Graphmatica Bug reports

Past bugs

• Bug 87: Copy graphs... Color uses default gray instead of black or white color scheme

  Symptom: Copy graphs color (both BMP and EMF formats) does not keep changes to predefined color schemes. The copied image always has the default colors (gray background) unless you create a custom color scheme.
  Cause: I removed a set of parentheses while doing code cleanup in version 2.0d that turned out to be necessary after all. The resulting logic error means that setting the colors to one of the alternate pre-defined schemes inadvertently sets a flag that indicates the monochrome mode is set (and Copy graphs Color tries to help by setting the colors back to the default).
  Workaround: Change any color setting so you are not using one of the predefined schemes. This issue does not arise when using a custom color scheme.
  Status: Fixed in 12/14 reissue of version 2.0e (patch version 2.0.5.13).

• Bug 86: new abs value operator has trouble with implicit multiplication and unary minus

  Symptom: Can't graph y=|x|-1 or y=2|x|
  Cause: Have to keep track of number of |'s seen and treat odd ones like "(", even ones like ")" when deciding whether "-" is minus or unary minus, and whether to insert implict "*"
  Workaround: Use abs(x) notation instead.
  Status: Fixed in 12/6 reissue of 2.0e (version 2.0.5.12)

• Bug 84: Annotations print in miniscule font size

  Symptom: Annotations print at unreadably small size
  Cause: Fonts were being scaled back down to screen resolution before annotations were drawn.
  Workaround: Copy to clipboard and print from Word or Paint.
  Status: Fixed for version 2.0e.

• Bug 83: Can't change data plot name

  Symptom: In version 2.0d, attempting to change the name of a data plot always reverts back to the original name.
  Cause: Check added in 2.0d to avoid marking document dirty when name was not changed inadvertently removed the code that was setting the name in the graph document data structure.
  Workaround: Change name by editing saved .gr file.
  Status: Fixed for version 2.0e.

• Bug 82: Curve fit doesn't come close enough to zero on missing coefficients

  Symptom: Attempting to curve fit from a small (5 point) data set with the exact equation y=2x produces approximation of y = 2x+2.1*10^-5
  Cause: Need to run more iterations on these equations to get to the point where the coefficients are close enough to zero to be considered noise and eliminated.
  Workaround: Adding more data points reduces the size of the error but does not eliminate it.
  Status: Fixed for version 2.0e.

• Bug 80: Point tables incomplete when increment smaller than pixel separation

  Symptom: Point tables appear to stop early when setting increment to .01 or smaller with a typical domain.
  Cause: The graphing loop could not keep up with the need to generate points, because it was calculating only one per pass (step size is adapted dynamically to maintain 1-2 pixel separation between plotted points).
  Workaround: Zoom in until desired increment matches separation between screen pixels.
  Status: Fixed for version 2.0e. Added code to evaluate additional points per graph step if necessary (although total number of points supported by tables is still limited to 1000).

• Bug 79: Inequalities which approach discontinuity too steeply shade wrong

  Symptom: y >= sqrt(5x-1)-2 draws with the shading flipped under the axis on the left side most of the time
  Cause: The steepness of the curve at the discontinuity, plus the round-off error inherent in using the sampled points from the point list to determine where to make guesses, led to cases where a point that was above the curve in physical space ended up rounding off to below the curve in logical space, resulting in an incorrect sample for the inequality boundary drawing code.
  Workaround: Resize the grid. In very occasional scenarios (or when the curve is squashed down so it's not so steep) the calculation works properly
  Status: Fixed for version 2.0e.

• Bug 77: Equation text for derviatives garbled

  Symptom: Finding derviative of "y=1/(3x^2-4x)" produces a correct graph labeled with an invalid equation.
  Cause: Introduced during conversion to unicode in 2.0d. The code that formats the equation is not properly ignoring old data from the temporary buffer
  Workaround: None
  Status: Fixed for version 2.0e.

• Bug 76: Status bar can partially obscure grid scollbar at certain font sizes

  Symptom: The status bar partly overlaps the bottom of the grid window.
  Cause: The size of the graph window was being determined based on the size of the client area minus the height of the status bar font plus a constant. It appears that the height of the whitespace in the status bar control varies somewhat based on font size, though, so if you made the font big enough the constant was no longer adequate.
  Workaround: None
  Status: Fixed for version 2.0e.

