Please submit all suggestions and bugs via github.
A data acquisition, manipulation, and analysis program for the X Window System.
The main window in Glove is a table which contains columns of numbers. From this table, you can manipulate the data in many ways.
To edit a cell, double-click on it with the mouse. If it is already selected you can edit it by hitting the return key. To accept the changes to the cell, you can either hit return, shift-return, tab, or shift-tab, or click the "Ok" button in the top left corner of the table. If you hit the return key, the changes will be accepted and the cell below the current one will then be in edit mode. Shift-return will accept the changes and begin editing on the cell above the current one. Tab and shift-tab will accept the changes and shift editing to the cell the the right/left of the current one. To cancel the changes you made, hit the escape key. All changes made can be undone/redone by selecting "Undo/Redo" from the "Edit" menu.
You can insert, delete or duplicate rows, columns, or ranges of cells by selecting the appropriate cells by click-dragging the mouse and choosing "Insert", "Delete", or "Duplicate" from the "Edit" menu. Rows or columns can be selected by clicking on the row or column header number corresonding the the desired row or column. To extend the selection, you can either shift-left-click or right-click. The selection can be cleared by hitting the Escape key. All changes made can be undone/redone by selecting "Undo/Redo" from the "Edit" menu.
After selecting a group of cells, you can cut or copy them by selecting "Cut" or "Copy" from the "Edit" menu. After cutting or copying the desired cells, you can paste them in in a few ways. If you have no cells selected when you paste, the pasted cells will be placed in a new column appended to the table. If you have a group of columns selected, the pasted cells will be placed in columns inserted before the selected columns. If you copied a single cell, you can select the destination cell and paste over it, thereby changing its value. If you copied a range of cells all contained within a single column, you can replace the same sized range of cells elsewhere, or insert them into a column by selecting the cell before which the pasted cells will be inserted. All changes made can be undone/redone by selecting "Undo/Redo" from the "Edit" menu.
Glove can generate a sequence of numbers for you with the "Generate column by range" and "Generate column by increment" menu items in the "Data" menu. The resultant column of numbers can either be appended to the table, or can replace a current column. This or any other column of numbers can then be transformed with the transform dialog which is found in the "Data" menu.
To start a new file, select "New" from the "File" menu. To open an existing file, select "Open" from the "File" menu. If the file is a Glove file, it will open immediately. If not, a dialog box will pop up asking which module to use to open it. If the file is a delimited text file, select "Text (delimited)" from the menu. You will then be presented with a delimiter dialog box in which you can specify how the file is delimited.
To save a file, select "Save" from the "File" menu. To save a file under a different name, select "Save as..." from the "File" menu. If you wish to save the file in a non-Glove format, use the desired module from the "Export modules..." menu.
To print, select "Print..." from the "File" menu. Select "Print setup" from the "File" menu to specify whether to print in landscape or portrait mode. In the print dialog box, you can choose whether to print to a file or to a printer.
To plot columns of data, select "Plot data..." from the "Data" menu. The plot dialog will ask which columns represent x, y and the errors in x and y if desired. Use the "Plot" menu in the dialog to specify which plot to add this curve to. By default, a new plot is created. If you have columns selected, they will by default specify the x and y axis and the errors in the y and x axis respectively.
To plot a vector field, select "Plot vector field..." from the "Data" menu. If you have columns selected, they will by default specify the x and y axis and dx and dy.
To change the settings of a curve, either select "Change curve options..." from the Options menu, double-click on the appropriate curve label in the legend, or right-click on the curve name in the legend.
You can change the scale of the plot in several ways. To view the scale dialog box, which allows you to change the scale in both axes, you can select "Change scale..." from the "Options" menu, or double-click on the appropriate axis. To scale without the dialog, click and drag the mouse, which will create a zoom rectangle. If the Meta key is pressed, the scale will be changed to show exactly the region contained by the zoom rectangle. If the Meta key is not pressed, a reasonable scale will be chosen around the area of interest. To return the scale to the default, select "Reset scale" from the "Options" menu.
The range specifies what portion of the curves will be drawn, and what portion of the curves will by used in the analysis. To set the range, zoom in to the desired range and select "Copy scale to range" from the "Options" menu. Alternately, you can specify a range in the "Edit Plot Range" dialog by selecting "Change range..." from the "Options" menu. You can then reset the scale by selecting "Reset scale" from the "Options" menu.
You can edit the plot labels either by selecting "Change labels..." from the "Options" menu or by double-clicking on the appropriate label.
Cursors are a useful way of determining the position of various features on a curve. If you view Dual cursors, the values of both cursors will be displayed along with the distance between them. To move a cursor, click and drag it. For fine adjustments, click and then use appropriate the arrow keys.
The values of the cursors can be used by an external module, for example to look up a value in an external table, by selected the appropriate module in the "Cursor module" sub-menu of the "Cursor" menu.
You can mark a particular value with a line, either vertical or horizontal, by selecting "Mark cursor" from the "Cursor" menu. If you select "Show mark window", a table of the current marks will be displayed at the bottom of the plot. This table can be printed to an encapsulated postscript (EPS) file by selecting "Print marks as EPS..." from the "Plot" menu.
To open the Fit window, in order to fit a curve, select "Open fit window" from the Analysis menu. Once fits have been calculated from the fit window, the fit parameters and residual plots can be viewed by selecting the appropriate entry in the "Fit parameters" and "Residual plots" sub-menus.
The fit window consists of a number of subwindows. On the left side, the available curves and fits are listed. After a fit is selected, its functional form is displayed in the bottom left corner of the window, and its parameters are listed in the upper right of the window. For some functional forms, like polynomial, the fit can be done with no input from the user. To fit this type of curve, either double click on the fit name itself (in the fit list), or select "Fit" from the "Fit" menu.
