Ignoring patterns in CDF callbacks (CCSinvokeCallbacks)

The CCScallbackPatternsToIgnore is quite simple, and so has to be POSIX-style regular expressions (rather than perl-compatible regular expressions) because the code also was designed to work with IC5141. However, you can have a list of expressions - so you can just set this (in C-like syntax - I tend to write code in LISP syntax much of the time, but it's probably a bit more accessible this way):

CCScallbackPatternsToIgnore=list("^schCDFCallback('Plot)" "^schCDFCallback('about)")

However, I don't think these are the right callback names. This is the function that is used in the Virtuoso Schematic Editor as the function called from the edit properties form (say) but handles interfacing with the CDF callback. You should check in the CDF for the component in question, and look for the Plot and about parameters, and see what the callback defined in the CDF (via Tools→CDF→Edit in the CIW). Those are the functions you should block, because it's these that the CCSinvokeCdfCallbacks will use - not the wrapper schCDFCallback.

Andrew

Thanks Andrew, following your advice I was able to find the actual callback statements from the CDF editor. Unfortunately I still have not been able to get the expected behavior out of the CCScallbackPatternsToIgnore command line inside the script. I tried to follow the example given in the already existing expression "^MYPDKNOT_Allowed.*"

(unless (boundp 'CCScallbackPatternsToIgnore)
  (setq CCScallbackPatternsToIgnore
        ("^PLOT_CALLBACK().*" "^info().*")))

As a result of your advice and my trials with this expression, the infobox and the plot have stopped showing up.I'm very hesitant to call it a victory, as it seems like even if I comment or delete the ignore callbacks function, I am unable to get the windows to open up again. I would expect that the previous behavior would return once I undo my changes, but alas it didn't and that has me a little confused. The CIW output looks like this when I first load the changed file and then run it with my cellview, in case if it is of any help:

load("./CCSinvokeCdfCallbacks.il")function CCSshouldCallbackBeExecuted redefined

function CCSgetLibCellFromCDF redefined

function CCScreateEffectiveCDFLookalike redefined

function CCSaddFormFieldsToEffectiveCDFLookalike redefined

function CCSinvokeObjCdfCallbacks redefined

function CCSinvokeInstCdfCallbacks redefined

*WARNING* defstruct: Redefining Structure CCScellCdfCache.

function make_CCScellCdfCache redefined

function copy_CCScellCdfCache redefined

function CCSgetCdfForCell redefined

function CCSgetCachedCdfgForm redefined

function CCSrestoreCellCdf redefined

function CCSconvertCdfToPcellParams redefined

function CCSinvokeCdfCallbacks redefined

t

CCSinvokeCdfCallbacks(cv)

*WARNING* Can't modify attributes of effective CDF data.

I did not make any changes inside the CDF editor, so I am a little puzzled where it could be trying to modify the attributes of the CDF data.

Thanks again for your previous answer, as it has allowed me to progress with this issue.

I wouldn't change the default line in the code - instead just (somewhere) set the variable as I suggested in the previous post. If you do it the way that you're doing it, it won't have a difference any time the code is loaded beyond the first time, because there's a check to see if the variable is set before setting it to the default.

Which version of CCSinvokeCdfCallbacks are you using? There's a version number at the top in the comments with a "SCCS" line. In the latest it is not using the effective CDF by default, so I would be slightly surprised by the warning message at the bottom of your last post.

Andrew

Hi Andew,

Thanks so much for your help, Adding everything in their own variable outside of the default function simplifies things a lot and seems to be working. The reason why I was getting that warning was because of another CDF callback which was trying to re-calculate some parameters inside the device, but it was failing to do so because it was trying to use text variables I had set to do so.

I now figured out also why I was getting the odd behavior where even commenting out the CDF callback ignore list was still not giving me back the original behavior where the info and plot functions were not showing up. It seems like at least in my case that when I set the callback ignore list to some variable (the functions to ignore), then comment it out and load it again, the original list still remains in the memory until I explicitly set it to an empty list and then load it. That way I am able to guarantee that the list is actually empty, and the behavior is as I expect.

Everything works now, but since you asked the SCCS version number is stated as "SCCS Info: @(#) CCSinvokeCdfCallbacks.il 11/01/22.10:52:57 1.20" In the code I am running.

Thanks again, this has been valuable for my learning.

Thanks for the update!

AB_1716362704952 said:

Everything works now, but since you asked the SCCS version number is stated as "SCCS Info: @(#) CCSinvokeCdfCallbacks.il 11/01/22.10:52:57 1.20" In the code I am running.

Yes, that's the latest version.

Andrew