Grid and List: Difference between revisions

From BR Wiki
Jump to navigation Jump to search
Line 247: Line 247:
===== Output  =====
===== Output  =====


[[Image:NEWC0037.jpg]]  
[[Image:Output2.jpg]]  


The third part of the program demonstrates use of read types ROWSUB and ROWCNT and selection type SEL. Line 1600 inputs the number of selected rows into AVAIL_ROWS. The user may select rows using the mouse or the keyboard and SHIFT and CTRL keys. Line 1700 redimensions matrix SUBSCR to the number of selected rows. Line 1800 inputs the subscripts of the selected rows into matrix SUBSCR. Line 1900 performs a HEADERS operation for a new LIST using the same matrices as were used in the previous LIST. Line 2000 populates the first row of the new LIST with the data from the first selected row from the previous LIST using the primary flag =. If user selects more than one row, then lines 2100 - 2500 add the data from the selected rows using the primary flag +.  
The third part of the program demonstrates use of read types ROWSUB and ROWCNT and selection type SEL. Line 1600 inputs the number of selected rows into AVAIL_ROWS. The user may select rows using the mouse or the keyboard and SHIFT and CTRL keys. Line 1700 redimensions matrix SUBSCR to the number of selected rows. Line 1800 inputs the subscripts of the selected rows into matrix SUBSCR. Line 1900 performs a HEADERS operation for a new LIST using the same matrices as were used in the previous LIST. Line 2000 populates the first row of the new LIST with the data from the first selected row from the previous LIST using the primary flag =. If user selects more than one row, then lines 2100 - 2500 add the data from the selected rows using the primary flag +.  

Revision as of 21:12, 13 March 2013

Grid and List (a.k.a. ListView) are Two dimensional controls, which means that they contain rows and columns. GRIDs are normally used for data entry, while LISTs are used only for selection. Outputting to these controls is done with PRINT FIELDS and inputting from them is done with Input Fields. Grid and List are also Widgets. Each field specification pertains to one *Control*. However, a single FIELD specification can pertain to a parenthesized group of arrays.

The INPUT FIELDS specification states Grid or List as the field type. The display area is specified in terms of rows and columns separated by a slash. The following parameter, instead of being leading field attributes, is a secondary keyword indicating the type (CNT/SUB/CELL/ROWCNT/ROWSUB/ROW) of output or input operation to be performed. The next parameter further qualifies the IO operation (CHG/SEL/ALL). Normally this trailing attribute is an Fkey value which is shifted right one parameter in this context.

It is useful to work with 2D controls in terms of rows versus cells, particularly when the columns are dissimilar. A complete set of row oriented parameters are provided for that purpose. Output operations support the mass populating of 2D controls row by row. And input operations can be row oriented as well. The keywords associated with row processing are R, ROWCNT, ROWSUB, and ROW.

As of Business Rules! versions 4.3+, arrays are automatically resized when receiving data from 2D INPUT operations. This also applies to grouped arrays. Automatic resizing only applies to one dimensional arrays and does not occur when INPUTing into two dimensional arrays. For example:

INPUT FIELDS "row,col,LIST rows/cols, ROW, ALL" : ( MAT Array1$, MAT Array2$, MAT Array3, MAT Array4, MAT Array5$ ) 

Where all arrays are one dimensional and may have the incorrect number of elements.

RINPUT does not work with 2D controls because the output statements are somewhat different from the input statements. The output consists of setting up the columns, including providing the column headings, and populating the control with data. The input consists of identifying what has changed in a manner that enables selective data retrieval and corresponding file updating.

If the user clicks on a column heading the GRID or LIST will be sorted on that column. Sorting is done in terms of rows. Such sorting of these controls has no affect on the BR program. The information returned to BR will be as though no sorting were performed. If a control's population is increased by populating with the plus (+) flag, the control will be resequenced back to its original order before the data is added. One way a program can restore the original displayed order of a GRID or LIST is to populate it incrementally (+) with no data. As of version 4.1 and higher, if a GRID input is attempted on a protected field, BR issues a Bell.

Shift+PgUp and Shift+PgDn selects within a List/Grid.

FKey Processing

An FKey value can be associated with a LIST or GRID control by specifying the FKey number during either output or input operations. Once an Fkey value is specified, the control retains the setting until it is reset. An FKey value can be cleared by specifying an Fkey value of minus one (-1).

When a LIST or GRID has an FKey value set processing is dependent on whether or not the 2D control is being processed by an INPUT FIELDS statement:

  • Displayed but inactive- Single clicking any cell produces the FKey interrupt.
  • Active (participating in Input Fields)- Double clicking any cell produces an FKey completion.