• Bug 75: Grid does not resize when height of edit or status bar font changes

  Symptom: If you make the equation editor or status bar font size larger, the grid does not immediately adjust and may temporarily be obscured by these elements of the screen.
  Cause: The Fonts dialog box was not requesting that the size of the window be recalculated under all appropriate circumstances.
  Workaround: Resize the window manually to force recalculation of window sizes
  Status: Fixed for version 2.0e.

• Bug 74: Crash when entering an ode with derivative raised to an even power

  Symptom: For example, dy^2=x crashes
  Cause: The equation parser was incorrectly marking the ode as a cartesian relation to be graphed as "double" equation, but the data structures were not set up properly to go along with that flag.
  Workaround: Isolate dy manually, e.g. dy = sqr(x) instead of dy^2=x
  Status: Fixed for version 2.0e.

• Bug 72: Crash attempting to change fonts for legends etc. with no printer available

  Symptom: Attempting to change any of the fonts that may be used in printed output (i.e. Legends, Axis Labels, Titles, ...) caused the program to crash before the Choose Font dialog box even came up.
  Cause: Expected information about the printer is not available from Windows when no printer is installed.
  Workaround: Install any printer driver, or re-enable the print spooler if you have a printer but have disabled it manually in the Windows control panel.
  Status: Fixed for version 2.0d.

• Bug 67: Implicit graphs with no solutions may display erroneous curves

  Symptom: x^2+4x+y^2-6y+14=0 draws a half-circle
  Cause: Implicit equations like this, where neither x nor y can be isolated symbolically, are graphed by ODE approximation. In this case, no initial values for the approximation can be found (because the equation has no real solutions); however the ODE is being graphed anyway.
  Workaround: None. If you find a graph looks suspicious, validate it by entering the coordinates of a point on it back into the equation and checking that they do solve the equation.
  Status: Fixed for release 2.0d.

• Bug 71: x=(-3)^y does not report discontinuous function error

  Symptom: y=(-3)^x correctly reports that it cannot be graphed correctly because it is a discontinuous function. x=(-3)^y was just not drawing anything.
  Cause: In the process of solving this equation for y, the discontinous function turned into a domain error over all of the domain, which is only reported when Warnings are enabled in the options panel.
  Workaround: None
  Status: Fixed for version 2.0d

• Bug 70: minus sign interpreted differently in y=-3^x than y=1-3^x

  Symptom: y=-3^x was interpreted as the discontinuous function y=(-3)^x instead of the well-defined -(3^x) as the standard order of operations implies it should be.
  Cause: an optimization I made for parsing negative numbers behaves equivalently in most situations but not here.
  Workaround: Add parentheses as necessary to enforce the desired meaning.
  Status: Fixed for version 2.0d

• Bug 69: Strict inequalities do not paint properly following reload

  Symptom: After reloading a saved .gr file, the curves of strict inequalities paint as solid lines instead of dashed, and always in graph color 1 (magenta in the default color scheme).
  Cause: The code for reloading the saved equation colors fails to account for the fact that these curves have negative color values to indicate the dashed line.
  Workaround: Delete the lines with negative numbers under the [eqcolors] section before reloading your .gr file. The graphs may be assigned new colors, but will graph in different colors and using dashed lines.
  Status: Fixed for release 2.0d.

• Bug 23: Complex "double" inequalities may not shade properly

  Symptom: Inequalities which are based on relations that require evaluation of both positive and negative roots (e.g. hyperbolae like x^2-y^2 < 6 or absolute values like abs x - abs y > 3) may have triangular pieces missing from the shaded region.
  Cause: The code which computes the region to fill is designed to work with single-valued functions. To display inequalities such as these, a number of rather unscientific heuristics are applied to determine how the regions for the two halves of the relation should be joined. There appear to be some errors here.
  Workaround: In some cases, you can obtain the correct fill behavior by manipulating the window size slightly. The resulting minor change in the grid range may cause the boundaries of the region to be computed correctly.

  If you notice this bug, please e-mail me the inequality you had trouble with. (You needn't describe the exact filling defect or send me a screen shot or anything, unless I cannot reproduce it here.) When I have some time to dedicate to understanding this bug I will attempt to tweak the code so that as many of these graphs fill properly as possible.
  Status: All cases I have observed are fixed in version 2.0c. If you see any other inequalities that do not shade properly, please report them to me.

