J_K9 @ Linux

HyperGet Has Changed

January 17, 2006 on 3:11 pm | In Linux, hyper-get |

The HyperGet project has changed. I haven’t had the time to update the official site yet, but I will as soon as I can.

The original design involved having two applications - one on the offline computer and another on the online one (as I described in an old post). But now, the design has changed radically. Having got in touch with Synaptic’s maintainer, Michael Vogt, we decided to work together to produce a more efficient and smaller HyperGet.

As of now, Xiro (the application for the online computer) will be the only one coded. The reason for this is that Michael has proposed that Synaptic could spit the URI’s needed for download onto the zip drive’s /hyper-get directory, and later copy the downloaded files to its cache and install them. So, here is a revised summary of usage:

1. User opens Synaptic on Ubuntu and selects the packages to be downloaded as per norm. Then, the user chooses to export their URI’s with the option “HyperGet”. Synaptic places the URI’s in a file named ‘uris.txt’ and copies that onto the specified zip drive, in a subdirectory labelled ‘hyper-get’. hyperget,jar is also copied into the hyper-get/ subdirectory.
2. The user unplugs the zip drive from the offline Ubuntu computer and takes it to an online computer - the Operating System running is not an issue. The user plugs in the drive and HyperGet is started up.¹
3. HyperGet finds the ‘uris.txt’ file and downloads the packages from the supplied URI’s. A status bar will show the progress of each individual file and the overall progress, and a message box will pop up to notify the user of completion once the files have all been downloaded. The packages will be placed within the ‘hyper-get’ subdirectory.
4. The user takes the zip drive back to the offline Ubuntu computer and plugs it in there. The user fires up Synaptic, and chooses to import packages with option “HyperGet”. Synaptic automatically finds the ‘hyper-get’ subdirectory on the zip drive and moves the packages to its cache. Synaptic verifies the md5sum’s to make sure they are correct and that the packages have not been interfered with, and then it installs them if everything is correct.

That is a run-down of what will happen. There is one tiny grey-spot though:
¹ - We still have not decided whether a user will download a “Bootstrapper” first, so that HyperGet is started as soon as the zip drive is plugged in, or whether the Bootstrapper should just be left out - because it will likely cause HyperGet’s cross-platform beauty to end. Or, the user could download prebuilt binaries for the online system. Or, the user could compile hyper-get.jar manually (following a README). This has still not been decided, but we will come to a conclusion soon - and I shall let you know what will happen.

Coding is currently going well, and certain modules of the application have already seen major progress. However, we are hoping for a release by 23^rd Feb. because that’s when Ubuntu 6.04 Dapper’s feature freeze date is. So: code and pray people, code and pray