Formerly, CURROW and CURCOL represented the character position where the cursor was when FIELDS processing is completed. This is still true except when the most recent field processed is a 2D control.

When FIELDS processing ends and control is returned to the program while the 'current' control is of type LIST or GRID, CURROW and CURCOL are set to the current cell row and column within the 2D control instead of the character position relative to the window.

Displayed but inactive- Single clicking any cell produces the FKey interrupt.

Active (participating in Input Fields)- Double clicking any cell produces an FKey completion.

Restoring a User Sorted 2D Control

In versions 4.3 and higher the syntax for sorting a listview is:

   PRINT FIELDS "nn,nn,GRID 10/40,SORT": { column number }

This has been extended to allow a numeric array instead of a scalar. If an array is given, it is assumed to be in the same format that SORT_ORDER returns. The new format is:

   PRINT FIELDS "nn,nn,GRID 10/40,SORT": { column-number | numeric-array }

Where the numeric-array has one element for each column left to right. BR will resort the columns in the requested order.

The numeric array values indicating the order of column sorting to be performed do not need to exactly match the standard format. e.g Fractions are allowed, the values can fall within any range, and there does not need to be trailing zero elements for unsorted columns.

A new 'Read Qualifier is available.

DISPLAYED_ORDER - indicates that the read operation is to not restore the data into it's original order before returning the results to the program, for example:

INPUT FIELDS "row,col,LIST rows/cols, ROWSUB, ALL, DISPLAYED_ORDER, NOWAIT": numeric-array 

This reads the original row subscripts for all rows in their present order. This qualifier works only with the ALL selection type. It may be used in conjunction with other qualifiers such as FKEY.

A multi column LISTview

01000 PRINT FIELDS "10,20,LIST 10/80,HEADERS,[HDRS][,fkey]": (MAT HEADINGS$,MAT WIDTHS, MAT FORMS$)

Note: Widths are expressed in character positions, not percentages like they are in BR_VB.

The [HDRS] notation refers to an optional CONFIG ATTRIBUTE HDRS specification for setting the appearance of the header row. In this case the brackets [] are required and the term HDRS may be any bracketed attribute name.

If a function key value (e.g. 1520) is given then when the control is not active, it may be clicked to trigger the specified interrupt similar to any other hot control. A function key interrupt is also triggered by double clicking during an Input operation.

MAT HEADINGS$ Specifies the Titles that will appear at the top of each column in the List or Grid. The format of this row may be specified by an optional parameter following HEADERS in the above example.
MAT WIDTHS Specifies the number of characters in each column. For example if four columns are specified and the widths are given as 10, 15, 10 and 5, then the 2D control occupies 40 character positions (plus a little for the column separators) irrespective of the number of columns specified for the display area. If needed, scroll bars are used to display wider controls within the displayed area.
MAT FORMS$ specifies the display characteristics of each column such as "C 12" or "PIC(z,zzz,zz#.##-)". A "P" following the display parameter will cause the field to be protected and no data entry will be allowed in GRID mode.


The FIELD FORM ARRAY ELEMENTS MAY ALSO INCLUDE A TRAILING COMMA FOLLOWED BY FIELD ATTRIBUTES (e.g. color) THAT PERTAIN TO THE COLUMN.

PRINT FIELDS "10,20,LIST 10/80, = [,fkey]": (MAT NAME$, MAT CITY$, MAT AGE, MAT WEIGHT)

In this example, the fourth element of the HEADERS FIELD_FORM$ array (above) could specify rounding of WEIGHT to two decimal places. If an fkey value is specified, then double-clicking (or single clicking if S is specified) a cell will generate the specified fkey interrupt.

Grid and List BR_VB Similarities

The BR VB exceptions are given here and a more complete description follows. These controls work like the BR_VB interface except headers now specify the form of each column, and a new input type has been added.

A multi column LISTview-

00100 PRINT FIELDS "10,20,LIST 10/80,HEADERS,[hdrs],[fkey]": (MAT HEADINGS$, MAT WIDTHS, MAT FORMS$) 

Note Widths are expressed in character positions, not percentages like they are in BR VB.

The populate operation-

00200 PRINT FIELDS "10,20,LIST 10/80,=": (MAT NAME$, MAT CITY$, MAT AGE, MAT WEIGHT)

Read-Ctl type: CNT (in place of CELLROWSUB) returns a single numeric value which is the number of subscripts available to read. In the case of grids, this is the number of cells. In the case of LISTviews, this is the number of rows.

