NAME
uudeview - a powerful decoder for binary files
SYNOPSIS
uudeview [-i] [-d] [-f] [-o] [-b1] [-t] [(+e|-e) extlist]
[-v] [-s] [-m] [-p path] [@file] file(s)
DESCRIPTION
uudeview decodes files that you have received in encoded
form via electronic mail or from the usenet, similar to
the standard uudecode(1) command, yet with more comfort
and flexibility. uudeview supports the uuencoding, xxen-
coding, Base64 and BinHex encoding methods, and is able to
handle split-files (which have been sent in multiple
parts) as well as multiple files at once, thus greatly
simplifying the decoding process. Usually, you will not
have to manually edit files to prepare them for decoding.
After invoking uudeview, it will scan all given files for
encoded data, sort them and their parts and then present
you with the list of files that seem like they can be
decoded properly. You can then pick files individually for
decoding.
OPTIONS
-i Disables interactivity. After scanning the files
and sorting everything out, the program will not
ask for whether a file shall be decoded or not, but
immediately batch-decodes everything possible.
-d Sets the program into desperate mode. It will then
offer you to decode incomplete files. This is use-
ful if you are missing the last part of a 50-parts
posting, but in most cases the desperately-decoded
files will simply be corrupt and unusable. The
degree of usefulness of an incomplete file depends
on the file type.
-f Uses fast mode for file scanning. The program
assumes that each input file holds at most one
part, which is usually true for files in a news
spool directory. This option breaks decoding of
input files with multiple articles. Also, certain
sanity checks are disabled, probably causing erro-
neous files to be presented for decoding. Some-
times you'll get error messages when decoding,
sometimes you'll just receive invalid files. Don't
use -f if you can't live with these problems.
-o Gives the OK to overwrite existing files when
decoding. The default is to prompt the user whether
to overwrite, rename or skip the file.
-v Disables verbosity. Normally, the program prints
some status messages while reading the input files,
which can be very helpful if something should go
wrong. Use if these messages disturb you.
-p path
Sets the path where decoded files shall be written
to. This must be a valid pathname, or you'll get
errors when trying to decode anything. Defaults to
the current working directory.
+e exts
Selects only the files with the given extensions
for decoding, others will be ignored. +e .gif.jpg
would decode all gif and jpeg files, but not tif or
other files. The list of extensions works case-
insensitive.
-e exts
The reverse of the above.
You will experience unwanted results if you try to mix +e
and -e options on the command line.
-b1 This changes uudeview's policy of finding a part
number on a subject line and may only be needed in
some rare cases when part numbers are found in ()
parentheses as well as in [] brackets, for example
in a series of multi-part postings. By default,
uudeview uses the numbers found in () parentheses
first. But if this number indicates the file's num-
ber in the series and the part number is given in
[] brackets, use this parameters to make the pro-
gram read the other number first. This does not
affect decoding of files with only one or neither
type of brackets. If you prefer, you can also use
the option as -b[]
-s Read "minus smartness". This option turns off auto-
matic part number detection from the subject line.
Try this option if uudeview fails to parse the sub-
ject line correctly and makes errors at guessing
part numbers, resulting in incorrect ordering of
the parts. With this option, parts are always put
together sequentially (so the parts must be cor-
rectly ordered in the input file). Also, with this
option, the program cannot detect that parts are
missing. Note: The correct part number found in
proper MIME files is still evaluated.
-m Ignore file mode. uuencoded and xxencoded files
have the original file permissions stored on the
begin line. If this option is not given, the
decoder tries to restore them. With this option,
the resulting permissions will always be 0666 minus
your umask.
-t Use plaintext messages. Usually, uudeview only pre-
sents encoded data for decoding. With this option
set, text parts from MIME messages and non-encoded
messages are also offered. Plaintext messages fre-
quently don't have an associated filename, so
they're assigned a unique name from a sequential
four-digit number.
file(s)
The files to be scanned for encoded files. You can
also give a single hyphen '-' to read from standard
input. Any number of files may be given, but there
is usually a limitation of 128 options imposed by
the shell. If you are composing the list of files
with wildcards, make sure you don't accidentally
feed the program with binary files. This will
result in undefined behaviour.
@file Makes uudeview read further options from the file.
Each line of the file must hold exactly one option.
The file is erased after the program finishes. This
feature may be used to specify an unlimited number
of files to be scanned. Combined with the powers of
find(1), entire directory trees (like the news
spool directory) can be processed.
DECODING
After all input files have been scanned, you are asked for
each file what do do with it. Of course, the usual answer
is to decode it, but there are other possibilities. You
can use the following commands (each command is a single
letter):
d (D)ecode the file and write the decoded file to
disk, with the given name.
y (Y)es does the same as (d).
x E(x)tract also decodes the file.
n Skips this file without decoding it.
b Steps back to the previous file.
i Displays info about the file, if present. If a mul-
tipart posting had a zeroeth part, it is printed,
otherwise the first part up to the encoded data is
printed.
e Execute a command. You can enter any arbitrary com-
mand, possibly using the current file as an argu-
ment. All dollar signs '$' in this command line are
replaced with the filename of the current file
(speaking correctly, the name of a temporary file).
