DBIx::Recordset - Perl extension for DBI recordsets


DBIx::Recordset - Perl extension for DBI recordsets


 use DBIx::Recordset;
 # Setup a new object and select some recods...
 *set = DBIx::Recordset -> Search ({'!DataSource' => 'dbi:Oracle:....',
                                    '!Table'      => 'users',
                                    '$where'      => 'name = ? and age > ?',
                                    '$values'     => ['richter', 25] }) ;
 # Get the values of field foo ...
 print "First Records value of foo is $set[0]{foo}\n" ;
 print "Second Records value of foo is $set[1]{foo}\n" ;
 # Get the value of the field age of the current record ...
 print "Age is $set{age}\n" ;
 # Do another select with the already created object...
 $set -> Search ({name => 'bar'}) ;
 # Show the result...
 print "All users with name bar:\n" ;
 while ($rec = $set -> Next)
    print $rec -> {age} ;
 # Setup another object and insert a new record
 *set2 = DBIx::Recordset -> Insert ({'!DataSource' => 'dbi:Oracle:....',
                                     '!Table'      => 'users',
                                     'name'        => 'foo',
                                     'age'         => 25 }) ;

 # Update this record (change age from 25 to 99)...
 $set -> Update ({age => 99}, {name => 'foo'}) ;


DBIx::Recordset is a perl module for abstraction and simplification of database access.

The goal is to make standard database access (select/insert/update/delete) easier to handle and independend of the underlying DBMS. Special attention is made on web applications to make it possible to handle the state-less access and to process the posted data of formfields, but DBIx::Recordset is not limited to web applications.

DBIx::Recordset uses the DBI API to access the database, so it should work with every database for which a DBD driver is available (see also DBIx::Compat).

Most public functions take a hash reference as parameter, which makes it simple to supply various different arguments to the same function. The parameter hash can also be taken from a hash containing posted formfields like those available with CGI.pm, mod_perl, HTML::Embperl and others.

Before using a recordset it is necessary to setup an object. Of course the setup step can be made with the same function call as the first database access, but it can also be handled separately.

Most functions which set up an object return a typglob. A typglob in Perl is an object which holds pointers to all datatypes with the same name. Therefore a typglob must always have a name and can't be declared with my. You can only use it as global variable or declare it with local. The trick for using a typglob is that setup functions can return a reference to an object, an array and a hash at the same time.

The object is used to access the object's methods, the array is used to access the records currently selected in the recordset and the hash is used to access the current record.

If you don't like the idea of using typglobs you can also set up the object, array and hash separately, or just set the ones you need.


Since most methods take a hash reference as argument, here is a description of the valid arguments first.

Setup Parameters

All parameters starting with an '!' are only recognized at setup time. If you specify them in later function calls they will be ignored. You can also preset these parameters with the TableAttr method of DBIx::Database. This allows you to presetup most parameters for the whole database and they will be use every time you create a new DBIx::Recordset object, without specifing it every time.

Specifies the database to which to connect. This information can be given in the following ways:
Same as the first parameter to the DBI connect function.

DBIx::Recordset object
Takes the same database handle as the given DBIx::Recordset object.

DBIx::Database object
Takes Driver/DB/Host from the given database object. See the DBIx::Database manpage for details about DBIx::Database object. When using more then one Recordset object, this is the most efficient method.

DBIx::Datasbase object name
Takes Driver/DB/Host from the database object which is saved under the given name ($saveas parameter to DBIx::Database -> new)

an DBI database handle
Uses given database handle.

Tablename. Multiple tables are comma-separated.

Username. Same as the second parameter to the DBI connect function.

Password. Same as the third parameter to the DBI connect function.

Reference to a hash which holds the attributes for the DBI connect function. See perldoc DBI for a detailed description.

Fields which should be returned by a query. If you have specified multiple tables the fieldnames should be unique. If the names are not unique you must specify them along with the tablename (e.g. tab1.field).

NOTE 1: Fieldnames specified with !Fields can't be overridden. If you plan to use other fields with this object later, use $Fields instead.

NOTE 2: The keys for the returned hash normally don't have a table part. Only the fieldname part forms the key. (See !LongNames for an exception.)

NOTE 3: Because the query result is returned in a hash, there can only be one out of multiple fields with the same name fetched at once. If you specify multiple fields with the same name, only one is returned from a query. Which one this actually is depends on the DBD driver. (See !LongNames for an exception.)

NOTE 4: Some databases (e.g. mSQL) require you to always qualify a fieldname with a tablename if more than one table is accessed in one query.

The TableFilter parameter specifies which tables should be honoured when DBIx::Recordset searches for links between tables (see below). When given as parameter to DBIx::Database it filters for which tables DBIx::Database retrieves metadata. Only thoses tables are used which starts with prefix given by !TableFilter. Also the DBIx::Recordset link detection tries to use this value as a prefix of table names, so you can leave out this prefix when you write a fieldname that should be detected as a link to another table.

When set to 1, the keys of the hash returned for each record not only consist of the fieldnames, but are built in the form table.field.

Fields which should be used for ordering any query. If you have specified multiple tables the fieldnames should be unique. If the names are not unique you must specify them among with the tablename (e.g. tab1.field).

NOTE 1: Fieldnames specified with !Order can't be overridden. If you plan to use other fields with this object later, use $order instead.

Condition which describes the relation between the given tables (e.g. tab1.id = tab2.id) (See also !TabJoin.)
  '!Table'       => 'tab1, tab2',
  '!TabRelation' => 'tab1.id=tab2.id',
  'name'         => 'foo'
  This will generate the following SQL statement:
  SELECT * FROM tab1, tab2 WHERE name = 'foo' and tab1.id=tab2.id ;

!TabJoin allows you to specify an INNER/RIGHT/LEFT JOIN which is used in a SELECT statement. (See also !TabRelation.)
  '!Table'   => 'tab1, tab2',
  '!TabJoin' => 'tab1 LEFT JOIN tab2 ON (tab1.id=tab2.id)',
  'name'     => 'foo'
  This will generate the following SQL statement:
  SELECT * FROM tab1 LEFT JOIN tab2 ON  (tab1.id=tab2.id) WHERE name = 
'foo' ;

Name of the primary key. When this key appears in a WHERE parameter list (see below), DBIx::Recordset will ignore all other keys in the list, speeding up WHERE expression preparation and execution. Note that this key does NOT have to correspond to a field tagged as PRIMARY KEY in a CREATE TABLE statement.

