[Yum-devel] [PATCH][RFC] Skip-broken support again

[Tim Lauridsen]

seth vidal wrote:
> On Tue, 2007-12-11 at 14:45 +0100, Florian Festi wrote:
>> Sorry. It looks like the skip broken code works exactly the opposite way I 
>> thought: It removes the packages that leads to the problem.
>>
>> I've commited two test cases that give an idea why this might not be the 
>> best strategy (hmm, may be I should add some doc strings...).
> 
> There's a lot of the test case you're describing in
> testAlternativePackageAvailable() that has much more to do with intent.
> 
> if I type
> 
> yum install foo bar-1.1
> 
> and skipbroken realizes bar-1.1 can't be installed and backtracks it out
> and replaces it with bar-ng. I, as the user, am going to be fairly
> pissed.
> 
> ..
> 
> -sv
> 
> 
> _______________________________________________
> Yum-devel mailing list
> Yum-devel at linux.duke.edu
> https://lists.dulug.duke.edu/mailman/listinfo/yum-devel

As it works now:

foo is the initial transaction.
the depsolver adds bar to the transaction.
bar fails because for baz is missing
skip-broken removes bar & foo and leaves an empty transaction.

As testcase suggest it should do this:

foo is the initial transaction.
the depsolver adds bar to the transaction.
bar fails because for baz is missing
skip-broken removes bar
the depsolver adds bar-ng and the transaction completes ok.

this can only happen if skip-broken is excluding bar from the package 
sack too, else the depsolver will pick up bar again, and then we got a 
loop and skip-broken will bail out and return the depsolver errors.

As far as i can see can see, the current way, is right at the moment, i 
don't give unpredicted results for the user.

Tim