Halyna R.

task-1/output/clean_users.json

@@ -0,0 +1,86 @@
+[
+  {
+    "id": 1,
+    "name": "Alice Johnson",
+    "email": "alice.johnson@company.com",
+    "department": "Engineering",
+    "salary": 85000
+  },
+  {
+    "id": 2,
+    "name": "Bob Smith",
+    "email": "bob.smith@company.com",
+    "department": "Unknown",
+    "salary": 72000
+  },
+  {
+    "id": 3,
+    "name": "Carol Williams",
+    "email": "carol.williams@company.com",
+    "department": "Engineering",
+    "salary": null
+  },
+  {
+    "id": 4,
+    "name": "David, Jr.",
+    "email": "david.brown@company.com",
+    "department": "Sales",
+    "salary": 68000
+  },
+  {
+    "id": 5,
+    "name": "Caf\u00e9 Owner",
+    "email": "eva@company.com",
+    "department": "Engineering",
+    "salary": 88000
+  },
+  {
+    "id": 6,
+    "name": "FRANK WILSON",
+    "email": "frank@company.com",
+    "department": "marketing",
+    "salary": 95000
+  },
+  {
+    "id": 7,
+    "name": "Grace Lee",
+    "email": "grace.lee@company.com",
+    "department": "Engineering",
+    "salary": null
+  },
+  {
+    "id": 9,
+    "name": "Henry Davis",
+    "email": "henry.davis@company.com",
+    "department": "Sales",
+    "salary": 82000
+  },
+  {
+    "id": 11,
+    "name": "Linda Taylor",
+    "email": "linda.t@company.com",
+    "department": "HR",
+    "salary": 55000
+  },
+  {
+    "id": 12,
+    "name": "Mike Brown",
+    "email": "mike.b@company.com",
+    "department": "Sales",
+    "salary": null
+  },
+  {
+    "id": 13,
+    "name": "Sarah Connor",
+    "email": "s.connor@sky.net",
+    "department": "Unknown",
+    "salary": -1
+  },
+  {
+    "id": 15,
+    "name": "John Doe",
+    "email": "john.doe@company.net",
+    "department": "Engineering",
+    "salary": 100000
+  }
+]

task-1/src/utils.py

@@ -13,23 +13,28 @@ def clean_name(raw: str) -> str:
 
     Returns the cleaned string. An empty input returns "".
     """
-    raise NotImplementedError("Implement clean_name (Task 1)")
+    return raw.strip()
 
 
 def clean_email(raw: str) -> str:
     """Lowercase the email, strip surrounding whitespace.
 
     Returns the cleaned string. An empty input returns "".
     """
-    raise NotImplementedError("Implement clean_email (Task 1)")
+    return raw.strip().lower()
 
 
 def clean_department(raw: str) -> str:
     """Return the department, or 'Unknown' if missing/empty.
 
     Strip whitespace; treat empty string as missing.
     """
-    raise NotImplementedError("Implement clean_department (Task 1)")
+    department = raw.strip()
+
+    if department == "":
+        return "Unknown"
+
+    return department
 
 
 def clean_salary(raw: str) -> int | None:
@@ -38,4 +43,15 @@ def clean_salary(raw: str) -> int | None:
     Handles inputs like "85000", "  95000", '"68,000"', "N/A", "".
     Returns None when the value cannot be parsed (missing or "N/A").
     """
-    raise NotImplementedError("Implement clean_salary (Task 1)")
+    salary = raw.strip()
+    if salary in ("", "N/A"):
+        return None
+
+    salary = salary.replace('"', "")
+    salary = salary.replace(",", "")
+    salary = salary.strip()
+
+    try:
+        return int(salary)
+    except ValueError:
+        return None

task-2/AI_DEBUG.md

@@ -11,19 +11,124 @@ Document one debugging session you had during Task 1 where you used an LLM
 