Name of the primary key. In contrast to !PrimKey this field is treated as an autoincrement field. If the database does not support autoincrement fields, but sequences the field is set to the next value of a sequence (see !Sequence and !SeqClass) upon each insert. If a !SeqClass is given the values are always retrived from the sequence class regardless if the DBMS supports autoincrement or not. The value from this field from the last insert could be retrieved by the function LastSerial.

Name of the sequence to use for this table when inserting a new record and !Serial is defind. Defaults to <tablename>_seq.

Name and Parameter for a class that can generate unique sequence values. This is a string that holds comma separated values. The first value is the class name and the following parameters are given to the new constructor. See also DBIx::Recordset::FileSeq and DBIx::Recordset::DBSeq.

Example: '!SeqClass' => 'DBIx::Recordset::FileSeq, /tmp/seq'

!WriteMode specifies which write operations to the database are allowed and which are disabled. You may want to set !WriteMode to zero if you only need to query data, to avoid accidentally changing the content of the database.

NOTE: The !WriteMode only works for the DBIx::Recordset methods. If you disable !WriteMode, it is still possible to use do to send normal SQL statements to the database engine to write/delete any data.

!WriteMode consists of some flags, which may be added together:

DBIx::Recordset::wmNONE (0)
Allow no write access to the table(s)

DBIx::Recordset::wmINSERT (1)

DBIx::Recordset::wmUPDATE (2)

DBIx::Recordset::wmDELETE (4)

DBIx::Recordset::wmCLEAR (8)
To allow DELETE for the whole table, wmDELETE must be also specified. This is necessary for assigning a hash to a hash which is tied to a table. (Perl will first erase the whole table, then insert the new data.)

DBIx::Recordset::wmALL (15)
Allow every access to the table(s)

Default is wmINSERT + wmUPDATE + wmDELETE

If present, this will cause DBIx::Recordset to store all rows which will be fetched between consecutive accesses, so it's possible to access data in a random order. (e.g. row 5, 2, 7, 1 etc.) If not specified, rows will only be fetched into memory if requested, which means that you will have to access rows in ascending order. (e.g. 1,2,3 if you try 3,2,4 you will get an undef for row 2 while 3 and 4 is ok) see also DATA ACCESS below.

By default, the hash returned by the setup function is tied to the current record. You can use it to access the fields of the current record. If you set this parameter to true, the hash will by tied to the whole database. This means that the key of the hash will be used as the primary key in the table to select one row. (This parameter only has an effect on functions which return a typglob.)

This parameter defines how empty and undefined values are handled. The values 1 and 2 may be helpful when using DBIx::Recordset inside a CGI script, because browsers send empty formfields as empty strings.
  1. 0 (default)
    An undefined value is treated as SQL NULL: an empty string remains an empty string.

  2. All fields with an undefined value are ignored when building the WHERE expression.

  3. All fields with an undefined value or an empty string are ignored when building the WHERE expression.

NOTE: The default for versions before 0.18 was 2.

Filters can be used to pre/post-process the data which is read from/written to the database. The !Filter parameter takes a hash reference which contains the filter functions. If the key is numeric, it is treated as a type value and the filter is applied to all fields of that type. If the key if alphanumeric, the filter is applied to the named field. Every filter description consists of an array with at least two elements. The first element must contain the input function, and the second element must contain the output function. Either may be undef, if only one of them are necessary. The data is passed to the input function before it is written to the database. The input function must return the value in the correct format for the database. The output function is applied to data read from the database before it is returned to the user.

     '!Filter'   => 
        DBI::SQL_DATE     => 
                sub { shift =~ /(\d\d)\.(\d\d)\.(\d\d)/ ; "19$3$2$1"},
                sub { shift =~ /\d\d(\d\d)(\d\d)(\d\d)/ ; "$3.$2.$1"}
        'datefield' =>
                sub { shift =~ /(\d\d)\.(\d\d)\.(\d\d)/ ; "19$3$2$1"},
                sub { shift =~ /\d\d(\d\d)(\d\d)(\d\d)/ ; "$3.$2.$1"}

Both filters convert a date in the format dd.mm.yy to the database format 19yymmdd and vice versa. The first one does this for all fields of the type SQL_DATE, the second one does this for the fields with the name datefield.

The !Filter parameter can also be passed to the function TableAttr of the DBIx::Database object. In this case it applies to all DBIx::Recordset objects which use these tables.

A third parameter can be optionally specified. It could be set to DBIx::Recordset::rqINSERT, DBIx::Recordset::rqUPDATE, or the sum of both. If set, the InputFunction (which is called during UPDATE or INSERT) is always called for this field in updates and/or inserts depending on the value. If there is no data specified for this field as an argument to a function which causes an UPDATE/INSERT, the InputFunction is called with an argument of undef.

During UPDATE and INSERT the input function gets either the string 'insert' or 'update' passed as second parameter.

This allows you to get a clear text description of a linked table, instead of (or in addition to) the !LinkField. For example, if you have a record with all your bills, and each record contains a customer number, setting !LinkName DBIx::Recordset can automatically retrieve the name of the customer instead of (or in addition to) the bill record itself.
  1. select additional fields
    This will additionally select all fields given in !NameField of the Link or the table attributes (see TableAttr).

  2. build name in uppercase of !MainField
    This takes the values of !NameField of the Link or the table attributes (see TableAttr) and joins the content of these fields together into a new field, which has the same name as the !MainField, but in uppercase.

  3. replace !MainField with the contents of !NameField
    Same as 2, but the !MainField is replaced with ``name'' of the linked record.

See also !Links and WORKING WITH MULTIPLE TABLES below

This parameter can be used to link multiple tables together. It takes a reference to a hash, which has - as keys, names for a special ``linkfield'' and - as value, a parameter hash. The parameter hash can contain all the Setup parameters. The setup parameters are taken to construct a new recordset object to access the linked table. If !DataSource is omitted (as it normally should be), the same DataSource (and database handle), as the main object is taken. There are special parameters which can only occur in a link definition (see next paragraph). For a detailed description of how links are handled, see WORKING WITH MULTIPLE TABLES below.