You should not background processes using this tem-
porary file, as programs might get confused if
their input file suddenly disappears.
l List a file. Use this command only if you know that
the file in question is a textfile, otherwise,
you'll get a load of junk.
r Rename. You can choose a different name for the
file in order to save it under this new name.
p Set the path where decoded files shall be written
to. This path can also be set with the -p command
line option.
q Quits the program immediately.
? Prints a short description of all these commands.
If you don't enter a command and simply hit return at the
prompt, the default command, decoding the file, is used.
RUNTIME MESSGAGES
In verbose mode (that is, if you didn't disable verbosity
with the -v option), progress messages will appear. They
are extremely helpful in tracing what the program does,
and can be used to figure out the reason why files cannot
be decoded, if you understand them. This section explains
how to interpret them. Understanding this section is not
essential to operate the program.
First, there are "Loading" messages, which begin with the
string "Loaded". Each line should feature the following
items:
Source File
The first item is the source file from which a part
was loaded. Many parts can be detected within a
single file.
Subject Line
The complete subject is reproduced in single
quotes.
Identifier
The program derives a unique identification for
this thread from the subject line, for grouping
articles that look like they belong to the same
file. The result of this algorithm is presented in
braces.
Filename
If a filename was detected on the subject line or
within the data (for example, on a begin line, or
as part of the Content-Type information).
Part Number
The part number derived from the subject line, or,
in the case of properly MIME-formatted messages,
from the "part" information.
Begin/End
If a "begin" or "end" token was detected, it is
printed here.
Encoding Type
If encoded data was detected within this part,
either "UUdata", "Base64", "XXdata" or "Binhex" is
printed here.
More messages are printed after scanning has completed. A
single line will be printed for each group of articles.
The contents of this line are best understood by looking
at an example. Here is one:
Found 'mailfile.gz' State 16 UUData Parts begin 1 2 3 4 5
end 6 OK
This indicates that the file mailfile.gz has been found.
The file was uuencoded ("UUData") and consists of 6 parts.
The "begin" token was found in the first part, and the
"end" token was found in the sixth part. Because it looks
like everything's there, this file is tagged as being
"OK". The State is a set of bits, where the following val-
ues may be or'ed:
1 Missing Part
2 No Begin
4 No End
8 No encoded data found.
16 File looks Ok
32 An error occured during decoding of the file.
64 File was successfully decoded.
NOTES
Because the program cannot receive terminal input when
reading from standard input, the interactivity is automat-
ically disabled in this case.
When MIME-style message headers are detected, the program
behaves nearly MIME-compliant. Nearly, because the stan-
dard does not allow a file to hold more than one messages,
but uudeview works without this restrictions. Actually, if
you guarantee this condition using the -f command line
switch, the program fully complies to RFC1521. (Even with
this switch set, all parts of a proper MIME multipart mes-
sage are handled.)
The scanner tends to ignore short Base64 data (less than
four lines) outside of MIME messages. Some checks for this
condition are used in desperate mode, but they may cause
misdetection of encoded data, resulting in some invalid
files.
Files are always decoded into a temporary file first, then
this file is copied to the final location. This is to pre-
vent accidentally overwriting existing files with data
that turns out too late to be undecodeable. Thus be care-
ful to have twice the necessary space available. Also,
when reading from standard input, all the data is dumped
to a temporary file before starting the usual scanning
process on that file.
uudeview tries to derive all necessary information from
the Subject: line if present. If it holds garbage, or if
the program fails to find a unique identification and the
part number there, uudeview might still be able to decode
the file using other heuristics, but you'll need major
luck then.
Yet this is only a concern with split-files. If all
encoded files only consist of single parts, don't worry.
If you rename, copy or link the program to uudecode, it
may act as a smart replacement for the standard, accepting
the same command-line options. This has not been well-
tested yet.
SEE ALSO
uuenview(1), uudecode(1), uuencode(1),
The uudeview homepage on the Web,
http://www.informatik.uni-frankfurt.de/~fp/uudeview/
BUGS
Reading from a file whose name starts with a hyphen '-' is
impossible.
Before reporting a bug, make sure the file can be decoded
by other means. I hate to receive bug-reports where it
turns out that uudeview just failed to decode complete
garbage.
If you think you've found a bug, email the source file (at
best, compress and encode the original file, don't just
include it) and a listing of the program's messages (from
verbose mode) to fp@informatik.uni-frankfurt.de.
The checksums found in BinHex data are currently ignored.
The program cannot fully handle partial multipart messages
(MIME-style multipart messages split over several mail
messages). The individual parts are recognized and con-
catenated, and the embedded multipart message is "decoded"
into a plain-text file, which must then be fed again to
uudeview. Don't worry, these kinds of messages are rare.