| 1 |
On 26/04/2015 10:17, Neil Bothwick wrote: |
| 2 |
> On Sun, 26 Apr 2015 09:35:57 +0200, Alan McKinnon wrote: |
| 3 |
> |
| 4 |
>> <x11-libs/libXfont-1.5.0 required by |
| 5 |
>> (x11-base/xorg-server-1.12.4-r4:0/1.12.4::gentoo, installed) |
| 6 |
>> |
| 7 |
>> This means that you have xorg-server-1.12.4-r4 installed which depends |
| 8 |
>> on libXfont with this limitation: <x11-libs/libXfont-1.5.0 |
| 9 |
>> |
| 10 |
>> You only get that sort of emerge output when portage is forced to |
| 11 |
>> install a package that is NOT latest due to some other package having a |
| 12 |
>> constraint on it dependencies. Look at eix for libXfont, there's a |
| 13 |
>> version 1.5.1 available but portage can't use it because of limitations |
| 14 |
>> from your current version of xorg-server. |
| 15 |
>> |
| 16 |
>> The same process applies to fontsproto as well. |
| 17 |
>> |
| 18 |
>> There is nothing for you to do with this output, it is informational |
| 19 |
>> (but not labelled as such so you can see it). Maybe run emerge without |
| 20 |
>> -v, see if that removes the output |
| 21 |
> |
| 22 |
> Running without -v doesn't help. I almost never use -v with emerge, it's |
| 23 |
> output is verbose enough as it is and burying useful messages in even |
| 24 |
> more trivial, not-for-users output makes decoding it even harder. |
| 25 |
> |
| 26 |
> emerge really needs an option to hide such "informational" messages. |
| 27 |
|
| 28 |
|
| 29 |
I disagree. emerge really needs to have it's output redesigned from |
| 30 |
scratch. Right now it arrives at the conclusion (the top) and dumps it's |
| 31 |
data tree bottom-up, apparently stopping halfway and never getting to |
| 32 |
output what the top is. |
| 33 |
|
| 34 |
It needs to print it's output top-down instead. Yeah gods, this is such |
| 35 |
a basic output design principle that I keep stating so I start to wonder |
| 36 |
if maybe I'm crazy after all.... |
| 37 |
|
| 38 |
Top-down output lets you easily add more and more output depending on |
| 39 |
the number of -v in the arguments! So much software out there does with |
| 40 |
apparent ease, I can't figure why the portage devs persist with the |
| 41 |
current obtuse method. Especially as portage HAS TO HAVE all necessary |
| 42 |
information available at the time |
| 43 |
|
| 44 |
|
| 45 |
|
| 46 |
> Even |
| 47 |
> if they really were informative, that are not necessary to maintaining a |
| 48 |
> functioning system. When I tell portage to update my system to the latest |
| 49 |
> suitable versions of everything, I have no need to be told "there is a |
| 50 |
> later version for libfoo, but you can't have it". I just want the latest |
| 51 |
> I can have. |
| 52 |
> |
| 53 |
> It doesn't help that these warning messages have a superficial |
| 54 |
> resemblance to blocker errors. |
| 55 |
> |
| 56 |
> emerge --just-tell-me-when-something-is-wrong @world please. |
| 57 |
> |
| 58 |
> |
| 59 |
|
| 60 |
|
| 61 |
-- |
| 62 |
Alan McKinnon |
| 63 |
alan.mckinnon@×××××.com |
Archives