Data Visualization: Audit chart choices

Review a set of planned charts and explain what is wrong with each, the way an editor reviews a report.

What to build

A program that checks chart choices against the rules of good practice.

Requirements

  • Each line holds a chart (pie, bar, line, scatter or histogram), a goal (compare, trend, distribution, relationship or composition), the number of categories and yes or no for negative values.
  • Problems, in this order: a pie whose goal is not composition: a pie only shows parts of a whole; a pie with more than 5 categories: too many slices for a pie (N); a pie with negative values: a pie cannot show negative values; a line whose goal is compare: a line suggests an order; use bars; a scatter whose goal is not relationship: a scatter plot shows relationships; a histogram whose goal is not distribution: a histogram shows a distribution; a bar chart with more than 12 categories: too many bars to read (N); show the top ones.
  • Print I. CHART for GOAL: ok or I. CHART for GOAL: followed by the problems joined by ; , with I counting from 1.

Input

Line 1: n. Then n lines CHART GOAL CATEGORIES NEGATIVES.

Output

n lines.

Constraints

1 ≤ n ≤ 50.

Example

Input:

4
pie composition 4 no
pie composition 9 yes
line compare 6 no
bar compare 20 no

Output:

1. pie for composition: ok
2. pie for composition: too many slices for a pie (9); a pie cannot show negative values
3. line for compare: a line suggests an order; use bars
4. bar for compare: too many bars to read (20); show the top ones

Why it works

Each rule comes from how people read charts: angles of many slices cannot be compared, lines imply order, and every chart type answers one kind of question, so a mismatch is worth flagging before anyone reads the chart.

Examples

Example 1 · Input
4
pie composition 4 no
pie composition 9 yes
line compare 6 no
bar compare 20 no
Expected output
1. pie for composition: ok
2. pie for composition: too many slices for a pie (9); a pie cannot show negative values
3. line for compare: a line suggests an order; use bars
4. bar for compare: too many bars to read (20); show the top ones
Example 2 · Input
2
scatter trend 0 no
histogram distribution 0 no
Expected output
1. scatter for trend: a scatter plot shows relationships
2. histogram for distribution: ok

Prompts inside input() are ignored when checking, so input("n: ") is fine.

Hints

Hint 1

Collect the problems for each chart in a list, then print them joined by "; " or "ok".

Hint 2

Check the problems in the order the requirements list them.

Solution

Show one correct solution (try it yourself first)
def problems(chart, goal, count, negatives):
    found = []
    if chart == "pie":
        if goal != "composition":
            found.append("a pie only shows parts of a whole")
        if count > 5:
            found.append(f"too many slices for a pie ({count})")
        if negatives:
            found.append("a pie cannot show negative values")
    if chart == "line" and goal == "compare":
        found.append("a line suggests an order; use bars")
    if chart == "scatter" and goal != "relationship":
        found.append("a scatter plot shows relationships")
    if chart == "histogram" and goal != "distribution":
        found.append("a histogram shows a distribution")
    if chart == "bar" and count > 12:
        found.append(f"too many bars to read ({count}); show the top ones")
    return found


n = int(input())
for i in range(1, n + 1):
    chart, goal, count, negatives = input().split()
    found = problems(chart, goal, int(count), negatives == "yes")
    print(f"{i}. {chart} for {goal}: {'; '.join(found) if found else 'ok'}")

Lesson: Choosing the right chart →