter out blanks/duds
elif token in ["$", "0x", "x"]:
return False
#can't have both "$" and "x" in the number
elif "$" in token and "x" in token:
return False
elif "x" in token and not "0x" in token: #it should be 0x
return False
elif "$" in token and not "x" in token:
token = token.replace("$", "0x")
offset = int(token, 16)
elif "0x" in token and not "$" in token:
offset = int(token, 16)
else: #might just be "1" at this point
token = token.lower()
#check if there are bad characters
for c in token:
if c not in valid:
return False
offset = int(token, 16)
if offset == None and bank == None:
return False
if bank == None:
bank = calculate_bank(offset)
returnable["bank"] = bank
returnable["offset"] = offset
returnable["address"] = calculate_pointer(offset, bank=bank)
return True
def test_calculate_bank(self):
self.failUnless(calculate_bank(0x8000) == 2)
self.failUnless(calculate_bank("0x9000") == 2)
self.failUnless(calculate_bank(0) == 0)
for address in [0x4000, 0x5000, 0x6000, 0x7000]:
self.assertRaises(Exception, calculate_bank, address)
def find_addresses():
"""returns a list of text pointers
useful for testing parse_text_engine_script_at
Note that this list is not exhaustive. There are some texts that
are only pointed to from some script that a current script just
points to. So find_all_text_pointers_in_script_engine_script will
have to recursively follow through each script to find those.
.. it does this now :)
"""
addresses = set()
# for each map group
for map_group in map_names:
# for each map id
for map_id in map_names[map_group]:
# skip the offset key
if map_id == "offset": continue
# dump this into smap
smap = map_names[map_group][map_id]
# signposts
signposts = smap["signposts"]
# for each signpost
for signpost in signposts:
if signpost["func"] in [0, 1, 2, 3, 4]:
# dump this into script
script = signpost["script"]
elif signpost["func"] in [05, 06]:
script = signpost["script"]
else: continue
# skip signposts with no bytes
if len(script) == 0: continue
# find all text pointers in script
texts = find_all_text_pointers_in_script_engine_script(script, smap["event_bank"])
# dump these addresses in
addresses.update(texts)
# xy triggers
xy_triggers = smap["xy_triggers"]
# for each xy trigger
for xy_trigger in xy_triggers:
# dump this into script
script = xy_trigger["script"]
# find all text pointers in script
texts = find_all_text_pointers_in_script_engine_script(script, smap["event_bank"])
# dump these addresses in
addresses.update(texts)
# trigger scripts
triggers = smap["trigger_scripts"]
# for each trigger
for (i, trigger) in triggers.items():
# dump this into script
script = trigger["script"]
# find all text pointers in script
texts = find_all_text_pointers_in_script_engine_script(script, pointers.calculate_bank(trigger["address"]))
# dump these addresses in
addresses.update(texts)
# callback scripts
callbacks = smap["callback_scripts"]
# for each callback
for (k, callback) in callbacks.items():
# dump this into script
script = callback["script"]
# find all text pointers in script
texts = find_all_text_pointers_in_script_engine_script(script, pointers.calculate_bank(callback["address"]))
# dump these addresses in
addresses.update(texts)
# people-events
events = smap["people_events"]
# for each event
for event in events:
if event["event_type"] == "script":
# dump this into script
script = event["script"]
# find all text pointers in script
texts = find_all_text_pointers_in_script_engine_script(script, smap["event_bank"])
# dump these addresses in
addresses.update(texts)
if event["event_type"] == "trainer":
trainer_data = event["trainer_data"]
addresses.update([trainer_data["text_when_seen_ptr"]])
addresses.update([trainer_data["text_when_trainer_beaten_ptr"]])
trainer_bank = pointers.calculate_bank(event["trainer_data_address"])
script1 = trainer_data["script_talk_again"]
texts1 = find_all_text_pointers_in_script_engine_script(script1, trainer_bank)
addresses.update(texts1)
script2 = trainer_data["script_when_lost"]
texts2 = find_all_text_pointers_in_script_engine_script(script2, trainer_bank)
addresses.update(texts2)
def find_addresses():
"""returns a list of text pointers
useful for testing parse_text_engine_script_at
Note that this list is not exhaustive. There are some texts that
are only pointed to from some script that a current script just
points to. So find_all_text_pointers_in_script_engine_script will
have to recursively follow through each script to find those.
.. it does this now :)
"""
addresses = set()
# for each map group
for map_group in map_names:
# for each map id
for map_id in map_names[map_group]:
# skip the offset key
if map_id == "offset": continue
# dump this into smap
smap = map_names[map_group][map_id]
# signposts
signposts = smap["signposts"]
# for each signpost
for signpost in signposts:
if signpost["func"] in [0, 1, 2, 3, 4]:
# dump this into script
script = signpost["script"]
elif signpost["func"] in [05, 06]:
script = signpost["script"]
else:
continue
# skip signposts with no bytes
if len(script) == 0: continue
# find all text pointers in script
texts = find_all_text_pointers_in_script_engine_script(
script, smap["event_bank"])
# dump these addresses in
addresses.update(texts)
# xy triggers
xy_triggers = smap["xy_triggers"]
# for each xy trigger
for xy_trigger in xy_triggers:
# dump this into script
script = xy_trigger["script"]
# find all text pointers in script
texts = find_all_text_pointers_in_script_engine_script(
script, smap["event_bank"])
# dump these addresses in
addresses.update(texts)
# trigger scripts
triggers = smap["trigger_scripts"]
# for each trigger
for (i, trigger) in triggers.items():
# dump this into script
script = trigger["script"]
# find all text pointers in script
texts = find_all_text_pointers_in_script_engine_script(
script, pointers.calculate_bank(trigger["address"]))
# dump these addresses in
addresses.update(texts)
# callback scripts
callbacks = smap["callback_scripts"]
# for each callback
for (k, callback) in callbacks.items():
# dump this into script
script = callback["script"]
# find all text pointers in script
texts = find_all_text_pointers_in_script_engine_script(
script, pointers.calculate_bank(callback["address"]))
# dump these addresses in
addresses.update(texts)
# people-events
events = smap["people_events"]
# for each event
for event in events:
if event["event_type"] == "script":
# dump this into script
script = event["script"]
# find all text pointers in script
texts = find_all_text_pointers_in_script_engine_script(
script, smap["event_bank"])
# dump these addresses in
addresses.update(texts)
if event["event_type"] == "trainer":
trainer_data = event["trainer_data"]
addresses.update([trainer_data["text_when_seen_ptr"]])
addresses.update(
[trainer_data["text_when_trainer_beaten_ptr"]])
trainer_bank = pointers.calculate_bank(
event["trainer_data_address"])
script1 = trainer_data["script_talk_again"]
texts1 = find_all_text_pointers_in_script_engine_script(
script1, trainer_bank)
addresses.update(texts1)
script2 = trainer_data["script_when_lost"]
texts2 = find_all_text_pointers_in_script_engine_script(
script2, trainer_bank)
addresses.update(texts2)
