props/physical: add printing of PlanGrams

[michae2]

opt: add Operator.CamelCase function

To make plangrams look more like opt and canned opt plans, we want to
use camel case operator names. Add a new function.

Epic: None
Release note: None

---

props/physical: add printing of PlanGrams

Add functions to print plangrams, and also clean up some of the structs
and comments.

Informs: #152053
Release note: None

[trunk-io]

😎 Merged successfully - details.

[cockroach-teamcity]

This change is 

[blathers-crl]

Detected infrastructure failure (matched: ). Automatically rerunning failed jobs. (run link)

[blathers-crl]

Detected infrastructure failure (matched: ). Automatically rerunning failed jobs. (run link)

[DrewKimball]

Nice!

@DrewKimball reviewed 6 files and all commit messages, and made 1 comment.
Reviewable status: complete! 1 of 0 LGTMs obtained (waiting on andyyang890 and ZhouXing19).

[andyyang890]

@andyyang890 made 3 comments.
Reviewable status: complete! 1 of 0 LGTMs obtained (waiting on michae2 and ZhouXing19).

---

pkg/sql/opt/optgen/cmd/optgen/ops_gen.go line 78 at r1 (raw file):

	for _, define := range g.sorted {
		fmt.Fprintf(&indexes, "%d, ", names.Len())
		fmt.Fprint(&names, define.Name)

Is it maybe worth adding a linter or some kind of validation that the source names are all camel case already? IIUC there's an implicit assumption of that here. If there's already some validation, would you mind adding a comment explaining where the validation happens?

---

pkg/sql/opt/props/physical/plangram.go line 144 at r2 (raw file):

		defer b.WriteRune('\n')
	}
	if p.root == nil {

nit: use p.Any() instead

---

pkg/sql/opt/props/physical/plangram.go line 164 at r2 (raw file):

			return
		}
		switch t := term.(type) {

Should we have a default case for unexpected types?

[michae2]

TFTRs!

@michae2 made 4 comments and resolved 3 discussions.
Reviewable status: complete! 0 of 0 LGTMs obtained (and 1 stale) (waiting on DrewKimball and ZhouXing19).

---

pkg/sql/opt/optgen/cmd/optgen/ops_gen.go line 78 at r1 (raw file):

Previously, andyyang890 (Andy Yang) wrote…

Is it maybe worth adding a linter or some kind of validation that the source names are all camel case already? IIUC there's an implicit assumption of that here. If there's already some validation, would you mind adding a comment explaining where the validation happens?

Good point, added a test.

---

pkg/sql/opt/props/physical/plangram.go line 144 at r2 (raw file):

Previously, andyyang890 (Andy Yang) wrote…

nit: use p.Any() instead

Done.

---

pkg/sql/opt/props/physical/plangram.go line 164 at r2 (raw file):

Previously, andyyang890 (Andy Yang) wrote…

Should we have a default case for unexpected types?

Done.

[michae2]

Thanks for the reviews!

/trunk merge