00110 INPUT FIELDS "10,20,LIST 10/80,ROWCNT,SEL": avail_rows ! # of selected rows
00120 MAT DATA$(3 * avail_rows)          ! redimension.. 3 cols x selected rows
00130 INPUT FIELDS "10,20,LIST 10/80,ROW,SEL,NOWAIT": MAT DATA$ ! read rows

Input Operations

When using INPUT FIELDS to read from a 2D control, the leading attributes specification states the type of read operation and the trailing attributes specification is the type of cell or row selection to be performed. The third parameter can optionally specify NOWAIT or an FKEY value. If it is an FKEY value, it signifies that an FKEY event should be triggered when, in navigation mode, a selection is made by double clicking or pressing the Enter key.

Syntax

Parameters

Quotation marks must surround the specifications.

Row and Column specify the space where the grid or list begins.

List with rows and columns separated by a slash determine how big the list or grid is going to be.

The following Read Types are valid for both grids and lists:

RowCnt The number of rows specified.
RowSub The subscripts of the specified rows.
Row Read all cells in each specified row.

These Read Types are valid for Grids only:

Cnt Specify the number of cells specified (see selection types below).
Sub Read the Cell Subscript Values (see example below).
Cell Read each cell specified.

For the "Sel-type" parameter, the following selection types are valid for both grids and lists:

Sel Read one or more selected items.
SelOne Select only one item.
All Read all items in the control (except headings).
Cur Current cell or row number.

This Selection Type is valid for Grids only:

Chg All items changed since the last '=' populate or the last CHG retrieval of cell/row contents.

The Fkey and Nowait parameters are optional. FKEY means that an FKEY event should be triggered when a selection is made by double clicking or pressing the Enter key, in navigation mode. Nowait simply means that it does not wait for user input.

Following the ending quotation mark, a colon sets off the name of the I/O List.

Examples

LIST
00210 INPUT FIELDS "10,20,LIST 10/80,ROWCNT,SEL,FKEY": avail_rows ! selected row cnt
00220 ! next INPUT operation does not wait for operator
00230 MAT subscr(avail_rows)          ! redimension with number of selected rows
00240 INPUT FIELDS "10,20,LIST 10/80,ROWSUB,SEL,NOWAIT": MAT subscr ! read subscripts
Uniform GRID

Contains one data array and multiple columns

00210 INPUT FIELDS "10,20,GRID 10/80,CNT,CHG": cells      ! # of changed cells
00220 MAT subscr(cells)                                         ! redimension
00230 INPUT FIELDS "10,20,GRID 10/80,SUB,CHG,NOWAIT": MAT subscr ! read subscripts
00240 MAT data$(cells)                                          ! redimension
00250 INPUT FIELDS "10,20,GRID 10/80,CELL,CHG,NOWAIT": MAT data$ ! read changes
Row Oriented GRID
00210 INPUT FIELDS "10,20,GRID 10/80,ROWCNT,CHG": rows   ! # of changed rows
00220 MAT subscr(rows)                                   ! redimension
00230 INPUT FIELDS "10,20,GRID 10/80,ROWSUB,CHG,NOWAIT": MAT subscr ! read subscripts
00240 MAT NAME$(rows) : MAT CITY$(rows) : MAT AGE(rows) : MAT WEIGHT(rows) ! redimension
00250 INPUT FIELDS "10,20,GRID 10/80,ROW,CHG,NOWAIT": (MAT NAME$, MAT CITY$,MAT AGE, MAT WEIGHT)   ! read changed rows

This brings us to the question of what is to be done with the information after it has been read. If it is to be stored in a file, then we should have included a hidden column with master file record numbers of the data in each row. This would support looping through the input array and rewriting the changed data.


While LIST subscripts are expressed in terms of rows, GRID subscripts may be either. In a four column five row GRID, the cell at row three column two has a subscript of ten.

Cell Subscript Values of a 5 x 4 GRID

1   2   3   4
5   6   7   8
9  10  11  12
13 14  15  16
17 18  19  20

The following example shows use of LIST and GRID controls. There are seven parts in this program.

The first part creates a LIST with 2 columns and 3 rows. On lines 300 - 500, column headings, widths, and form specifications are assigned to the corresponding matrices. Lines 600 and 700 place data into the matrices to be printed in the LIST. The HEADERS operation on line 800 sets the column headings, widths and form specs. Line 900 populates the LIST by row, which is the default.

