Data transformation

Enables extraction of dictionary keys whose values satisfy a positive threshold.

Data analysis: selecting feature names with non‑zero importance; Configuration loading: keeping only enabled options; Reporting: listing items with a count above zero

Python scripts, data‑science notebooks, web‑application back‑ends that manipulate dictionaries

[key for key, value in dict_var.items() if value > 0]

[... for ... in ... .items() if ... > 0]

key: any hashable object, value: numeric type (int or float) > 0, dict_var: dict mapping keys to numeric values

• dict.items()
• if condition
• list comprehension
• > operator

• Use a for‑loop with append – more verbose
• Use filter() with lambda – less readable
• Use dict comprehension to build a new dict – returns a dict instead of a list of keys

Omitting the .items() call – iterates over keys only, causing value to be a key; Using mutable default values in the dict – leads to unexpected shared state; Forgetting the if clause – returns all keys regardless of value

list comprehension vs explicit for‑loop; filter() function vs comprehension; dict comprehension for building a filtered dict instead of a list of keys

Coming from JavaScript: may try to use Object.keys().filter(...) which returns strings, but Python's comprehension expects iteration over items; Coming from SQL: assuming the comprehension can be executed lazily like a query – it materializes a list immediately

• list comprehensions
• dictionary filtering
• Python idiomatic loops

Do not use when the dictionary is extremely large and memory is a concern – a generator expression would be better; The comprehension runs in O(n) time and creates a new list, which can affect performance for huge inputs; If the dictionary contains non‑numeric values, the > 0 comparison will raise a TypeError

Produces a concise, readable list of relevant keys, reducing boilerplate and potential off‑by‑one errors in manual loops

Think of the comprehension as a sieve that lets only the keys with heavy enough values slip through.

The result is a list, not a view; if you need a lazy sequence, wrap the comprehension in a generator expression or use (key for ...) syntax

Consider using a generator expression (key for k, v in mapping.items() if v > 0) for lazy evaluation, or a dict comprehension {k: v for k, v in mapping.items() if v > 0} to obtain a filtered dictionary.