Home > Community > Forums > Custom IC Design > how does hiCreateInst work for pcells?

Email

* Required Fields

Recipients email * (separate multiple addresses with commas)

Your name *

Your email *

Message *

Contact Us

* Required Fields
First Name *

Last Name *

Email *

Company / Institution *

Comments: *

 how does hiCreateInst work for pcells? 

Last post Mon, Oct 30 2006 10:21 AM by archive. 5 replies.
Started by archive 27 Oct 2006 01:26 PM. Topic has 5 replies and 1818 views
Page 1 of 1 (6 items)
Sort Posts:
  • Fri, Oct 27 2006 1:26 PM

    • archive
    • Top 75 Contributor
    • Joined on Fri, Jul 4 2008
    • Posts 88
    • Points 4,930
    how does hiCreateInst work for pcells? Reply

    Whenever a user begins to instantiate a pcell, the system creates a temporary scratch cell in that library, where the temp cell is named like: zpcellSCRATCH.  Any changes to fields in the instantiate form triggers a re-evaluation of the layout in the scratch cell.  This cell is then used as the basis for the complex "rubberband shadow" that the user sees while moving the cursor around in the target layout before the location is designated.

    What skill function is used to mimic this interactive copy between the scratch cell and the target layout?

    Here's the background:  I have a similar situation where I would like a user to be able to copy predetermined objects from a temporary, scratch cellView, and the user must be unaware of the source cellView.  To the user, it must appear that they are just instantiating data.  They must be unaware of the scratch cellView, and that they are actually copying data from it.  However, I would like the user to have the visual feedback of the copy function.

    I think I just want to mimic the instantiate function as far how it handles pcells.  Any thoughts on how to do this?  Is there a better way?

    Thanks!


    Originally posted in cdnusers.org by m27315
    • Post Points: 5
  • Sun, Oct 29 2006 11:14 PM

    • archive
    • Top 75 Contributor
    • Joined on Fri, Jul 4 2008
    • Posts 88
    • Points 4,930
    RE: how does hiCreateInst work for pcells? Reply

    It doesn't do an "interactive copy" between the zpcellSCRATCH cellView. That is only used to generate the "ghost" image used during the enterFunction - once the location of the instance has been chosen by the user, it is created using dbCreateParamInst.

    From SKILL you can use various enterFunctions (e.g. enterPoint, enterPath, enterBox etc) which provide some level of rubberbanding/ghosting. However, there is no SKILL access to a general ghosting approach as done with the create instance command.
    That is only available in the underlying code at the C level.

    So I'm afraid you won't be able to get the ghosting the way you want. That doesn't mean you can't build a scratch cellView and then copy the contents using dbCopyFig() etc to the destination cellView when done - but you'll need to use one of the existing enterFunctions to allow the user to interact with your program.

    Best Regards,

    Andrew.


    Originally posted in cdnusers.org by adbeckett
    • Post Points: 0
  • Mon, Oct 30 2006 7:12 AM

    • archive
    • Top 75 Contributor
    • Joined on Fri, Jul 4 2008
    • Posts 88
    • Points 4,930
    RE: how does hiCreateInst work for pcells? Reply

    Actually, there is a command called geHiDragFig that allows you to implement the ghosting functionality. It is a little flaky, but it can be made to work. I recommend you dive into the cdsdoc to get the details.

    geHiDragFig(
    w_windowID
    r_form
    t_srcPrompt
    t_dstPrompt
    t_cmdName
    t_SelSet
    g_useMultiWindow
    [ l_firstPoint ]
    )

    Used in conjunction with other SKILL functions to implement the dragging of selected database figures.


    Originally posted in cdnusers.org by dmay
    • Post Points: 5
  • Mon, Oct 30 2006 7:41 AM

    • archive
    • Top 75 Contributor
    • Joined on Fri, Jul 4 2008
    • Posts 88
    • Points 4,930
    RE: how does hiCreateInst work for pcells? Reply

    Good point. I'd completely forgotten about geHiDragFig. That would be a way of doing it.

    Regards,

    Andrew.


    Originally posted in cdnusers.org by adbeckett
    • Post Points: 0
  • Mon, Oct 30 2006 10:19 AM

    • archive
    • Top 75 Contributor
    • Joined on Fri, Jul 4 2008
    • Posts 88
    • Points 4,930
    RE: how does hiCreateInst work for pcells? Reply

    Posted By n/a on 10/30/2006 7:12 AM

    Actually, there is a command called geHiDragFig that allows you to implement the ghosting functionality. It is a little flaky, but it can be made to work. I recommend you dive into the cdsdoc to get the details.


    Wow!  That's exactly what I needed!  So, here's my implementation, using the function you mentioned:
    dstPt = cadadr(geHiDragFig(dstWinId 0 "" "Enter Destination Point" "Move" srcObjectIds t (0:0)))
    This "looks" like you are doing a copy/move/instantiation from a source cell to the desired destination window.

    Fantastic help!  Thanks!!!


    Originally posted in cdnusers.org by m27315
    • Post Points: 0
  • Mon, Oct 30 2006 10:21 AM

    • archive
    • Top 75 Contributor
    • Joined on Fri, Jul 4 2008
    • Posts 88
    • Points 4,930
    RE: how does hiCreateInst work for pcells? Reply

    Sorry about the Re-Posting of this reply. My old cndusers account Dislay Name contained my email address which is good news for spammers, but not for me. The old post will be deleted.


    Actually, there is a command called geHiDragFig that allows you to implement the ghosting functionality. It is a little flaky, but it can be made to work. I recommend you dive into the cdsdoc to get the details.

    geHiDragFig(
    w_windowID
    r_form
    t_srcPrompt
    t_dstPrompt
    t_cmdName
    t_SelSet
    g_useMultiWindow
    [ l_firstPoint ]
    )

    Used in conjunction with other SKILL functions to implement the dragging of selected database figures.


    Originally posted in cdnusers.org by dmay
    • Post Points: 0
Page 1 of 1 (6 items)
Sort Posts:
Started by archive at 27 Oct 2006 01:26 PM. Topic has 5 replies.