00100 dim HEADINGS$(2), WIDTHS(2), FORMS$(2), NAMES$(3)*28, CITIES$(3)*18, DATA$(1)*80, SUBSCR(1)
00200 print NEWPAGE
00300 let HEADINGS$(1)="Name": let HEADINGS$(2)="City"
00400 let WIDTHS(1)=30 : let WIDTHS(2)=20
00500 let FORMS$(1)="CC 28" : let FORMS$(2)="CC 18"
00600 let NAMES$(1)="Stalin" : let NAMES$(2)="Napoleon" : let NAMES$(3)="Roosevelt"
00700 let CITIES$(1)="Moscow" : let CITIES$(2)="Paris" : let CITIES$(3)="Washington"
00800 print fields "1,1,list 8/60,headers": (MAT HEADINGS$,MAT WIDTHS,MAT FORMS$)
00900 print fields "1,1,list 8/60,=R": (MAT NAMES$,MAT CITIES$)
01000 print fields "9,1,C 50" : "Press enter to insert at the end of list"
01100 let KSTAT$(1)
Output

The second part of the program demonstrates the use of the primary flag +, which adds to the end of any previously populated data. Line 01200 redimensions matrices NAMES$ and CITIES$ to allow room for one extra item in each of them. Line 01300 places new data into the matrices to be printed in the LIST.
Line 10400 adds the new data to the LIST.

01200 mat NAMES$(UDIM(NAMES$)+1) : mat CITIES$(UDIM(CITIES$)+1)
01300 let NAMES$(4)="Churchill" : let CITIES$(4)="London"
01400 print fields "1,1,list 8/60,+": (MAT NAMES$(4:4),MAT CITIES$(4:4))
01500 print fields "9,1,C 50" : "Select rows to be read into a matrix"
Output

The third part of the program demonstrates use of read types ROWSUB and ROWCNT and selection type SEL. Line 1600 inputs the number of selected rows into AVAIL_ROWS. The user may select rows using the mouse or the keyboard and SHIFT and CTRL keys. Line 1700 redimensions matrix SUBSCR to the number of selected rows. Line 1800 inputs the subscripts of the selected rows into matrix SUBSCR. Line 1900 performs a HEADERS operation for a new LIST using the same matrices as were used in the previous LIST. Line 2000 populates the first row of the new LIST with the data from the first selected row from the previous LIST using the primary flag =. If user selects more than one row, then lines 2100 - 2500 add the data from the selected rows using the primary flag +.

01600 input fields "1,1,list 8/60,ROWCNT,SEL": AVAIL_ROWS
01700 mat SUBSCR(AVAIL_ROWS)
01800 input fields "1,1,list 8/60,rowsub,sel,nowait": MAT SUBSCR
01900 print fields "12,1,list 8/60,headers": (MAT HEADINGS$,MAT WIDTHS,MAT FORMS$)
02000 print fields "12,1,list 8/60,=": (MAT NAMES$(SUBSCR(1):SUBSCR(1)),MAT CITIES$(SUBSCR(1):SUBSCR(1)))
02100 if UDIM(SUBSCR) > 1 then
02200   for I = 2 to UDIM(SUBSCR)
02300     print fields "12,1,list 8/60,+": ( MAT NAMES$(SUBSCR(I):SUBSCR(I)),MAT CITIES$(SUBSCR(I):SUBSCR(I)))
02400   next I
02500 end if
02600 print fields "9,1,C 50" : "Press enter to insert at beginning of list"
Output

File:NEWC0038.jpg

The fourth part of the program demonstrates the use of the primary flag -, which inserts in the beginning of any previously populated data. Line 2800 redimensions matrices NAMES$ and CITIES$ to allow room for one extra item in each of them. Line 2900 places new data into the matrices to be printed in the LIST. Line 3000 adds the new data to the beginning of the LIST ahead of previously populated data.

02700 let KSTAT$(1)
02800 mat NAMES$(UDIM(NAMES$)+1): mat CITIES$(UDIM(CITIES$)+1)
02900 let NAMES$(5)="Castro" :! let CITIES$(5)="Havana"
03000 print fields "1,1,list 8/60,-": (MAT NAMES$(5:5),MAT CITIES$(5:5))
03100 print fields "9,1,C 60" : "Press enter to populate list by column"
Output

File:NEWC0039.jpg

The fifth part of the program, more specifically, Line 3300, demonstrates the use of the secondary flag C to populate the LIST by column. The primary flag = is also used in order to replace any previously populated data.

03200 let KSTAT$(1)
03300 print fields "1,1,list 8/60,=C": (MAT NAMES$,MAT CITIES$)
Output

File:NEWC0040.jpg

The sixth part of the program creates a GRID. The HEADERS operation on line 3600 uses the same HEADINGS$, WIDTHS, and FORMS$ as the previously constructed LISTs. Line 3700 sets CURFLD to be on the sixth cell of the GRID (this is discussed in the next section). Line 3800 populates the GRID. The user may change the contents of the cells.

