Skip to content

M
mariadb
Commit 801ecd07 authored by unknown's avatar unknown
Browse files
Options

Merge work:/home/bk/mysql into prima.mysql.fi:/home/my/bk/mysql

parents 4c5b71f3 81fa7543
Show whitespace changes
Inline Side-by-side
Showing with 336 additions and 171 deletions
tests/mail_to_db.pl
View file @ 801ecd07
#!/usr/bin/perl #!/usr/bin/perl -w
# Copyright Abandoned 1998 TCX DataKonsult AB & Monty Program KB & Detron HB # Copyright Abandoned 1998 TCX DataKonsult AB & Monty Program KB & Detron HB
# This file is public domain and comes with NO WARRANTY of any kind # This file is public domain and comes with NO WARRANTY of any kind
# #
# This program is brought to you by Janne-Petteri Koilo with the # This program is brought to you by Janne-Petteri Koilo with the
# administration of Michael Widenius. # administration of Michael Widenius.
#
# Rewritten with a lot of bug fixes by Jani Tolonen and Thimble Smith
# 15.12.2000
#
# This program takes your mails and puts them into your database. It ignores # This program takes your mails and puts them into your database. It ignores
# messages with the same from, date and message text. # messages with the same from, date and message text.
# You can use mail-files that are compressed or gzipped and ends with # You can use mail-files that are compressed or gzipped and ends with
...@@ -13,287 +16,372 @@ ...@@ -13,287 +16,372 @@
use DBI; use DBI;
use Getopt::Long; use Getopt::Long;
$VER = "1.6"; $| = 1;
$VER = "2.0";
$opt_db = "mail";
$opt_table = "mails"; $opt_help = 0;
$opt_version = 0;
$opt_debug = 0;
$opt_host = undef();
$opt_port = undef();
$opt_socket = undef();
$opt_db = undef();
$opt_table = undef();
$opt_user = undef();
$opt_password = undef();
$opt_max_mail_size = 65536; $opt_max_mail_size = 65536;
$opt_db_engine = "mysql"; $opt_create = 0;
$opt_host = "localhost"; $opt_test = 0;
$opt_user = $opt_password = ""; $opt_no_path = 0;
$opt_help = $opt_version = $opt_test=0; $opt_stop_on_error = 0;
GetOptions("help","version","user=s","password=s", my ($dbh, $progname, $mail_no_from_f, $mail_no_txt_f, $mail_too_big,
"db_engine=s","db=s","host=s","max_mail_size=s","test") || usage(); $mail_forwarded, $mail_duplicates, $mail_no_subject_f, $mail_inserted);
usage($VER) if ($opt_help || $opt_version || !$ARGV[0]); $mail_no_from_f = $mail_no_txt_f = $mail_too_big = $mail_forwarded =
$mail_duplicates = $mail_no_subject_f = $mail_inserted = 0;
%months= ('Jan' => 1, 'Feb' => 2, 'Mar' => 3, 'Apr' => 4, 'May' => 5, my %months = ('Jan' => 1, 'Feb' => 2, 'Mar' => 3, 'Apr' => 4, 'May' => 5,
'Jun' => 6, 'Jul' => 7, 'Aug' => 8, 'Sep' => 9, 'Oct' => 10, 'Jun' => 6, 'Jul' => 7, 'Aug' => 8, 'Sep' => 9, 'Oct' => 10,
'Nov' => 11, 'Des' => 12); 'Nov' => 11, 'Dec' => 12);
$count_no_from = $count_no_txt = $count_too_big = 0; $progname = $0;
$count_forwarded_msgs = $count_duplicates = $no_subject = 0; $progname =~ s/.*[\/]//;
$inserted_mails = 0;
$dbh=0;
$dbh = DBI->connect("DBI:$opt_db_engine:$opt_db:$opt_host",$opt_user, main();
$opt_password,{ PrintError => 0}) || die $DBI::errstr;
if (!$opt_test) ####
{ #### main sub routine
create_table_if_needed($dbh); ####
}
foreach (@ARGV) sub main
{ {
if (/^(.*)\.(gz|Z)$/) #checks if the file is compressed or gzipped my ($connect_arg, @args, $ignored, @defops, $i);
if (defined(my_which("my_print_defaults")))
{ {
open(FILE, "zcat $_ |"); @defops = `my_print_defaults mail_to_db`;
process_mail_file($dbh,$1); chop @defops;
splice @ARGV, 0, 0, @defops;
} }
else else
{ {
open(FILE,$_); print "WARNING: No command 'my_print_defaults' found; unable to read\n";
process_mail_file($dbh,$_); print "the my.cnf file. This command is available from the latest MySQL\n";
print "distribution.\n";
} }
