On Mon, Oct 26, 2015 at 01:56:48PM -0400, Richard Heck wrote: > On 10/26/2015 01:31 AM, Scott Kostyshak wrote: > >On Mon, Oct 26, 2015 at 03:50:43AM +0100, Uwe Stöhr wrote: > >>The attached patch adds support for the modernCV features introduced by > >>version 1.5 from April 2013. > >> > >>I think it is safe enough to support 2 years old features and would > >>therefore like to put it in. > >Do we have any policy on this? What have we done in the past? > > It's a long-standing problem to know what versions of these packages to > support.
I see. It would be nice to have a rule so we don't have to have the same discussion multiple times. We could add such a rule to Development.lyx. Of course, the rule could be violated for some cases if there is an argument. What Uwe suggests seems reasonable to me at first thought for the rule: for a major release, we can have our layouts depend on the previous version of TeX Live. Currently there is TeX Live 2015 so anything that is supported in TeX Live 2014 we can depend on starting for a major release. Does anyone else have an opinion? > It might be worth renaming the old layout to moderncv-14.layout or > something, in case people don't have the new version. This seems like a good idea to me. And then we could expire the old version the next major release? > >>Since a style is added, this would be a simple fileformat change. > >> > >>OK to go on? > >Uwe have you tried to run the tests yet? > > Any format change will lead to test failures. We know that, right? But this > will have a null forward conversion, I believe. And if it does then simply > updating the format numbers in the test files should be enough, right? I was talking about the export tests. The export tests are a quick and easy way to see if changing either a layout file or a .lyx file leads to possible compilation issues (which would suggest there could be a bug in the commit). It is possible to run just the tests related to the code you touched. In this case: ctest -R "modernCV" > I'm a bit with Uwe on this. I think the tests are important, but in the past > people who know about tex2lyx (i.e., not me) have taken care of that part of > it. I guess the thought is that Uwe, or I, should be able to ask someone > else to handle that side of things. If this is what is decided, I would at least prefer that it all be done in one commit. And further, I think that everyone should be able to run the tex2lyx tests. If they fail, then do not commit. Ask for help, like Richard suggests. But ask for help before committing. Scott
