Closed jbwhit closed 7 years ago
Well, it's not a solution to the conda-forge real problem posted, but adding conda-forge (going w/ the default instead of conda-forge channel) as suggested in https://github.com/conda-forge/conda-forge.github.io/issues/232 fixes the problem superficially.
So conda-forge
doesn't provide rpy2
. That has to be an issue with package from defaults
then. @mingwandroid, could you please take a look at this?
Has this issue been resolved? I'm getting the same error with the following versions: Python 3.5.2 :: Anaconda 4.2.0 (x86_64) jupyter notebook 4.2.3 rpy2 2.8.3
I'm getting the same issue. My setup:
# Sever: Ubuntu 14.04 x86_64.
# Anaconda (for python3.5) install
wget https://repo.continuum.io/archive/Anaconda3-4.2.0-Linux-x86_64.sh
bash Anaconda3-4.2.0-Linux-x86_64.sh -b -p `pwd`"/anaconda3"
conda update conda
# Creating environment
conda create -y -n test_py2 python=python=2.7.12
conda install -y -n test_py2 -c conda-forge jupyter_contrib_nbextensions
conda install -c r rpy2
I get a similar error as @jbwhit when I try %load_ext rpy2.ipython
. Here's the traceback:
ImportError Traceback (most recent call last)
<ipython-input-1-691c6d73b073> in <module>()
----> 1 get_ipython().magic(u'load_ext rpy2.ipython')
/ebio/abt3_projects/methanogen_host_evo/anaconda3/envs/py2_ley0.4/lib/python2.7/site-packages/IPython/core/interactiveshell.pyc in magic(self, arg_s)
2156 magic_name, _, magic_arg_s = arg_s.partition(' ')
2157 magic_name = magic_name.lstrip(prefilter.ESC_MAGIC)
-> 2158 return self.run_line_magic(magic_name, magic_arg_s)
2159
2160 #-------------------------------------------------------------------------
/ebio/abt3_projects/methanogen_host_evo/anaconda3/envs/py2_ley0.4/lib/python2.7/site-packages/IPython/core/interactiveshell.pyc in run_line_magic(self, magic_name, line)
2077 kwargs['local_ns'] = sys._getframe(stack_depth).f_locals
2078 with self.builtin_trap:
-> 2079 result = fn(*args,**kwargs)
2080 return result
2081
<decorator-gen-62> in load_ext(self, module_str)
/ebio/abt3_projects/methanogen_host_evo/anaconda3/envs/py2_ley0.4/lib/python2.7/site-packages/IPython/core/magic.pyc in <lambda>(f, *a, **k)
186 # but it's overkill for just that one bit of state.
187 def magic_deco(arg):
--> 188 call = lambda f, *a, **k: f(*a, **k)
189
190 if callable(arg):
/ebio/abt3_projects/methanogen_host_evo/anaconda3/envs/py2_ley0.4/lib/python2.7/site-packages/IPython/core/magics/extension.pyc in load_ext(self, module_str)
35 if not module_str:
36 raise UsageError('Missing module name.')
---> 37 res = self.shell.extension_manager.load_extension(module_str)
38
39 if res == 'already loaded':
/ebio/abt3_projects/methanogen_host_evo/anaconda3/envs/py2_ley0.4/lib/python2.7/site-packages/IPython/core/extensions.pyc in load_extension(self, module_str)
81 if module_str not in sys.modules:
82 with prepended_to_syspath(self.ipython_extension_dir):
---> 83 __import__(module_str)
84 mod = sys.modules[module_str]
85 if self._call_load_ipython_extension(mod):
/ebio/abt3_projects/methanogen_host_evo/anaconda3/envs/py2_ley0.4/lib/python2.7/site-packages/rpy2/ipython/__init__.py in <module>()
----> 1 from .rmagic import load_ipython_extension
/ebio/abt3_projects/methanogen_host_evo/anaconda3/envs/py2_ley0.4/lib/python2.7/site-packages/rpy2/ipython/rmagic.py in <module>()
50 # numpy and rpy2 imports
51
---> 52 import rpy2.rinterface as ri
53 import rpy2.robjects as ro
54 import rpy2.robjects.packages as rpacks
/ebio/abt3_projects/methanogen_host_evo/anaconda3/envs/py2_ley0.4/lib/python2.7/site-packages/rpy2/rinterface/__init__.py in <module>()
90 del(os)
91
---> 92 from rpy2.rinterface._rinterface import (baseenv,
93 emptyenv,
94 endr,
ImportError: libicuuc.so.54: cannot open shared object file: No such file or directory
Any ideas on what's going on? Thanks
Any ideas on what's going on? Thanks
Like it is said above, conda-forge
does not package r2py
(yet) and there is no way to ensure that the defaults
version of it will work with the rest of the conda-forge
stack.
