Update README code snippets; reflects syntax changes.

[Mx7f]

The example code was based on syntax used in the pre-Eigen days.

---

This change is 

[SeanCurtis-TRI]

Seems an accurate depiction of the state of the art, but it highlights some unfortunate architecture.

Reviewed 1 of 1 files at r1.
Reviewable status: all files reviewed, 1 unresolved discussion (waiting on @Mx7f)

---

README.md, line 59 at r1 (raw file):

// BVHModel is a template class for mesh geometry, for default OBBRSS template is used
typedef BVHModel<OBBRSSf> Model;
std::shared_ptr<Model> geom = std::make_shared<Model>();

BTW Why this particular change? It seems the old example was still "correct" (if antiquated).

---

README.md, line 154 at r1 (raw file):

// a) a callback to collision or distance; 
// b) an intermediate data to store the information generated during the broadphase computation
// For a), FCL's test framework provides default callbacks for both collision and distance.

This is very unfortunate. Test should be a code sink not a code source. This points to a defect. If we want people using this default, it shouldn't be defined in test.

The default callbacks should probably be moved into some common space that tests and programmers can use.

[Mx7f]

I agree; but is there anything I should change in this pull request? It seems like having accurate documentation is an improvement over incorrect documentation of the same architecture; and I am an FCL beginner, so I do not wish to make architectural changes at this time.

Reviewable status: all files reviewed, 1 unresolved discussion (waiting on @SeanCurtis-TRI)

---

README.md, line 59 at r1 (raw file):

Previously, SeanCurtis-TRI (Sean Curtis) wrote…

BTW Why this particular change? It seems the old example was still "correct" (if antiquated).

I'm new to fcl, but the only use I know of for the BVHModel class is as input to a CollisionObject, which requires it to be a shared_ptr (see line 86). The variable name change was just to use the same name here and on line 83.

---

README.md, line 154 at r1 (raw file):

Previously, SeanCurtis-TRI (Sean Curtis) wrote…

This is very unfortunate. Test should be a code sink not a code source. This points to a defect. If we want people using this default, it shouldn't be defined in test.

The default callbacks should probably be moved into some common space that tests and programmers can use.

I agree; but was concerned foremost with documenting whats currently going on (I'm trying to spin up fcl for a project).

[SeanCurtis-TRI]

I've gone ahead and created an issue. However, this PR represents a net improvement and it's good to go.

Reviewable status: complete! all files reviewed, all discussions resolved

---

README.md, line 59 at r1 (raw file):

Previously, Mx7f (Michael Mara) wrote…

I'm new to fcl, but the only use I know of for the BVHModel class is as input to a CollisionObject, which requires it to be a shared_ptr (see line 86). The variable name change was just to use the same name here and on line 83.

I'd forgotten that. Thanks.

---

README.md, line 154 at r1 (raw file):

Previously, Mx7f (Michael Mara) wrote…

I agree; but was concerned foremost with documenting whats currently going on (I'm trying to spin up fcl for a project).

No worries.