} GetOptions("help","version","host=s","port=i","socket=s","db=s","table=s",
$dbh->disconnect if (!$opt_test); "user=s","password=s","max_mail_size=i","create","test",
"no_path","debug","stop_on_error")
$ignored = $count_no_from + $count_no_txt + $count_too_big + $count_duplicates + $no_subject; || die "Wrong option! See $progname --help\n";
print "Mails inserted:\t\t\t$inserted_mails\n";
print "Mails ignored:\t\t\t$ignored\n";
print "Mails without \"From:\" -field:\t$count_no_from\n";
print "Mails without message:\t\t$count_no_txt\n";
print "Too big mails (> $opt_max_mail_size):\t$count_too_big\n";
print "Duplicate mails:\t\t$count_duplicates\n";
print "Forwarded mails:\t\t$count_forwarded_msgs\n";
print "No subject:\t\t\t$no_subject\n";
print "Mails altogether:\t\t";
print $inserted_mails+$ignored;
print "\n";
exit(0);
sub usage usage($VER) if ($opt_help || $opt_version || (!$ARGV[0] && !$opt_create));
{
my($VER)=@_;
$0 =~ s/.\/(.+)/$1/; # Check that the given inbox files exist and are regular files
if ($opt_version) for ($i = 0; defined($ARGV[$i]); $i++)
{ {
print "$0 version $VER\n"; die "FATAL: Can't find inbox file: $ARGV[$i]\n" if (! -f $ARGV[$i]);
} }
else
{
print <<EOF;
$0 version $VER
Usage: $0 [options] file1 [file2 file3 ...] $connect_arg = "DBI:mysql:";
push @args, "database=$opt_db" if defined($opt_db);
push @args, "host=$opt_host" if defined($opt_host);
push @args, "port=$opt_port" if defined($opt_port);
push @args, "mysql_socket=$opt_socket" if defined($opt_socket);
push @args, "mysql_read_default_group=mail_to_db";
$connect_arg .= join ';', @args;
$dbh = DBI->connect("$connect_arg", $opt_user, $opt_password)
|| die "Couldn't connect: $DBI::errstr\n";
Description: Inserts mails from file(s) into a database die "You must specify the database; use --db=" if (!defined($opt_db));
die "You must specify the table; use --table=" if (!defined($opt_table));
Options: create_table($dbh) if ($opt_create);
--help show this help and exit
--version shows the version of the program foreach (@ARGV)
--db_engine=... database server (default: $opt_db_engine) {
--db=... database to be used (default: $opt_db) # Check if the file is compressed
--host=... hostname to be used (default: $opt_host) if (/^(.*)\.(gz|Z)$/)
--password=... user password for the db server {
--user=... username for the db server open(FILE, "zcat $_ |");
--max_mail_size=# max size of a mail to be inserted into the db. process_mail_file($dbh, $1);
mail will be ignored if it exceeds this size }
(default $opt_max_mail_size) else
--test Don\'t connect to the database, just write the {
queries to stdout open(FILE, $_);
EOF process_mail_file($dbh, $_);
}
} }
$dbh->disconnect if (!$opt_test);
$ignored = ($mail_no_from_f + $mail_no_subject_f + $mail_no_txt_f +
$mail_too_big + $mail_duplicates);
print "Mails inserted:\t\t\t$mail_inserted\n";
print "Mails ignored:\t\t\t$ignored\n";
print "Mails without \"From:\" -field:\t$mail_no_from_f\n";
print "Mails without message:\t\t$mail_no_txt_f\n";
print "Mails without subject:\t\t$mail_no_subject_f\n";
print "Too big mails (> $opt_max_mail_size):\t$mail_too_big\n";
print "Duplicate mails:\t\t$mail_duplicates\n";
print "Forwarded mails:\t\t$mail_forwarded\n";
print "Total number of mails:\t\t";
print $mail_inserted + $ignored;
print "\n";
exit(0); exit(0);
} }
sub create_table_if_needed ####
{ #### table creation
my ($dbh)=@_; ####
my ($sth,$create);
$sth = $dbh->prepare("select count(*) from $opt_table") or die $dbh->errstr; sub create_table
if (!$sth->execute) {
{ my ($dbh) = @_;
$create = "CREATE TABLE $opt_table (msg_nro mediumint unsigned not null "; my ($sth, $query);
$create .= "auto_increment, date DATETIME NOT NULL, time_zone CHAR(6) ";
$create .= "NOT NULL, mail_from char(120) not null, reply char(120), "; $query = <<EOF;
