Having trouble with cfscript to uncompress files - experts only

ASKER

Correction:

The pdf file will be in the format like:

dop99998888777744445555.pdf

And I'll need to get just the numbers from out of that string.

ASKER CERTIFIED SOLUTION

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial

ASKER

@duncancumming:

Yes the 'untarring' part is working fine. It's putting together the rest of the cfscript/cfml that I'm having trouble with. This is what I have now:

That throws an error "object instantiation exception". Where is the error in that code and is there a better way of doing that? thanks,

-roger

Please Wait...<Br>
<Cfflush> 
<!--- Get file names of TARs to be untarred --->
<cfdirectory action="list" name="getPPS" directory="#ppsdir#" filter="dop*.tar" />
<cfset tarFileCount = getPPS.RecordCount>


<cfset tarFileList = quotedvaluelist(getPPS.name, ",")>

   <cfscript>
		TarEntry = createobject("java", "org.apache.tools.tar.TarEntry");
		TarInputStream = createobject("java", "org.apache.tools.tar.TarInputStream");
		FileInputStream = createobject("java", "java.io.FileInputStream");
		FileOutputStream = createobject("java", "java.io.FileOutputStream");
		FileIO = createobject("java", "java.io.File");
		for(file_idx=1;file_idx lte listlen(tarFileList);file_idx++)
		{
			tarfile = listgetat(tarFileList,file_idx,",");
			tarfile_quotefree = expandpath(ReReplace(tarfile,"'","","ALL"));
			outdir = expandpath("unpacked");
			tin = TarInputStream.init(FileInputStream.init(tarfile_quotefree));
			te = tin.getNextEntry();
			while (structkeyexists(variables, 'te'))
			{
         			destpath = outdir & "\" & te.getName();
            			if (te.isDirectory())
                			destpath.mkdir();
                		else
                		{
                			fos = FileOutputStream.init(destpath);
                    		tin.copyEntryContents(fos);
                    		fos.close();
                		}
                		te = tin.getNextEntry();
				}
			}
				tin.close();
		
	</cfscript>

<cfdirectory action="list" name="getPdfNames" directory="unpacked" filter="DOP*.pdf">
<cfset lstCertNum = quotedvaluelist(getPdfNames.name)>
<cfdump var="#lstCertNum#">

<cfabort>

Open in new window

That error could be happening on any of these lines:
TarEntry = createobject("java", "org.apache.tools.tar.TarEntry");
TarInputStream = createobject("java", "org.apache.tools.tar.TarInputStream");
FileInputStream = createobject("java", "java.io.FileInputStream");
FileOutputStream = createobject("java", "java.io.FileOutputStream");
FileIO = createobject("java", "java.io.File");

or maybe this one:
tin = TarInputStream.init(FileInputStream.init(tarfile_quotefree));

or even
fos = FileOutputStream.init(destpath);

The line number of the error message should tell you which one is the problem. Alternatively comment them all out and uncomment each one individually until that error is thrown.

That REReplace isn't necessary, this would also work:
tarfile_quotefree = expandpath(Replace(tarfile,"'","","ALL"));

SOLUTION

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial

ASKER

OK folks this is perplexing:

I'm getting an object intantiation exception and like agx says it's generic, the actual error is Caused by: java.io.FileNotFoundException: \\xxxxx\xxxx\xxxxx\xxxxx\xxxxxx\xxxxxxxxx\DOP0323100001.tar (The system cannot find the file specified).

But the problem is that when I do a cfdump at the very begining, after line 5, it shows that there are 3 files as there should be (with the exact same name DOP0323100001.tar and so on), but when it come down to executing the following line (which is in bold meaning it errored out at that line), it gives me that error message:

 Object Instantiation Exception.
An exception occurred when instantiating a Java object. The class must not be an interface or an abstract class. Error: ''.
 
The error occurred in \\xxxxxxxx\xxxxxx\xxxxxxxx\webroot\imaging\xxxxxxx\step2.cfm: line 35

33 : 			outdir = expandpath("unpacked");
34 : 
35 : 			tin = TarInputStream.init(FileInputStream.init(tarfile_quotefree));
36 : 
37 : 			te = tin.getNextEntry();

Open in new window

ASKER

Allright, I've made some progress but hit a snag. I've managed to unpack all the tar files into a directory. Now I want to create a list or array containing the names of all pdf files in that list. Any suggestions on how best to do that?

<cfdirectory action="list" name="getPdfNames" directory="#unpacked#">
<cfset lstCertNum = getPdfNames.RecordCount>



<cfset arrCertNum = ArrayNew(1)>

<cfloop query="getPdfNames" startrow="1" endrow="#lstCertNum#" index="idx">
<cfset arrCertNum[idx] = ArrayAppend(arrCertNum,#name#)>
</cfloop>

> does not work, obviously I'm doing something wrong

Assuming you really need an array... the "index" attribute isn't allowed with a query loop.
So remove that. Also, arrayAppend doesn't return an array. It modifies the array in place.
Just use

<cfloop query="getPdfNames" startrow="1" endrow="#getPdfNames.recordCount#">
<cfset arrayAppend(arrCertNum, name)>
</cfloop>

> the actual error is Caused by: java.io.FileNotFoundException
> but when it come down to executing the following line ....
> it gives me that error message:

FileNotFoundException doesn't always mean the file doesn't exist. It can also mean the file can't be accessed. For example if the code opened certain types of file objects, but forget to close them, the file might be left in an inaccessible state.

> destpath = outdir & "\" & te.getName();

Since you already know the file names, you could also store them while inside your "unzip" loop instead.

As CFDirectory returns a query object, you can just do this:

<cfdirectory action="list" name="getPdfNames" directory="#unpacked#">
<cfset lstFiles = ValueList(getPdfNames.name)>

If you'd rather they were in an array, do:
<cfset arrFiles = ListToArray(ValueList(getPdfNames.name))>

> As CFDirectory returns a query object, you can just do this:

Yes, there are easier ways to get the results. Technically they don't even need to do a cfdirectory.
They could just store the file names as they're unzipping them. But any of the options mentioned
would work.

> <cfset arrFiles = ListToArray(ValueList(getPdfNames.name))>

It may not apply here, but be careful to select the right delimiter with list functions. The default is a ","
which isn't always safe choice. File names/paths can contain commas, which would cause the final
results to be wrong.

ASKER

The response time was woeful, by the time I got a response, I'd already solved the problem and moved on to the next one. I believe one of the main reasons for using this venue is to get answers relatively quickly which is as important if not more, of getting an answer. The whole purpose is defeated if that first part is not the case.