Add //model/@placement_frame to proposal

[azeey]

Resolves gazebosim/sdformat#335

Preview: http://sdformat.org/tutorials?tut=composition_proposal&cat=pose_semantics_docs&branch=model_placement_frame#1-4-4-placement-frame-model-placement_frame-and-include-placement_frame

---

This change is 

[EricCousineau-TRI]

with two small comments

Reviewed 1 of 1 files at r1.
Reviewable status: all files reviewed, 1 unresolved discussion (waiting on @azeey and @scpeters)

---

composition/proposal.md, line 478 at r1 (raw file):

~~~xml
<model name="super_model">
  <model name="table" placement_frame="bottom_left_leg">

BTW In an example here or in a unittest in code, should we note the interaction between placement_frame and @canonical_link?

(i.e. they are similarly-scoped aspects, but orthogonal in application... or something less obtuse than that :P)

---

composition/proposal.md, line 510 at r1 (raw file):

  </include>
</model>
~~~

nit Consider mentioning that we originally thought about just having //include/placement_frame, but found it'd be hard to implement w/o it being a proper model attribute.

[scpeters]

Reviewed 1 of 1 files at r1.
Reviewable status: all files reviewed, 3 unresolved discussions (waiting on @azeey and @EricCousineau-TRI)

---

composition/proposal.md, line 484 at r1 (raw file):

      <pose relative_to="bottom_center">0 0 2 0 0 0</pose>
    </frame>
    <link name="bottom_left_leg"/>

I would add a pose offset with non-zero values of x and y translation for this example

---

composition/proposal.md, line 486 at r1 (raw file):

    <link name="bottom_left_leg"/>
  </model>
  <model name="mug" placement_frame="bottom_center">

having frames named bottom_center in both models makes it a little confusing. we could add some comments clarifying scope, or use different names just for clarity (center_bottom?)

---

composition/proposal.md, line 510 at r1 (raw file):

Previously, EricCousineau-TRI (Eric Cousineau) wrote…

nit Consider mentioning that we originally thought about just having //include/placement_frame, but found it'd be hard to implement w/o it being a proper model attribute.

this could be an "alternative considered"

[azeey]

I turned on the spell checker before I pushed and found some typos. Fixed in f0457c9

Reviewable status: 0 of 1 files reviewed, 2 unresolved discussions (waiting on @EricCousineau-TRI and @scpeters)

---

composition/proposal.md, line 478 at r1 (raw file):

Previously, EricCousineau-TRI (Eric Cousineau) wrote…

BTW In an example here or in a unittest in code, should we note the interaction between placement_frame and @canonical_link?

(i.e. they are similarly-scoped aspects, but orthogonal in application... or something less obtuse than that :P)

Done. 0fcaee0

---

composition/proposal.md, line 484 at r1 (raw file):

Previously, scpeters (Steve Peters) wrote…

I would add a pose offset with non-zero values of x and y translation for this example

I added a pose to the model in 0fcaee0. Is that what you had in mind?

---

composition/proposal.md, line 486 at r1 (raw file):

Previously, scpeters (Steve Peters) wrote…

having frames named bottom_center in both models makes it a little confusing. we could add some comments clarifying scope, or use different names just for clarity (center_bottom?)

I removed bottom_center from table and used bottom_left_leg as the relative_to for specifying top_center. 0fcaee0

---

composition/proposal.md, line 510 at r1 (raw file):

Previously, scpeters (Steve Peters) wrote…

this could be an "alternative considered"

Done. 0fcaee0

[scpeters]

Reviewed 1 of 1 files at r2.
Reviewable status: all files reviewed, 1 unresolved discussion (waiting on @azeey)

---

composition/proposal.md, line 484 at r1 (raw file):

Previously, azeey (Addisu Z. Taddese) wrote…

I added a pose to the model in 0fcaee0. Is that what you had in mind?

LGTM

---

composition/proposal.md, line 516 at r2 (raw file):

will have that pose. This is a very practical means of setting the location of
models. The `//model/@canonical_link` attribute on the other hand specifies to
which link the implicit model frame is attached. This does not affect the

nit: "specifies to which link" -> "specifies the link to which"

[azeey]

Reviewable status: 0 of 1 files reviewed, all discussions resolved (waiting on @scpeters)

---

composition/proposal.md, line 516 at r2 (raw file):

Previously, scpeters (Steve Peters) wrote…

nit: "specifies to which link" -> "specifies the link to which"

Done. 579eb16

[scpeters]

Reviewed 1 of 1 files at r3.
Reviewable status: complete! all files reviewed, all discussions resolved