如何使用shell腳本在行和列中轉換我的文件

Question

我有一個文件格式如下。任何人都可以在列中轉換它？ 我已經嘗試過下面的awk命令，但是如果一個客戶有多個主機名，它會創建更多的4列。

awk '/"customer_name":/{if (x)print x;x="";}{x=(!x)?$0:x","$0;}END{print x;}' filename 

輸入：

customer_name: "abc" 
    "HostName": "tm-1" 
    "LastDayRxBytes": 0 
    "Status": "offline" 
    "HostName": "tm-2" 
    "LastDayRxBytes": 0 
    "Status": "offline" 
    "HostName": "tm-3" 
    "LastDayRxBytes": 0 
    "Status": "offline" 
    "HostName": "new-va-threat-01" 
    "LastDayRxBytes": 0 
    "Status": "offline" 
customer_name: "xyz" 
    "HostName": "tm-56" 
    "LastDayRxBytes": 10708747 
    "Status": "ok" 
customer_name: "def" 
customer_name: "uvw" 
    "HostName": "tm-23" 
    "LastDayRxBytes": 34921829912 
    "Status": "ok" 
customer_name: "new cust" 
    "HostName": "tm-1-3" 
    "LastDayRxBytes": 33993187093 
    "Status": "ok" 
customer_name: "a12 d32 ffg" 
customer_name: "bcd abc" 
customer_name: "mno opq" 
customer_name: "abc dhg pvt ltd." 
    "HostName": "tm-10" 
    "LastDayRxBytes": 145774401010 
    "Status": "ok" 
    "HostName": "tm-ngtm-13" 
    "LastDayRxBytes": 150159680874 
    "Status": "ok" 
    "HostName": "new-ngtm-11" 
    "LastDayRxBytes": 207392526747 
    "Status": "ok" 
    "HostName": "old-ngtm-06" 
    "LastDayRxBytes": 17708734533 
    "Status": "ok" 
    "HostName": "tm-08" 
    "LastDayRxBytes": 559289251 
    "Status": "ok" 
    "HostName": "tm-12" 
    "LastDayRxBytes": 534145552271 
    "Status": "ok" 

我希望它在柱被打印和行爲：

Column 1    Column 2    Column 3    Column 4 
CustName    Host     Last RX    Status 
abc     tm-1     0     offline 
abc     tm-2     0     offline 
abc     tm-3     0     offline 
abc     new-va-threat-01  0     offline 
xyz     tm-56    10708747    ok 
def      
uvw     tm-23    34921829912   ok 
new_cust    tm-1-3    33993187093   ok 
a12 d32 ffg 
acd abc 
mno opq 
abc dhg pvt ltd.  tm-10    145774401010   ok 
abc dhg pvt ltd.  tm-ngtm-13   150159680874   ok 
abc dhg pvt ltd.  new-ngtm-11   207392526747   ok 
abc dhg pvt ltd.  old-ngtm-06   17708734533   ok 
abc dhg pvt ltd.  tm-08    559289251   ok 
abc dhg pvt ltd.  tm-12    534145552271   ok 

Answer 1

我會寫這個

awk -F": " -v OFS="\t" ' 
    BEGIN {print "CustName", "Host", "Last RX", "Status"} 
    { 
     gsub(/"/,"") 
     sub(/^[[:blank:]]+/,"") 
    } 
    $1 == "customer_name" { 
     if ("customer_name" in data && !have_data) 
      print data["customer_name"] 
     have_data = 0 
    } 
    { 
     data[$1] = $2 
    } 
    ("HostName" in data) && ("LastDayRxBytes" in data) && ("Status" in data) { 
     print data["customer_name"], data["HostName"], data["LastDayRxBytes"], data["Status"] 
     delete data["HostName"] 
     delete data["LastDayRxBytes"] 
     delete data["Status"] 
     have_data = 1 
    } 
' file | column -s $'\t' -t 