03400 print fields "23,1,C 50" : "Press enter to continue": let KSTAT$(1)
03500 print NEWPAGE
03600 print fields "1,1,grid 8/60,headers": (MAT HEADINGS$,MAT WIDTHS,MAT FORMS$)
03700 let CURFLD (1,6)
03800 print fields "1,1,grid 8/60,=": (MAT NAMES$,MAT CITIES$)
Output

File:NEWC0041.jpg

The seventh part of the program demonstrates the use of read types CNT, CELL, and SUB and selection type CHG. Line 3900 counts the number of changed cells and inputs that number into variable CELLS. Line 4000 redimensions the matrix DATA$ to the number of changed cells. Line 4100 inputs the changed cells into DATA$. Line 4200 prints DATA$. Line 4300 redimensions matrix SUBSCR to the number of changed cells. Line 4400 inputs the subscripts of the changed cells into matrix SUBSCR. Line 4500 prints matrix SUBSCR.

03900 input fields "1,1,grid 8/60,cnt,chg": CELLS
04000 mat SUBSCR(CELLS)
04100 input fields "1,1,grid 8/60,sub,chg,nowait": MAT SUBSCR
04200 mat DATA$(CELLS)
04300 input fields "1,1,grid 8/60,cell,chg,nowait": MAT DATA$
04400 print MAT DATA$
Output

File:NEWC0041.jpg

Output of line 04400

File:NEWC0042.jpg

ListView

00200 INPUT FIELDS "10,20,LIST 10/80,ROWCNT,SEL,Fkey": avail_rows ! selected row cnt
 - next INPUT operation does not wait for operator -
00210 MAT subscr(avail_rows)          ! redimension.. number of selected rows
00220 INPUT FIELDS "10,20,LIST 10/80,ROWSUB,SEL,NOWAIT": MAT subscr ! read subscripts
Output Operations

The HEADERS operation sets the column headings and widths. The corresponding input/output list value must be a parenthesized group of three arrays. e.g.

00250 PRINT FIELDS "10,20,LIST 10/80,HEADERS": (MAT HEADINGS$, MAT WIDTHS, MAT FIELD_FORMS$)
     - or -
00250 PRINT FIELDS "10,20,GRID 10/80,HEADERS,[hdrs],1520": (MAT HEADINGS$, MAT WIDTHS,MAT FIELD_FORMS$)

The [hdrs] notation refers to an optional CONFIG ATTRIBUTE [HDRS] specification for setting the appearance of the header row. In this case the [] brackets are required.

If a function key value (e.g. 1520) is given then when the control is not active, it may be clicked to trigger the specified interrupt similar to any other hot control.

MAT HEADINGS$ Contains the column titles that will be displayed at the top of each column. The font, color and shading of these titles can be set through the [HDRS] or similar substitution attribute.

MAT WIDTHS specifies DISPLAYED Column Widths and is expressed as the number of character positions occupied by each column. Scrollbars are provided as needed to honor overall control size specified in the FIELDS specification.

For example, if four columns are specified and the widths are given as 10, 15, 10 and 5, then the 2D control occupies 40 character positions (plus a little for the column separators) irrespective of the number of columns specified for the display area. If needed, scroll bars are used to display wider controls within the displayed area.

Displayed widths of zero characters are allowed. This enables the use of hidden columns for storing things like record numbers and record keys.

MAT FIELD_FORMS$ provides the BR FORM for each column. e.g. C 15 stipulates a maximum field CAPACITY of 15. The actual displayed length is a function of the grid size and the column relative width.

The field form array elements may also include a comma followed by leading field attributes (e.g. color) pertaining to the column.

The [hdrs] notation refers to an optional CONFIG ATTRIBUTE [HDRS] specification for setting the appearance of the header row. In this case the [] brackets are required and the term HDRS may be any bracketed attribute name.

If a function key value (e.g. 1520) is given then when the control is not active, it may be clicked to trigger the specified interrupt similar to any other hot control. A function key interrupt is also triggered by double clicking during an Input operation.

The number of columns must be set with HEADERS prior to Populating a control (loading data into it).

The populate operation loads data into the control. In the following example four columns are loaded:

03010 PRINT FIELDS "10,20,LIST 10/80, =": (MAT NAME$, MAT CITY$,MAT AGE, MAT WEIGHT)
     - or -
03020 PRINT FIELDS "10,20,GRID 10/80, =": (MAT NAME$, MAT CITY$, MAT AGE, MAT WEIGHT)

In this example, the fourth element of the HEADERS FIELD_FORM$ array (above) could specify rounding of WEIGHT to two decimal places. If an fkey value is specified, then double-clicking (or single clicking if S is specified) a cell will generate the specified fkey interrupt.

