telescope -r should always display the input address

[gsingh93]

telescope -r addr will skip displaying addr if the value at that address is equal to the previous few values:

pwndbg> p &target_address
$1 = (unsigned long *) 0x5555555582a0 <target_address>
pwndbg> telescope -r &target_address
00:0000│     0x555555558208 (buf+360) ◂— 0x0
... ↓        4 skipped
05:0028│ rax 0x555555558230 (buf+400) ◂— 0x80
06:0030│     0x555555558238 (buf+408) ◂— 0x0
... ↓        13 skipped

We should make this command always display the address that the user asked for:

pwndbg> telescope -r &target_address
00:0000│     0x555555558208 (buf+360) ◂— 0x0
... ↓        4 skipped
05:0028│ rax 0x555555558230 (buf+400) ◂— 0x80
06:0030│     0x555555558238 (buf+408) ◂— 0x0
... ↓        12 skipped
13:0098│     0x5555555582a0 (target_address) ◂— 0x0

[lonnywong]

I found that telescope behaves strangely if set history remove-duplicates unlimited.

The check_repeated depends on the history of gdb.

pwndbg/pwndbg/commands/__init__.py

Lines 113 to 148 in e456b31

	def check_repeated(self, argument, from_tty):
	"""Keep a record of all commands which come from the TTY.
	Returns:
	True if this command was executed by the user just hitting "enter".
	"""
	# Don't care unless it's interactive use
	if not from_tty:
	return False
	lines = gdb.execute("show commands", from_tty=False, to_string=True)
	lines = lines.splitlines()
	# No history
	if not lines:
	return False
	last_line = lines[-1]
	number, command = last_line.split(None, 1)
	try:
	number = int(number)
	except ValueError:
	# Workaround for a GDB 8.2 bug when show commands return error value
	# See issue #523
	return False
	# A new command was entered by the user
	if number not in Command.history:
	Command.history[number] = command
	return False
	# Somehow the command is different than we got before?
	if not command.endswith(argument):
	return False
	return True

If remove-duplicates is set, the check_repeated is not working normally. That's what I mentioned in #1072 (comment)

I can't find another way to implement the check_repeated. I'm removing the set history remove-duplicates unlimited from my ~/.gdbinit.

[disconnect3d]

I'd say we should handle the two cases as separate bugs:

1. The telescope -r should always include/display the input address (and not collapse it if its value is the same as previous few ones)
2. The telescope behavior when set history remove-duplicates unlimited is on should be investigated further I guess.