|
When reading Usenet news on the OIT machines (with 'rn',
'trn', or 'tin), you will find some groups which contain
images or sound files. Since these are binary files and
usually large, they are dealt with in a special way on
Usenet.
For starters, the binary format must be translated into a
non-binary format. This is usually done via the encoding
scheme called "" the file. After this step, the file is then
evaluated for size. There is an unwritten rule to limit the
size of your posts on Usenet to 1000 bytes or less per post,
so large files are usually split across several posts. They
will usually be labeled as " 1 of 3", " 2 of 3", or
ŕ/3", Ŗ/3", etc. so you know in what order to
reassemble the parts.
You can do this several ways, from the tedious and
time-consuming, individual by-hand treatment to the fully
automated way. This depends upon how the file is posted and
what newsreader you're using. Most posts on Usenet are merely
multi-part uuencoded files, or multi-part shell scripts, each
part containing one of the uuencoded parts. The rules for
extracting these are similar. Other posts might be poorly
done or using a different technique, so may have to be dealt
with by hand.
Using 'rn' and 'trn':
When using the 'rn' (ReadNews) or 'trn' (Threaded ReadNews)
newsreaders, you can use the 'extract' function to do most of
the work. At the " of article, what next" prompt, you simply
type the following:
<beginning-of-range>-<end-of-range>:e directory
Say, for example, the file you wish to extract is in Usenet
posts numbers 110, 111, 112 and 113. (This is assuming a
four-part file, with all the parts in the correct order.) You
want the resulting file to end up in the /tmp directory,
because you don't know if your account will hold the whole
file. You would then issue the command:
110-113:e /tmp
If the file is posted as a series of uuencoded sections or
these sections wrapped in a shell script, the 'extract'
function will handle it properly. You'll see a message as
each news post is dealt with, and when done you'll be back at
the " of article" prompt. At this point, the file should be
in the directory you specified, and you can examine it, view
it, play it, copy it, whatever you want.
Many times, the posts will arrive at your site out of order.
(This is due to the nature of the Internet, where there is no
single path from point A to point B.) Say, for example, they
showed up in this order:
110 Part 3/4
111 Part 1/4
112 Part 2/4
113 Part 4/4
If you decoded this using the range 110-113, it would be
assembled out of order, and the resulting file would be
worthless. You can specify any order of news items when you
extract; the command:
111,112,110,113:e /tmp
will extract them in the correct order and leave the
resulting file in /tmp.
Using 'tin':
When you first get into 'tin' and into a specific newsgroup,
you'll first see a list of the articles and subjects. You
need to "" the article(s) you wish to save/extract; to do so,
select each one in turn and press the "" key. You should see
a number appear next to each tagged item, starting with
ŕ", then Ŗ" and so forth. When you've tagged the
articles in the correct order, then choose "" for save. The
next prompt at the bottom of the screen will be:
Save a)rticle, t)hread, h)ot, p)attern, T)agged articles,
q)uit: T
The "" at the end is the default choice, only to deal with
tagged articles. Either type "" here and press Enter, or just
press Enter if the "" is already in place. The next prompt
you'll see will be for the filename you wish to create from
the tagged news items:
Save filename [test.save]>
The name "" is the default used if no other name is
specified. The next prompt will be how to process the tagged
items:
Process n)one, s)har, u)ud, l)ist zoo, e)xt zoo, L)ist
zip, E)xt zip, q)uit: n
This is the prompt where you tell it what you want to do with
the tagged items. If it's a simple series of uuencoded items,
choose "" to uudecode them; if it's a shell script, type ""
to run the script; if it's a zoo archive, you can use ""
(list contents) or "" (extract contents); if it's a zip
archive, you can use "" (list contents) or "" (extract
contents). If the tagged items are merely a series of text
posts, you can choose "" for no processing; this is the
default action. Once you choose the action, the newsreader
will extract the items using the action chosen. You may get
the prompt:
Delete saved files that have been post processed? (y/n): y
which means it's asking you if you want it to remove the
intermediate files it created in the process of creating the
final file. You will usually answer "" to this question.
After this point, you'll be placed back at the list of
articles, and you can go on to read or decode more items.
When you save/uudecode a file, it will be placed in whatever
directory you specify. By default, it will end up in the
directory where you invoked your newsreader, so it's probably
in your home directory. You can specify this explicitly when
giving the filename to save to; when you get the prompt:
Save filename [test.save]>
you can specify the tilde ("~"):
Save filename [test.save]> ~/saved.file
which will put it in your home directory, or you can do
something like:
Save filename [test.save]> /tmp/saved.file
if you want to place it somewhere which will not impact your
quota. (This is useful for intermediate processing of files
if you're close to your quota limit - copy the final versions
of files back to your directory space when you've verified
that it's what you want. Keep in mind that any files in the
/tmp directory will be erased, probably overnight.)
You can probably find public-domain utilities in the
newsgroups:
alt.binaries.pictures.utilities and/or
alt.graphics.pixutils
Look for a FAQ (Frequently Asked Questions) posting, which
will probably have utilities broken down by platform, or tell
you where to get such a list.
|