Permissible leading attribute values are:

Primary Flags
= Replace any previous data
+ Add to any previously populated data (this allows loading in chunks)
- Insert data ahead of previously populated data (4.16+)

Secondary Flags

R Load one row at a time (the default - use grouped IO parens)
C Load one column at a time - This is for loading multiple columns of the same data type
L Provide the FKEY (see INPUT below) or Enter interrupt if the user presses up arrow or page up in the first field, or down arrow or page down in the last field. Note that this is not specified in the individual field leading attributes.
S Single click to activate an Enter or FKEY event (otherwise a double click is required) (4.17+)


Note that the following example will NOT work-

00100 PRINT FIELDS "10,20,LIST 10/80,=C": MAT NAME$,MAT CITY$,MAT AGE,MAT WEIGHT

The reason is that only MAT NAME$ will reach the list. MAT CITY$, MAT AGE and MAT WEIGHT will be associated with subsequent fields. Multiple arrays provided to a single control must be grouped with parentheses.

Populating a two-dimensional object by row with grouped IO means NAME$(1) will go into (1,1), AGE(1) will go into (1,2) and WGT(1) will go into (1,3) and so on. If a single array (MAT DATA$) is specified instead of a group, MAT DATA$ is applied horizontally instead of vertically. So DATA$(1) - DATA$(3) will be the first row and DATA$(4) - DATA$(6) will be the next row and so on.

Populating a two dimensional grid by column with an array named DATA$ means that DATA$(1) goes in (1,1) and DATA$(2) goes in (2,1) and DATA$(3) goes in (3,1). Therefore the first however many values of DATA refer to the first column and the second however many values of DATA refer to the second column. So if there are 3 columns and UDIM(DATA$) = 90 then DATA$(1)-DATA$(30) is the first column, and DATA$(31) - DATA$(60) is the second column and DATA$(61) - DATA$(90) is the last column.

e.g.

MAT NAME$ = George, Peter, Tom

MAT CITY$ = Dallas, Detroit, Denver

MAT AGE$ = 42, 23, 35

MAT WEIGHT$ = 180, 212, 193

00200 PRINT FIELDS "10,20,GRID 10/80, =": (MAT NAME$, MAT CITY$,MAT AGE$, MAT WEIGHT$)

Peter, Detroit, 23, 212

Tom, Denver, 35, 193

00200 PRINT FIELDS "10,20,GRID 10/80, =C": (MAT NAME$, MAT CITY$,MAT AGE$, MAT WEIGHT$)

Dallas, Detroit, Denver

42, 23, 35

180, 212, 193

As you can see using C with grouped arrays is counter-intuitive and doesn't fit well. C is most useful with a single array that should be loaded vertically down several columns.

Uniform Grid - One Data Array - Multiple Columns

00310 INPUT FIELDS "10,20,GRID 10/80,CNT,CHG": cells      ! # of changed cells
00320 MAT subscr(cells)                                   ! redimension
00330 INPUT FIELDS "10,20,GRID 10/80,SUB,CHG,NOWAIT": MAT subscr ! read subscripts
00340 MAT data$(cells)                                    ! redimension
00350 INPUT FIELDS "10,20,GRID 10/80,CELL,CHG,NOWAIT": MAT data$ ! read changes

Row Oriented Grid

00410 INPUT FIELDS "10,20,GRID 10/80,ROWCNT,CHG": rows   ! # of changed rows
00420 MAT subscr(rows)                                   ! redimension
00430 INPUT FIELDS "10,20,GRID 10/80,ROWSUB,CHG,NOWAIT": MAT subscr ! read subscripts

-- read changed rows --

00440 MAT NAME$(rows) : MAT CITY$(rows) : MAT AGE(rows) : MAT WEIGHT(rows)
00450 INPUT FIELDS "10,20,GRID 10/80,ROW,CHG,NOWAIT": (MAT NAME$, MAT CITY$,MAT AGE, MAT WEIGHT)

This opens the question of what is to be done with the information after it has been read. If it is to be stored in a file then we should have included a hidden column with master file record numbers of the data in each row. This would support looping through the input array and rewriting the changed data.

Valid read types

GRIDs only-
  1. CNT- Specify the number of cells specified (see selection types below).
  2. SUB- Read the Cell Subscript Values (see example below).
  3. CELL- Read each cell specified.
Grids and Lists
  1. COLCNT - The number of columns established by the header arrays. e.g. INPUT FIELDS "row,col,LIST rows/cols, COLCNT, ALL" : numeric-variable
  2. ROWCNT - The number of rows specified.
  3. ROWSUB - The subscripts of the specified rows.
  4. ROW - Read all cells in each specified row.
  5. SORT_ORDER - Provides a value of zero for each unsorted column and gives the ascending sequence of column sorts that have occurred. If a column has been reversed (double sorted) it's value will be negative.
Valid selection types are
  1. SEL- Read one or more selected items.
  2. SELONE - Select only one item.
  3. ALL- Read all items in the control (except headings).
  4. CUR- Current cell or row number.
  5. RANGE - Specifies which portion of a 2D control is to be input.
  6. CELL_RANGE – A special type of output range.