• Bug 65: equations using free vars only in user functions don't accept domains

  Symptom: Define function f(t) = a * t y=f(x) {a:1,3,1} fails to parse with "Your equation included a variable or function that Graphmatica does not support, or you mis-typed a function name. Please edit your equation and try again."
  Cause: The code to inspect functions to see if they used the free variables a, b, and c was not modified to search the expressions for user-defined functions.
  Workaround: Include "+ 0a" in equation to trick parser into accepting domain for 'a' without changing the graph
  Status: Fixed in version 2.0c

• Bug 63: Paste Data Plot crashes when no current data plot available

  Symptom: The program crashes when Edit/Paste Data Plot is chosen before the data plot editor is shown.
  Cause: The main menu was passing in a null data plot in this case.
  Workaround: Show the data plot editor prior to attempting paste.
  Status: Fixed for version 2.0b

• Bug 61: program crashes when AutoRange activated with a flow-field on screen

  Symptom: With AutoRange enabled, plot an ODE flow field, followed by a cartesian equation that is mostly off-screen. The program crashes attempting to analyze the range of on-screen equations to reset the range of the y axis.
  Cause: An incorrect comparison was causing the program to try to evaluate the range of the flow field as well, which fails due to differences in the way these types of equations are represented internally.
  Workaround: Avoid using AutoRange with flow fields on screen.
  Status: Fixed for version 2.0b

• Bug 60: Program crashes loading data plot on startup

  Symptom: Double-clicking on a .gr file with a data plot (or specifying one in a command line by hand) crashes loading the file.
  Cause: Code to reset the plot number was looking at CurrentDoc before it was initialized.
  Workaround: Start the program first, then load the .gr file by drag n drop or the File/Open menu item.
  Status: Fixed for version 2.0a

• Enh 59: Size of dots on dotted grid background not affected by gridlinewidth

  Symptom: The size of the background dots on the graph paper doesn't change regardless of what you set the gridlinewidth parameter to. It should.
  Status: Fixed for version 2.0a

• Bug 58: Set Initial Value doesn't work for ODEs of > 1st order

  Symptom: The rubber band for the slope seems to get stuck at the first point clicked on--subsequent Set Initial Value calls don't change it. The rubber band is not erased properly when done. And it doesn't seem to change the IV of the equation either.
  Cause: The code that tracks the progress in picking an IV was not updated properly when the new code to handle the always-on coordinate cursor was added.
  Workaround: Enter IV manually.
  Status: Fixed for version 2.0a

• Bug 57: Set Initial Value for systems of ODEs leaves marks on screen

  Symptom: The 2nd and 3rd IVs in the system might be displayed a pixel to the side of where they ought to be, leaving little oval marks on the screen when they are erased.
  Cause: The mouse capture area was being set to a zero-width rectangle, which doesn't work. It needed to have a width of 1 to constrain the cursor to the vertical line of t=initial selection properly.
  Workaround: Enter IV manually.
  Status: Fixed for version 2.0a

• Bug 55: Should not be able to set grid element colors same as background

  Symptom: Legends did not appear because user had selected the same color as the background for them.
  Cause: The Colors dialog box does not attempt to validate selections, so changing just the background color to white leaves the legends unreadable.
  Workaround: Don't do that.
  Status: Fixed for version 2.0a

• Bug 54: y = (1/a)^x {a: 1,3,1} fails as discontinuous

  Symptom: The above curve should work, but is reported to be a case of b^x, b < 0.
  Cause: Code designed to determine the domain of exponentials like this was failing due to division by zero because it was run before 'a' was set.
  Workaround: use y = a ^-x {a: 1,3,1} instead.
  Status: Fixed for version 2.0a

• Bug 53: Really long curves don't graph fully.

  Symptom: For example, y=sin(x) + 0.5*cos(x * 100) gets cut off, at various x coords depending on fineness value
  Cause: This is a curve with really long arc length when drawn with fineness high enough to show the detail. The program has a limitation of 4000 points stored per equation; curves that require more than that will not draw properly when you refresh the screen.
  Workaround: None
  Status: Fixed for version 2.0a

• y=int x does not finish graphing

  Symptom: The entire graph is drawn, but the program stays in "graph" mode.
  Cause: A change avoid missing steep, narrow lobes of polynomials by detecting changes from positive to negative slope was unintentionally catching graphs with zero slope as well.
  Workaround: Abort graphing manually once it is done.
  Status: Fixed in version 2.0.