In this case what you are experiencing is the mismatch if the icu
version. defaults
uses 54.*
while conda--forge uses 56.*
.
FYI there is an effort underway to add R to conda-forge, but may take a bit of time to complete.
I'm glad to hear that there's an effort to add R to conda-forge.
In the meantime, this is such an interesting thread.
I think I understand the problem: using conda-forge installs the versions of each package it knows about, including icu
on version 56.*
. Then, when installing a package that conda-forge doesn't know about (rpy2
for example), it will download it, but that package might no longer work because it relies on the conda defaults
channel, which points to 54.*
in this case.
Now, I can understand that this is unsolvable in the sense conda-forge can't reason about dependencies for things they don't know about, but to mark the issue as closed
seems incorrect. It may be unsolvable, but this issue is a real one. People who have added the conda-forge channel will have otherwise working installs broken (in a really hard to diagnose way).
I'm not sure what to do in the meantime. Perhaps put a warning somewhere to watch out, and that this kind of thing can happen?
IMHO there should be a bug called "conda defaults and conda-forge are incompatible" and all these things should be marked as duplicates of that and then closed.
Indeed. I'll do it soon.
So... I'm not sure how to get rpy2 working. Loading rpy2.ipython still throws the error: ImportError: libicuuc.so.54: cannot open shared object file: No such file or directory
. My conda environment has the following r-related packages:
r 3.3.2 r3.3.2_0 r
r-assertthat 0.1 r3.3.2_4 r
r-base 3.3.2 1 conda-forge
r-bh 1.62.0_1 r3.3.2_0 r
r-boot 1.3_18 r3.3.2_0 r
r-class 7.3_14 r3.3.2_0 r
r-cluster 2.0.5 r3.3.2_0 r
r-codetools 0.2_15 r3.3.2_0 r
r-colorspace 1.3_1 r3.3.2_0 r
r-dbi 0.5_1 r3.3.2_0 r
r-dichromat 2.0_0 r3.3.2_2 r
r-digest 0.6.10 r3.3.2_0 r
r-dplyr 0.5.0 r3.3.2_0 r
r-foreign 0.8_67 r3.3.2_0 r
r-ggplot2 2.2.0 r3.3.2_0 r
r-gtable 0.2.0 r3.3.2_0 r
r-kernsmooth 2.23_15 r3.3.2_0 r
r-labeling 0.3 r3.3.2_2 r
r-lattice 0.20_34 r3.3.2_0 r
r-lazyeval 0.2.0 r3.3.2_0 r
r-magrittr 1.5 r3.3.2_2 r
r-mass 7.3_45 r3.3.2_0 r
r-matrix 1.2_7.1 r3.3.2_0 r
r-mgcv 1.8_16 r3.3.2_0 r
r-munsell 0.4.3 r3.3.2_0 r
r-nlme 3.1_128 r3.3.2_0 r
r-nnet 7.3_12 r3.3.2_0 r
r-plyr 1.8.4 r3.3.2_0 r
r-r6 2.2.0 r3.3.2_0 r
r-rcolorbrewer 1.1_2 r3.3.2_3 r
r-rcpp 0.12.8 r3.3.2_0 r
r-recommended 3.3.2 r3.3.2_0 r
r-reshape2 1.4.2 r3.3.2_0 r
r-rpart 4.1_10 r3.3.2_0 r
r-scales 0.4.1 r3.3.2_0 r
r-spatial 7.3_11 r3.3.2_0 r
r-stringi 1.1.2 r3.3.2_0 r
r-stringr 1.1.0 r3.3.2_0 r
r-survival 2.40_1 r3.3.2_0 r
r-tibble 1.2 r3.3.2_0 r
r-tidyr 0.6.0 r3.3.2_0 r
readline 6.2 0
requests 2.12.4 py35_0
rpy2 2.8.5 py35r3.3.2_2 r
If the issue is that I'm using the conda-forge version of r-base
, then how do I change to a different version of r-base
? I can't seem to changer-base
to another channel (e.g., r
):
$ conda install --override-channels -c r r-base
Fetching package metadata .....
Solving package specifications: ..........
