使用命令行工具合并 2 个文件

Question

我有 2 个这样的 csv 文件：

id, name, job
1, bob, fireman
3, alice, nurse
7, peter, policeman
...

并且：

id, name, age
2, john, 26
4, craig, 32
5, mary, 45
6, lucy, 23
...

如您所见，它们都是按 id 排序的，第一个 csv 中缺少的 id 实际上在第二个 csv 中。

是否可以通过命令行工具（例如 awk 或类似工具将这 2 个 csv 合并成一个这样的文件？

id, name, job, age
1, bob, fireman,
2, john, , 26
3, alice, nurse,
4, craig, , 32
...

非常感谢您的帮助？

Answer 1

应该这样做：

awk -F, -v OFS=, 'FNR==NR && FNR>1 {a[]=[=10=];c++;next} FNR>1{$NF=" ,"$NF;a[]=[=10=];c++} END {print "id, name, job, age";for (i=1;i<=c;i++) print a[i]}' file1 file2
id, name, job, age
1, bob, fireman
2, john, , 26
3, alice, nurse
4, craig, , 32
5, mary, , 45
6, lucy, , 23
7, peter, policeman

工作原理：

awk -F, -v OFS=, '              # Set input and output Field separator to ","
FNR==NR && FNR>1 {              # For first file except first record do:
    a[]=[=11=]                    # Store records inn to array "a"
    c++                         # Increment "c" for every record
    next}                       # Skip to next record
FNR>1 {                         # For second file except first record do:
    $NF=" ,"$NF                 # Replace last record with an extra ","
    a[]=[=11=]                    # Store records inn to array "a"
    c++}                        # Increment "c" for every record
END {                           # When all file is read do:
    print "id, name, job, age"  # Print header
    for (i=1;i<=c;i++)          # Loop "c" times
        print a[i]}             # Print records
' file1 file2                   # Read the files

FNR==NR经常在读取多个文件时用来区分要处理哪个文件

使用命令行工具合并 2 个文件

Merge 2 files using command line tools

csv

awk

sed

command-line-interface