Link Parameters

The !MailField parameter holds a fieldname which is used to retrieve a key value for the search in the linked table from the main table. If omitted, it is set to the same value as !LinkedField.

The fieldname which holds the key value in the linked table. If omitted, it is set to the same value as !MainField.

This specifies the field or fields which will be used as a ``name'' for the destination table. It may be a string or a reference to an array of strings. For example, if you link to an address table, you may specify the field ``nickname'' as the name field for that table, or you may use ['name', 'street', 'city'].

Look at !LinkName for more information.

You can give an SQL Statement (or an array reference of SQL statements), that will be executed every time, just after an connect to the db. As third possibilty you can give an hash reference. After every successful connect, DBIx::Recordset excutes the statements, in the element which corresponds to the name of the driver. '*' is executed for all drivers.

Specifies default values for new rows that are inserted via hash or array access. The Insert method ignores this parameter.

Setting this parameter to zero will cause DBIx::Recordset to not tie the returned rows to an DBIx::Recordset::Row object and instead returns an simple hash. The benefit of this is that it will speed up things, but you aren't able to write to such an row, nor can you use the link feature with such a row.

Set the debug level. See DEBUGGING.

Only for tieing a hash! Gives an where expression (either as string or as hashref) that is used to prefetch records from that database. All following accesses to the tied hash only access this prefetched data and don't execute any database queries. See !Expires how to force a refetch. Giving a '*' as value to !PreFetch fetches the whole table into memory.
 The following example prefetches all record with id < 7:
 tie %dbhash, 'DBIx::Recordset::Hash', {'!DataSource'   =>  $DSN,
                                        '!Username'     =>  $User,
                                        '!Password'     =>  $Password,
                                        '!Table'        =>  'foo',
                                        '!PreFetch'     =>  {
                                                             '*id' => '<',
                                                             'id' => 7
                                        '!PrimKey'      =>  'id'} ;
 The following example prefetches all records:
 tie %dbhash, 'DBIx::Recordset::Hash', {'!DataSource'   =>  $DSN,
                                        '!Username'     =>  $User,
                                        '!Password'     =>  $Password,
                                        '!Table'        =>  'bar',
                                        '!PreFetch'     =>  '*',
                                        '!PrimKey'      =>  'id'} ;

Only for tieing a hash! If the values is numeric, the prefetched data will be refetched is it is older then the given number of seconds. If the values is a CODEREF the function is called and the data is refetched is the function returns true.

Only for tieing a hash! Gives an reference to an function that is called when more then one record for a given hash key is found to merge the records into one. The function receives a refence to both records a arguments. If more the two records are found, the function is called again for each following record, which is already merged data as first parameter.
 The following example sets up a hash, that, when more then one record with the same id is
 found, the field C<sum> is added and the first record is returned, where the C<sum> field
 contains the sum of B<all> found records:
 tie %dbhash, 'DBIx::Recordset::Hash', {'!DataSource'   =>  $DSN,
                                        '!Username'     =>  $User,
                                        '!Password'     =>  $Password,
                                        '!Table'        =>  'bar',
                                        '!MergeFunc'    =>  sub { my ($a, $b) = @_ ; $a->{sum} += $b->{sum} ; },
                                        '!PrimKey'      =>  'id'} ;

Where Parameters

The following parameters are used to build an SQL WHERE expression

Give an SQL WHERE expression literaly. If $where is specified, all other where parameters described below are ignored. The only expection is $values which can be used to give the values to bind to the placeholders in $where

Values which should be bound to the placeholders given in $where.
 *set = DBIx::Recordset -> Search ({'!DataSource' => 'dbi:Oracle:....',
                                    '!Table'      => 'users',
                                    '$where'      => 'name = ? and age > ?',
                                    '$values'     => ['richter', 25] }) ;

NOTE: Filters defined with !Filter are not applied to these values, because DBIx::Recordset has no chance to know with values belongs to which field.

Value for field. The value will be quoted automatically, if necessary. The value can also be an array ref in which case the values are put together with the operator passed via $valueconj (default: or)
  'name' => [ 'mouse', 'cat'] will expand to name='mouse' or name='cat'

Value for field. The value will always be quoted. This is only necessary if DBIx::Recordset cannot determine the correct type for a field.

Value for field. The value will never be quoted, but will converted a to number. This is only necessary if DBIx::Recordset cannot determine the correct type for a field.

Value for field. The value will not be converted in any way, i.e. you have to quote it before supplying it to DBIx::Recordset if necessary.

Values for multiple fields. The value must be in one/all fields depending on $compconj Example: '+name|text' => 'abc' will expand to name='abc' or text='abc'

'or' or 'and' (default is 'or'). Specifies the conjunction between multiple fields. (see above)

Regular expression for splitting a field value in multiple values (default is '\t') The conjunction for multiple values could be specified with $valueconj. By default, only one of the values must match the field.
 'name' => "mouse\tcat" will expand to name='mouse' or name='cat'
 NOTE: The above example can also be written as 'name' => [ 'mouse', 'cat']

'or' or 'and' (default is 'or'). Specifies the conjunction for multiple values.

'or' or 'and' (default is 'and') conjunction between fields

Default operator if not otherwise specified for a field. (default is '=')

  • Operator for the named field
     'value' => 9, '*value' => '>' expand to value > 9

    Could also be an array ref, so you can pass different operators for the values. This is mainly handy when you need to select a range

        $set -> Search  ({id          => [5,    7   ],
                         '*id'        => ['>=', '<='],
                         '$valueconj' => 'and'})  ;
      This will expanded to "id >= 5 and id <= 7"

    NOTE: To get a range you need to specify the $valueconj parameter as and because it defaults to or.

    $expr can be used to group parts of the where expression for proper priority. To specify more the one sub expression, add a numerical index to $expr (e.g. $expr1, $expr2)
        $set -> Search  ({id          => 5,
                         '$expr'      => 
                            'name'  => 'Richter',
                            'country' => 'de',
                            '$conj'   => 'or'
                          }) ;
        This will expand to
            (name = 'Richter' or country = 'de') and id = 5

  • Search parameters

    First row to fetch. The row specified here will appear as index 0 in the data array.

