從CSV到MySQL數據庫使用Perl

Question

我試圖從CSV一些數據上傳到MySQL數據庫 - 但它不工作

下面一個是我的代碼

#!/usr/bin/perl -w 
use DBI; 
use strict; 
use TEXT::CSV; 
use warnings; 

my $driver  = "mysql"; 
my $database  = "test"; 
my $host   = "localhost" 
my $databaseport = "3306"; 
my $userid  = "root"; 
my $password  = "password"; 
my $csv   = "C:/Perl/scripts/table.csv"; 

my $dsn   = "dbi:mysql:dbname=$databasename;host=$dbhost;port=$dbport;"; 

open (CSV, "$csv") or die "Couldn't open csvfile: $!"; 
my $dbh = DBI->connect($dsn, $userid, $password,{ RaiseError => 1}) 
or die "Could not connect to database! $DBI::errstr"; 
{ 
local $/ = undef; 
    $dbh->do("INSERT INTO student (stud_id,stud_name,dept_id,stud_mark,stud_address) 

    values (?, ?, ?, ?, ?)", undef, <CSV>); 
} 
$dbh->disconnect; 
close CSV; 

Answer 1

有一些問題在這裏。我會列出那些會首先給你錯誤信息的人。

• 沒有模塊TEXT :: CSV。雖然有一個叫做Text::CSV。
• 您在查詢中使用了5個佔位符，但您要通過鑽石運算符<CSV>傳遞csv文件的第一行。這會給出錯誤信息。

然後你的邏輯有問題。您將完整的文件傳遞給數據庫（作爲第一個參數）。那沒有意義。您需要輸入split或使用Text :: CSV來執行此操作並逐行讀取文件。

此外，現在使用帶有三個參數的open並使文件句柄爲詞法是一種很好的做法。

我已經把所有這些都寫成了自制CSV處理的例子。如果您的文件更復雜，請閱讀Text :: CSV並使用它。

use DBI; 
use strict; 
use warnings; 

my $csv   = "C:/Perl/scripts/table.csv"; 
# omitted settings here ... 

my $dbh = DBI->connect($dsn, $userid, $password,{ RaiseError => 1}) 
    or die "Could not connect to database! $DBI::errstr"; 
open (my $fh, '<', $csv) 
    or die "Couldn't open csvfile: $!"; 

# prepare statement handle for reuse in the loop 
my $sth = $dbh->prepare(qq{ 
    INSERT INTO student(stud_id,stud_name,dept_id,stud_mark,stud_address) 
    VALUES (?, ?, ?, ?, ?)}); 

# read the file line by line 
while (my $line = <$fh>) { 
    chomp $line; # remove newline 
    $sth->execute(split /;/, $line); # assuming the separator is a semicolon 
} 

close $fh; 
# DB handle will disconnect implicitly on end of program 

正如你所看到的，我決定先用prepare the statement重新使用它。這在循環中節省了大量時間，因爲DB會記住該語句。

Answer 2

從列表上下文中的文件句柄（即代碼中的<CSV>位）讀取從文件中讀取所有行並將它們作爲列表返回。因此，您的?, ?, ?, ?佔位符每個都會從文件中獲得整行（包括最後的換行符）。對於某些字段（可能是dept_id？），這可能不是有效值，因此INSERT語句失敗。

雖然實際上，您也將$/設置爲undef，這使得它更加令人尷尬。 $/在閱讀文本文件時改變了Perl新概念。將它設置爲undef意味着Perl會將整個文件視爲單行。

我們猜測，您要做的是一次讀取一行CSV文件，並將每個文件抽入數據庫。

#!/usr/bin/perl 
use strict; 
use warnings; 
use DBI; 
use Text::CSV; # case-sensitive! 

my $driver  = "mysql"; 
my $database  = "test"; 
my $host   = "localhost" 
my $databaseport = "3306"; 
my $userid  = "root"; 
my $password  = "password"; 
my $csv   = "C:/Perl/scripts/table.csv"; 

# Connect to database. 
my $dsn = "dbi:mysql:dbname=$databasename;host=$dbhost;port=$dbport;"; 
my $dbh = DBI->connect($dsn, $userid, $password,{ RaiseError => 1}) 
    or die "Could not connect to database! $DBI::errstr"; 

# DBI can be more efficient if you prepare the SQL query once, and then 
# execute it multiple times, rather than calling `do` for each insert. 
my $sth = $dbh->prepare(<<'SQL'); 
    INSERT INTO student (stud_id,stud_name,dept_id,stud_mark,stud_address) 
    VALUES (NULL, ?, ?, ?, ?)" 
SQL 

# Open the CSV file.A 
open my $CSV, '<', $csv 
    or die "Couldn't open csvfile: $!"; 

# Create an instance of Text::CSV. 
my $reader = Text::CSV->new; 

# Use Text::CSV to read a line. 
while (my $row = $reader->getline($CSV)) 
{ 
    # Insert into database. 
    $sth->execute(@$row); 
} 

# Clean up (optional; Perl will do this when your script ends anyway). 
$dbh->disconnect; 
close $CSV;