How to separate this

[timle]

This is the content of "dirsize.txt"
-rwxrwxrwx 1 owner group 563682 Sep 19 11:22 filename1.pdf
-rwxrwxrwx 1 owner group 729569 Sep 19 12:40 filename2.pdf

How do I use separate to get the "filename.pdf part to use for my script

thank you


ReadFile>%dir1%\dirsize.txt,data
Separate>data,Dates,filename

[Me_again]

Something like this should work for you:

Code: Select all

ReadFile>c:\temp\dirsize.txt,dirdata
//First separate the file into lines
Separate>dirdata,%CRLF%,dirfiles
If>dirfiles_count=0,end
//Loop through the lines
Let>k=0
Repeat>k
Let>k=k+1
//Then separate the line using a space for separator
Separate>dirfiles_%k%, ,filenames
//The file name is the 9th data element
MDL>filenames_9
Until>k,dirfiles_count
Label>end

[timle]

-rwxrwxrwx 1 owner group 12456361 Sep 19 11:22 filename1.pdf
-rwxrwxrwx 1 owner group 729569 Sep 19 12:40 filename2.pdf

The result is not always at the 9th data element, if there are inconsistent in spacing in the 5th element.
is there a way to delete all the extra space and replace with just one space before we separate them out,
(it did not show in my post, but the real file has more space between the word "group" and the numbers)

thanks

[Marcus Tettmar]

Filename is always the last item. Therefore:

Code: Select all

Let>str=-rwxrwxrwx 1 owner group   563682 Sep 19 11:22   filename1.pdf
Separate>str,SPACE,parts
Let>filename=parts_%parts_count%
MessageModal>filename

[Bob Hansen]

Nice solution Marcus.

Here is another option:
Let>str=-rwxrwxrwx 1 owner group 563682 Sep 19 11:22 filename1.pdf
Position>filename1.pdf,%str%,1,StartPos
MidStr>%str%,%StartPos%,20,filename
MessageModal>%filename%

But your solution is much better, do not need to know the name of the filename.

[timle]

but if I also wants to get the "byte" after the "group"
how can I do that

thanks

[Me_again]

If you know that the problem is double spaces just use StringReplace> to replace any double spaces with single spaces.

[Bob Hansen]

Use the code from Marcus, with a simple change:

Let>str=-rwxrwxrwx 1 owner group 563682 Sep 19 11:22 filename1.pdf
Separate>str,SPACE,parts
Let>groupID=%parts_5%
MessageModal>%groupID%

Check out the Help section for the Separate command to understand how the segments are numbered and counted as variables.

[jpuziano]

Filename is always the last item. Therefore:

Code: Select all

Let>str=-rwxrwxrwx 1 owner group   563682 Sep 19 11:22   filename1.pdf
Separate>str,SPACE,parts
Let>filename=parts_%parts_count%
MessageModal>filename

There's a problem. If there are spaces within the filenames like this...

• -rwxrwxrwx 1 owner group 563682 Sep 19 11:22 file name 1.pdf
  -rwxrwxrwx 1 owner group 729569 Sep 19 12:40 file name 2.pdf

...then the above code would return "1.pdf" and "2.pdf" instead of the full filename.

Here's one solution...

Since there will never be a colon character : in the filename, we can break up the line using : as the delimiter instead of SPACE. The right-most piece will always start with the colon from the time value... like this for example:

• :22 file name 1.pdf

So now all we have to do is throw away the first few characters... and what's left is the filename. Here's the code:

Code: Select all

Let>str=-rwxrwxrwx 1 owner group   563682 Sep 19 11:22   file name 1.pdf
Separate>str,:,parts
Let>filename=parts_%parts_count%
MidStr>filename,6,1000,just_the_filename
Message>just_the_filename

Note I didn't bother to get the length of the string for use in the MidStr> command, I just plunked in 1000. This is assuming any filename ever encountered will be less than 1000 chars long which I'd say is a safe bet... and the command doesn't complain.

Now as for parsing the "bytes after the group", here's one solution...