    Maximum number of rows to fetch. Every attempt to fetch more rows than specified here will return undef, even if the select returns more rows.

    Add the number supplied with $max to $start. This is intended to implement a next button.

    Subtract the number supplied with $max from $start. This is intended to implement a previous button.

    Fieldname(s) for ordering (ORDER BY) (must be comma-separated, could also contain USING)

    Fieldname(s) for grouping (GROUP BY) (must be comma-separated, could also contain HAVING).

    String which is appended to the end of a SELECT statement, can contain any data.

    Fields which should be returned by a query. If you have specified multiple tables the fieldnames should be unique. If the names are not unique you must specify them along with the tablename (e.g. tab1.field).

    NOTE 1: If !fields is supplied at setup time, this can not be overridden by $fields.

    NOTE 2: The keys for the returned hash normally don't have a table part. Only the fieldname part forms the key. (See !LongNames for an exception.)

    NOTE 3: Because the query result is returned in a hash, there can only be one out of multiple fields with the same name fetched at once. If you specify multiple fields with same name, only one is returned from a query. Which one this actually is, depends on the DBD driver. (See !LongNames for an exception.)

    Name of primary key. DBIx::Recordset assumes that if specified, this is a unique key to the given table(s). DBIx::Recordset can not verify this. You are responsible for specifying the right key. If such a primary exists in your table, you should specify it here, because it helps DBIx::Recordset optimize the building of WHERE expressions.

    See also !primkey

    Execute parameters

    The following parameters specify which action is to be executed:

    search data

    update record(s)

    insert record

    delete record(s)

    setup empty object



    The data which is returned by a Select or a Search can be accessed in two ways:

    1.) Through an array. Each item of the array corresponds to one of the selected records. Each array-item is a reference to a hash containing an entry for every field.

    $set[1]{id} access the field 'id' of the second record found
    $set[3]{name} access the field 'name' of the fourth record found

    The record is fetched from the DBD driver when you access it the first time and is stored by DBIx::Recordset for later access. If you don't access the records one after each other, the skipped records are not stored and therefore can't be accessed anymore, unless you specify the !StoreAll parameter.

    2.) DBIx::Recordset holds a current record which can be accessed directly via a hash. The current record is the one you last accessed via the array. After a Select or Search, it is reset to the first record. You can change the current record via the methods Next, Prev, First, Add.

    $set{id} access the field 'id' of the current record
    $set{name} access the field 'name' of the current record

    Instead of doing a Select or Search you can directly access one row of a table when you have tied a hash to DBIx::Recordset::Hash or have specified the !HashAsRowKey Parameter. The hashkey will work as primary key to the table. You must specify the !PrimKey as setup parameter.

    $set{4}{name} access the field 'name' of the row with primary key = 4


    One way to update/insert data into the database is by using the Update, Insert or Execute method of the DBIx::Recordset object. A second way is to directly assign new values to the result of a previous Select/Search.

    Example: # setup a new object and search all records with name xyz *set = DBIx::Recordset -> Search ({'!DataSource' => 'dbi:db:tab', '!PrimKey => 'id', '!Table' => 'tabname', 'name' => 'xyz'}) ;

      #now you can update an existing record by assigning new values
      #Note: if possible, specify a PrimKey for update to work faster
      $set[0]{'name'} = 'zyx' ;
      # or insert a new record by setting up an new array row
      $set[9]{'name'} = 'foo' ;
      $set[9]{'id'}   = 10 ;
      # if you don't know the index of a new row you can obtain
      # one by using Add
      my $i = $set -> Add () ;
      $set[$i]{'name'} = 'more foo' ;
      $set[$i]{'id'}   = 11 ;
      # or add an empty record via Add and assign the values to the current
      # record
      $set -> Add () ;
      $set{'name'} = 'more foo' ;
      $set{'id'}   = 11 ;
      # or insert the data directly via Add
      $set -> Add ({'name' => 'even more foo',
                    'id'   => 12}) ;
      # NOTE: up to this point, NO data is actually written to the db!
      # we are done with that object,  Undef will flush all data to the db
      DBIx::Recordset::Undef ('set') ;

    IMPORTANT: The data is not written to the database until you explicitly call flush, or a new query is started, or the object is destroyed. This is to keep the actual writes to the database to a minimum.


    DBIx::Recordset has some nice features to make working with multiple tables and their relations easier.


    First, you can specify more than one table to the !Table parameter. If you do so, you need to specify how both tables are related. You do this with !TabRelation parameter. This method will access all the specified tables simultanously.

    Join Example:

    If you have the following two tables, where the field street_id is a pointer to the table street:

      table name
      name      char (30),
      street_id  integer
      table street
      id        integer,
      street    char (30),
      city      char (30)

    You can perform the following search:

      *set = DBIx::Recordset -> Search ({'!DataSource' => 'dbi:drv:db',
                         '!Table'      => 'name, street',
                         '!TabRelation'=> 'name.street_id = street.id'}) ;

    The result is that you get a set which contains the fields name, street_id, street, city and id, where id is always equal to street_id. If there are multiple streets for one name, you will get as many records for that name as there are streets present for it. For this reason, this approach works best when you have a 1:1 relation.

    It is also possible to specify JOINs. Here's how:

      *set = DBIx::Recordset -> Search ({
                '!DataSource' => 'dbi:drv:db',
                '!Table'   => 'name, street',
                '!TabJoin' => 'name LEFT JOIN street ON (name.street_id=street.id)'}) ;

    The difference between this and the first example is that this version also returns a record even if neither table contains a record for the given id. The way it's done depends on the JOIN you are given (LEFT/RIGHT/INNER) (see your SQL documentation for details about JOINs).