CustName   Host    Last RX  Status 
abc    tm-1    0    offline 
abc    tm-2    0    offline 
abc    tm-3    0    offline 
abc    new-va-threat-01 0    offline 
xyz    tm-56    10708747  ok 
def 
uvw    tm-23    34921829912 ok 
new cust   tm-1-3   33993187093 ok 
a12 d32 ffg 
bcd abc 
mno opq 
abc dhg pvt ltd. tm-10    145774401010 ok 
abc dhg pvt ltd. tm-ngtm-13  150159680874 ok 
abc dhg pvt ltd. new-ngtm-11  207392526747 ok 
abc dhg pvt ltd. old-ngtm-06  17708734533 ok 
abc dhg pvt ltd. tm-08    559289251  ok 
abc dhg pvt ltd. tm-12    534145552271 ok 

Answer 2

Perl來救援：

perl -lne ' 
    if (/customer_name: "(.*)"/) { 
     print $h{name} unless $h{printed} || !%h; 
     undef $h{printed} if $1 ne $h{name}; 
     $h{name} = $1; 
    } else { 
     /"([^"]+)": "?([^"]+)"?/ and $h{$1} = $2; 
     $h{printed} = print join "\t", 
      @h{qw{ name HostName LastDayRxBytes Status }} 
      if "Status" eq $1; 
    } 
    END { print $h{name} unless $h{printed} || !%h } 
    ' < input_file 

• 的%h散列用於收集有關要打印的行的信息。
• 讀取客戶名稱時，如果還沒有打印過，則會打印先前的客戶名稱。同樣的情況發生在輸入的最後，以打印可能的最後一個客戶而沒有詳細信息。
• 讀取狀態時會打印一行。

Answer 3

GNU AWK溶液：

$ cat tst.awk 
BEGIN { 
    RS="customer_name: " 
    pr("Column1", "Column2", "Column3", "Column4") 
    pr("Custname", "Host", "Last RX", "Status") 
} 
match($0, /"([^"]+)"/, cust) { 
    printed=0 
    str=substr($0, RLENGTH+2) 
    while (match(str, /"HostName":\s"([^"]+)"\s+"LastDayRxBytes":\s(\S+)\s+"Status":\s"([^"]+)"\s/, col)){ 
     str=substr(str, RLENGTH+3) 
     pr(cust[1], col[1], col[2], col[3]) 
     printed=1 
    } 
    if (!printed) pr(cust[1]) 
} 
function pr(cust,host,rx,status) { 
    printf "%-16s\t%-16s\t%-16s\t%-10s\n", cust, host, rx, status 
} 

基於示例性輸入，可以使用正則表達式處理這一個和匹配功能也是如此。測試它：

$ awk -f tst.awk input.txt 
Column1    Column2    Column3    Column4 
Custname   Host    Last RX    Status 
abc     tm-1    0     offline 
abc     tm-2    0     offline 
abc     tm-3    0     offline 
abc     new-va-threat-01 0     offline 
xyz     tm-56    10708747   ok 
def 
uvw     tm-23    34921829912   ok 
new cust   tm-1-3    33993187093   ok 
a12 d32 ffg 
bcd abc 
mno opq 
abc dhg pvt ltd. tm-10    145774401010  ok 
abc dhg pvt ltd. tm-ngtm-13   150159680874  ok 
abc dhg pvt ltd. new-ngtm-11   207392526747  ok 
abc dhg pvt ltd. old-ngtm-06   17708734533   ok 
abc dhg pvt ltd. tm-08    559289251   ok 
abc dhg pvt ltd. tm-12    534145552271  ok 

說明：

• 記錄分隔符RS設置在customer_name:，所以$ 0包含每個客戶的所有主機，RX和狀態信息。
• 第一場比賽與正則表達式"([^"]+)"將捕獲客戶
• 第二場比賽與正則表達式"HostName":\s"([^"]+)"\s+"LastDayRxBytes":\s(\S+)\s+"Status":\s"([^"]+)"\s將捕獲主機名，rx和狀態。
• 如果第二場比賽成功，請縮短您想在下一場比賽中使用的字符串。

我知道，這不是awk way的處理方式，但是輸入的常規格式再次允許這個 - 非常簡潔 - 基於正則表達式的解決方案。