Class TreeTableFormat

  • All Implemented Interfaces:
    Serializable, Cloneable, Localized

    public class TreeTableFormat
    extends TabularFormat<TreeTable>
    A parser and formatter for Tree­Table instances. This formatter is given an arbitrary number of Table­Columns to use during the formatting. The first column is taken as the node label. If a Tree­Table is formatted with only that column, then the String result is like the following example:
    Node #1
      ├─Node #2
      │   └─Node #4
      └─Node #3
    If the same Tree­Table is formatted with two columns, then the String result is like the following example:
    Node #1……………………… More #1
      ├─Node #2…………… More #2
      │   └─Node #4… More #4
      └─Node #3…………… More #3
    This representation can be printed to the console output (for example) if the stream uses a monospaced font and supports Unicode characters.
    Customization
    Some formatting characteristics (indentation width, column where to draw the vertical line below nodes) can be modified by calls to the setter methods defined in this formatter. In particular, the dots joining the node labels to their values can be specified by the column separator pattern. The default pattern is "?……[…] ", which means "If the next value is non-null, then insert the "……" string, repeat the '…' character as many time as needed (may be zero), and finally insert a space".
    Safety against infinite recursivity
    Some Tree­Table implementations generate the nodes dynamically as wrappers around Java objects. Such Java objects may contain cyclic associations (A contains B contains C contains A), which result in a tree of infinite depth. Some examples can been found in ISO 19115 metadata. This Tree­Table­Format class contains a safety against such cycles. The algorithm is based on the assumption that for each node, the values and children are fully determined by the user object, if non-null. Consequently for each node C to be formatted, if the user object of that node is the same instance (in the sense of the == operator) than the user object of a parent node A, then the children of the C node will not be formatted.
    Since:
    0.3
    See Also:
    Serialized Form

    Defined in the sis-utility module

    • Constructor Detail

      • TreeTableFormat

        public TreeTableFormat​(Locale locale,
                               TimeZone timezone)
        Creates a new tree table format.
        Parameters:
        locale - the locale to use for numbers, dates and angles formatting, or null for the root locale.
        timezone - the timezone, or null for UTC.
    • Method Detail

      • getColumns

        public TableColumn<?>[] getColumns()
        Returns the table columns to parse and format, or null for the default list of columns. The default is:
        Returns:
        the table columns to parse and format, or null for the default.
      • setColumns

        public void setColumns​(TableColumn<?>... columns)
                        throws IllegalArgumentException
        Sets the table columns to parse and format. A null value means to use the default list of columns, as defined in the get­Columns() method.
        Parameters:
        columns - the table columns to parse and format, or null for the default.
        Throws:
        Illegal­Argument­Exception - if the given array is empty, contains a null element or a duplicated value.
      • getIndentation

        public int getIndentation()
        Returns the number of spaces to add on the left margin for each indentation level. The default value is 4.
        Returns:
        the current indentation.
      • setIndentation

        public void setIndentation​(int indentation)
                            throws IllegalArgumentException
        Sets the number of spaces to add on the left margin for each indentation level. If the new indentation is smaller than the vertical line position, then the later is also set to the given indentation value.
        Parameters:
        indentation - the new indentation.
        Throws:
        Illegal­Argument­Exception - if the given value is negative.
      • getVerticalLinePosition

        public int getVerticalLinePosition()
        Returns the position of the vertical line, relative to the position of the root label. The default value is 2, which means that the vertical line is drawn below the third letter of the root label.
        Returns:
        the current vertical line position.
      • setVerticalLinePosition

        public void setVerticalLinePosition​(int verticalLinePosition)
                                     throws IllegalArgumentException
        Sets the position of the vertical line, relative to the position of the root label. The given value can not be greater than the indentation.
        Parameters:
        vertical­Line­Position - the new vertical line position.
        Throws:
        Illegal­Argument­Exception - if the given value is negative or greater than the indentation.
      • getNodeFilter

        public Predicate<TreeTable.Node> getNodeFilter()
        Returns the filter that specify whether a node should be formatted or ignored. This is the predicate specified in the last call to set­Node­Filter(Predicate). If no filter has been set, then this method returns null.
        Returns:
        a filter for specifying whether a node should be formatted, or null if no filtering is applied.
        Since:
        1.0
      • setNodeFilter

        public void setNodeFilter​(Predicate<TreeTable.Node> filter)
        Sets a filter specifying whether a node should be formatted or ignored. Filters are tested at formatting time for all children of the root node (but not for the root node itself). Filters are ignored at parsing time.
        Parameters:
        filter - filter for specifying whether a node should be formatted, or null for no filtering.
        Since:
        1.0
      • parse

        public TreeTable parse​(CharSequence text,
                               ParsePosition pos)
                        throws ParseException
        Creates a tree from the given character sequence, or returns null if the given text does not look like a tree for this method. This method can parse the trees created by the format(…) methods defined in this class.
        Parsing rules
        • Each node shall be represented by a single line made of two parts, in that order:
          1. white spaces and tree drawing characters ('│', '├', '└' or '─');
          2. string representations of node values, separated by the colunm separator.
        • The number of spaces and drawing characters before the node values determines the node indentation. This indentation does not need to be a factor of the get­Indentation() value, but must be consistent across all the parsed tree.
        • The indentation determines the parent of each node.
        • Parsing stops at first empty line (ignoring whitespaces), or at the end of the given text.
        Error index
        If the given text does not seem to be a tree table, then this method returns null. Otherwise if parsing started but failed, then:
        Specified by:
        parse in class Compound­Format<Tree­Table>
        Parameters:
        text - the character sequence for the tree to parse.
        pos - the position where to start the parsing.
        Returns:
        the parsed tree, or null if the given character sequence can not be parsed.
        Throws:
        Parse­Exception - if an error occurred while parsing a node value.
      • createFormat

        protected Format createFormat​(Class<?> valueType)
        Creates a new format to use for parsing and formatting values of the given type. This method is invoked by the first time that a format is needed for the given type. Subclasses can override this method if they want to configure the way dates, numbers or other objects are formatted. See parent class documentation for more information.

        The implementation in Tree­Table­Format differs from the default implementation in the following aspects:

        Overrides:
        create­Format in class Compound­Format<Tree­Table>
        Parameters:
        value­Type - the base type of values to parse or format.
        Returns:
        the format to use for parsing of formatting values of the given type, or null if none.
      • writeColumnSeparator

        protected void writeColumnSeparator​(int nextColumn,
                                            TableAppender out)
        Writes characters between columns. The default implementation applies the configuration specified by Tabular­Format​.set­Column­Separator­Pattern(String) as below:
        out​.append(before­Fill); out​.next­Column(fill­Character); out​.append(column­Separator);
        The output with default values is like below:
        root
          └─column0…… column1…… column2…… column3
        Subclasses can override this method if different column separators are desired. Note however that doing so may prevent the parse(…) method to work.
        Parameters:
        next­Column - zero-based index of the column to be written after the separator.
        out - where to write the column separator.
        Since:
        1.0
        See Also:
        Table­Appender​.next­Column(char)