    If you have 1:n relations between two tables, the following may be a better way to handle it:

      *set = DBIx::Recordset -> Search ({'!DataSource' => 'dbi:drv:db',
                         '!Table'      => 'name',
                         '!Links'      => {
                            '-street'  => {
                                '!Table' => 'street',
                                '!LinkedField' => 'id',
                                '!MainField'   => 'street_id'
                        }) ;

    After that query, every record will contain the fields name and street_id. Additionally, there is a pseudofield named -street, which could be used to access another recordset object, which is the result of a query where street_id = id. Use

      $set{name} to access the name field
      $set{-street}{street} to access the first street (as long as the
                                        current record of the subobject isn't
      $set{-street}[0]{street}      first street
      $set{-street}[1]{street}      second street
      $set{-street}[2]{street}      third street
      $set[2]{-street}[1]{street} to access the second street of the
                                        third name

    You can have multiple linked tables in one recordset; you can also nest linked tables or link a table to itself.

    NOTE: If you select only some fields and not all, the field which is specified by '!MainField' must be also given in the '!Fields' or '$fields' parameter.

    NOTE: See also Automatic detection of links below


    In the LinkName feature you may specify a ``name'' for every table. A name is one or more fields which gives a human readable ``key'' of that record. For example in the above example id is the key of the record, but the human readable form is street.

      *set = DBIx::Recordset -> Search ({'!DataSource' => 'dbi:drv:db',
                         '!Table'      => 'name',
                         '!LinkName'   => 1,
                         '!Links'      => {
                            '-street'  => {
                                '!Table' => 'street',
                                '!LinkedField' => 'id',
                                '!MainField'   => 'street_id',
                                '!NameField'   => 'street'
                        }) ;

    For every record in the table, this example will return the fields:

      name  street_id  street

    If you have more complex records, you may also specify more than one field in !NameField and pass it as an reference to an array e.g. ['street', 'city']. In this case, the result will contain

      name  street_id  street  city

    If you set !LinkName to 2, the result will contain the fields

      name  street_id  STREET_ID

    where STREET_ID contains the values of the street and city fields joined together. If you set !LinkName to 3, you will get only

      name  street_id

    where street_id contains the values of the street and city fields joined together.

    NOTE: The !NameField can also be specified as a table attribute with the function TableAttr. In this case you don't need to specify it in every link. When a !NameField is given in a link description, it overrides the table attribute.

    Automatic detection of links

    DBIx::Recordset and DBIx::Database will try to automatically detect links between tables based on the field and table names. For this feature to work, the field which points to another table must consist of the table name and the field name of the destination joined together with an underscore (as in the above example name.street_id). Then it will automatically recognized as a pointer to street.id.

      *set = DBIx::Recordset -> Search ({'!DataSource' => 'dbi:drv:db',
                                         '!Table'      => 'name') ;

    is enough. DBIx::Recordset will automatically add the !Links attribute. Additionally, DBIx::Recordset adds a backlink (which starts with a star ('*')), so for the table street, in our above example, there will be a link, named *name, which is a pointer from table street to all records in the table name where street.id is equal to name.street_id.

    You may use the !Links attribute to specify links which can not be automatically detected.

    NOTE: To specify more then one link from one table to another table, you may prefix the field name with an specifier followed by two underscores. Example: first__street_id, second__street_id. The link (and backlink) names are named with the prefix, e.g. -first__street and the backlink *first__name.


    The DBIx::Database object gathers information about a datasource. Its main purpose is to create, at startup, an object which retrieves all necessary information from the database. This object detects links between tables and stores this information for use by the DBIx::Recordset objects. There are additional methods which allow you to add kinds of information which cannot be retreived automatically.


      $db = DBIx::Database -> new ({'!DataSource'   =>  $DSN,
                                    '!Username'     =>  $User,
                                    '!Password'     =>  $Password,
                                    '!KeepOpen'     => 1}) ;
       *set = DBIx::Recordset -> Search ({'!DataSource'   =>  $db,
                                          '!Table'        =>  'foo',
                                         })  ;

    new ($data_source, $username, $password, \%attr, $saveas, $keepopen)

    Specifies the database to which to connect. Driver/DB/Host. Same as the first parameter to the DBI connect function.

    Username (optional)

    Password (optional)

    Attributes (optional) Same as the attribute parameter to the DBI connect function.

    Name for this DBIx::Database object to save as. The name can be used in DBIx::Database::Get, or as !DataSource parameter in call to the DBIx::Recordset object.

    This is intended as mechanism to retrieve the necessary metadata; for example, when your web server starts (e.g. in the startup.pl file of mod_perl). Here you can give the database object a name. Later in your mod_perl or Embperl scripts, you can use this metadata by specifying this name. This will speed up the setup of DBIx::Recordset object without the need to pass a reference to the DBIx::Database object.

    Normaly the database connection will be closed after the metadata has been retrieved from the database. This makes sure you don't get trouble when using the new method in a mod_perl startup file. You can keep the connection open to use them in further setup calls to DBIx::Recordset objects. When the database is not kept open, you must specify the !Password parameter each time the recordset has to be reopend.

    same as setup parameter !TableFilter

    same as setup parameter !DoOnConnect

    If set, forces DBIx::Database to undef any preexisting database handle and call connect in any case. This is usefull in together with Apache::DBI. While the database connection are still kept open by Apache::DBI, Apache::DBI preforms a test if the handle is still vaild (which DBIx::Database itself wouldn't).

    You also can specify a hashref which can contain the following parameters:

    !DataSource, !Username, !Password, !DBIAttr, !SaveAs, !KeepOpen, !TableFilter, !DoOnConnect, !Reconnect

    $db = DBIx::Database -> DBHdl

    returns the database handle (only if you specify !KeepOpen when calling new).

    $db = DBIx::Database -> Get ($name)

    $name = The name of the DBIx::Database object you wish to retrieve

    Get a DBIx::Database object which has already been set up based on the name.

    $db -> TableAttr ($table, $key, $value)

    get and/or set an attribute for an specfic table.

    Name of table(s). You may use '*' instead of the table name to specify a default value which applies to all tables for which no other value is specified.

    key to set/get

    if present, set key to this value

    $db -> TableLink ($table, $linkname, $value)

    Get and/or set a link description for an table. If no $linkname is given, returns all links for that table.

    Name of table(s)

    Name of link to set/get

    if present, this must be a reference to a hash with the link decription. See !Links for more information.

    $db -> MetaData ($table, $metadata, $clear)

    Get and/or set the meta data for the given table.

    Name of table(s)

    If present, this must be a reference to a hash with the new metadata. You should only use this if you really know what you are doing.

