Add test for bulk-insert with special characters (tabs, newlines, backslashes)

t/pt-archiver/bulk_insert_special_chars.t

@@ -0,0 +1,75 @@
+#!/usr/bin/env perl
+
+# Test that pt-archiver --bulk-insert correctly escapes and preserves data
+# containing tabs, newlines, and backslashes (no field misalignment).
+# See: fix for escape() order (backslash first, then newline, then tab).
+
+BEGIN {
+   die "The PERCONA_TOOLKIT_BRANCH environment variable is not set.\n"
+      unless $ENV{PERCONA_TOOLKIT_BRANCH} && -d $ENV{PERCONA_TOOLKIT_BRANCH};
+   unshift @INC, "$ENV{PERCONA_TOOLKIT_BRANCH}/lib";
+};
+
+use strict;
+use warnings FATAL => 'all';
+use English qw(-no_match_vars);
+use Test::More;
+
+use PerconaTest;
+use Sandbox;
+require "$trunk/bin/pt-archiver";
+
+my $dp  = new DSNParser(opts=>$dsn_opts);
+my $sb  = new Sandbox(basedir => '/tmp', DSNParser => $dp);
+my $dbh = $sb->get_dbh_for('source');
+
+if ( !$dbh ) {
+   plan skip_all => 'Cannot connect to sandbox source';
+}
+
+my $cnf = "/tmp/12345/my.sandbox.cnf";
+
+$sb->wipe_clean($dbh);
+$sb->load_file('source', 't/pt-archiver/samples/bulk_insert_special_chars.sql');
+
+# Rows with special characters: tab, newline, backslash, quotes.
+# Using prepared statement so we pass actual characters (not SQL-escaped).
+my @test_rows = (
+   [ 1, 'John "Doe"', "Software Engineer\t\n", 123, 'Senior "Developer"' ],
+   [ 2, 'Alice', "Data Scientist\nwith newline\n", 456, 'Lead Analyst' ],
+   [ 3, 'Bob', "Engineer with tab\tcharacter\n", 789, 'Manager' ],
+   [ 4, 'Charlie', "Backslash\\Test\n", 101, 'Director' ],
+   [ 5, 'Eve', "Quote\'s Test\\t\n", 202, 'Consultant' ],
+);
+
+my $ins = $dbh->prepare(
+   'INSERT INTO bulk_escape.source (id, name, job, stu_id, title) VALUES (?, ?, ?, ?, ?)'
+);
+for my $row ( @test_rows ) {
+   $ins->execute(@$row);
+}
+
+# Archive from source to dest using --bulk-insert (purge source).
+my $output = output(
+   sub { pt_archiver::main(
+      qw(--where 1=1 --bulk-insert --bulk-delete --limit 100 --statistics),
+      '--source', "L=1,D=bulk_escape,t=source,F=$cnf",
+      '--dest',   "D=bulk_escape,t=dest") },
+);
+
+# Verify row counts.
+my $src_count = $dbh->selectrow_array('SELECT COUNT(*) FROM bulk_escape.source');
+my $dst_count = $dbh->selectrow_array('SELECT COUNT(*) FROM bulk_escape.dest');
+is($src_count, 0, 'Source table is purged');
+is($dst_count, scalar(@test_rows), 'All rows archived to dest');
+
+# Compare archived data with expected (order by id).
+my $archived = $dbh->selectall_arrayref(
+   'SELECT id, name, job, stu_id, title FROM bulk_escape.dest ORDER BY id'
+);
+is_deeply($archived, \@test_rows, 'Bulk-insert preserves tabs, newlines, backslashes (no field misalignment)');
+
+$sb->wipe_clean($dbh);
+ok($sb->ok(), 'Sandbox servers') or BAIL_OUT(__FILE__ . ' broke the sandbox');
+done_testing;
+exit;

t/pt-archiver/samples/bulk_insert_special_chars.sql

@@ -0,0 +1,22 @@
+-- Schema for testing pt-archiver --bulk-insert with special characters
+-- (tabs, newlines, backslashes) to ensure no field misalignment.
+DROP DATABASE IF EXISTS `bulk_escape`;
+CREATE DATABASE `bulk_escape`;
+
+CREATE TABLE `bulk_escape`.`source` (
+  `id` int unsigned NOT NULL AUTO_INCREMENT,
+  `name` varchar(500) DEFAULT NULL,
+  `job` varchar(450) DEFAULT NULL,
+  `stu_id` int DEFAULT NULL,
+  `title` varchar(450) DEFAULT NULL,
+  PRIMARY KEY (`id`)
+) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4;
+
+CREATE TABLE `bulk_escape`.`dest` (
+  `id` int unsigned NOT NULL AUTO_INCREMENT,
+  `name` varchar(500) DEFAULT NULL,
+  `job` varchar(450) DEFAULT NULL,
+  `stu_id` int DEFAULT NULL,
+  `title` varchar(450) DEFAULT NULL,
+  PRIMARY KEY (`id`)
+) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4;