Revision: 56457
http://sourceforge.net/p/brlcad/code/56457
Author: starseeker
Date: 2013-08-02 02:57:08 +0000 (Fri, 02 Aug 2013)
Log Message:
-----------
Add examples illustrating the various behaviors of the -l option for comb.
Modified Paths:
--------------
brlcad/trunk/doc/docbook/system/mann/en/comb.xml
Modified: brlcad/trunk/doc/docbook/system/mann/en/comb.xml
===================================================================
--- brlcad/trunk/doc/docbook/system/mann/en/comb.xml 2013-08-01 22:51:25 UTC
(rev 56456)
+++ brlcad/trunk/doc/docbook/system/mann/en/comb.xml 2013-08-02 02:57:08 UTC
(rev 56457)
@@ -176,8 +176,8 @@
<listitem>
<para></para>
<para>
- Suppose we have a situation where region2.r is unioned into
region1.r, and both
- region1.r and region2.r have the region flag set, as seen in the
tree below.
+ Suppose we have a situation where
<emphasis>region2.r</emphasis> is unioned into <emphasis>region1.r</emphasis>,
and both
+ <emphasis>region1.r</emphasis> and
<emphasis>region2.r</emphasis> have the region flag set, as seen in the tree
below.
</para>
<literallayout>
@@ -195,11 +195,11 @@
This geometry has an error in that one
region is not supposed to be combined into another using boolean
operations.
Further, suppose that one or
- more assemblies use region2.r and expect it to be a region - i.e.,
region2.r needs
+ more assemblies use <emphasis>region2.r</emphasis> and expect it to be a
region - i.e., <emphasis>region2.r</emphasis> needs
to remain a region with the same contents.
- To fix this problem without disturbing other geometry using region2.r,
first
- make a combination below region2.r using the <option>w</option> flag to
<command>comb</command>:
+ To fix this problem without disturbing other geometry using
<emphasis>region2.r</emphasis>, first
+ make a combination below <emphasis>region2.r</emphasis> using the
<option>w</option> flag to <command>comb</command>:
</para>
<literallayout>
@@ -213,8 +213,8 @@
</literallayout>
<para>
- The next step is to adjust the region1.r tree definition to reference
the
- non-region combination instead of region2.r:
+ The next step is to adjust the <emphasis>region1.r</emphasis>
tree definition to reference the
+ non-region combination instead of <emphasis>region2.r</emphasis>:
</para>
<literallayout>
@@ -235,12 +235,12 @@
<para>
Because the tree is manipulated as a Tcl list, we need glob
compatibility mode to be
- off on the MGED command line. Once the tree is re-defined, region1.r
no longer has
- region2.r below it, but still defines the same physical volume via
region2.r.c.
+ off on the MGED command line. Once the tree is re-defined,
<emphasis>region1.r</emphasis> no longer has
+ <emphasis>region2.r</emphasis> below it, but still defines the same
physical volume via <emphasis>region2.r.c</emphasis>.
</para>
<para>
- Optionally, the name region2.r.c can be changed to the name region2.c
with
+ Optionally, the name <emphasis>region2.r.c</emphasis> can be changed
to the name <emphasis>region2.c</emphasis> with
the <command>mvall</command> command:
</para>
@@ -302,8 +302,8 @@
</literallayout>
<para>
- Note that sph2.c is used in both spheres.c and misc.c, but sph1.c
is only used in spheres.c
- In the <command>tops</command> command output, misc.c and
spheres.c are the only top level objects. Targeting spheres.c with the
<option>f</option> option:
+ Note that <emphasis>sph2.c</emphasis> is used in both
<emphasis>spheres.c</emphasis> and <emphasis>misc.c</emphasis>, but
<emphasis>sph1.c</emphasis> is only used in <emphasis>spheres.c</emphasis>
+ In the <command>tops</command> command output,
<emphasis>misc.c</emphasis> and <emphasis>spheres.c</emphasis> are the only top
level objects. Targeting <emphasis>spheres.c</emphasis> with the
<option>f</option> option reworks the tree as follows:
</para>
<literallayout>
@@ -325,8 +325,8 @@
</literallayout>
<para>
- Notice that while spheres.c has indeed been flattened and no
longer has the intermediate combinations over its solids,
- misc.c is still intact. Notice also that while sph2.c was
preserved, since it is used by misc.c, sph1.c was removed
+ Notice that while <emphasis>spheres.c</emphasis> has indeed been
flattened and no longer has the intermediate combinations over its solids,
+ <emphasis>misc.c</emphasis> is still intact. Notice also that
while <emphasis>sph2.c</emphasis> was preserved, since it is used by
<emphasis>misc.c</emphasis>, <emphasis>sph1.c</emphasis> was removed
and does not show up in the <command>tops</command> command's
output since it was no longer used by any object in the
database.
</para>
@@ -336,7 +336,158 @@
</example>
+ <example><title>Example Of Comb Region Lifting.</title>
+ <variablelist>
+ <varlistentry>
+ <term>
+ The <option>l</option> option is useful in cases where many
<option>w</option>
+ type operations are needed.
