On Thu, Dec 01, 2016 at 09:39:42AM -0500, Aaron M. Ucko wrote:
> Eriberto <eribe...@eriberto.pro.br> writes:
> > There are three problems here.
> Thanks for taking a look!
You are welcome. :-)
> > 1. The current FTP address is
> > ftp://ftp.ncbi.nih.gov/toolbox/ncbi_tools/old/, not
> > ftp://ftp.ncbi.nlm.nih.gov/toolbox/ncbi_tools/old.
> Hm? Either name is valid, and I'd say the variant with .nlm is
Doesn't matter anyway - see below.
> $ dig @ns.nih.gov ftp.ncbi.nih.gov CNAME
> ftp.ncbi.nih.gov. 86400 IN CNAME ftp.ncbi.nlm.nih.gov.
> > 2. Now, this FTP site only accepts passive mode.
> That shouldn't be relevant if we're switching to http(s) anyway.
> > 3. The FTP scan via uscan seems accept files only (not directory
> > names). However, this site has http too.
> Good point. Better yet, it has https.
Yes - but at least we do now have a working watch file that enabled me
downloading the new upstream source. As I wrote in another mail I
failed injecting it into the packaging Git. I admit I'd really
appreciate if you could take over from the current state of the Git
repository since I consider you way more competent than me for the job.