Why Every Full-Stack Developer Should Learn Pandas in 2026

import pandas as pd

df = pd.read_csv("customers.csv")

print(df.shape)          # (rows, columns)
print(df.dtypes)         # column types
print(df.head())         # first 5 rows
print(df.isna().sum())   # missing values per column

import pandas as pd

df = pd.read_csv("customers.csv")

# Standardise column names: " Email Address " -> "email_address"
df.columns = df.columns.str.strip().str.lower().str.replace(" ", "_")

# Normalise text and types
df["email_address"] = df["email_address"].str.strip().str.lower()
df["signup_date"] = pd.to_datetime(df["signup_date"], errors="coerce")
df["city"] = df["city"].str.title()

# Remove exact duplicate emails, keep the first
df = df.drop_duplicates(subset="email_address", keep="first")

import pandas as pd

REQUIRED = {"name", "email", "signup_date"}

def validate_upload(df: pd.DataFrame):
    df.columns = df.columns.str.strip().str.lower()
    missing = REQUIRED - set(df.columns)
    if missing:
        return None, [f"Missing column: {c}" for c in sorted(missing)]

    df["email"] = df["email"].astype(str).str.strip().str.lower()
    df["signup_date"] = pd.to_datetime(df["signup_date"], errors="coerce")

    bad_email = ~df["email"].str.contains(r"^[^@\s]+@[^@\s]+\.[^@\s]+$", regex=True)
    bad_date = df["signup_date"].isna()
    dupes = df.duplicated(subset="email", keep="first")

    errors = []
    for idx in df.index[bad_email]:
        errors.append(f"Row {idx + 2}: invalid email")
    for idx in df.index[bad_date]:
        errors.append(f"Row {idx + 2}: unreadable date")
    for idx in df.index[dupes]:
        errors.append(f"Row {idx + 2}: duplicate email")

    valid = df[~(bad_email | bad_date | dupes)]
    return valid, errors

# pip install fastapi uvicorn pandas python-multipart
import io
import pandas as pd
from fastapi import FastAPI, UploadFile
from validator import validate_upload

app = FastAPI()

@app.post("/customers/upload")
async def upload_customers(file: UploadFile):
    content = await file.read()
    df = pd.read_csv(io.BytesIO(content))
    valid, errors = validate_upload(df)
    if valid is None:
        return {"saved": 0, "errors": errors}
    # save valid rows with your ORM or df.to_sql(...)
    return {"saved": len(valid), "errors": errors[:50]}

import pandas as pd
from sqlalchemy import create_engine

engine = create_engine("postgresql+psycopg://user:pass@localhost/shop")

def monthly_revenue():
    df = pd.read_sql("SELECT created_at, amount FROM orders WHERE status = 'paid'", engine,
                     parse_dates=["created_at"])
    monthly = (
        df.set_index("created_at")["amount"]
          .resample("MS").sum()
          .reset_index()
    )
    monthly["month"] = monthly["created_at"].dt.strftime("%Y-%m")
    return monthly[["month", "amount"]].to_dict(orient="records")

# -> [{"month": "2026-09", "amount": 125000.0}, ...]

# pip install openpyxl
import io
import pandas as pd
from fastapi import FastAPI
from fastapi.responses import StreamingResponse

app = FastAPI()

@app.get("/orders/export")
def export_orders():
    df = pd.DataFrame([{"id": 1, "customer": "Asha", "amount": 1200}])  # replace with your query
    buffer = io.BytesIO()
    df.to_excel(buffer, index=False, sheet_name="Orders")
    buffer.seek(0)
    return StreamingResponse(
        buffer,
        media_type="application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
        headers={"Content-Disposition": "attachment; filename=orders.xlsx"},
    )

BULK IMPORT CHECKLIST

[ ] Downloadable sample template (CSV / XLSX)
[ ] Max file size and row limit defined
[ ] Required columns checked
[ ] Types coerced: dates, numbers, booleans
[ ] Duplicates handled (reject or merge?)
[ ] Errors returned with spreadsheet row numbers
[ ] Valid rows saved in a single transaction
[ ] Import logged: user, time, rows saved, rows rejected

REPORT SPEC

Report name     : ____________________
Requested by    : ____________________
Filters         : date range / status / city
Grouping        : month / category / customer
Metrics         : sum / count / average of ____
Output          : JSON for chart / table / XLSX
Expected rows   : ~____ (decide SQL vs pandas)
Refresh         : live / cached daily

# test_validator.py
import pandas as pd
from validator import validate_upload

def test_missing_column_is_reported():
    df = pd.DataFrame({"name": ["A"], "email": ["a@x.com"]})
    valid, errors = validate_upload(df)
    assert valid is None
    assert errors == ["Missing column: signup_date"]

def test_bad_email_is_rejected():
    df = pd.DataFrame({"name": ["A", "B"],
                       "email": ["a@x.com", "not-an-email"],
                       "signup_date": ["2026-01-05", "2026-01-06"]})
    valid, errors = validate_upload(df)
    assert len(valid) == 1
    assert "Row 3: invalid email" in errors