$create .= "mail_to TEXT, cc TEXT, sbj char(200), txt MEDIUMTEXT NOT "; CREATE TABLE $opt_table
$create .= "NULL, file char(32) noT NULL, hash INT NOT NULL, key "; (
$create .= "(msg_nro), primary key (mail_from, date, time_zone, hash))"; mail_id MEDIUMINT UNSIGNED NOT NULL auto_increment,
$sth = $dbh->prepare($create) or die $dbh->errstr; date DATETIME NOT NULL,
$sth->execute() or die $dbh->errstr; time_zone VARCHAR(20),
} mail_from VARCHAR(120) NOT NULL,
reply VARCHAR(120),
mail_to TEXT,
cc TEXT,
sbj VARCHAR(200),
txt MEDIUMTEXT NOT NULL,
file VARCHAR(64) NOT NULL,
hash INTEGER NOT NULL,
KEY (mail_id),
PRIMARY KEY (mail_from, date, hash))
TYPE=MyISAM COMMENT=''
EOF
$sth = $dbh->prepare($query) or die $DBI::errstr;
$sth->execute() or die "Couldn't create table: $DBI::errstr\n";
} }
####
#### inbox processing
####
sub process_mail_file sub process_mail_file
{ {
my ($dbh,$file_name)= @_; my ($dbh, $file_name) = @_;
my (%values,$type,$check); my (%values, $type, $check);
%values=(); $type=""; $file_name =~ s/.*[\/]// if ($opt_no_path);
$check=0;
%values = ();
$type = "";
$check = 0;
while (<FILE>) while (<FILE>)
{ {
chop; chop;
if ($type ne "message") if ($type ne "message")
{ {
if (/^Reply-To: (.*)/i) # finding different fields from file if (/^Reply-To: (.*)/i)
{ {
$type="reply"; $type = "reply";
$values{$type}= $1; $values{$type} = $1;
} }
elsif (/^From: (.*)/i) elsif (/^From: (.*)/i)
{ {
$type="from"; $type = "from";
$values{$type}= $1; $values{$type} = $1;
} }
elsif (/^To: (.*)/i) elsif (/^To: (.*)/i)
{ {
$type="to"; $type = "to";
$values{$type}= $1; $values{$type} = $1;
} }
elsif (/^Cc: (.*)/i) elsif (/^Cc: (.*)/i)
{ {
$type="cc"; $type = "cc";
$values{$type}= $1; $values{$type} = $1;
} }
elsif (/^Subject: (.*)/i) elsif (/^Subject: (.*)/i)
{ {
$type="subject"; $type = "subject";
$values{$type}= $1; $values{$type} = $1;
} }
elsif (/^Date: (.*)/i) elsif (/^Date: (.*)/i)
{ {
date_parser($1,\%values); date_parser($1, \%values, $file_name);
$type="rubbish"; $type = "rubbish";
} }
elsif (/^[\w\W-]+:\s/) elsif (/^[\w\W-]+:\s/)
{ {
$type="rubbish"; $type = "rubbish";
} }
elsif ($_ eq "") elsif ($_ eq "")
{ {
$type="message"; $type = "message";
$values{$type}=""; $values{$type} = "";
} }
else else
{ {
s/^\s*/ /; s/^\s*/ /;
$values{$type}.= $_; $values{$type} .= $_;
} }
} }
elsif ($check!=0 && $_ ne "") # in case of forwarded messages elsif ($check != 0 && $_ ne "") # in case of forwarded messages
{ {
$values{$type}.= "\n" . $_; $values{$type} .= "\n" . $_;
$check--; $check--;
} }
elsif (/^From .* \d\d:\d\d:\d\d\s\d\d\d\d$/) elsif (/^From .* \d\d:\d\d:\d\d\s\d\d\d\d$/)
{ {
$values{'hash'}= checksum("$values{'message'}"); $values{'hash'} = checksum("$values{'message'}");
update_table($dbh,$file_name,\%values); update_table($dbh, $file_name, \%values);
%values=(); $type=""; %values = ();
$check=0; $type = "";
$check = 0;
} }
elsif (/-* forwarded message .*-*/i) # in case of forwarded messages elsif (/-* forwarded message .*-*/i) # in case of forwarded messages
{ {
$values{$type}.= "\n" . $_; $values{$type} .= "\n" . $_;
$check++; $check++;
$count_forwarded_msgs++; $mail_forwarded++;
} }
else else
{ {
$values{$type}.= "\n" . $_; $values{$type} .= "\n" . $_;
} }
} }
$values{'hash'}= checksum("$values{'message'}"); $values{'hash'} = checksum("$values{'message'}");
update_table($dbh,$file_name,\%values); update_table($dbh, $file_name, \%values);
} }
######## ####
#### get date and timezone
# converts date to the right form ####
sub date_parser sub date_parser
{ {
my ($date_raw,$values)=@_; my ($date_raw, $values, $file_name, $tmp) = @_;
$date_raw =~ /\s*(\d{1,2}) (\w+) (\d{2,4}) (\d+:\d+:\d+)\s*([\w-+]{3-5})?/; # If you ever need to change this test, be especially careful with
