Issue 136

[tmarble]

Please see the commit message and additional comments in #136

[pmdoll]

• Is there a benefit to keep pamela symbols as symbols in IR? i.e In IR, why change field :a to a?
• Also, we seemed to have lost pclass info. The new version does not has :pclass this for the field.
• Do we expect fields to have multiple references? I see :names [a] instead of :field a

[tmarble]

@pmdoll

1. Use of symbols for fields is consistent everywhere
2. A field-ref is a specialization of symbol-ref where 'this is elided from the name vector because fields references always start from this class.
3. Yes. One may refer to a method of pclass initialized in a field like field1.methodAor the field of another pclass like field2.field3

[pmdoll]

syntax error is reported as grammar is ambigous.

(defpclass a []
  :fields {:breathing true})

[tmarble]

@dcerys i believe the regressions you have found have been addressed in https://github.com/tmarble/pamela/commit/e6469721385450e9f1ed3bbb5d08016cd8d356fa

[tmarble]

@dcerys I have just pushed an additional commit to address the issues from yesterday's telecon (please see the commit message).

[pmdoll]

Removal of plantid will have ripple effects to other components.

[tmarble]

@pmdoll do you want to keep :plantid in addition to OR in replacement of :plant-id ?

[dcerys]

It is good to inch forward on the name harmonization. Since we agreed to do this in a backwards-compatible way, we should preserve the old attributes while adding the new ones. This will result in some temporary redundancy, but will maintain short-term compatibility.

So, we should do this for :plant-id, :plant-interface, and :plant-part. All new/updated code should be using these new attributes.

[pmdoll]

Since this is a trivial change, I suggest we leave it as is.

[tmarble]

@dcerys I'm not super clear from your previous comments what needs to change in order for the PR to be merged? The formal args for the video function are: [client server display] and we know that argsmap is NOT guaranteed to be in positional argument order (hence our new design for args-map). Please advise.

[tmarble]

@dcerys please review this latest commit.