savannah-hackers
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[savannah-help-public] Gnnuspeech release for savannah.gnu.org/projects/


From: David Hill
Subject: [savannah-help-public] Gnnuspeech release for savannah.gnu.org/projects/gnuspeech
Date: Sun, 20 Sep 2015 19:03:27 -0700

Dear people,

I am preparing the Gnuspeech release and have read the Information for "Maintainers of GNU Software: Automated Upload Procedures" (http://www.gnu.org/prep/maintain/html_node/Automated-Upload-Registration.html#Automated-Upload-Registration, etc).

I have registered.

I just want to check the subsequent procedures.

I intend, very soon, to upload a defined triplet for a release, suitably signed etc, which is fine. I already have made one .directive.asc file and can make different/additional ones if necessary (see below). Making the other two files (.sig and the main material) for the triplet should be no problem, but I still have some questions

I presume I do not have to use "gnupload" or "ncftpput-ftp" scripts, but can simply create and name the three triplet files, including the main .tar.gz file containing the material for distribution, using any suitable tools (e.g. the git archive tool), provided they all have the correct form of file-name; and that I can then use basic anonymous ftp to upload the triplet -- the main, the .sig file and the directive.asc file -- to ftp.gnu.org. The instructions say says "At the same time". That means one after the other within a very short time -- right?

You don't expect uploaders to use sftp -- right?

The main file should contain the directory structure similar to the repo, packed and compressed. It should unpack to be a non-bare repo clone -- right?

I understand that it is OK to put (a) binary version(s) in the distribution .tar.bz file, so it wouldn't be identical to the repo on the savannah site, which does not have binaries -- but see the next paragraph.

Then there's the fact that we have two repos under the gnuspeech project umbrella: gnuspeech and gnuspeechsa. As this is partly to avoid people having to download the whole of gnuspeech as well as gnuspeechsa if only gnspeechsa is wanted (and vice-versa), should I be uploading two triplets, one for each repo (sub-project). And should I put the binaries in a third triplet rather than as part of the two sets of source & manual material? This would perhaps raise naming problem for the binaries. If separation is the preferred straregy, what two names would be appropriate for the two seats of binaries.

When the release has been accepted, I should send a note out to the mailing list telling people where to find it, with perhaps some other comments. Any suggestions as to what to include? And as a matter of interest, where will people find the release?

I think that covers the things I'd like to know before doing the release. Any comments/advice/corrections/additions would be welcome.

Many thanks for your help.

All good wishes.

david

---------


reply via email to

[Prev in Thread] Current Thread [Next in Thread]