+ </term>
+
+ <listitem>
+ <para></para>
+ <para>Suppose a geometry has the following structure:</para>
+ <literallayout>
+ <computeroutput>
+<prompt>mged></prompt> <userinput>make sph1.s sph</userinput>
+<prompt>mged></prompt> <userinput>make sph2.s sph</userinput>
+<prompt>mged></prompt> <userinput>r r1 u sph1.s</userinput>
+<prompt>mged></prompt> <userinput>r r2 u sph2.s</userinput>
+<prompt>mged></prompt> <userinput>comb assembly u r1 u r2</userinput>
+<prompt>mged></prompt> <userinput>r r3 u r1 - r2</userinput>
+<prompt>mged></prompt> <userinput>tops</userinput>
+assembly/ r3/R
+
+<prompt>mged></prompt> <userinput>tree assembly</userinput>
+assembly/
+ u r1/R
+ u sph1.s
+ u r2/R
+ u sph2.s
+<prompt>mged></prompt> <userinput>tree r3</userinput>
+r3/R
+ u r1/R
+ u sph1.s
+ - r2/R
+ u sph2.s
+ </computeroutput>
+ </literallayout>
+
+ <para>
+ There are regions below region <emphasis>r3</emphasis>, which is
not good modeling practice.
+ The <option>l</option> option applied to <emphasis>r3</emphasis>
makes <emphasis>r3</emphasis>
+ a region that no longer has regions below it, while at the same
time preserving the
+ geometric volumes defined by all existing combs and preserving
the meaning of
+ the <emphasis>assembly</emphasis> region definitions:
+ </para>
+
+ <literallayout>
+ <computeroutput>
+<prompt>mged></prompt> <userinput>comb -l r3</userinput>
+<prompt>mged></prompt> <userinput>tree assembly</userinput>
+assembly/
+ u r1/R
+ u r1.c/
+ u sph1.s
+ u r2/R
+ u r1.c/
+ u sph1.s
+<prompt>mged></prompt> <userinput>tree r3</userinput>
+r3/R
+ u r1.c/
+ u sph1.s
+ - r2.c/
+ u sph2.s
+
+ </computeroutput>
+ </literallayout>
+ </listitem>
+ </varlistentry>
+ </variablelist>
+
+ </example>
+
+
+ <example><title>Example Tree Structure That Will Prevent Successful Region
Lifting.</title>
+
+ <variablelist>
+ <varlistentry>
+ <term>
+ While the <option>l</option> option can handle many tree
configurations, there are
+ some situations where manual intervention is required. This
example illustrates
+ one such case.
+ </term>
+
+ <listitem>
+ <para></para>
+ <para>Suppose a geometry has the following structure:</para>
+ <literallayout>
+ <computeroutput>
+<prompt>mged></prompt> <userinput>make sph1.s sph</userinput>
+<prompt>mged></prompt> <userinput>make sph2.s sph</userinput>
+<prompt>mged></prompt> <userinput>make sph3.s sph</userinput>
+<prompt>mged></prompt> <userinput>r r1 u sph1.s</userinput>
+<prompt>mged></prompt> <userinput>r r2 u sph2.s</userinput>
+<prompt>mged></prompt> <userinput>comb subassembly u r1 u r2</userinput>
+<prompt>mged></prompt> <userinput>r r3 u sph3.s - subassembly</userinput>
+<prompt>mged></prompt> <userinput>comb assembly u subassembly</userinput>
+<prompt>mged></prompt> <userinput>tops</userinput>
+assembly/ r3/R
+
+<prompt>mged></prompt> <userinput>tree assembly</userinput>
+assembly/
+ u subassembly/
+ u r1/R
+ u sph1.s
+ u r2/R
+ u sph2.s
+<prompt>mged></prompt> <userinput>tree r3</userinput>
+r3/R
+ u sph3.s
+ - subassembly/
+ u r1/R
+ u sph1.s
+ u r2/R
+ u sph2.s
+
+ </computeroutput>
+ </literallayout>
+
+ <para>
+ This geometry has regions below region <emphasis>r3</emphasis>,
which is not ideal. However, attempting to
+ use the <option>l</option> option on <emphasis>r3</emphasis> will
produce the following error:
+ </para>
+
+ <literallayout>
+ <computeroutput>
+<prompt>mged></prompt> <userinput>comb -l spheres.c</userinput>
+Comb region lift failed - the following combs in the tree contain
+regions and are also used outside the tree of r3:
+
+subassembly, containing region r1
+subassembly, containing region r2
+
+The above combs must be reworked before region lifting the tree
+of r3 can succeed.
+ </computeroutput>
+ </literallayout>
+
+ <para>
+ While <emphasis>r3</emphasis> has problems, the geometry tree
under <emphasis>assembly</emphasis> is entirely correct. For
+ <emphasis>r3</emphasis> to become a toplevel region with no
regions under it, the definition of <emphasis>subassembly</emphasis>
+ would have to change too. Because
<emphasis>subassembly</emphasis> is used in the definition of
<emphasis>assembly</emphasis>,
+ as well as the definition of <emphasis>r3</emphasis>, changing
<emphasis>subassembly</emphasis>'s contents would destroy the meaning of
+ the (valid) <emphasis>assembly</emphasis> comb. Hence, the
command fails and does not change the geometry.
+ </para>
+ </listitem>
+ </varlistentry>
+ </variablelist>
+
+ </example>
+
+
+
</refsection>
<refsection xml:id="author"><title>AUTHOR</title><para>BRL-CAD
Team</para></refsection>
This was sent by the SourceForge.net collaborative development platform, the
world's largest Open Source development site.
------------------------------------------------------------------------------
Get your SQL database under version control now!
Version control is standard for application code, but databases havent
caught up. So what steps can you take to put your SQL databases under
version control? Why should you start doing it? Read more to find out.
http://pubads.g.doubleclick.net/gampad/clk?id=49501711&iu=/4140/ostg.clktrk
_______________________________________________
BRL-CAD Source Commits mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/brlcad-commits