# the timezone; it may be just a number (-0600), or just a name (EET), or
# both (-0600 (EET), or -0600 (EET GMT)), or without parenthesis: GMT.
# You probably should use a 'greedy' regexp in the end
$date_raw =~ /^\D*(\d{1,2})\s+(\w+)\s+(\d{2,4})\s+(\d+:\d+)(:\d+)?\s*(\S+.*)?/;
$values->{'date'}=$3 . "-" . $months{$2} . "-" . "$1 $4"; if (!defined($1) || !defined($2) || !defined($3) || !defined($4) ||
$values->{'time_zone'}=$5; !defined($months{$2}))
{
if ($opt_debug || $opt_stop_on_error)
{
print "FAILED: date_parser: 1: $1 2: $2 3: $3 4: $4 5: $5\n";
print "months{2}: $months{$2}\n";
print "date_raw: $date_raw\n";
print "Inbox filename: $file_name\n";
}
exit(1) if ($opt_stop_on_error);
}
$tmp = $3 . "-" . $months{$2} . "-" . "$1 $4";
$tmp.= defined($5) ? $5 : ":00";
$values->{'date'} = $tmp;
print "INSERTING DATE: $tmp\n" if ($opt_debug);
$values->{'time_zone'} = $6;
} }
######### ####
#### Insert to table
# this is runned when the whole mail is gathered. ####
# this actually puts the mail to the database.
sub update_table sub update_table
{ {
my($dbh,$file_name,$values)=@_; my($dbh, $file_name, $values) = @_;
my($query); my($q);
if (! defined($values->{'subject'}) || !defined($values->{'to'})) if (!defined($values->{'subject'}) || !defined($values->{'to'}))
{ {
$no_subject++; $mail_no_subject_f++;
return; # Ignore these return; # Ignore these
} }
$values->{'message'} =~ s/^\s*//; #removes whitespaces from the beginning $values->{'message'} =~ s/^\s*//; #removes whitespaces from the beginning
$values->{'message'} =~ s/\s*$//; #removes whitespaces from the end $values->{'message'} =~ s/\s*$//; #removes whitespaces from the end
$query = "insert into $opt_table values (NULL,'" . $values->{'date'};
$query .= "','" . $values->{'time_zone'} . "',";
$query .= (defined($values->{'from'}) ? $dbh->quote($values->{'from'}) : "NULL") . ",";
$query .= (defined($values->{'reply'}) ? $dbh->quote($values->{'reply'}) : "NULL") . ",";
$query .= (defined($values->{'to'}) ? $dbh->quote($values->{'to'}) : "NULL") . ","; $q = "INSERT INTO $opt_table (";
$query .= (defined($values->{'cc'}) ? $dbh->quote($values->{'cc'}) : "NULL") . ","; $q.= "mail_id,";
$query .= $dbh->quote($values->{'subject'}) . ","; $q.= "date,";
$query .= $dbh->quote($values->{'message'}) . "," . $dbh->quote($file_name); $q.= "time_zone,";
$query .= ",'" . $values->{'hash'} . "')"; $q.= "mail_from,";
$q.= "reply,";
if (length($values->{'message'}) > $opt_max_mail_size) #disables big message $q.= "mail_to,";
$q.= "cc,";
$q.= "sbj,";
$q.= "txt,";
$q.= "file,";
$q.= "hash";
$q.= ") VALUES (";
$q.= "NULL,";
$q.= "'" . $values->{'date'} . "',";
$q.= (defined($values->{'time_zone'}) ?
