Bash variables, if, and a loop

02Read one construct at a time

In note_label='red blue', the assignment has no spaces around =. The straight quotes keep the two words together as its value. Later, "$note_label" expands that value as one argument. Quoting an expansion preserves its argument boundary instead of allowing word splitting or filename expansion.

The loop for tone in amber violet silver; do ...; done assigns each listed value to tone and runs the body once per value. do starts the body and done ends it. A semicolon or newline separates shell commands. "$tone" passes the current value to printf as one argument.

The command if [ -n 'practice' ]; then ...; fi tests whether a string is nonempty. The [ command receives -n and the string as arguments, then a closing ] argument; those spaces matter. A successful test gives status zero, so the commands after then run. fi ends this condition.

A command after done is outside the loop. Its printed line does not create another iteration. Make one trace row per assignment, then a separate row for the later condition. Count body executions and output lines separately.

Use task-specific variable names. HOME and PATH already affect home-directory expansion and command lookup, so do not replace them for an exercise. The supplied script only prints text. Review source before using any later script that can change files or contact another system.

note_label='red blue'

note_label = red blue

not yet passed

not yet printed

03Try tracing then running

The complete script source supplies the activity evidence. It uses three tone values and then a separate conditional command. The accompanying transcript retains output recorded earlier with Bash 5.2.21; compare your prediction with that record without treating it as a new run.

On paper, mark each assignment, the line printed by the body, and the later nonempty test. If running it, use Bash explicitly after reading all source. UTF-8 describes the text encoding; LF means the Linux-style line ending. These details help avoid confusing a text-format problem with a permission problem.

#!/bin/bash
# Original harmless trace.
for tone in amber violet silver; do
  printf '%s\n' "$tone"
done
if [ -n 'practice' ]; then
  printf '%s\n' 'nonempty'
fi

Trace the loop body separately from later commands

Read trace-loop.sh first. Make a row for each assigned tone value, then
a separate row for the condition after done. The record below retains
earlier output from Bash 5.2.21. Comparing your prediction with that
record does not mean you have executed the script yourself.

trace-loop.sh stdout observed with Bash 5.2.21:
amber
violet
silver
nonempty
Three loop iterations; the conditional line is outside the loop.
Separate question loop: for tone in red blue green white; do ...; done
Assume no nested loop, break, return or exit. Count body runs only.

1. Read the complete trace-loop.sh source. Make a paper row for each tone assignment, its printed value and the conditional test after the loop. Expected: three body runs.
2. Predict the separate conditional line. Expected: nonempty after amber, violet and silver because the test's string is nonempty. That fourth output line is not a fourth loop iteration.
3. If doing live work, save the exact source as a UTF-8 file with LF newlines under your owned lab and run bash with its quoted actual path, such as bash ~/Downloads/trace-loop.sh. The command interprets the reviewed source; no execute-bit change is needed.
4. Compare actual stdout with the four predicted lines. If you see a parsing issue, inspect quotes, newlines and which interpreter ran; do not add permissions or sudo to fix syntax.
5. Keep the source and trace note. No background job, file write or remote action occurs, so execution needs no operational cleanup. Paper-only work is marked execution deferred.

04Optional challenge

Allow about ten extra minutes. Move the conditional inside the loop on paper. Predict each printed line in order and compare body count with output count. Explain which boundary changed; you do not need to edit or run another script.

05Fresh cases for the checks

Use these fictional cases for the checks below. They have different inputs from the worked examples. Read them here or make notes; no installation or external service is needed.

parcel_label='copper grey'
printf '%s\n' "$parcel_label"
Predict the value arguments after the format string, not the format string itself.
Source-reading case only; no execution is recorded.

for badge in "" teal "" coral; do
  if [ -n "$badge" ]; then
    printf '%s\n' "$badge"
  fi
done
printf '%s\n' 'complete'
Each quoted empty string is a listed loop value. Predict stdout; no run is recorded.

for stop_name in "north wing" central "south wing" annex courtyard; do
  printf '%s\n' "$stop_name"
done
printf '%s\n' 'route-end'
No nested loop, early exit or other control flow is present.

room_label='east hall'
Original command: printf '%s\n' "$room_label"
Changed command:  printf '%s\n' $room_label
Assume ordinary Bash whitespace word splitting. The value contains no filename-pattern characters.
For this printf format, each resulting value argument is printed on its own line.
Only the quotes around the expansion are removed. Predict value arguments and stdout.

stamp=''
if [ -n "$stamp" ]; then
  printf '%s\n' 'inside'
fi
printf '%s\n' 'closed'
Compared with a nonempty stamp, this card changes only its value to the empty string.

06Explain arguments, iterations and output

Optional self-check: explain why a multiword variable value can be one argument, why a loop visit need not print, and why a command after done does not add a loop iteration.

Review your explanation: connect double quotes to preserved argument boundaries; evaluate -n using the current string; mark do/done and then/fi boundaries; and distinguish value arguments, body runs and output lines. Keep predictions separate from any actual run. This explanation is self-reviewed.