[Solution] Python dict.has_key() Deprecated — Use ‘in’ Operator

The dict.has_key() method was deprecated in Python 2.6 and removed in Python 3. The preferred way to check for key existence in a dictionary is the in operator, which is more Pythonic, readable, and consistent with other container types in Python.

What You’ll See

If you run Python 2 code with has_key() under Python 3:

config = {"host": "localhost", "port": 8080}
if config.has_key("host"):
    print("Host found")

You get:

AttributeError: 'dict' object has no attribute 'has_key'

Why Deprecated

The has_key() method was removed for several reasons:

The in operator is more general: The same in keyword works for checking membership in lists, sets, tuples, strings, and dictionaries. has_key() was dict-specific.
Readability: if "key" in d reads naturally in English. if d.has_key("key") is more verbose.
Consistency: Other data structures in Python use in for membership testing. Having a separate method for dicts was inconsistent.
Performance: in and has_key() have identical performance, so there is no reason to prefer the method.

Old Code (Deprecated)

# Check if key exists
config = {"host": "localhost", "port": 8080}
if config.has_key("host"):
    print("Host is", config["host"])

# Check and set default
if not settings.has_key("timeout"):
    settings["timeout"] = 30

# Count keys from a list
keys_to_check = ["name", "email", "age"]
user = {"name": "Alice", "email": "alice@example.com"}
found = 0
for key in keys_to_check:
    if user.has_key(key):
        found += 1

# Nested check
data = {"users": {"count": 42}}
if data.has_key("users") and data["users"].has_key("count"):
    print("User count:", data["users"]["count"])

New Code (Replacement)

# Check if key exists — use 'in' operator
config = {"host": "localhost", "port": 8080}
if "host" in config:
    print("Host is", config["host"])

# Check and set default — use dict.setdefault() or 'in'
if "timeout" not in settings:
    settings["timeout"] = 30

# Or use setdefault() which does both in one call
settings.setdefault("timeout", 30)

# Count keys from a list
keys_to_check = ["name", "email", "age"]
user = {"name": "Alice", "email": "alice@example.com"}
found = sum(1 for key in keys_to_check if key in user)

# Nested check
data = {"users": {"count": 42}}
if "users" in data and "count" in data["users"]:
    print("User count:", data["users"]["count"])

# Or use get() for safer nested access
count = data.get("users", {}).get("count", 0)
print("User count:", count)

Python 2/3 Compatibility

If you need code that works under both Python 2 and Python 3:

# The 'in' operator works in both Python 2.6+ and Python 3
config = {"host": "localhost"}

# This works in both versions
if "host" in config:
    print(config["host"])

# For Python 2.5 and earlier, use dict.get()
# This works in all Python 2 and Python 3 versions
if config.get("host") is not None:
    print(config["host"])

The in operator was added for dicts in Python 2.6, so using it is safe for any Python version you are likely to encounter.

Migration Steps

Find all has_key() calls:

grep -rn "\.has_key(" --include="*.py" /path/to/project/

Replace d.has_key(k) with k in d. Note the reversed order — the key comes first with the in operator.
For negative checks, convert not d.has_key(k) to k not in d.
Consider using dict.get() when you want to check for a key and use a default value:

# Before
if d.has_key("key"):
    val = d["key"]
else:
    val = "default"

# After
val = d.get("key", "default")

Use dict.setdefault() for the check-and-set pattern.
Run 2to3 to automate the conversion:

2to3 -f has_key -w /path/to/project/

Run your test suite to verify all dictionary lookups produce correct results.