Am 14.07.2016 um 09:33 schrieb Steve Cassidy: > Hi Bjorn, thanks for the response, very happy to go with the future > way of doing things. I did see the references to conda but couldn’t > quite decipher what was needed. > > Are you saying that I should create a new recipe in bioconda for my > dependency?
Yes :) > Does it matter that this is not in the least ‘bio’ - > these are for our language analysis tools, in the first instance this > is a tool to pull data from our language repository. No this does not matter, just put it in :) > How does Galaxy know about the bioconda channel - is there a > configuration setting pointing to the channels to find dependencies > in? Yes, please look in your galaxy.ini and search for conda. > Could we create a language-conda channel to house language > dependencies? Does that question make sense (not totally up on conda > yet). Yes this question makes sense, but than you have this additional overhead of maintaining the build-infrastructure, building a community around it, merging PR ... and so on. From my experience this is quite a bit of work! Cheers, Bjoern > Steve > > — Department of Computing, Macquarie University > http://web.science.mq.edu.au/~cassidy > >> On 14 Jul 2016, at 5:00 PM, Björn Grüning >> <bjoern.gruen...@gmail.com> wrote: >> >> Hi Steve, >> >> are you fine restricting your tools to a recent Galaxy version, >> let's say 6 month old? >> >> If you forget about tool_depenencies.xml files, completely forget >> these! Conda packaging is the way to go and way more easy for you. >> >> Let me know and I will give a short introduction or follow the >> description on https://github.com/bioconda/bioconda-recipes to >> create a PR and a package for your tool. >> >> This can be used by Galaxy automatically. Thanks, Bjoern >> >> Am 14.07.2016 um 03:49 schrieb Steve Cassidy: >>> I’m just getting my head around tool dependencies and in doing so >>> trying to push my tools to a local toolshed which I can handily >>> run with the planemo/server docker image via Kitematic. >>> >>> That toolshed is set up with an admin user >>> plan...@test.com<mailto:plan...@test.com> so I have a stanza in >>> my .planemo.yml: >>> >>> shed_username: "stevecassidy" sheds: http://localhost:32768/: >>> key: "" email: "plan...@test.com<mailto:plan...@test.com>" >>> password: "planemo" >>> >>> Now I try to create a repository in the toolshed with planemo: >>> >>> $ planemo shed_create --shed_target http://localhost:32768/ >>> Repository created cd >>> '/Users/steve/projects/hcsvlab-galaxy/tools/alveo' && git >>> rev-parse HEAD cd >>> '/Users/steve/projects/hcsvlab-galaxy/tools/alveo' && git diff >>> --quiet Repository [alveoimport] does not exist in the targeted >>> Tool Shed. >>> >>> >>> Note the conflicting messages here. The repository is created >>> (and I can confirm that it is there) but then I get the message >>> that the repo does not exist in the Toolshed. After a bit of >>> digging it turns out that this is because the repo is owned by >>> the user ‘planemo’ rather than my username as declared in >>> .planemo.yml. If I add —owner plane to the command then it works >>> fine. >>> >>> So I’m not sure if this is a bug in the shed_create code or a >>> weird interaction with the setup on planemo/server. >>> >>> Steve >>> >>> — Department of Computing, Macquarie University >>> http://web.science.mq.edu.au/~cassidy >>> >>> >>> >>> ___________________________________________________________ >>> Please keep all replies on the list by using "reply all" in your >>> mail client. To manage your subscriptions to this and other >>> Galaxy lists, please use the interface at: >>> https://lists.galaxyproject.org/ >>> >>> To search Galaxy mailing lists use the unified search at: >>> http://galaxyproject.org/search/mailinglists/ > ___________________________________________________________ Please keep all replies on the list by using "reply all" in your mail client. To manage your subscriptions to this and other Galaxy lists, please use the interface at: https://lists.galaxyproject.org/ To search Galaxy mailing lists use the unified search at: http://galaxyproject.org/search/mailinglists/