Other functional forms require a first guess to be supplied by the user. These guesses are typed into the "Fit values" input fields beside the appropriate "Parameter name". When a value for each paramter has been input, the values of the parameters can be tested by selecting "Test start values" from the "Fit" menu. A curve corresponding to the current parameters will then be drawn in the plot. Further changes made to the "Fit values" will automatically be reflected in the test curve. Once a reasonable approximation to the data is attained, select "Fit" from the "Fit" menu.
Once a good fit has been obtained, it can be plotted on the original plot by selecting "Plot" from the "Fit" menu.
There are a number of ways to add new fit functions. A polynomial fit can be added by selecting "New polynomial fit" from the "Fit" menu. A non-linear fit can be added by selecting "New non-linear fit" from the "Fit" menu.
For increased performance, a new fit function can be added by creating a
new fit module. An example of such a module can be found in the
"fitmodule" directory of the distribution. Once a module is created,
it should be placed in the user's home directory, in the
~/.glove/fitdfmodule/ directory. It will then be displayed (after
restarting Glove), in the "Fit" list.
Modules are intended to provide extended functionality to Glove at the user's control. The protocols are very simple, and involve reading and writing to standard in/out. Since Glove handles this through pipes, the modules may be written in any language.
Below is an enum containing relevant values for the modules:
enum
{
// File import/export module values
kGloveDataError = 0,
kGloveMatrixDataFormat = 1,
kGloveRaggedDataFormat = 2,
// General module values
kGloveOK = 3,
kGloveFail = 4,
kGloveFinished = 5,
// Data module values
kGloveRequiresData = 6,
kGloveRequiresNoData = 7,
kGloveDataDump = 8,
kGloveDataControl = 9,
// Fit module values
kGloveXY = 10,
kGloveXYdY = 11,
kGloveXYdX = 12,
kGloveXYdYdX = 13
};
Import modules should be placed in the importmodule directory of one of
your module directories (by default /usr/local/lib/glove
and ~/.glove).
The module is called with the name of the file to be imported as the argument.
The module then verifies that the file is in the appropriate format. If it
is not, the module prints the value kGloveDataError. If the file is in the
proper format for the module, it prints kGloveMatrixDataFormat if the file
is in a matrix format, i.e., all columns have the same number of rows. It
prints kGloveRaggedDataFormat for ragged format. For matrix format, the
module then prints the column count, followed by the data given row by row,
i.e., c1r1, c2r1, c3r1, c4r1, c1r2, …. For ragged format, the module then
prints the column count, followed by the data column by column. Before
each column, the number of rows in that column are printed.
Export modules should be placed in the exportmodule directory of one of
your module directories (by default /usr/local/lib/glove
and ~/.glove).
The module is called with the name of the file to be exported as the argument.
The module then prints the format that it desires, either matrix
(kGloveMatrixDataFormat) or ragged (kGloveRaggedDataFormat).
For matrix format, Glove will then write the column count, the row count, and the data
given row by row, i.e., c1r1, c2r1, c3r1, c4r1, c1r2, …. For ragged
format, Glove will write out the column count data column by column. Before
each column, the number of rows in that column are written.
Data modules should be placed in the datamodule directory of one of
your module directories (by default /usr/local/lib/glove
and ~/.glove).
Data modules serve two purposes. For transformations that require more than a basic equation, a data module can be used to transform an arbitrary number of columns in an arbitrary way. A data module can also act as an acquisition module, serving as an interface to data acquisition hardware.
The module should begin by indicating whether or not it requires data. This
is done by printing either kGloveRequiresData or
kGloveRequiresNoData. If
the module requires data, it will then print the number of columns that it
requires. Glove will then write out the number of rows, and the values
given row by row.
The module now prints whether it will just be printing all of the data
row by row with kGloveDataDump, or printing individual values, specified
by row, column, and value, with kGloveDataControl. The module then prints
the number of columns of data it will be printing.
The module is now ready to print the data. Each line of data is preceded
by either kGloveFinished, kGloveOK,
or kGloveFail. If the module is using
kGloveDataDump, then the module should then print all of the current row's
values. If the module is using kGloveDataControl, it should print the row,
column, and value of its data. When the module is finished, it should
print kGloveFinished.
Cursor modules should be placed in the cursormodule directory of one of
your module directories (by default /usr/local/lib/glove
and ~/.glove).
After the module process is started, Glove passes three booleans (T/F),
XCursorVisible(), YCursorVisible(),
and DualCursorsVisible(). It will then
write the valid values in the order x1, x2, y1, y2. If dual cursors are not
on, for example, neither x2 or y2 will be written.
The module must now indicate if there was an error. It prints a line
containing either kGloveOK, or kGloveFail followed
immediately by an error
string. Once this line is printed, the cursor module simply prints whatever
information it wants to print, which will be appended to the session.
Fit modules should be placed in the fitmodule directory of one of
your module directories (by default /usr/local/lib/glove
and ~/.glove).
After starting the module process, Glove must send the data to be fit to
the module. Before the data is sent, Glove writes the type from,
kGloveXYdYdX, kGloveXYdX, kGloveXYdY,
and kGloveXY. This is followed by the
number of data points in the data set to be fit. Then Glove send the data
in order x, y, dy, dx, but only if it is appropriate. For example, if there
are no x errors, the data will be sent x, y, dy only.
The module must now indicate if there was an error. It prints a line
containing either kGloveOK, or kGloveFail
followed immediately by an error
string. Once this line is printed, the module prints the number of
parameters, whether or not it has errors (T/F) and whether or not it has a
goodness of fit value, all in one line. The module then prints the
individual parameter's name and value, each on a single line.