Partially decode (-p)
Brought to you by:
liyang
Menu
▾
▴
#4 Partially decode (-p)
A switch to partially decode a segment of a multi-part
encoding. This should work as follows.
If -p as set, then yydecode should NOT attempt to
assemble the files, but instead decode to a part-file.
The part-file would be the filename with part<x>
appended. If there is only 1 part in the encoding, then
the file is written like normal.
yydecode -p FILES
would yield:
file1.part1
file1.part3
file2
file3.part4
...
These files can then - at leisure - be combined by
concatenating the files in the correct order manually.
In the above, file2 is 1 file encoded in a single
encoded file.
The switch should also work with the proposed -a
(auto-rename) switch, so that if eg. file1.part1
exists, then it would make a file1.part1.<counter> file
instead.
This would allow partially decoding a set of files,
then assembling them when the missing parts arrive.
The part-files should NOT be filled with the missing
previous parts as is the case when doing normal decode
of a single part.
Logged In: YES
user_id=36015
This is /not/ going to be trivial to implement, if I decide
to do it at all... I'm just not convinced it's all that
useful given the difficulty/annoyingness of implementation!
The existing design is ill-suited to do what you're after,
I'm afraid, so it's either very ugly hacks or a complete
overhaul. (Neither of which I'm willing to undertake at the
minute...)
Why can't you keep the encoded parts you have as is, wait
for the rest to come through and decode the whole thing at once?
If you have the encoded parts as separate files (i.e. one
file per article posted), then you can pass them through
yydecode one by one, so it doesn't get any chance to
assemble them. You'll need to use yydecode as a filter, to
prevent it from seeking to the `correct' place in the output
file, thus:
yydecode < encoded.part4 > decoded.part4
I'm sure this can all be automated with a simple shell script...
(Patches would be accepted though. ^_-)