(grid only)

  1. CHG- All items changed since the last '=' populate or the last CHG retrieval of cell/row contents.

While LISTview subscripts are expressed in terms of rows, grid subscripts may be either. In a four column five row grid, the cell at row three column two has a subscript of ten.

Cell Subscript Values of a 5 x 4 grid
1  2  3  4
5  6  7  8
9 10 11 12
13 14 15 16
17 18 19 20
Grid Data Validation

BR Processing Improvements- GRIDs are now validated as each cell is exited instead of when control is passed to the BR program after all data is entered.

Color and Font changes in Cells

The attributes that determine font, color and style in each cell can be set for an entire column by including these parameters in the heading FORM array. Individual cells can then be changed using a PRINT statement.

The format of the print statement is

00100 PRINT #WINNO, fields "2,2,LIST 10/60,ATTR":(mat start, mat end, mat attribute$)

Where

mat start contains the cell number(s) where the attribute chane should begin,
mat end contins the cell nuber where the attribute should stop applying
mat attribute$ contains the attibute specificatin trhat should be applied to the cell rnage(s) specified.

If the 2d control is a GRID, then mat Start and mat End refer to starting and ending Cell Numbers. If the 2d control is a listview, then mat Start and mat End refer to starting and ending Row Numbers.

The attributes specified for any COLUMN can be overridden on a cell basis by specifying the starting cell number, then ending cell number and the overriding attributes in three arrays that are printed to the grid window with the same grid specificatins and the key word "ATTR".

02420 PRINT #BLISTWIN,FIELDS BLISTSPEC$&",ATTR": (MAT BROWS,MAT BROWE,MAT BATT$)

Where BLISTWIN is the window number, BLISTSPEC$ is the grid specification "GRID 10/40" for example, BROWS is an array holding the starting cell number, BROWE is an array holding the ending cell number, and BATT$ is an array holding the overriding attributes. See also 2D Control Coloring. In a list the attributes of the first cell in the row controls the appearance of the entire row.

Numeric Column Sorting

2D controls now facilitate numeric column sorting. This works well in conjunction with the new DATE field format (see release notes 4.16) where the data is stored as day of century, but is displayed as a formatted date. It also works with all numeric columns.

GRID Numeric Validation

Entry of numeric data redisplays upon exit of each cell with the ARROW key.

Range Input

The following examples are used in versions 4.3 and higher. In these examples BR will redimension the receiving arrays as needed:

 INPUT FIELDS "row,col,LIST rows/cols, CELL, RANGE" :
            start, end, MAT Data$

This reads the specified range of cells. BR redimensions MAT Data$ as needed. Note that CELL may now be used with LIST. Previously, LISTs were only addressable by row.

 INPUT FIELDS "row,col,LIST rows/cols, ROW, RANGE" :
            (start:=7), (end:=11), ( MAT Array1$, MAT Array2, MAT Array3$ )

This reads the cells in rows 7 through 11. The recieving arrays are redimensioned as appropriate.

 INPUT FIELDS "row,col,GRID rows/cols, ROW, RANGE" :
            MAT start, MAT end, ( MAT Data1$, MAT Data2$, MAT Data3 )

This reads one or more ranges of rows.

A more detailed example of this is:

 100 -- create and populate a LIST control --
 ........
 200 MAT START(3) : MAT END(3)
 210 READ MAT START, MAT END
 220 DATA 7,21,33
 230 DATA 11,21,38
 240 INPUT FIELDS "row,col,LIST rows/cols, ROW, RANGE" : MAT START,
           MAT END, ( MAT Array1$, MAT Array2, MAT Array3$ )

This reads 12 rows of data ( row 7-11, row 21 and rows 33-38 ).

Range Output

The following examples are valid for versions 4.3 and higher. By default, RANGE output refers to rows. The special keyword CELL_RANGE is used to denote the specification of cell ranges. Additionally, the use of scalars versus arrays for start and end values determines important characteristics of the output process.


Using Scalars For Range Specification

 PRINT FIELDS "7,8,GRID 10/75, RANGE": start, end, MAT Data$

This replaces the values in rows numbered 'start' through 'end' with the data in MAT Data$. The size of MAT Data$ must be a multiple of the number of columns in the GRID or an error is generated.

