=head1 NAME urpmi.addmedia - adds a new rpm medium to be used by urpmi =head1 SYNOPSIS urpmi.addmedia [options] [with ] =head1 DESCRIPTION urpmi.addmedia is used to add a new media to be used by urpmi, so it can find new rpms at the specified location. Currently supported media types are: local drives, removable drives (such as CDs), and networked media via different protocols (http, ftp, ssh and rsync). One of the following modes must be chosen: =over =item B [I] > http://>:>@>/> [with >] where > is your reference for the medium (e.g. "updates"). > is optional if B<--distrib> is given in the options list. >/> is the location of the media directory on the net (e.g. C). > is the pathname where to find the hdlist of the rpms. The location is given relative to >. B is used by default to download files, B may be used if B is not installed or if or B<--wget> is given in I. Example: urpmi.addmedia http http://jpackage.sf.net/rpm/free with hdlist.cz ftp: =item B [I] > ftp://>:>@>/> [with >] The same as for http. Add your login and password if required (note that you don't need to specify B as login for anonymous access ftp servers). If B is used to download files with a firewall you may have to ensure that the B option is on in B (that's usually the default). Example: urpmi.addmedia ftp ftp://a:a@leia//export/media/main with media_info/hdlist.cz =item B [I] > ssh://>@>/> [with >] The same as for http and ftp; add your login and password if required. It is strongly recommended to export the public key in order not to have to type your password all the time. B and B will be used to get files. Example: urpmi.addmedia -h ssh ssh://fpons@bi/c/i586/media/main =item B [I] > rsync://>@>:>/> [with >] The same as for http; you can use the >::> syntax too. Example: urpmi.addmedia rsync rsync://ftp.orst.edu::mandrake-devel/contrib/ppc with ../../cooker/ppc/media/media_info/hdlist2.cz =item B [I] > file://> [with >] where > is your reference for the medium (e.g. Myrpms). > is the location of the media directory on your machine (e.g. C). =item B [I] > removable://> [with >] > is the location of the media directory on your machine when the removable device is mounted. This works for removable drives, but also for ISO images mounted on a loop device. urpmi will then be able to re-mount the ISO image to the specified path (but the ISO image must be mounted when you run C). =back =head1 OPTIONS =over =item B<--wget> Use wget only for downloading distant files. By default curl is used if available. =item B<--curl> Use curl only for downloading distant files. This is the default if curl is available. =item B<--curl-options> I<'options'> =item B<--rsync-options> I<'options'> =item B<--wget-options> I<'options'> Specify additional command-line options to be passed to curl, rsync or wget when retrieving files. If several options are to be passed, separate them with spaces and enclose them in quotes. =item B<--limit-rate I> Try to limit the download speed, I is given in bytes/sec. This option is not active by default. =item B<--proxy> I Use specified HTTP proxy. =item B<--proxy-user> I Use specified user and password to use for proxy authentication. Specifying B<--proxy-user=ask> will cause C to prompt for a username and a password. =item B<--update> Add a medium which will be taken into account by B or by C when looking for updates. =item B<--probe-synthesis> Try to find and use synthesis file if B is not given. =item B<--probe-hdlist> Try to find and use hdlist file if B is not given. =item B<--no-probe> Do not try to find any synthesis or hdlist file. So, if you use a network media (http, ftp, etc.), specifying the location of a synthesis or hdlist with B is mandatory. On the other hand, if you use a local media C will extract all RPM headers and construct an hdlist by itself. =item B<--distrib> Retrieve a set of media from a distribution. Typically, the URL provided to C will represent the parent directory of a directory B, which in turn will contain various subdirectories for each medium of the distribution. > is combined with medium names found to create newer medium names in the urpmi database. The B option, is not necessary, as hdlists found on the installation medium are used instead automatically. =item B<--from> I Use specified URL for list of mirrors. The default is taken from the file F, or F if this file doesn't exist. =item B<--virtual> Creates a virtual medium. This only works for C and C protocols, and no files will be created in F. The medium is always up-to-date and so it does not need to be updated by C. =item B<--norebuild> Don't try to rebuild the hdlist files from the RPMs if the original hdlist wasn't readable or was corrupted. =item B<--raw> Add the new media in the urpmi configuration file, but don't update it nor proceed to any download. The media, to be usable, will need to be updated with C. =item B<--nopubkey> Don't import pubkey of added media. =item B<-c> Clean headers cache directory in F. =item B<-h> Try to find and use synthesis or hdlist file. =item B<-f> Force rebuild of base files, or of hdlist files when specified twice. =back =head1 Variables Beginning with urpmi 4.6.16, you can use variables in media URLs (for example F). The variables supported so far are: =over =item B<$ARCH> The architecture (as given in F) =item B<$RELEASE> The OS release (as given in F; its value is B on a cooker system.) =item B<$HOST> The canonical hostname of the machine urpmi runs on. =back =head1 SEE ALSO urpmi(8), urpmi.update(8), urpmi.removemedia(8), urpmf(8), urpmq(8), urpmi.files(5). =head1 Author Pascal Rigaux (original author), FranEois Pons, Rafael Garcia-Suarez, (current maintainer)