    Clears the metadata for the given table, The next call to DBIx::Database -> new will recreate the metadata. Useful if your table has changed (e.g. by ALTER TABLE).

    $db -> AllTables

    This returns a reference to a hash of the keys to all the tables of the datasource.

    $db -> AllNames ($table)

    Returns a reference to an array of all fieldnames for the given table.

    $db -> AllTypes ($table)

    Returns a reference to an array of all fieldtypes for the given table.

    $db -> do ($statement, $attribs, \%params)
    Same as DBI. Executes a single SQL statement on the open database.

    $db -> CreateTables ($dbschema, $schemaname, $user, $setpriv, $alterconstraints)

    The CreateTables method is used to create an modify the schema of your database. The idea is to define the schema as a Perl data structure and give it to this function, it will compare the actual schema of the database with the one provided and creates new tables, new fields or drop fields as neccessary. It also sets the permission on the tables and is able to create indices for the tables. It will never drop a whole table! NOTE: Create tables cannot deteminate changes of the datatype of a fields, because DBI is not able to provide this information in a standart way.

    Either the name of a file which contains the schema or a array ref. See below how this schema must look like.

    schemaname (only used for Oracle)

    User that should be granted access. See !Grant parameter.

    If set to true, access privilegs are revoked and granted again for already existing tables. That is necessary when $user changes.

    If set to true contrains are cleared/set for already existing fields. DBI doesn't provide a database independ way to check which contrains already exists.

    Schema definition

    If give as a filename, the file must contain an hash %DBDefault and an array @DBSchema. The first gives default and the second is an array of hashs. Every of this hash defines one table.


      %DBDefault =
        '!Grant' => 
      @DBSchema = (
        '!Table' => 'language',
        '!Fields' => 
            'id'            => 'char (2)',
            'directory'     => 'varchar(40)',
            'name'          => 'varchar(40)',
            'europe'        => 'bool', 
        '!PrimKey' => 'id',
        '!Default' =>
            'europe'    => 1,
        '!Init' =>
            {'id' => 'de', 'directory' => 'html_49', 'name' => 'deutsch'},
            {'id' => 'en', 'directory' => 'html_76', 'name' => 'english'},
            {'id' => 'fr', 'directory' => 'html_31', 'name' => 'french'},
       '!Index' =>
            'directory' => '',

    The hash which defines a table can have the following keys:

    Gives the table name

    Array with field names and types. There a some types which a translated database specifc. You can define more database specific translation in Compat.pm.

    If an autoincrementing integer. For databases (like Oracle) that doesn't have such a datatype a sequence is generated to provide the autoincrement value and the fields will be of type integer.

    variables length text with up to 255 characters

    variables length text

    Name of the primary key

    Can contain the same key as the table definintion, but is only executed for a specifc database.


        '!For' => { 
            'Oracle' => {
                '!Constraints' =>
                    'web_id'           => ['foreign key' => 'REFERENCES web (id)'],
                    'prim__menu_id'    => ['!Name'       => 'web_prim_menu_id',
                                           'foreign key' => 'REFERENCES menu (id)',
                                           'not null'    => ''],

    Used to define contraints. See example under !For.
    !Name => <name>
    <constraint> => <second part>
    Used to initialy populate the table.

    Used to set a default value for a field, when the table is created. This doesn't have any affect for further INSERTs/UPDATEs.

    Give the rights that should be grant to $user

    Gives the names for the fields for which indices should be created. If the second parameter for an index is not empty, it gives the index name, otherwise a default name is used.

    $db -> DropTables ($schemaname, $user)

    Drops all tables. Use with care!

    schemaname (only used for Oracle)

    User that should be revoked access. See !Grant parameter.


    In SQL all names (field/tablenames etc.) should be case insensitive. Various DBMS handle the case of names differently. For that reason DBIx::Recordset translates all names to lower case, ensuring your application will run with any DBMS, regardless of whether names are returned in lower/uppercase by the DBMS. Some DBMS are case-sensitive (I know at least Sybase, depending on your collate settings). To use such a case-sensitive DBMS, it is best to create your database with all names written in lowercase. In a situation where this isn't possible, you can set $PreserveCase to 1. In this case DBIx::Recordset will not perform any case translation. NOTE: $PreserveCase is still experimental and may change in future releases.

    FETCHSIZE / $FetchsizeWarn

    Some operations in Perl (i.e. foreach, assigning arrays) need to know the size of the whole array. When Perl needs to know the size of an array it call the method FETCHSIZE. Since not all DBD drivers/DBMS returns the number of selected rows after an SQL SELECT, the only way to really determine the number of selected rows would be to fetch them all from the DBMS. Since this could cause a lot of work, it may be very inefficent. Therefore DBIx::Recordset by default calls die() when Perl calls FETCHSIZE. If you know your DBD drivers returns the correct value in $sth -> rows after the execution of an SELECT, you can set $FetchsizeWarn to zero to let FETCHSIZE return the value from $sth -> rows. Setting it to 1 will cause DBIx::Recordset to only issue a warning, but perform the operation.

    NOTE: Since I don't have enough experience with the behaviour of this feature with different DBMS, this is considered experimental.


    DBIx::Recordset is able to write a logfile so you can see what's happening inside. There are two public variables and the !Debug parameter used for this purpose:

    $DBIx::Recordset::Debug or !Debug
    Debuglevel 0 = off 1 = log only errors 2 = show connect, disconnect and SQL Statements 3 = some more infos 4 = much infos

    $DBIx::Recordset::Debug sets the default debug level for new objects, !Debug can be used to set the debuglevel on a per object basis.

    The filehandle used for logging. The default is STDOUT, unless you are running under HTML::Embperl, in which case the default is the Embperl logfile.

        # open the log file
        open LOG, ">test.log" or die "Cannot open test.log" ;
        # assign filehandle
        *DBIx::Recordset::LOG = \*LOG ; 
        # set debugging level
        $DBIx::Recordset::Debug = 2 ;
        # now you can create a new DBIx::Recordset object


    Since one possible application of DBIx::Recordset is its use in a web-server environment, some attention should paid to security issues.

    The current version of DBIx::Recordset does not include extended security management, but some features can be used to make your database access safer. (More security features will come in future releases.)

