UBK:ITER

This tag acts much like a for or foreach, over a range of numbers or over an array, specified in a variety of ways.

If used to loop over a range of numbers, two new environment variables are set:

• ITER-START: the start value
• ITER-CURRENT: the current value

and the attributes that the tag handles are the followings: this modality is used if both inizio and fine are specified; no other attribute is considered.

attribute	required	default	domain	notes
inizio	X		int	the start value
fine	X		int	the end value
step		1	int	the increment for the index

If used to loop over an array, two new environment variables are set:

• ITER-CURRENT-KEY: the index or the current key of the array
• ITER-CURRENT: the current value

and the attributes that the tag handles are the followings: this modality is used if one of array or dati are specified; no other attribute is considered.

attribute	required	default	domain	notes
array			string	the name of a global variable that actually contains the array to loop over
dati			mixed	any expression returning a string suitable to define an array (for example: dati="%implode(array(1,2,3,4,5))", dati="'one' => 1,'two' => 2")

UBK:RIGHE

This tag loops over the current result set, repeating its content for every tuple. Some attributes can be specified to loop over just a part of the result set:

this modality is used if page-size is specified; if neither start nor page is specified, then the loop starts from the first record (like if page was set to 0).

attribute	required	default	domain	notes
page-size	X		int	the number of records to display, starting from the first
start			int	the record number to start from
page		0	int	the page number (assumed containing page-size record) to start from

UBK:RIGHE-CORRELATE

This tag acts much like ubk:righe, but changing the result set to loop over, that is the set of tuple in another table somehow related to current tuple.

attribute	required	default	domain	notes
tabella	X		string	the new table / view to loop over
campo-id-tabella		tabella's primary key	string	the name of the field in tabella used as id
campo-id-link		campo-id-tabella	string	the name of the field in tabella used to find related tuples
campo-id		campo-id-link	string	the name of the field in current table used to find related tuples
filtro			SQL	additional filter for tabella's tuples
ordine			SQL	additional sort clause for tabella's tuples
colonne			column|column[,column]	which columns should be fetched: specifying less columns in wide tables speeds up the process
distinct		FALSE	FALSE, TRUE	should tuples be forced to be distinct

A common and minimal use of this tag is the following:

<ubk:righe-correlate tabella="xxx" campo-id-link="i_yyy_id">
....
</ubk:righe-correlate>

When primary key in tabella and foreign key in current table have same name, you can just specify tabella. Default filtering is done like this:

SELECT * FROM tabella WHERE campo-id-link = current_table[ campo-id ]

that usually is equivalent to

SELECT * FROM tabella WHERE primary-key = current_table[ tabella.primary-key ]

Some attributes can be specified to loop over just a part of the result set:

this modality is used if page-size is specified; if neither start nor page is specified, then the loop starts from the first record (like if page was set to 0).

attribute	required	default	domain	notes
page-size	X		int	the number of records to display, starting from the first
start			int	the record number to start from
page		0	int	the page number (assumed containing page-size record) to start from

UBK:RIGHE-RICORSIVE

This tags loops over a table with tree structure, i.e. whos tuples contains a link to a "parent" tuple.

attribute	required	default	domain	notes
tabella	X			the name of the database table containing values to display
campo-id-figlio		tabella's id field		the name of the database field containing ids (primary key)
campo-id-padre	X			the name of the database field containing the link to the parent node
nodi-aperti		null	array	keys in this array contains ids of nodes whose children will be displayed; if no array is provided, all nodes will be printed; if an empty array is provided, only roots will be printed
filtro			SQL	filter condition on root nodes (those having campo-id-padre NULL)
filtro-generale			SQL	filter condition on internal nodes (those having campo-id-padre NOT NULL)
ordine		campo-id-padre asc	SQL	sort clause for records; default clause is always pre-pended to any sort clause specified
indentazione	X		int	indentation amount for nested nodes, expressed in um-ind
um-ind		px	px, em, pt, ...	CSS indentation unit measure
incremento-ind		0	int	indentation increment for each level
start-ind		0	int	starting value for indentation accumulator
metodo-ind		padding-left	padding-left, margin-left, ...	CSS indentation method
child-display		block	block, inline, ...	CSS display value for node's children block
tag		div	HTML tag	tag used to build node and children blocks
js-over			js	javascript to be executed on mouse over node
js-out			js	javascript to be executed on mouse out node
js-click			js	javascript to be executed on node click
class			string	the css class to be applied to the node

During iteration, these environment variables are set:

• RIGHE-RICORSIVE-LIVELLO: carries the nesting level, starting from 0 (root)
• RIGHE-RICORSIVE-INDICE: a progressive index, incremented for every node printed out
• RIGHE-RICORSIVE-FIGLI: the amount of current node's children

Initial filter condition is built in this way:

• if both filtro and filtro-generale are set, then condition is : (filtro and campo-id-padre is null) or (filtro-generale and campo-id-padre is not null)
• if only filtro is set, then condition is: (filtro and campo-id-padre is null)
• if only filtro-generale is set then condition is filtro-generale

Then an additional node condition , to build whole filter condition:

• if at least filtro is set: (condition) and (campo-id-padre in nodi-aperti)
• if filtro-generale is set: (condition) and ((campo-id-padre is null) or (campo-id-padre in nodi-aperti))
• else: (campo-id-padre is null) or (campo-id-padre in nodi-aperti)

Seems a bit odd, but this ensure minimal tuple selection over the whole tree.