Re: [Bug-wget] Development Ideas - Request 2 New Features

[Emm]

Hi, I too was looking for such a feature... although Feature 1 isn't
really necessary as we can already do that by using the --header
option and I would prefer to use it this way instead of a specific
option because it allows for flexibility, by using either of the range
syntaxes present in the HTTP specifications. Not only that, but let's
say that the web server doesn't support properly the method
implemented in Wget... that would leave only the --header option as a
possible solution.

The second feature is nice... and that would be really useful. So I
too vote for it.

However, by using the header = Range: option... there's a small
problem which I have been able to work around, but not ideally.

I'm used to download from megashares... which sometimes offers me
250MB download per sessions, while other times 500MB. And mostly I
download some shows that don't air properly in my country... it's sad,
but at least the internet offers some solutions, with each episode
having around 1.09 GB. The thing is if I use a proxy and download in
parallel, by using the option header = Range:
bytes=524288000-1048576000 I need to have a file of 500 MB already
present to be able to write to it... and my solution was to make an
empty 500MB file and "continue" the download, after which I would
truncate the file and leave only the useful data. If I attempt to
download to an empty file, Wget just stops. I guess Wget checks if a
file is already present and what size it is for the range specified
for the 206 response and downloads only then.

Basically, I would like to be able to tell Wget, download from bytes x
to y to a new file, suppressing the file verification parts... through
an option maybe or by internally verifying if the range header was set
by the user, implying it's specifically meant to do that, thus
allowing the download.