On Sat, 2004-10-30 at 14:12 -0400, Beartooth wrote: > On Fri, 29 Oct 2004 22:03:59 -0400, William Hooper wrote: > > > a) The yum.conf contains no "code", and doesn't require the ability to > > write "code" to edit it. If you have the ability to type > > http://fedora.redhat.com to get Fedora, you have the ability to put a > > URL in the yum.conf. > > All right, so I don't know code from gibberish; I'm not surprised. > All I really know is that that file is not in English -- nor any > other language I can read. But you're still overestimating me. (Thanks for > the compliment.) I can type that on either of two machines, and do a > download; but I've had miserable results trying to burn CDs -- and haven't > even dared try installing anything straight off the net. > > The first time I got Fedora, I was sent CDs by an electronic friend. The > second time I used ones out of a book I bought. > > I have yet to get more off the Net than updates in any way to be able to > use it. (I had FC1 running, after a lot of help; gave up on it when I > got a flat panel monitor; and got it again when that monitor showed up > in the list under Main Menu > System Settings > Display.) > > As for putting a URL into a .conf file : the only editor I know is pico; > until I learned about the -w switch (You do call it a switch, don't you?), > I couldn't preserve line length, indentation, etc., etc. I learned about > -w only by asking dumb questions in places like this. Beforehand, no, I > couldn't put a URL into a file -- not so that it would work. The yum configuration file is fairly simple, to look at, with a little bit of familiarity and deductive reasoning. You have on the first line [main] this depicts a group. so the lines below it are relevant to the group "main". In this case, the "group" controls how yum, in general works. Some entries in here are quite self explanitory, for example logfile=/var/log/yum.log This tells yum where to record it's log information. Other lines are FAR from obvious, for example tolerant=1, I havn't the foggiest what this means, on checking the man page, this doesn't tell me either. Only place to find out what that line means is here, or IRC or some where else you may gain the ear of the experienced. We soon hit another line that is as follows [base] As we had assertained with [main] we now have another "group". How do we know what this group is for? Looking down we see several of these entries, with things like "name", "baseurl" and some other entries. For me, it told me that we are now looking at "repositories", groups where packages can be obtained from. A simple repository will look like this [group name, internal to yum] name=Happy freindly name for user to see, when information is printed to the screen baseurl=<http/ftp/file location of "yum" files available to install> That is it. Both the information in [] and the "name=" is completely arbitrary, and has no "real" meaning other then information. The "baseurl" must have valid information, and that is the location of the yum files. To my knowledge, a "yum repository" must contain several peices of information. They are as follows (for the newest version of yum, slightly different to the one your using on FC2). 1. Header files: There is one of these for each rpm package, it contains certain information, I believe things like dependancy information. 2. RPM files: The actual packages to be installed 3. Several xml files who's names I care little to remember. These are there to speed up the use of yum. If your really curious some one here who knows more about them can tell you. This "should" give you enough to start writing your own yum.conf files (or in the case of the new yum, adding your own "repo configuration files" under /etc/yum.repos.d/<indevidual repo files>.repo Did that help at all? > > b) Downloading and using a random yum.conf is just as bad as running any > > random e-mail attachment on Windows. > > Well, it was hardly random, having been authored for this list by a > regular here. And it works. And hasn't broken anything. > > >> What would you test?? > > > > Speed of the repo, that it is reachable, etc. > > > >> I did use wget once -- with someone walking me through, step by step. > > > > wget it a very useful tool, and easy to use. Start with "man wget". > > I did. And despaired. Wget's man page is one of the worst: you have to > have an advanced CS degree just to be able to keep your place in it, > let alone find anything you can understand. > > Don't get me wrong: I'm sure such man pages are very useful to many here; > but after more time and effort spent on linux than most people I know > would dream of, I still can't begin to make any use of most. That's why I > label so many of my posts VDQ : Very Dumb Question. > > I run linux not only because it's not MS, but because I know virtually > anything *can* be tweaked to suit the way I use it -- *after* I've learned > to use it, if that's within my capability -- and someone online will > probably tell me how to do the tweak if I try to find it and don't, or > find it and don't understand it. > > How many languages would anyone here use, if before you learned any you > had to be able to compare and contrast the history of Balto-Fennic versus > Indo-European grammar and phonology? I can do that -- but I didn't get to > the ability by learning it *before* all the languages I know. Man pages > are great for those who were present at their creation, and have followed > along during their growth -- or so I presume, secondhand -- but many, such > as wget, are a barrier and not an aid to the uninstructed. I hear you here, some man files are absolutely horrible to try and understand, try learning regular expressions, or have a look at man awk or man sed. Trust me, man wget is a pussy cat ;) Hopefully, you have the feeling that you can come to the community and seek assistance, so that we can all get through it ;) -- Douglas Furlong <douglas.furlong@xxxxxxxxxxx>
