技术咨询：如何实现脚本每次运行时内部变量值不同及Bash脚本问题

Let's tackle your two questions one by one, with practical examples you can adapt directly.

1. Getting Dynamic Expected Values Each Time the Script Runs #

If you need your script's internal variable to pull a different expected value on every run, here are three common approaches depending on your use case:

Option 1: Random Values (for non-deterministic ranges) #

If the expected value just needs to be a random number within a specific range (e.g., 70-100% for your file check), use Bash's built-in RANDOM variable:

# Generate a random expected percentage between 70 and 100
expected_percent=$(( RANDOM % 31 + 70 ))  # 31 = 100-70+1 to include both endpoints

Option 2: Cycle Through a Predefined List #

If you have a fixed set of expected values you want to rotate through (e.g., 50%, 75%, 90%, 100%), track the last used index in a file to pick the next value each run:

# Define your list of expected values
expected_values=(50 75 90 100)
# File to store the last index we used (persists between runs)
index_tracker="/tmp/expected_value_index.txt"

# Initialize the tracker if it doesn't exist
[[ ! -f $index_tracker ]] && echo 0 > "$index_tracker"

# Get current index and corresponding value
current_index=$(cat "$index_tracker")
current_expected=${expected_values[$current_index]}

# Update index for next run (cycle back to 0 when we reach the end)
next_index=$(( (current_index + 1) % ${#expected_values[@]} ))
echo "$next_index" > "$index_tracker"

# Now use $current_expected in your script logic
echo "Today's expected percentage: $current_expected%"

Option 3: Pull from a Dynamic Local Source #

If the expected value is managed externally (e.g., updated by another script or user), read it from a local file:

# Path to a file that holds the current expected value
expected_value_file="/path/to/expected_value.txt"

# Read the value (make sure the file only contains the number/percentage)
current_expected=$(cat "$expected_value_file")

2. Bash Script for File Check, Status Tracking, and HTML Email #

Let's build a complete script that addresses your requirements: checks folder file count every 30 minutes, tracks verification status, appends data to a log, and sends an HTML table via email.

Full Script Example #

#!/bin/bash

# --------------------------
# Configuration (Edit These)
# --------------------------
EXPECTED_TOTAL=100          # Total number of files we expect
CUTOFF_TIME="2024-05-20 18:00:00"  # Deadline (YYYY-MM-DD HH:MM:SS)
WATCH_DIR="/path/to/your/target/folder"
DATA_LOG="/tmp/file_check_history.csv"
EMAIL_TO="your.email@example.com"
EMAIL_SUBJECT="File Ingestion Status Update"
EXPECTED_THRESHOLD=80       # % of files needed to mark as verified

# --------------------------
# Initialize Log File
# --------------------------
# Add CSV headers if log doesn't exist
if [[ ! -f $DATA_LOG ]]; then
    echo "Timestamp,File Count,Percentage,Verification Status" > "$DATA_LOG"
fi

# --------------------------
# Calculate Current Metrics
# --------------------------
# Count files (adjust find command if you need to filter by file type, e.g., -name "*.csv")
CURRENT_COUNT=$(find "$WATCH_DIR" -type f | wc -l)

# Calculate percentage (avoid division by zero)
if [[ $EXPECTED_TOTAL -eq 0 ]]; then
    CURRENT_PERCENT="0%"
else
    CURRENT_PERCENT="$(( (CURRENT_COUNT * 100) / EXPECTED_TOTAL ))%"
fi

# --------------------------
# Determine Verification Status
# --------------------------
CURRENT_TIME=$(date +"%Y-%m-%d %H:%M:%S")
VERIFICATION_STATUS="未验证"

# Mark as verified if threshold is met OR deadline has passed
if [[ ${CURRENT_PERCENT%?} -ge $EXPECTED_THRESHOLD ]] || [[ "$CURRENT_TIME" > "$CUTOFF_TIME" ]]; then
    VERIFICATION_STATUS="已验证"
fi

# --------------------------
# Append Data to Log
# --------------------------
echo "$CURRENT_TIME,$CURRENT_COUNT,$CURRENT_PERCENT,$VERIFICATION_STATUS" >> "$DATA_LOG"

# --------------------------
# Generate HTML Table
# --------------------------
HTML_TABLE="<html><body>
<h3>File Ingestion Status History</h3>
<table border='1' cellpadding='5' cellspacing='0'>
<thead>
<tr bgcolor='#f0f0f0'>
<th>Timestamp</th>
<th>Files Received</th>
<th>Percentage of Total</th>
<th>Verification Status</th>
</tr>
</thead>
<tbody>"

# Read log lines (skip header) and build table rows
tail -n +2 "$DATA_LOG" | while IFS=, read -r TS COUNT PERCENT STATUS; do
    # Highlight verified rows for clarity
    if [[ "$STATUS" == "已验证" ]]; then
        HTML_TABLE+="<tr bgcolor='#d4edda'>"
    else
        HTML_TABLE+="<tr>"
    fi
    HTML_TABLE+="<td>$TS</td><td>$COUNT</td><td>$PERCENT</td><td>$STATUS</td></tr>"
done

HTML_TABLE+="</tbody></table></body></html>"

# --------------------------
# Send HTML Email
# --------------------------
# Use mailx (adjust command if your system uses sendmail or another tool)
echo -e "Content-Type: text/html\n\n$HTML_TABLE" | mailx -s "$EMAIL_SUBJECT" "$EMAIL_TO"

Key Notes: #

• Status Logic: The script marks status as "已验证" if either the file percentage meets/exceeds your threshold, or the deadline has passed (whichever comes first).
• Log Persistence: All runs append data to a CSV log, so the HTML table will grow with each check.
• Email Formatting: The HTML table includes subtle styling (gray header, green background for verified rows) to make it easier to read.
• Cron Job Setup: To run every 30 minutes, add this to your crontab (run crontab -e):

  */30 * * * * /path/to/your/script.sh >> /tmp/script_run_log.txt 2>&1
  

Troubleshooting Tips: #

• Ensure mailx is installed (on Debian/Ubuntu: sudo apt install mailutils; on RHEL/CentOS: sudo yum install mailx).
• Verify the find command counts the correct files (add filters like -name "*.csv" if needed).
• Double-check date formats for CUTOFF_TIME (must match date +"%Y-%m-%d %H:%M:%S" output).

内容的提问来源于stack exchange，提问作者shruk