Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
    Victor Nițu
    @nightsh
    I'm not a Travis expert in any way, but I'm out of ideas
    Felipe Vieira Falcão
    @felipevieira
    already tried it. no success
    tox uses its own virtualenv to install dependecies
    installing numpy on travis would enable the lib outside tox virtualenv
    if I'm correct
    Victor Nițu
    @nightsh
    yes, that makes sense
    what about this in tox.ini?
    # [...]
    [testenv]
    deps =
      numpy
      -r{toxinidir}/requirements.txt
      pytest
      mock
    # [...]
    Felipe Vieira Falcão
    @felipevieira
    doesn't work as well, deps groups all depedencies listed in a unique pip command
    Victor Nițu
    @nightsh
    oh, bummer
    Felipe Vieira Falcão
    @felipevieira
    will try it tho
    Felipe Vieira Falcão
    @felipevieira
    yep, no success. frustrating
    Vitor Baptista
    @vitorbaptista
    @felipevieira Have you tried this?
    ```
    deps =
        numpy
        nose >= 1.2.1
        coverage
        text-unidecode
        twython
        pyparsing
        python-crfsuite
        rednose
    
    changedir = nltk/test
    commands =
        ; scipy and scikit-learn requires numpy even to run setup.py so
        ; they can't be installed in one command
    
        pip install scipy scikit-learn
        ; python runtests.py --with-coverage --cover-inclusive --cover-package=nltk --cover-html --cover-html-dir={envdir}/docs []
        python runtests.py []
    They install numpy in the deps, and then install the other ones via the commands, so numpy would already be installed here
    Felipe Vieira Falcão
    @felipevieira
    @vitorbaptista hmm in this case I would do pip install -r requirements.txt via commands, right? I've just commited a similar approach and got the tests ok. This approach you've just sent seems more elegant and should work. Will use it. Thanks
    Vitor Baptista
    @vitorbaptista
    @felipevieira hmmm... yes, that should work. Add numpy and the other dev requirements on the deps section and, on the commands, do the pip install -r {toxinidir}/requirements.txt
    That's a bit hacky but I don't see another way
    Unless something like this works:
    deps =
        numpy
        -r{toxinidir}/requirements.txt
    commands =
        py.test
    Worth a try, I'd guess
    Felipe Vieira Falcão
    @felipevieira
    already tried it, as I've discussed with @nightsh earlier. doesnt work because deps groups all depedencies listed in a unique pip command
    the last hack works tho
    Vitor Baptista
    @vitorbaptista
    @felipevieira Happy to move forward if it works, just add a comment on our tox.ini so others reading it understand what's going on
    Felipe Vieira Falcão
    @felipevieira
    :+1:
    Georgiana Bere
    @georgiana-b
    @victorbaptista It appears that a data contribution can have both a data_url(uploaded file) and a url (external link) now. Is this intended functionality?
    Vitor Baptista
    @vitorbaptista
    @georgiana-b Yes. The names are confusing, but the separation is needed, as a user can upload a document and add a link to where she found it, for example
    So data_url is the actual file on S3 (if it exists), and url is the link the user added
    Georgiana Bere
    @georgiana-b
    @vitorbaptista So, when this kind of data contribution is processed, should we create 2 documents or just 1 document with the data_url (S3 link) prioritized?
    Vitor Baptista
    @vitorbaptista
    Good question. We don't have a way to map this on our API database now, so it would be a single document with the data_url prioritized.
    Georgiana Bere
    @georgiana-b
    :+1:
    Alessia
    @alessiabardi
    Hi all, I am Alessia Bardi from the OpenAIRE technical team
    I have a couple of questions regarding the API
    oh, I see none of you is available now, I will write you an email then. Talk to you soon
    stevestork22
    @stevestork22
    Hi All, Im a research coordinator with very limited dev experience. Been playing around with the terminal & npm commands. Cant get my local host to pull in data from opentrials. Not sure how to set up my files. The README instructs me to alter the .env file, but im not sure how. Anyone willing to offer their expertise?
    Rufus Pollock
    @rufuspollock
    @stevestork22 hey there - can you give more detail. I'd also recommend asking on okfn/chat channel (I know, I know - another channel but that is a more central place for open knowledge community tech/data projects!)
    Vitor Baptista
    @vitorbaptista
    @stevestork22 Hi Steve, I imagine you're more interested in the data itself than OpenTrials as a platform? If that's the case, my recommendation would be to download the latest data dump from https://explorer.opentrials.net/data and loading that into a local PostgreSQL database. That'll allow you to use SQL to dig into our data.
    Rufus Pollock
    @rufuspollock
    @vitorbaptista as part of Open Trials did we get a database of drugs with their trade names? And can i find that dump somewhere?
    Vitor Baptista
    @vitorbaptista
    @rufuspollock We have some information on drugs, but not a good and clean dataset. They'll mostly be in the interventions table (look for type == 'drug'), and the fda_applications (that would have all drugs approved for the USA)
    You can download a dump of the DB via https://explorer.opentrials.net/data
    There're some drug datasets available, like https://www.drugbank.ca/ (non-commercial license), but we haven't extracted any of them
    Rufus Pollock
    @rufuspollock
    @vitorbaptista do we have anything tracking the places we looked for this data?
    Vitor Baptista
    @vitorbaptista
    @rufuspollock Unfortunately not
    Rufus Pollock
    @rufuspollock
    @vitorbaptista could you drop your comments into datasets/registry#222
    Vitor Baptista
    @vitorbaptista
    @rufuspollock I checked the issue and the links and I don't know anything else from what was already said. I've pinged Seb Bacon on the issue asking if he knows of anything else
    Slythfox
    @slythfox
    Just came here to say that https://opentrials.net/ is giving a HTTP 301. Is this intended?
    Paul Walsh
    @pwalsh
    @slythfox that is fixed. thanks for the report.
    Raul Figueira
    @raulfmiranda
    What is the OpenTrials endpoint? Can anyone give me a complete url that gets a json data from OpenTrials API ?