Template for table style
To enable table style:
tbl: #include %table-template.red
style 'table tbl
After that table style can be used in layout, as e.g.
view [table]
This will create an empty table with default size of 317x217 and grid 3x8. Default cell size is 100x25. Both vertical and horizontal scrollers are always included. Scrollers are 17 points thick.
Specifying size for table will fill the extra space with additional cells.
view [table 717x517]]
This will create an empty table with 7x20 grid.
Grid size of table can be specified separately, e.g. following will create empty table with 10 columns and 50 rows, but in previous boundaries.
view [table 717x517 data 10x50 options [auto-col: #[true]]
When instead of grid size a block is presented as data, this block is interpreted as table. Block should consist of row blocks of equal size. E.g.
view [table 717x517 data [["" A B][1 First Row][2 Second Row]]]
Values are formed to be presented in table.
Instead of giving data directly as block, file name or url may be specified, to be loaded as table, e.g.
view [
table 717x515
data https://support.staffbase.com/hc/en-us/article_attachments/360009197011/username-password-recovery-code.csv
options [delimiter: #";"]
]
Non-standard delimiter (standard is comma) can currently be specified for urls only.
If you have set up sqlite, data source may be specified as sql query, e.g.
sql-query: func [sql][
out: copy ""
call/output rejoin [{sqlite3 -csv -header C:\Users\Toomas\sqlite\chinook.db "} sql {"}] out
load/as out 'csv
]
view [
table 717x517 with [
data: sql-query "select TrackId, Name, Title, Composer from tracks inner join albums on tracks.AlbumID = albums.AlbumId"
]
]
Auto-headers
It is possible to add automatcally indexed column (and/or row) to you tabel by specifying corresponding option. When auto-col is set to true an extra column will be created, automatically enumerated. By this the original order of rows can be restored whenever necessary:
view [table 717x517 data 10x50 options [auto-col: #[true]]]
Both row and column headers can be specified in this manner:
view [table 717x517 data 10x50 options [auto-col: #[true] auto-row: #[true]]]
Auto-generated header refer to the position of given row/col in original data. Index row/col itself has index 0.
Another way to auto-generate indexes is by option sheet:
view [table options [sheet: #[true]]]
This way both column and row headers will always be generated. But they are different, referenig to actual visual order of rows/cols, not to their order in original data. Therefore also when sorting (see below) headers generated with sheet option will not change, but headers generated with auto-col and auto-row will be reordered.
Row and column sizes can be changed by dragging on cell border. If holding down control while dragging, sizes of all following rows/columns will be changed too. If ctrl-dragging on first row/column, default size is changed.
Scrolling will move the whole grid together with selection. Wheeling will sroll table vertically by 3 rows, with ctrl-down by page. Shift-wheel scrolls table horizontally.
Navigation by keys moves selection, extending it with shift-down. Moving selection outside of visual borders will automatically scroll table if not in end already. Selection is also moved by clicking on cells (if not on/near border), extending selection with shift-down and/or ctrl-down.
Freezing of rows and/or columns is enabled from local menu. Right-click on row/column/cell which you want to freeze and choose "Freeze" from submenu of cell, row or column. Frozen rows/columns are dark-colored. Freezing can be repeated, e.g. if table is scrolled after previous freezing. "Unfreeze" removes all freezing correspondingly from cells/rows/columns. To unfreeze, it is not necessary to place mouse on corresponding row/column as when freezing.
Sorting is currently possible by single column only. With mouse on column to be sorted by, select "Up" or "Down" from Column->Sort submenu. Table is sorted from non-frozen rows downward. If table is created from csv file, then all data is of type string!, and sorted accordingly. To sort by loaded values, choose Column->Sort->Loaded... or convert column to different type before sorting (see below). The only way to restore original order currently is to sort by column that holds original order (e.g. with options/auto-index set to true) or to select "Unsort" from menu.
Filtering is so far also possible by single column at time only. As with sorting, only non-frozen rows are considered. Select Column->Filter from local menu and enter selection criteria. There is special format for criteria. It may start with operator, e.g. < 100 (provided data is of corresponding type - see type changing below), or with function name expecting data as its first argument, but without specifying this argument, e.g. find charset [#"a" #"e" #"i" #"o" #"u"]. Missing argument will be inserted automatically. Code may start also with a solitary set-word, which will reference data in edited field (but it should not be specified). So you can use the word in further code, e.g. string-field: all [find string-field #"a" 20 < length? string-field]. Repeated filtering will consider already filtered rows only (but it is buggy :)). To remove filter choose Column->Unfilter from local menu.
There is very limited basic regex func ~ for compact comparison of strings, e.g. ~ "^^2" is ~same as parse [#"2" to end].
Copying/pasting
Selected cells can be copied or cut, and then pasted in same or different configuration. If pasted when single cell is selected, selection is pasted in same configuration as it was copied, otherwise it is pasted in order as cells are selected. Order of selection of original cells is also significant.
Cell editing is activated with double-click or enter on cell or from local menu, and committed with enter.
Column editing can be activated from local menu, opened on column you are going to edit. Editing command is entered into code window and it has special format too: in case of a single command it takes a command with arguments, but with main argument right after the command name omitted. E.g. when you have opened CSV file with one column as formed float, but with commas as thousand-separators, and you want to remove these commas before you can change type of the column, then choose Column->Edit and enter replace/all comma ""