Earliest ver supporting DIRECTORY/SIZE=UNITS=BYTES

[Jon Pevzner]

Hi All,

Can someone please provide this info for me?

I have a VB FTP app that needs to retrieve the filesize in bytes from many different VMS versions.

Thanks,
Jon

[John Abbott_2 · Accepted solution]

Hi Jon,

Alpha VMS V7.3-2 see http://h71000.www7.hp.com/doc/732FINAL/aa-rv8xa-te/aa-rv8xa-te.PDF - section 2.

[John Abbott_2]

Jon,

Not sure what you're trying to do, but check out the vmsfaq http://h71000.www7.hp.com/faq/vmsfaq.pdf
for additional information. You might want to consider just accepting the value in blocks and converting it into bytes (1 block = 512 bytes) if you really need to. Then there's used space vs. allocated (cluster size affect) and of course CD media. Again faq is a good read.

John.
PS. Welcome to the OpenVMS forum!

[Jon Pevzner]

Thanks, John

The links are very helpful.

My app was designed to FTP files periodically from Unix servers. Because the files are always the same name (they get ovewritten daily) I was comparing file size to see if the file is needed. Blocks is not really specific enough as many of these files are really close in size. My best bet is just to ignore the size and download it anyway, that way nothing will be missed.

Tahnks again for your help.

Regards,
Jon

[Steven Schweda]

Why not look at the date/time instead of the
size? For example, Wget's "-N" option, as I
recall, tries to look at both date/time and
size (when it can get a reliable size).
From "wget -h":

-N, --timestamping
don't re-retrieve files unless newer than local.

Besides, UNITS=BYTES will only be converting
blocks to (K/M/G) bytes, so you still won't
be getting a byte-accurate size for a file
that way. VMS does not maintain
byte-accurate file sizes.

[Jon Pevzner]

Hi Steven,

I didn't know that it was only going to convert to the block's byte size... kinda useless, it seems.

Anyway, reconfiguring to use date/time would be too much programming, so I decided to let it download the files (there are at most 20 and not very large), and later in the routine read the file to compare to existing files and save it if necessary. It only added a few seconds to the process and works just fine.

Thanks for the info,
Jon