 <!-- Paste the full traceback or describe the wrong behaviour. Include the
 exact error message and the line of your code that triggered it. -->
+While working on Task 1, my first version of `clean_salary()` tried to convert the salary directly to an integer after removing commas and quotes. It worked for values like `85000` and `"68,000"`, but it crashed when the CSV contained an unexpected salary format with a period inside the value.
+
+My first version was:
+
+def clean_salary(raw: str) -> int | None:
+    """Parse a messy salary cell into an int.
+
+    Handles inputs like "85000", "  95000", '"68,000"', "N/A", "".
+    Returns None when the value cannot be parsed (missing or "N/A").
+    """
+    salary = raw.strip()
+    if salary in ("", "N/A"):
+        return None
+
+    salary = salary.replace('"', "")
+    salary = salary.replace(",", "")
+    salary = salary.strip()
+
+    return int(salary)
+
+When I ran the cleaner:
+
+python3 src/cleaner.py --input data/messy_users.csv --output output/clean_users.json
+
+I got this traceback:
+
+Traceback (most recent call last):
+  File "G:\Halyna_work\traineeship_Amsterdam\Data-track\week-1\c55-data-week1\task-1\src\cleaner.py", line 58, in <module>
+    main(args.input, args.output)
+  File "G:\Halyna_work\traineeship_Amsterdam\Data-track\week-1\c55-data-week1\task-1\src\cleaner.py", line 46, in main
+    cleaned = [c for row in reader if (c := clean_row(row)) is not None]
+                                            ^^^^^^^^^^^^^^
+  File "G:\Halyna_work\traineeship_Amsterdam\Data-track\week-1\c55-data-week1\task-1\src\cleaner.py", line 38, in clean_row
+    "salary": clean_salary(row.get("salary", "")),
+              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+  File "G:\Halyna_work\traineeship_Amsterdam\Data-track\week-1\c55-data-week1\task-1\src\utils.py", line 55, in clean_salary
+    return int(salary)
+           ^^^^^^^^^^^
+ValueError: invalid literal for int() with base 10: '62.000'
+
+The problem was that int() can only parse strings that look like valid integers, for example "62000". It cannot parse "62.000".
 
 ## The Prompt
 
 <!-- The exact text you sent to the LLM. Include the code you pasted with
 it. -->
+I am working on a beginner Python CSV cleaning task.
+
+The cleaner reads rows from messy_users.csv and calls this helper function:
+
+def clean_salary(raw: str) -> int | None:
+    salary = raw.strip()
+
+    if salary == "":
+        return None
+
+    if salary.lower() == "n/a":
+        return None
+
+    salary = salary.replace('"', "")
+    salary = salary.replace(",", "")
+
+    return int(salary)
+
+The assignment says:
+- salaries like "85000", " 95000", and '"68,000"' should become integers
+- empty values and "N/A" should become None
+- if the value cannot be parsed, the function should return None instead of crashing
+
+When I run the script, I get this error:
+
+ValueError: invalid literal for int() with base 10: '62.000'
+
+How should I fix clean_salary so the script keeps running?
 
 ## The Solution
 
 <!-- What did the LLM suggest? Did it work on the first try, or did you
 need a follow-up? -->
+The AI suggested wrapping the final int(salary) conversion in a try/except ValueError block. It also suggested keeping the cleaning steps for whitespace, quotes, and commas before the conversion.
+
+The fixed function was:
+
+def clean_salary(raw: str) -> int | None:
+    salary = raw.strip()
+    if salary in ("", "N/A"):
+        return None
+
+    salary = salary.replace('"', "")
+    salary = salary.replace(",", "")
+    salary = salary.strip()
+
+    try:
+        return int(salary)
+    except ValueError:
+        return None
+
+This worked after I reran the command:
+
+python3 src/cleaner.py --input data/messy_users.csv --output output/clean_users.json
+
+The script no longer crashed. Instead, rows with invalid salary formats were still included in the output, but their "salary" value became null in the JSON file.
 
 ## Reflection
 
 <!-- A few sentences on: did you understand WHY the original code was
 broken, or did you just accept the fix? What would you do differently next
 time? -->
+I understood why the original code was broken. The mistake was assuming that every non-empty and non-N/A salary could be converted with int(). That assumption was too strong because real CSV data can contain unexpected formats.
+
+I did not need to change cleaner.py, because the orchestration logic was already correct. The bug belonged inside the helper function that parsed one field.
+
+Next time, I would test helper functions separately before running the whole script. For example, I would manually check:
+
+clean_salary("85000")
+clean_salary('"68,000"')
+clean_salary("N/A")
+clean_salary("")
+clean_salary("62.000")
+
+That would catch the parsing problem earlier and make the traceback easier to understand.