 PRINT FIELDS "7,8,LIST 10/75, RANGE": start, end, (MAT NAME$,
                                         MAT CITY$, MAT AGE, MAT WEIGHT)

This replaces the values in ROWs numbered 'start' through 'end' with the data from MATs NAME$, CITY$, AGE and WEIGHT. The data arrays must all be dimensioned the same.

Insertion and Deletion

The number of rows being output do not need to match the number of rows being replaced. To delete a range of rows, output one or more grouped arrays with zero elements.

Examples- The following examples are valid for versions 4.3 and higher. Using the following statement various scenarios are described.

 PRINT FIELDS "7,8,LIST 10/75, RANGE": start, end, (MAT NAME$,
                                         MAT CITY$, MAT AGE, MAT WEIGHT)
 start=7, end=11, and the arrays have been DIMed to nine elements
 Result- Nine rows replace five, and the total content of the control is
         expanded by 4 rows.


 start=7, end=11, and the arrays are DIMed to zero elements
 Result- Five rows are deleted, and the total size of the control is
         reduced by 5 rows.
 start=7, end=0 (anything less than 7), and the arrays are DIMed to
          support three rows
 Result- Three rows are inserted ahead of row seven and the total content
         of the control is expanded by three rows
 start=5000, end={any value}, the control only has 482 rows, and the source
         arrays are DIMed to support eleven rows
 Result- Eleven rows are appended to the end of the control and become
         rows 483 through 493.

Outputting Ranges of Cells

The following examples are valid for versions 4.3 and higher.

Ranges of cells may be output in conjunction with the CELL_RANGE keyword.

 PRINT FIELDS "7,8,LIST 10/75, CELL_RANGE": start, end, (MAT NAME$,
                                         MAT CITY$, MAT AGE, MAT WEIGHT)
                               - or -
 PRINT FIELDS "7,8,GRID 10/75, CELL_RANGE": start, end, MAT Data$

In this case, start and end specify cells instead of rows. If insertion or deletion is indicated by dimensioning the data arrays to greater or fewer elements than are being replaced, then the data must be a multiple of the number of columns. Insertion and deletion is only valid in terms of rows, even when cell subscripts are used to specify ranges. In such cases, if the cell subscripts are not on row boundaries, an error is generated.

 PRINT FIELDS "7,8,GRID 10/75, CELL_RANGE": start, start, Data$

In this example, the value in one cell is replaced with the content of a scalar.

Using Arrays For Range Specification

If the start and end specifications are arrays, denoting multiple ranges, there must be a one to one correspondence between the number of rows specified and those in the data. This method implies replacement only and insertion or deletion is not allowed.

The data flow that this feature was designed to support is one where the user is presented with a LIST or GRID where multiple rows have been either selected or changed before returning control to the program and the program is responding by updating something on those rows.

The program begins by presenting a 2D control to the user and reading the the control with type ROWSUB or SUB. Type SUB only works for GRIDs where all colmns have the same data type. Of course the subscripts are read into a numeric array which BR redimensions as appropriate. Then the program reads the changed or selected data with NOWAIT. (This resets the CHG flags in the control.) The program then changes either row (ROWSUB) or cell (SUB) data and outputs the results using the subscript array as both the start and end specification. Other scenarios are possible but this is the primary intended use.

The following examples are valid for versions 4.3 and higher. An example of this is:

 100 -- create and populate a GRID --
 ........
 200 INPUT FIELDS "row,col,GRID rows/cols,ROWSUB,CHG": MAT Rowsubs
        (Reading subscripts does not reset the CHG flags in the control.)
 210 INPUT FIELDS "row,col,GRID rows/cols,ROW,CHG,NOWAIT": ( MAT Data1$,
                                 MAT Data2, MAT Data3$ )
         BR redimensions the receiving arrays as needed.
        (Reading the data also resets the CHG flags in the control.)
 220 -- process the changed rows now present in the data arrays --
 ........
 300 PRINT FIELDS "row,col,GRID rows/cols,RANGE": MAT Rowsubs,
                  MAT Rowsubs, ( MAT Data1$, MAT Data2, MAT Data3$ )

This outputs the updated rows.


unsorted

4.2!!!

The NoSort parameter is used to prevent users from sorting columns of a Grid or Listview.

For the statement: PRINT FIELDS "10,20,GRID 10/80,HEADERS,[hdrs],1520": (MAT HEADINGS$, MAT WIDTHS, MAT FIELD_FORMS$) The field attribute "^nosort" appearing in the MAT FIELD_FORM$ prevents the sorting of a grid or listview in response to the user clicking on the corresponding column header. This does not prevent programs from sorting on those columns.

Available as of BR! 4.20E.