    First of all, use the security feature of your database. Assign the web server process as few rights as possible.

    The greatest security risk is when you feed DBIx::Recordset a hash which contains the formfield data posted to the web server. Somebody who knows DBIx::Recordset can post other parameters than those you would expect a normal user to post. For this reason, a primary issue is to override all parameters which should never be posted by your script.

    Example: *set = DBIx::Recordset -> Search ({%fdat, ('!DataSource' => ``dbi:$Driver:$DB'', '!Table' => ``$Table'')}) ;

    (assuming your posted form data is in %fdat). The above call will make sure that nobody from outside can override the values supplied by $Driver, $DB and $Table.

    It is also wise to initialize your objects by supplying parameters which can not be changed.

    Somewhere in your script startup (or at server startup time) add a setup call:

     *set = DBIx::Recordset-> Setup ({'!DataSource'  =>  "dbi:$Driver:$DB",
                                                    '!Table'          =>  "$Table",
                                                    '!Fields'         =>  "a, b, c"}) ;

    Later, when you process a request you can write:

     $set -> Search (\%fdat) ;

    This will make sure that only the database specified by $Driver, $DB, the table specified by $Table and the Fields a, b, and c can be accessed.

    Compatibility with different DBD drivers

    I have put a great deal of effort into making DBIx::Recordset run with various DBD drivers. The problem is that not all necessary information is specified via the DBI interface (yet). So I have made the module DBIx::Compat which gives information about the difference between various DBD drivers and their underlying database systems. Currently, there are definitions for:

    DBD::Oracle (requires DBD::Oracle 0.60 or higher)
    DBIx::Recordset has been tested with all those DBD drivers (on Linux 2.0.32, except DBD::ODBC, which has been tested on Windows '95 using Access 7 and with MS SQL Server).

    If you want to use another DBD driver with DBIx::Recordset, it may be necessary to create an entry for that driver. See perldoc DBIx::Compat for more information.


    The following are some examples of how to use DBIx::Recordset. The Examples are from the test.pl. The examples show the DBIx::Recordset call first, followed by the generated SQL command.

     *set = DBIx::Recordset-> Setup ({'!DataSource'  =>  "dbi:$Driver:$DB",
                                                '!Table'      =>  "$Table"}) ;

    Setup a DBIx::Recordset for driver $Driver, database $DB to access table $Table.

