AWK脚本中Hash table相关语法错误排查求助

报错信息 #

awk: syntax error at source line 30 source file test.awk
context is
dealer_schedule[last_name] = >>> [ <<< date, time, am_pm, first_name, last_name]
awk: illegal statement at source line 30 source file test.awk
awk: syntax error at source line 33 source file test.awk

错误原因汇总 #

你的脚本问题主要来自awk 3.4的旧版语法限制，以及几个基础语法误用：

• awk 3.4不支持**数组字面量[元素1,元素2]**的赋值方式，这是新版gawk才有的特性
• 不存在assoc()函数，awk的关联数组直接使用即可，无需显式初始化
• 文件读取不能用open()和for (line in 文件对象)，要改用awk标准的getline方式
• 逻辑运算符需用&&而非and，旧版awk不支持and作为逻辑判断符
• split()函数用法错误，正确格式是split(字符串, 目标数组, 分隔符)，返回值是元素数量，不能赋值给数组名
• 数组索引未正确初始化，比如mylie_info[i]里的i没有自增逻辑
• 文件关闭要用close("文件名")，不是对象调用.close()方法

修正后的完整脚本 #

#!/usr/bin/awk -f

BEGIN {
    # 读取Mylie的信息到关联数组
    read_mylie_info("Mylie.txt", mylie_info)
}

{
    # 从文件名提取前4位作为日期
    date = substr(FILENAME, 1, 4)
    
    # 分割行数据到fields数组
    split($0, fields, "|")
    
    time = fields[2]
    am_pm = fields[3]
    first_name = fields[4]
    last_name = fields[5]
    
    # 用复合键模拟多维数组存储经销商排班
    if (!(last_name in dealer_record_count)) {
        dealer_record_count[last_name] = 0
    }
    dealer_record_count[last_name]++
    idx = dealer_record_count[last_name]
    dealer_schedule[last_name, idx, "date"] = date
    dealer_schedule[last_name, idx, "time"] = time
    dealer_schedule[last_name, idx, "am_pm"] = am_pm
    dealer_schedule[last_name, idx, "first"] = first_name
    dealer_schedule[last_name, idx, "last"] = last_name
    
    # 匹配Mylie的游玩时间，统计经销商次数
    for (i in mylie_info) {
        if (date == mylie_info[i, "date"] && time == mylie_info[i, "time"] && am_pm == mylie_info[i, "am_pm"]) {
            dealer_counts[last_name]++
        }
    }
}

END {
    # 输出经销商统计结果
    print "dealer,count" > "Notes_Dealer_Analysis.txt"
    for (dealer in dealer_counts) {
        print dealer "," dealer_counts[dealer] >> "Notes_Dealer_Analysis.txt"
    }
    close("Notes_Dealer_Analysis.txt")
    
    # 输出Mylie游玩时在岗的经销商排班
    print "date,time,am_pm,first_name,last_name" > "Dealers_working_during_losses.txt"
    for (dealer in dealer_record_count) {
        for (idx=1; idx<=dealer_record_count[dealer]; idx++) {
            # 检查当前经销商的这条记录是否匹配Mylie的时间
            for (i in mylie_info) {
                if (dealer_schedule[dealer, idx, "date"] == mylie_info[i, "date"] && 
                    dealer_schedule[dealer, idx, "time"] == mylie_info[i, "time"] && 
                    dealer_schedule[dealer, idx, "am_pm"] == mylie_info[i, "am_pm"]) {
                    print dealer_schedule[dealer, idx, "date"] "," \
                          dealer_schedule[dealer, idx, "time"] "," \
                          dealer_schedule[dealer, idx, "am_pm"] "," \
                          dealer_schedule[dealer, idx, "first"] "," \
                          dealer_schedule[dealer, idx, "last"] >> "Dealers_working_during_losses.txt"
                    break
                }
            }
        }
    }
    close("Dealers_working_during_losses.txt")
}

# 读取Mylie信息的函数
function read_mylie_info(filename, arr,    line, fields, i) {
    i = 0
    while ((getline line < filename) > 0) {
        split(line, fields, "|")
        i++
        arr[i, "date"] = fields[1]
        arr[i, "time"] = fields[2]
        arr[i, "am_pm"] = fields[3]
    }
    close(filename)
}

关键修正点说明 #

1. 数组存储方式调整：旧版awk不支持多维数组的直接字面量赋值，改用关联数组+复合键（比如dealer_schedule[last_name, idx, "date"]）来模拟多维存储，兼容所有awk版本。
2. 文件读取逻辑替换：用getline循环读取文件内容，这是awk标准的文件读取方式，避免了旧版不支持的文件对象遍历语法。
3. 逻辑运算符替换：把and换成&&，符合awk的标准语法规范。
4. split函数修正：正确使用split($0, fields, "|")，不再错误地将返回值赋值给数组名。
5. 数组索引初始化：在读取Mylie信息时，用计数器i来维护每条记录的索引，避免无意义的空索引问题。
6. 文件关闭修正：使用close("文件名")来关闭打开的文件，避免资源泄漏。

内容的提问来源于stack exchange，提问作者CApple