71 — Chart: a bar chart in the terminal

Guide 63 multiplied strings into rows, and guide 66 sorted records by a score. A bar chart puts both together: read values from JSON, then turn each value into a row of # blocks whose length is scaled to the largest value. Numbers become a picture you can read at a glance.

#Steps

1. Create chart.json — a list of records, each with a label and a value:

   [
     {"label": "Jan", "value": 12},
     {"label": "Feb", "value": 29},
     {"label": "Mar", "value": 18},
     {"label": "Apr", "value": 41},
     {"label": "May", "value": 24},
     {"label": "Jun", "value": 35}
   ]
   

   copyrun it →

   The labels become row names; the values become bar lengths.

2. Load the list with json_load from guide 14. Collecting the values into their own list makes max easy to call:

   use file latest
   rows = json_load("chart.json")
   values = []
   for row in rows:
       values.append(row["value"])
   show max(values)
   

   copyrun it →

   It prints 41 — the largest value, which sets the scale.

3. A bar is a fraction of the longest bar. max_value is the largest value; int(value / max_value * 20) is that value's share of 20 blocks:

   value = 29
   max_value = 41
   length = int(value / max_value * 20)
   show length
   

   copyrun it →

   It prints 14, because 29 is roughly 14/20 of 41.

4. "#" * length builds the bar itself — the string multiplication from guide 63:

   length = 14
   show "#" * length
   

   copyrun it →

   It prints fourteen # blocks. Repeating a string is how a number becomes a picture.

5. The full program loads the data, finds the scale, and draws one labeled bar per record. Save chart.nme:

   # chart.nme — a horizontal bar chart in the terminal.
   # Run: nme r chart
   # chart.json must be in the same folder.
   # Each bar is # blocks scaled to the largest value.
   
   use file latest
   
   rows = json_load("chart.json")
   
   values = []
   for row in rows:
       values.append(row["value"])
   
   max_value = max(values)
   show f"largest value: {max_value}"
   show ""
   
   for row in rows:
       label = row["label"]
       length = int(row["value"] / max_value * 20)
       bar = "#" * length
       show f"{label:>3}: {bar} {row['value']}"
   

   copyrun it →

   {label:>3} pads every label to three columns so all bars start at the same place. length and bar are recalculated for every row.

6. Run it with the data file present:

   nme r chart
   

   copy

   largest value: 41
   
   Jan: ##### 12
   Feb: ############## 29
   Mar: ######## 18
   Apr: #################### 41
   May: ########### 24
   Jun: ################# 35
   

   copy

   April, the largest value, fills all 20 blocks; January, the smallest, fills 5. Every bar is a fraction of the same scale, so the chart is honest.

7. Korean writes the same steps with 파일 사용 최신, json읽기, 말해, and 막대. The full Korean program is in the Korean guide.

#Try it yourself

Add {"label": "Jul", "value": 47} to chart.json and rerun — the scale moves to 47 and every other bar shrinks. Then change 20 to 40 in the length formula for a wider chart, or add a second "@" bar under each row.

#What you learned

• A record's value becomes a bar length; max(values) sets the scale.
• int(value / max_value * 20) turns a value into a count of blocks.
• "#" * length repeats one character into a bar.
• {label:>3} aligns labels so all bars start at the same column.