• Errors in second order ODE examples file

  Symptom: Two equations in the demo file ODE2.GR are incorrect (or the comment on them is incorrect, depending on your point of view).
  Cause: I was not paying attention and neglected a factor of two in the formula for simple harmonic motion.
  Workaround: You can download the corrected ODE2.GR here.
  Status: Fixed in version 2.0.

• Problems with families of curves when decimal separator is ","

  Symptom: After drawing the first graph in the family, the program stops with the error "Your equation included a variable or function that Graphmatica does not support, or you mis-typed a function name. Please edit your equation and try again."
  Reported:2/16/00
  Cause: When the decimal separator is set to "," the parser becomes confused unless you add spaces between the terms in a domain. Unfortunately, when the program defines the variable "a" for you, it does not do this.
  Workaround: Specify the domain for "a" by hand like so "{a:1, 5, 2}".
  Status: Fixed in French version 1.60e and in version 2.0.

• Flow fields drawn incorrectly for non-square grids

  Symptom:The program draws too many or too few slope vectors in the vertical dimension.
  Reported:2/9/00
  Cause: The flow field graphing code was unprepared for a non-square aspect ratio.
  Workaround: Square up the grid.
  Status: Fixed in version 2.0.

• Program crashes when choosing Redraw/Last Graph menu item

  Symptom:Choosing Redraw/Last Graph when there are no graphs crashes the prorgam.
  Cause:That is not really a valid time to be using that menu item.
  Workaround: Don't do that.
  Status: Fixed in version 2.0.

• Multiple unary minus signs handled incorrectly

  Symptom: y=--x graphed the same as y=-x.
  Cause: There was a bug in the code to scan negative numbers.
  Workaround: Don't enter silly equations like y=--x.
  Status: Fixed in verison 1.60e.

• Very small domain not respected in log-log graphing

  Symptom: Plotting a function with a domain with a small (much less than 1) endpoint on a logarithmic scale will often plot past the upper bound of the domain.
  Cause: The margin of error for the calculation to determine whether the function is within the bounds of the domain is not being adjusted properly to account for the logarithmic step rate.
  Workaround: None available.
  Status: Fixed in verison 1.60e.

• Set Domain feature may leave extraneous "x" or "y" in equation

  Symptom: After using the Point Evaluate or Critical Points dialog box, using the Set Domain function to select a new domain using the mouse will put an "x" or "y" in between the equation and domain instead of a space.
  Cause: Temporary variables in these dialog boxes were improperly assigned the same memory as the constant string for a single space, and overwriting it.
  Workaround: Enter the domain manually, or edit the equation to remove the offending character.
  Status: Fixed in verison 1.60d.

• Polar/parametric graphs not back-tracking properly when first on-screen point graphed

  Symptom: For example, some of the segments in r = a(pi - t) {a: - 15, 15, .12} stop short of the top/bottom of the screen. (This bug does not affect Cartesian graphs.)
  Cause: When the first part of the domain for a function is not visible on screen, the program is designed to scan forwards quickly until it finds an on-screen point, and then back-track if necessary to the first point in the domain which is visible. However, unwanted interaction between the back-tracking and automatic fineness adjustment code was causing the back-tracking to stop prematurely when the fineness had to be increased quite a bit to get a point close enough to the first to draw a smooth graph
  Status: Fixed in verison 1.60d.

• Graphs of odd powers of y only draw the positive half of the function

  Symptom: For example, x=y^3 only draws graph for x > zero.
  Cause: When isolating y in equations like this I neglected to check whether the resulting fractional power of x was odd or even, and therefore conservatively assume that it is even and therefore not defined for x < 0
  Workaround: Isolate y by hand (y=x^1/3) or add another term with y in it to force the equation to be evaluated as a function of y, not x (x=y^3 + 0y)
  Status: Fixed in verison 1.60d.

• Integrate curve shades wrong side of curve in certain zoomed-in situations

  Symptom: When finding the area under a curve, and the x axis is not on screen, Integrate curve may shade the area between the curve and infinity instead of between the curve and the axis.
  Cause: The calculation of the physical coordinate for y=0 can overflow if the axis is too far away from the edge of the screen, causing it to have the wrong sign.
  Workaround: Don't zoom in so far that the x axis goes off the screen when displaying integration regions.
  Status: Fixed in verison 1.60d.

