Re: [Rdkit-devel] RDKit 2018.03.1 Build and Installation

2018-05-22 Thread alb greg
Hi Greg, I'd like to know one thing : is it mandatory to use python 3 to install RDKit (as in your dockerfile) or is there a similar process in python 2.7 ? BR 2018-05-17 19:20 GMT+02:00 alb greg : > Thanks, Greg! > > Installation is okay now but ctest ends with these 2 failed

Re: [Rdkit-devel] RDKit 2018.03.1 Build and Installation

2018-05-17 Thread alb greg
hat is what ends up being used to build the RDKit. > You have a copy of boost 1.67 installed; this is too new for the version > of cmake you have (and has additional problems anyway); the system boost > solves this problem. > > On Wed, May 16, 2018 at 3:03 PM alb greg > wrot

Re: [Rdkit-devel] RDKit 2018.03.1 Build and Installation

2018-05-16 Thread alb greg
The system boost? You're talking about libboost? 2018-05-16 14:54 GMT+02:00 Greg Landrum : > cmake is finding boost v1.67, not the system boost. > > > On Wed, May 16, 2018 at 2:34 PM alb greg > wrote: > >> Ubuntu 16.04 >> >> *CmakeMessages :* >

Re: [Rdkit-devel] RDKit 2018.03.1 Build and Installation

2018-05-16 Thread alb greg
untu 16.04 here: > https://github.com/rdkit/rdkit_containers/blob/master/ > docker/ubuntu_xenial/Dockerfile > > Aside from the RDkit itself, this uses only software that is part of the > Ubuntu distribution (including the Ubuntu-supplied versions of boost and > Python). > > -greg >

Re: [Rdkit-devel] RDKit 2018.03.1 Build and Installation

2018-05-16 Thread alb greg
Thanks for your help Andrew! My cmake is in version 3.5.1 2018-05-03 19:09 GMT+02:00 Andrew Dalke : > On May 3, 2018, at 18:44, alb greg wrote: > > I installed cmake, python-dev, libsqlite3-dev, NumPy and Boost as > required. > > > > I tried to specify options like boos

[Rdkit-devel] RDKit 2018.03.1 Build and Installation

2018-05-03 Thread alb greg
Hello everyone, I'm an IT and I'm writing in connection with the installation process of RDKit. I have to install the latest version of RDKit (2018.03.1) on several workstations running on Ubuntu 16.04 but the building sequence systematically went wrong on the "cmake .." step (see logs below). I