It looks like you're new here. If you want to get involved, click one of these buttons!
I was trying to follow up on some comments about how
this function could be used to make "asserted" objects
by "text tagging" a shape. But I can't find any mention of
this function in the DRC manual sections of the online
documentation.
Now most of my interest has been in the context of
connectivity (replace complex "recognition logic" by
simple text-assertion logic) but I would employ it for
device-specific rules (and checking for un-asserted
devices) in DRC if I were able.
1) Is this function available to DRC scripts?
2) Where would I find documentation or examples?
Comments
Hi Jim,
sorry, I went into the future already. I'm upgrading the DRC engine currently. "with_text" is a new enhancement as the DRC engine is going to support text layers as native objects.
So far, texts are kind of "hidden" objects inside "normal" input layers. They can be extracted from original layers and turned into small, dot-like shapes which can be used in "interacting", like this:
which gives (layer 100 is the bold-edged shape):
It's worth noting that "texts" can only operate on original, unmodified layers. The documentation for "texts" is here: https://www.klayout.de/doc-qt5/about/drc_ref_layer.html#k_108
Best regards,
Matthias
Matthias,
What would be the "modern" way to write it ?
Laurent
Hi Laurent,
The "modern" way is to use "text layers". To input a text layer, you use "labels" in the same way you would use "input". But "labels" generates a text layer instead of a polygon layer.
A text layer can participate in a variety of operations with polygon and edge layers:
So the modern way simply is:
Matthias
Hi Matthias,
Has there been any documentation of this function, with
emphasis on DRC scripting (examples?)? Connectivity
by tagged polygons too? Parametrics like LF, WF, NF, M
(for MOSFETs) by text or by "measure bars" on special
(non-mask) layers?
Coming up on a need to embellish a runt DRC deck
(add a raft of special geom devices) and I would much
prefer to tag, than try to develop bulletproof recognition
and derivation by layer logic.
Hi @dick_freebird,
There is not really a particular documentation I'm afraid. The "texts" function is described here: https://www.klayout.de/doc-qt5/about/drc_ref_layer.html#h2-3353
The (somewhat) new thing is that texts form a special kind of layer with "point-like" objects. This layer can be filtered and participates in some polygon operations - for example boolean "AND" and "interact", as in the example above.
It is also a hierarchical layer, so you can use it in "deep" context.
"Tagging" - as I understand it - basically just means select + interact like in the above code:
Matthias