Code: Select all

Let>str=-rwxrwxrwx 1 owner group   563682 Sep 19 11:22   file name 1.pdf
Separate>str,group   ,parts
Let>line_starting_with_bytes=parts_%parts_count%
Separate>line_starting_with_bytes,SPACE,parts
Let>filename=parts_1
MessageModal>filename

Note in the first Separate> command, there are three spaces after group... so your delimiter is "group ". That saves you from having to Trim the result string... i.e. it will start with the actual bytes value. It also assumes all your lines will have three spaces between the end of "group" and the beginning of the bytes value. As long as that is always true, this will alway work.

After that, you just separate again, using SPACE as the delimiter only this time, what you are after is the first piece... not the last. Try it out, see if it works for you and let us know.

Can anyone find a shorter solution to the above?

P.S. Thanks Marcus for the Blog post on MS version 11 - much appreciated and looking forward to the beta!

[Bob Hansen]

Just an idea, no code here yet.....

Two methods of dealing with the spaces:
1. Use VB RegEx to replace multiple spaces with a single space, will take care of any number of spaces between strings
OR
2. Use multiple lines of StringReplace, replacing double spaces with a single space. Maybe three lines in a row to take care of four or five spaces in a row.

Then use code from Marcus

[jpuziano]

Just an idea, no code here yet.....

Two methods of dealing with the spaces:
1. Use VB RegEx to replace multiple spaces with a single space, will take care of any number of spaces between strings
OR
2. Use multiple lines of StringReplace, replacing double spaces with a single space. Maybe three lines in a row to take care of four or five spaces in a row.

Then use code from Marcus

Hi Bob,

What would happen if there were three spaces in a row within the filename... like this for example:

file^^^name.txt

...where each ^ represents a space

If you do either 1 or 2 above, you would be altering the actual filename and perhaps not able to parse out the real filename which is "file^^^name.txt" (contains 3 spaces) not "file^name.txt" (contains only 1 space).

The code I posted also works for filenames that contain multiple spaces in a row:

Code: Select all

Let>str=-rwxrwxrwx 1 owner group   563682 Sep 19 11:22   filenamecontains   3spaces.pdf
Separate>str,:,parts
Let>filename=parts_%parts_count%
MidStr>filename,6,1000,just_the_filename
Message>just_the_filename

[Marcus Tettmar]

Regex to get the bytes:

Code: Select all

//DOES THE REGEX
VBSTART
Function regExSearch(strPattern,str)
  Set regEx = New RegExp ' Create regular expression.
  regEx.Pattern = strPattern ' Set pattern.
  regEx.IgnoreCase = True ' Make case insensitive. Default=False
  Set matches = RegEx.Execute(str)
  List = ""
  For each match in matches
  	 List = List & match.value & ";"
  Next
  if List <> "" then
    regExSearch = Mid(List,1,Len(List)-1)
  end if
End Function
VBEND

Let>line=-rwxrwxrwx 1 owner group 563682 Sep 19 11:22 filename1.pdf
VBEval>regExSearch("group[\x20]*([0-9]{1,})\b","%line%"),bytes
StringReplace>bytes,group,,bytes
StringReplace>bytes, ,,bytes
MDL>bytes

This of course is a *nix directory listing. How is it being produced, because if you have control over this you could specify the ls parms to return the list in a different format that may not require parsing. E.g. the "ls" command can be set to only return the filenames. If this is from an FTP listing, FTPGetDirList can also be set to return only filenames.

[Bob Hansen]

For jpuziano

Hi Bob,

What would happen if there were three spaces in a row within the filename... like this for example:

You are right if the filename might be split. But I did not see any indication from timle, or his samples, that "filename" would sometimes have a space character in the middle. My assumption was that filename.ext would not be split. Thanks for pointing out what would happen with a bad assumption. But, if the assumption is good, I think the multiple lines would work OK.

[timle]

I used tips from everyone to make a script to check the ftp to see if any files coming in and when it stop growing then download the file to local drive then remove it from the FTP
Thanks for all your help