     $set -> Select () ;
     SELECT * from <table> ;
     $set -> Select ({'id'=>2}) ;
     is the same as
     $set1 -> Select ('id=2') ;
     SELECT * from <table> WHERE id = 2 ;
     $set -> Search({ '$fields' => 'id, balance AS paid - total ' }) ;
     SELECT id, balance AS paid - total FROM <table>
     $set -> Select ({name => "Second Name\tFirst Name"}) ;
     SELECT * from <table> WHERE name = 'Second Name' or name = 'First Name' ;
     $set1 -> Select ({value => "9991 9992\t9993",
                           '$valuesplit' => ' |\t'}) ;
     SELECT * from <table> WHERE value = 9991 or value = 9992 or value = 9993 ;
     $set -> Select ({'+name&value' => "9992"}) ;
     SELECT * from <table> WHERE name = '9992' or value = 9992 ;
     $set -> Select ({'+name&value' => "Second Name\t9991"}) ;
     SELECT * from <table> WHERE (name = 'Second Name' or name = '9991) or
                                (value = 0 or value = 9991) ;
     $set -> Search ({id => 1,name => 'First Name',addon => 'Is'}) ;
     SELECT * from <table> WHERE id = 1 and name = 'First Name' and addon = 'Is' ;
     $set1 -> Search ({'$start'=>0,'$max'=>2, '$order'=>'id'})  or die "not ok 
    ($DBI::errstr)" ;
     SELECT * from <table> ORDER BY id ;
     B<Note:> Because of the B<start> and B<max> only records 0,1 will be returned
     $set1 -> Search ({'$start'=>0,'$max'=>2, '$next'=>1, '$order'=>'id'})  or die "not ok 
    ($DBI::errstr)" ;
     SELECT * from <table> ORDER BY id ;
     B<Note:> Because of the B<start>, B<max> and B<next> only records 2,3 will be 
     $set1 -> Search ({'$start'=>2,'$max'=>1, '$prev'=>1, '$order'=>'id'})  or die "not ok 
    ($DBI::errstr)" ;
     SELECT * from <table> ORDER BY id ;
     B<Note:> Because of the B<start>, B<max> and B<prev> only records 0,1,2 will be 
     $set1 -> Search ({'$start'=>5,'$max'=>5, '$next'=>1, '$order'=>'id'})  or die "not ok 
    ($DBI::errstr)" ;
     SELECT * from <table> ORDER BY id ;
     B<Note:> Because of the B<start>, B<max> and B<next> only records 5-9 will be 
     *set6 = DBIx::Recordset -> Search ({  '!DataSource'   =>  "dbi:$Driver:$DB",
                                                            '!Table'            =>  "t1, t2",
                                                            '!TabRelation'  =>
                                            '!Fields'       =>  'id, name, text',
                                            'id'            =>  "2\t4" }) or die "not ok 
    ($DBI::errstr)" ;
     SELECT id, name, text FROM t1, t2 WHERE (id=2 or id=4) and t1.value=t2.value ;
     $set6 -> Search ({'name'            =>  "Fourth Name" }) or die "not ok 
    ($DBI::errstr)" ;
     SELECT id, name, text FROM t1, t2 WHERE (name = 'Fourth Name') and 
     $set6 -> Search ({'id'            =>  3,
                      '$operator'     =>  '<' }) or die "not ok ($DBI::errstr)" ;
     SELECT id, name, text FROM t1, t2 WHERE (id < 3) and t1.value=t2.value ;
     $set6 -> Search ({'id'            =>  4,
                      'name'          =>  'Second Name',
                      '*id'           =>  '<',
                      '*name'         =>  '<>' }) or die "not ok ($DBI::errstr)" ;
     SELECT id, name, text FROM t1, t2 WHERE (id<4 and name <> 'Second Name') and 
    t1.value=t2.value ;
     $set6 -> Search ({'id'            =>  2,
                      'name'          =>  'Fourth Name',
                      '*id'           =>  '<',
                      '*name'         =>  '=',
                      '$conj'         =>  'or' }) or die "not ok ($DBI::errstr)" ;
     SELECT id, name, text FROM t1, t2 WHERE (id<2 or name='Fourth Name') and 
    t1.value=t2.value ;
     $set6 -> Search ({'+id|addon'     =>  "7\tit",
                      'name'          =>  'Fourth Name',
                      '*id'           =>  '<',
                      '*addon'        =>  '=',
                      '*name'         =>  '<>',
                      '$conj'         =>  'and' }) or die "not ok ($DBI::errstr)" ;
     SELECT id, name, text FROM t1, t2 WHERE (t1.value=t2.value) and (  ((name <> 
    Name)) and (  (  id < 7  or  addon = 7)  or  (  id < 0  or  addon = 0)))
     $set6 -> Search ({'+id|addon'     =>  "6\tit",
                      'name'          =>  'Fourth Name',
                      '*id'           =>  '>',
                      '*addon'        =>  '<>',
                      '*name'         =>  '=',
                      '$compconj'     =>  'and',
                      '$conj'         =>  'or' }) or die "not ok ($DBI::errstr)" ;
     SELECT id, name, text FROM t1, t2 WHERE (t1.value=t2.value) and (  ((name = 
    Name)) or (  (  id > 6 and addon <> 6)  or  (  id > 0 and addon <> 0))) ;
     *set7 = DBIx::Recordset -> Search ({  '!DataSource'   =>  "dbi:$Driver:$DB",
                                        '!Table'        =>  "t1, t2",
                                        '!TabRelation'  =>  "t1.id=t2.id",
                                        '!Fields'       =>  'name, typ'}) or die "not ok 
    ($DBI::errstr)" ;
     SELECT name, typ FROM t1, t2 WHERE t1.id=t2.id ;
     %h = ('id'    => 22,
          'name2' => 'sqlinsert id 22',
          'value2'=> 1022) ;
     *set9 = DBIx::Recordset -> Insert ({%h,
                                        ('!DataSource'   =>  "dbi:$Driver:$DB",
                                         '!Table'        =>  "$Table[1]")}) or die "not ok 
    ($DBI::errstr)" ;
     INSERT INTO <table> (id, name2, value2) VALUES (22, 'sqlinsert id 22', 1022) ;
     %h = ('id'    => 22,
          'name2' => 'sqlinsert id 22u',
          'value2'=> 2022) ;
     $set9 -> Update (\%h, 'id=22') or die "not ok ($DBI::errstr)" ;
     UPDATE <table> WHERE id=22 SET id=22, name2='sqlinsert id 22u', value2=2022 ;
     %h = ('id'    => 21,
          'name2' => 'sqlinsert id 21u',
          'value2'=> 2021) ;
     *set10 = DBIx::Recordset -> Update ({%h,
                                        ('!DataSource'   =>  "dbi:$Driver:$DB",
                                         '!Table'        =>  "$Table[1]",
                                         '!PrimKey'      =>  'id')}) or die "not ok 
    ($DBI::errstr)" ;
     UPDATE <table> WHERE id=21 SET name2='sqlinsert id 21u', value2=2021 ;
     %h = ('id'    => 21,
          'name2' => 'Ready for delete 21u',
          'value2'=> 202331) ;
     *set11 = DBIx::Recordset -> Delete ({%h,
                                        ('!DataSource'   =>  "dbi:$Driver:$DB",
                                         '!Table'        =>  "$Table[1]",
                                         '!PrimKey'      =>  'id')}) or die "not ok 
    ($DBI::errstr)" ;
     DELETE FROM <table> WHERE id = 21 ;
     *set12 = DBIx::Recordset -> Execute ({'id'  => 20,
                                       '*id' => '<',
                                       '!DataSource'   =>  "dbi:$Driver:$DB",
                                       '!Table'        =>  "$Table[1]",
                                       '!PrimKey'      =>  'id'}) or die "not ok 
    ($DBI::errstr)" ;
     SELECT * FROM <table> WHERE id<20 ;
     *set13 = DBIx::Recordset -> Execute ({'=search' => 'ok',
                        'name'  => 'Fourth Name',
                        '!DataSource'   =>  "dbi:$Driver:$DB",
                        '!Table'        =>  "$Table[0]",
                        '!PrimKey'      =>  'id'}) or die "not ok ($DBI::errstr)" ;
     SELECT * FROM <table>  WHERE   ((name = Fourth Name))
     $set12 -> Execute ({'=insert' => 'ok',
                        'id'     => 31,
                        'name2'  => 'insert by exec',
                        'value2'  => 3031,
     # Execute should ignore the following params, since it is already setup
                        '!DataSource'   =>  "dbi:$Driver:$DB",
                        '!Table'        =>  "quztr",
                        '!PrimKey'      =>  'id99'}) or die "not ok ($DBI::errstr)" ;
     SELECT * FROM <table> ;
     $set12 -> Execute ({'=update' => 'ok',
                        'id'     => 31,
                        'name2'  => 'update by exec'}) or die "not ok ($DBI::errstr)" ;
     UPDATE <table> SET name2=update by exec,id=31 WHERE id=31 ;
     $set12 -> Execute ({'=insert' => 'ok',
                        'id'     => 32,
                        'name2'  => 'insert/upd by exec',
                        'value2'  => 3032}) or die "not ok ($DBI::errstr)" ;
     INSERT INTO <table> (name2,id,value2) VALUES (insert/upd by exec,32,3032) ;
     $set12 -> Execute ({'=delete' => 'ok',
                        'id'     => 32,
                        'name2'  => 'ins/update by exec',
                        'value2'  => 3032}) or die "not ok ($DBI::errstr)" ;
     DELETE FROM <table> WHERE id=32 ;


    As far as possible for me, support will be available via the DBI Users' mailing list. (dbi-user@fugue.com)


    G.Richter (richter@dev.ecos.de)


    Perl(1) =item DBI(3) =item DBIx::Compat(3) =item HTML::Embperl(3) http://perl.apache.org/embperl/ =item Tie::DBI(3) http://stein.cshl.org/~lstein/Tie-DBI/

     DBIx::Recordset - Perl extension for DBI recordsets