• Edge of screen not handled correctly for "vertical" graphs

  Symptom: Vertical lines (x=n) missing several pixels at top of screen; inequalities like x > n may not fill correctly
  Cause: The graph loop for these equations is exiting prematurely before the last several pixels have been calculated.
  Workaround: As above, you may be able to obtain the correct fill behavior by resizing the window slightly.
  Status: Fixed in patch release 1.60c.

• Integrate curve does not handle large, off-screen regions properly

  Symptom: Program may crash when drawing integration region where most of the region is not visible on the screen
  Cause: The calculation for the number of points required in the region to shade was exceeding the maximum allowed and wrapping around to negative numbers, causing invalid memory allocation.
  Workaround: Only integrate over a range which is completely visible on-screen.
  Status: Fixed in patch release 1.60c.

• [Windows only] Copy graphs (WMF format) contains unwanted raster data

  Symptom: Importing metafiles from Graphmatica into AutoCad fails with message "Raster data in the metafile has been ignored"
  Cause: An operation to clear the drawing surface for bitmap copy was incorrectly included in the metafile copy as well.
  Workaround: Unknown.
  Status: Fixed in patch release 1.60c.

• [Windows only] Default fonts not TrueType in low-resolution modes

  Symptom: Side labels are not rotated properly, or text is not scaled correctly when printed.
  Cause: When running in less than 1024x768 resolution, the program selects "Courier" instead of "Courier New" as the default font for grid text and labels. This is a screen font and may not scale or rotate properly like a TrueType font.
  Workaround: Select any font you wish for the affected text using the Fonts dialog box (View menu). Then Save Setup Info from the file menu to make this selection the new default. Graphmatica will only let you choose TrueType fonts when you select a font manually so any choice is OK.
  Status: Fixed in verison 1.60a.

• [Windows 95 only] Mouse capture not released when switching to another application

  Symptom: If you switch to another application while in the coordinate cursor, or in domain/tangent line/integral selection mode, the mouse pointer continues to be restricted to the borders of the current grid in Graphmatica.
  Cause: Historically, there has never been any code to handle this scenario, because it was impossible to encounter (in Windows 3.1, you cannot switch applications using Alt+Tab when mouse capture is on).
  Workaround: Switch back to Graphmatica and click on the grid to exit the coordinate cursor or other feature which invoked the mouse capture.
  Status: Fixed in verison 1.60a.

• Point tables may erroneously show values where function not defined

  Symptom: At fractional intervals between points, the point tables may show a value for discontinuous functions like y=(x^2-2)/(x-1) at the point of the discontinuity. They should show "Error" where the function is not defined.
  Cause: Due to normal roundoff error the point calculated was actually not precisely at the singularity, so the function really was defined. However, this is disguised by the fact that the value of the independent variable is rounded back to the precise value of the singularity when it is printed.
  Workaround: Zoom out. Rounding error is not usually significant when the interval between points in the table is an integer value.
  Status: Fixed in verison 1.60a.

• [Windows only] Cannot position annotations with the keyboard

  Symptom: When placing annotations, the arrow keys have no effect.
  Cause: The keyboard focus is not being set properly to the graph window.
  Workaround: You must place annotations using the mouse.
  Status: Fixed in verison 1.60a.

• [Windows 95 only] Toolbar buttons overlap equation editor when IE 4.0 installed

  Symptom: See picture.
  Cause: Internet Explorer 4.0 surreptitiously "upgrades" some system DLL(s) with versions that appear to be incompatible with programs built to use the original toolbar controls. In particular, the toolbar seems to require an explicit resize message before I can calculate its size correctly in order to position the editor window properly.
  Workaround: Avoid or un-install IE4.
  Status: Sort-of fixed in verison 1.60a. But it appears that Microsoft has fought back with yet another new version of the toolbar control, which still makes space on the buttons for unwanted text unless I add an API call which didn't exist in the original version of this control. Sheesh. But I think I have this fixed for real in version 1.60e.

• [Windows only] Polar graph paper with gridlines is broken

  Symptom: Polar graph paper displays as a single "blob" when you have selected gridlines or solid gridlines detail level
  Cause: I seem to have broken this when I modified the way the way graph paper is drawn when you have selected "dots" detail level so that the dots scale to a reasonable size when printed. This is causing the ellipses which are drawn for polar paper to be filled rather than outlined.
  Workaround: Select the Dots detail level when using polar graph paper.
  Status: Fixed in verison 1.60a.