("'" . $values->{'time_zone'} . "',") : "NULL,");
$q.= defined($values->{'from'}) ? $dbh->quote($values->{'from'}) : "NULL";
$q.= ",";
$q.= defined($values->{'reply'}) ? $dbh->quote($values->{'reply'}) : "NULL";
$q.= ",";
$q.= defined($values->{'to'}) ? $dbh->quote($values->{'to'}) : "NULL";
$q.= ",";
$q.= defined($values->{'cc'}) ? $dbh->quote($values->{'cc'}) : "NULL";
$q.= ",";
$q.= $dbh->quote($values->{'subject'});
$q.= ",";
$q.= $dbh->quote($values->{'message'});
$q.= ",";
$q.= $dbh->quote($file_name);
$q.= ",";
$q.= "'" . $values->{'hash'} . "'";
$q.= ")";
# Don't insert mails bigger than $opt_max_mail_size
if (length($values->{'message'}) > $opt_max_mail_size)
{ {
$count_too_big++; $mail_too_big++;
} }
elsif ($values->{'from'} eq "") #disables mails with no from field # Don't insert mails without 'From' field
elsif ($values->{'from'} eq "")
{ {
$count_no_from++; $mail_no_from_f++;
} }
elsif ($opt_test) elsif ($opt_test)
{ {
print "$query\n"; print "$q\n";
$inserted_mails++; $mail_inserted++;
} }
elsif ($values->{'message'} eq "") #disables mails with no message text # Don't insert mails without the 'message'
elsif ($values->{'message'} eq "")
{ {
$count_no_msg_text++; $mail_no_txt_f++;
} }
elsif ($dbh->do($query)) elsif ($dbh->do($q))
{ {
$inserted_mails++; $mail_inserted++;
} }
elsif (!($dbh->errstr =~ /Duplicate entry /)) #disables duplicates # This should never happen. This means that the above q failed,
# but it wasn't because of a duplicate mail entry
elsif (!($DBI::errstr =~ /Duplicate entry /))
{ {
die "Aborting: Got error '" . $dbh->errstr ."' for query: '$query'\n"; die "FATAL: Got error :$DBI::errstr\nAttempted query was: $q\n";
} }
else else
{ {
$count_duplicates++; $mail_duplicates++;
print "Duplicate mail: query: $q\n" if ($opt_debug);
} }
$query=""; $q = "";
} }
####
########## #### In case you have two identical messages we wanted to identify them
#### and remove additionals; We do this by calculating a hash number of the
# In case you have two identical messages we wanted to identify them #### message and ignoring messages with the same from, date and hash.
# and remove additionals; We do this by calculating a hash number of the #### This function calculates a simple 32 bit hash value for the message.
# message and ignoring messages with the same from, date and hash. ####
# This function calculates a simple 32 bit hash value for the message.
sub checksum sub checksum
{ {
...@@ -308,3 +396,80 @@ sub checksum ...@@ -308,3 +396,80 @@ sub checksum
} }
return $crc; return $crc;
} }
####
#### my_which is used, because we can't assume that every system has the
#### which -command. my_which can take only one argument at a time.
#### Return values: requested system command with the first found path,
#### or undefined, if not found.
####
sub my_which
{
my ($command) = @_;
my (@paths, $path);
return $command if (-f $command && -x $command);
@paths = split(':', $ENV{'PATH'});
foreach $path (@paths)
{
$path = "." if ($path eq "");
$path .= "/$command";
return $path if (-f $path && -x $path);
}
return undef();
}
####
#### usage and version
####
sub usage
{
my ($VER)= @_;
if ($opt_version)
{
print "$progname version $VER\n";
}
else
{
print <<EOF;
$progname version $VER
Description: Insert mails from inbox file(s) into a table.
This program can read group [mail_to_db] from the my.cnf
file. You may want to have db and table set there at least.
Usage: $progname [options] file1 [file2 file3 ...] [>& /path/to/log.txt]
or: $progname [options] --create [file1 file2...] [>& /path/to/log.txt]
Options:
--help Show this help and exit.
--version Show the version number and exit.
--debug Print some extra information during the run.
--host=... Hostname to be used. (Using: $opt_host)
--port=# TCP/IP port to be used with connection. (Using: $opt_port)
--socket=... MySQL UNIX socket to be used with connection.
(Using: $opt_socket)
--db=... Database to be used. (Using: $opt_db)
--table=... Table name for mails. (Using: $opt_table)
--user=... Username for connecting. (Using: $opt_user)
--password=... Password for the user.
--max_mail_size=# Maximum size of a mail.
Beware of the downside letting this variable be too big;
you may easily end up inserting a lot of attached
binary files (like MS Word documents etc), which take
space, make the database slower and are not really
searchable anyway. (Default: $opt_max_mail_size)
--create Create the mails table. This can be done with the first run.
--test Dry run. Print the queries and the result as it would be.
--no_path When inserting the file name, leave out any paths of
the name.
--stop_on_error Stop the run, if an unexpected, but not fatal error occurs
during the run. Without this option some fields may get
unwanted values. --debug will also report about these.
EOF
}
exit(0);
}
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7