Fix whitespace preservation behavior with TinyXML2

[mjcarroll]

This addresses a change in how whitespace is preserved/collapsed between
tinyxml and tinyxml2. The change primarily impacts SDF files that have
newlines, which is frequently done for aesthetics with include uris.

In this case, we use an alternate constructor for TinyXML2 that will
collapse whitespace by default.

Note that there is a performance impact of this behavior, which causes
the parsing to run essentially twice.

More information on the behavior may be found here: https://leethomason.github.io/tinyxml2/

Closes #322

Signed-off-by: Michael Carroll michael@openrobotics.org

[codecov-commenter]

Codecov Report

Merging #359 into sdf10 will increase coverage by 0.00%.
The diff coverage is 93.33%.

@@           Coverage Diff           @@
##            sdf10     #359   +/-   ##
=======================================
  Coverage   87.32%   87.32%           
=======================================
  Files          60       60           
  Lines        9214     9216    +2     
=======================================
+ Hits         8046     8048    +2     
  Misses       1168     1168           

Impacted Files	Coverage Δ
src/parser.cc	78.55% <93.33%> (+0.04%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 095d88a...544fa8a. Read the comment docs.

[scpeters]

I'm going to test a narrower approach that calls sdf::trim on the contents of <uri> tags

[scpeters]

I'm going to test a narrower approach that calls sdf::trim on the contents of <uri> tags

narrower approach in #365

[scpeters]

I think this is the right approach. I'll close #365

[scpeters]

@osrf-jenkins run tests please

[azeey]

Do we need to make the same changes in the URDF parser (src/parser_urdf.cc)?

[mjcarroll]

Do we need to make the same changes in the URDF parser (src/parser_urdf.cc)?

I don't believe so, because my understanding was that URDF only uses attributes in tags, versus the actual content of tags:

<pose rpy="0 0 0" xyz="0 0 0" />
vs
<pose>0 0 0 0 0 0</pose>

Is that correct?

[azeey]

Do we need to make the same changes in the URDF parser (src/parser_urdf.cc)?

I don't believe so, because my understanding was that URDF only uses attributes in tags, versus the actual content of tags:

<pose rpy="0 0 0" xyz="0 0 0" />
vs
<pose>0 0 0 0 0 0</pose>

Is that correct?

I think you're right about that, but if the URDF contains Gazebo extensions (eg. https://answers.gazebosim.org//question/7082/using-sdf-tags-in-urdf-gazebo-extension-directly/), it could have values in tags.

[mjcarroll]

URDF contains Gazebo extensions

Failed to account for that. I will update and add a test for it.

[mjcarroll]

I added a test case in 2721714, it actually didn't require any adjustment for me in the actual URDF parser.

I think this is because we trim the values that we parse from URDF in the process of changing to SDF.

In the case of this test, the tinyxml2::XMLDocument doc (parsed URDF) is:

<robot name="test">
    <link name="link">
        <inertial>
            <mass value="0.1"/>
            <origin rpy="1.570796326794895 0 0" xyz="0.123456789123456 0 0.0"/>
            <inertia ixx="0.01" ixy="0" ixz="0" iyy="0.01" iyz="0" izz="0.01"/>
        </inertial>
        <visual>
            <geometry>
                <sphere radius="1.0"/>
            </geometry>
        </visual>
        <collision>
            <geometry>
                <sphere radius="1.0"/>
            </geometry>
        </collision>
    </link>
    <gazebo reference="link">
        <material>
      Gazebo/Orange
    </material>
        <mu1>
      100
    </mu1>
        <mu2>

      1000

    </mu2>
    </gazebo>
</robot>

And then sdfXml (the converted SDF) is

<sdf version="1.7">
    <model name="test">
        <link name="link">
            <inertial>
                <pose>0.123456789123456 0 0 1.570796326794895 0 0</pose>
                <mass>0.1</mass>
                <inertia>
                    <ixx>0.01</ixx>
                    <ixy>0</ixy>
                    <ixz>0</ixz>
                    <iyy>0.01</iyy>
                    <iyz>0</iyz>
                    <izz>0.01</izz>
                </inertia>
            </inertial>
            <collision name="link_collision">
                <pose>0 0 0 0 0 0</pose>
                <geometry>
                    <sphere>
                        <radius>1</radius>
                    </sphere>
                </geometry>
                <surface>
                    <contact>
                        <ode/>
                    </contact>
                    <friction>
                        <ode>
                            <mu>100</mu>
                            <mu2>1000</mu2>
                        </ode>
                    </friction>
                </surface>
            </collision>
            <visual name="link_visual">
                <pose>0 0 0 0 0 0</pose>
                <geometry>
                    <sphere>
                        <radius>1</radius>
                    </sphere>
                </geometry>
                <material>
                    <script>
                        <name>Gazebo/Orange</name>
                        <uri>file://media/materials/scripts/gazebo.material</uri>
                    </script>
                </material>
            </visual>
            <gravity>true</gravity>
            <velocity_decay/>
        </link>
    </model>
</sdf>

[azeey]

Looks good! And thanks for adding the URDF Test.