[Question] node + continuity group explanations from `artifacts.py`

[jGaboardi]

@martinfleis & @anastassiavybornova Can you check/confirm my understanding of these?

Drop/add lines within artifacts when...

• artifacts.n1_g1_identical()
  • ... dealing with 1 node and 1 continuity group ({C, E, S})
• artifacts.nx_gx_identical()
  • ... dealing with nodes >= 2 and identical continuity groups >= 2
• artifacts.nx_gx()
  • ... dealing with nodes >= 2 and continuity groups >= 2
• artifacts.nx_gx_cluster()
  • ... dealing with clusters of nodes >= 2 and continuity groups >= 2

[jGaboardi]

@martinfleis Is my understanding here correct?

[martinfleis]

• Yes
• Might be my Friday eve brain but I have no clue. It was some special case but can't get and example to pop in my head. @anastassiavybornova can you save me?
• Yes
• Yes

[anastassiavybornova]

@martinfleis @jGaboardi

my first intuition is also that the "identical" refers to identical continuity groups (ike, 4CCC, or 5EE, or 3SSS), and then nx_gx deals with "mixed" continuity groups.

my second intutition is that "identical" refers to: number of nodes == number of continuity groups (ie there are no additional "to be preserved" nodes that we have to deal with).

the fact that we are using "identical" in "n1_g1_identical" is in my view a hint towards the 2nd intuition.

@martinfleis what does your saturday brain say?

i now have a slight discomfort of how we're using "x" in the function name cause it stands at first use for number of nodes (nx) and then at second use for number of continuity grops (gx) however they are not necessarily equal (but can be). so maybe instead of the above function names with "identical" and "x" we can use:

• n1_g1
• nx_gx (x==x)
• nx_gy (x!=y)

[martinfleis]

my first intuition is also that the "identical" refers to identical continuity groups (ike, 4CCC, or 5EE, or 3SSS), and then nx_gx deals with "mixed" continuity groups.

My Saturday brain is convinced that this is the case.

[jGaboardi]

So "first intuition" is the winner?

[jGaboardi]

so maybe instead of the above function names with "identical" and "x" we can use:

• n1_g1
• nx_gx (x==x)
• nx_gy (x!=y)

@anastassiavybornova This is clean. I like it.

@martinfleis Do you have any hesitations for renaming like so?

[martinfleis]

Yeah, the first is correct. I don't have how we name it.

[jGaboardi]

I'll move forward with the following translation:

• n1_g1_identical --> n1_g1()
• nx_gx_identical --> nx_gx() -- (x==x)
• nx_gx --> nx_gy() -- (x!=y)
• nx_gx_cluster --> nx_gy_cluster() -- (x!=y)

[martinfleis]

That switches the nx_gx name to a different function. Dangerous, can we not?

[anastassiavybornova]

@jGaboardi if the first (but NOT the second) intuition is correct, then renaming nx_gx_identical to nx_gx (x==x) is not correct (cause identical refers to the continuity groups, and not to the number of continuity groups vs nodes). so in addition to @martinfleis understandable concerns above it also introduces my false 2nd intuition here. therefore:

how about instead:

• n1_g1_identical --> n1_g1
• nx_gx_identical --> nx_gy_samecontinuity or smth similar
• nx_gx --> nx_gy
• nx_gx_cluster --> nx_gy_cluster

option b would be to learn to live with the fact that x can have identical or differing values :D and keep the naming as is now (documenting what exactly we mean)

[jGaboardi]

That switches the nx_gx name to a different function. Dangerous, can we not?

Good point. I misunderstood. 😵‍💫

[jGaboardi]

renaming nx_gx_identical to nx_gx (x==x) is not correct (cause identical refers to the continuity groups, and not to the number of continuity groups vs nodes)

Got it. I think I was looking at this too long yesterday.

I like these!

• n1_g1_identical --> n1_g1
• nx_gx_identical --> nx_gy_samecontinuity or smth similar
• nx_gx --> nx_gy
• nx_gx_cluster --> nx_gy_cluster

How about nx_gx_identical --> nx_gy_eq_cgroup? Is that too long? "equivalent continuity group"

[martinfleis]

nx_gx_identical is not for the same continuity group. It is used when there are at least two nodes, one or more continuity groups but all edges have the same position in their respective continuity groups. i.e they all are E (ending), or S (single). It does not mean that all edges belong to a single continuity group.

I don't think the current naming is problematic and would probably prefer not to touch it. We can be explicit in docstring. And keep in mind that all this is private.

[jGaboardi]

nx_gx_identical is not for the same continuity group. It is used when there are at least two nodes, one or more continuity groups but all edges have the same position in their respective continuity groups. i.e they all are E (ending), or S (single). It does not mean that all edges belong to a single continuity group.

I don't think the current naming is problematic and would probably prefer not to touch it. We can be explicit in docstring. And keep in mind that all this is private.

Reasonable