Tree data structure?

[jges]

I have a hierarchy of files (files from a CAD program that represent the structure of a physical assembly) that I need to process. For example:

C:\path\A.prt
  C:\path\B.prt
    C:\path\C.prt
    C:\path\D.prt
  C:\path\E.prt
  C:\path\F.prt

I need to process files C and D before I can process B, then I can process B, E, and F; and finally file A. This is a small example, in actual use it can range from no children to hundreds of children many levels deep.

I tried using an array, but had a difficult time keeping track of parents and if I had hit the last level or not.

Seems like a good time to use a tree structure, but I don't have any experience with them. Any advice on how to set one up and use it? Or an alternative suggestion?

 

[PHV]

I tried using an array
What about a Scripting.Dictionary object ?

Hope This Helps, PH.
FAQ219-2884
FAQ181-2886

 

[tsuji]

[0] To adequately treat tree-structure like this, you can use a quick , but, I don't think dirty, xml approach. It always offers an advantage of separating structure from processing logic.

[1] First make out a text (xml) file (c:\path\xyz.xml, say) adequately reflect the tree of parts and their files to process. My sketch herein suppose the root being a wrapping element without operational implication.
[tt]
<root>
<A file="c:\path\a.prt">
<E file="c:\path\e.prt" />
<F file="c:\path\f.prt" />
<B file="c:\path\b.prt">
<C file="c:\path\c.prt" />
<D file="c:\path\d.prt" />
</B>
</A>
</root>
[/tt]
[2] The processing logic would be this: process the file located in the file attribute from the deepest level to the level 1 (after the root, here element A, say). The script will show you how those files be obtained in the proper order. You'll find part-B will only be processed with part-C and part-D processed, automatically.
[tt]
[green]xfile="c:\path\xyz.xml"[/green]

set oparser=createobject("msxml2.domdocument.6.0")
with oparser
.async=false
.validateonparse=false
.resolveexternals=true
end with

bret=oparser.load(xfile)
if not bret then
wscript.echo oparser.parseerror.errorcode & vbcrlf & oparser.parseerror.reason
set oparser=nothing
wscript.quite 9
end if

with oparser
.setProperty "SelectionLanguage", "XPath"

set cnodes=.selectNodes("//*[count(descendant::*)=0]")

maxlevel=0
for each onode in cnodes
n=onode.selectNodes("ancestor::*").length
if n>maxlevel then maxlevel=n
next
set cnodes=nothing

for i=maxlevel to 1 step -1 'root node is not operational, it is just a pure wrapper
set cnodes=.selectNodes("//*[count(ancestor::*)=" & i & "]")
for each onode in cnodes
wscript.echo "processing level: " & i & vbcrlf & onode.tagname & vbcrlf & onode.getAttribute("file")
'you actual file processing script here, and you've got all the data needed.
next
set cnodes=nothing
next
end with
set oparser=nothing
[/tt]

 

[jges]

Thanks for the quick replies.

The xml route looks promising, I'm going to experiment with that.