# All requested packages already installed.
# packages in environment at /ebio/abt3_projects/methanogen_formate/anaconda3/envs/py3_ley0.4:
#
r-base 3.3.2 1 conda-forge
I can't even search for different versions of r-base
with anaconda search
:
$ anaconda search r-base
Using Anaconda API: https://api.anaconda.org
Run 'anaconda show <USER/PACKAGE>' to get more details:
Packages:
Name | Version | Package Types | Platforms
------------------------- | ------ | --------------- | ---------------
[ERROR]
Traceback (most recent call last):
File "/ebio/abt3_projects/methanogen_formate/anaconda3/envs/py3_ley0.4/bin/anaconda", line 6, in <module>
sys.exit(binstar_client.scripts.cli.main())
File "/ebio/abt3_projects/methanogen_formate/anaconda3/envs/py3_ley0.4/lib/python3.5/site-packages/binstar_client/scripts/cli.py", line 88, in main
description=__doc__, version=version)
File "/ebio/abt3_projects/methanogen_formate/anaconda3/envs/py3_ley0.4/lib/python3.5/site-packages/binstar_client/scripts/cli.py", line 66, in binstar_main
return args.main(args)
File "/ebio/abt3_projects/methanogen_formate/anaconda3/envs/py3_ley0.4/lib/python3.5/site-packages/binstar_client/commands/search.py", line 15, in search
pprint_packages(packages, access=False)
File "/ebio/abt3_projects/methanogen_formate/anaconda3/envs/py3_ley0.4/lib/python3.5/site-packages/binstar_client/utils/pprint.py", line 70, in pprint_packages
pprint_package(package, access, full_name, revision=revisions)
File "/ebio/abt3_projects/methanogen_formate/anaconda3/envs/py3_ley0.4/lib/python3.5/site-packages/binstar_client/utils/pprint.py", line 34, in pprint_package
package['conda_platforms'] = ', '.join(package['conda_platforms'])
TypeError: sequence item 2: expected str instance, NoneType found
Right now, I'm kind of stuck. Any help would be greatly appreciated!
Right now, I'm kind of stuck. Any help would be greatly appreciated!
@nick-youngblut we cannot help you with a software that is not in conda-forge. Please read https://github.com/conda-forge/conda-forge.github.io/issues/234#issuecomment-27574955
You need to rely on defaults
only for that.
You need to remove conda-forge
from your channels list in ~/.condarc
then do conda update all
, or just conda install -c r r-base
to install what you need.
You should use conda search -c r r-base
for the search you are attempting (though this does look like you've run into a bug in anaconda-client
).
Thanks for the help! Removing conda-forge
from my channels in my ~/.condarc
and then running conda install -c r r-base
worked.
I'm finding that conda can be tricky to use: there's always the pitfall of installing the 'wrong' version of a package, and then it can be hard to change the package version (e.g., having to manually remove a channel from ~/.condarc
). Still, for most bioinformatics software, conda (& especially bioconda) makes installation extremely simple.
Despite this issue being closed, I thought I would chime in since things are changed since @nick-youngblut experience given that the bioconda team seems to have gotten updated R into the conda-forge. Given that, it is preferable NOT to use the 'r' channel at all, and rather to have conda-forge come before default. With that setup, so far, for me, many bioconda installs are, as nick wrote "extremely simple".
I'm having this same issue, and I agree with the following from above:
but to mark the issue as closed seems incorrect. It may be unsolvable, but this issue is a real one. People who have added the conda-forge channel will have otherwise working installs broken (in a really hard to diagnose way).
As far as I can tell, if a user needs rpy2, and anything in conda-forge, this issue occurs. rpy2 is an essential part of the Python data stack for many. This problem means that I can't use conda at all for my current project, which seems to be a serious issue.
I'm happy to provide more information if needed on my specific case. But I think ignoring this issue outright is biting a lot of folks using conda for science.
Thanks for the offer of information, but I'm fairly sure it's just same old vanilla binary incompatibility.
If (as I do) you would like to be able to use Rpy2
in conda-forge
you can contribute a recipe, hopefully with my patches so that it works on Windows correctly. I would do it myself but I'm a bit busy:
https://github.com/conda/conda-recipes/tree/master/python/rpy2
Cheers!
It has been a while that conda-forge has rpy2
.
I'd like to know if that works for you there.
Ah well there you go!
Thanks @ocefpaf. I wonder if the problem is now that defaults has R 3.4.1 and conda-forge doesn't (or does it?)
Thanks @ocefpaf. I wonder if the problem is now that defaults has R 3.4.1 and conda-forge doesn't (or does it?)
We do https://github.com/conda-forge/r-base-feedstock/blob/master/recipe/meta.yaml#L3
We may need to re-render rpy2
though. I'll check that soon.
Ah, in that case @bgruening of @johanneskoester, if you have time, you may want to merge across https://github.com/conda/conda-recipes/tree/master/r-packages/r-rjava (and the rest?!).
Hm, so I'm trying to figure out precisely what's going wrong. I also need r-dbplyr which isn't on conda-forge, and I'm thinking that may change something conda-forge's rpy2 needs. Here is the full environment I'm trying to create (for completeness and other folks Googling similar things):
$ cat ~/.condarc
channels:
- bioconda
- conda-forge
- defaults
$ conda create -n fwpdy11 python=3.6 r-base=3.4.1 rpy2 r-tidyverse r-dbplyr r-devtools gsl=1.16 fwdpy11 pylibseq ipython jupyter numpy pandas tqdm msprime scipy
Then I try to get this working from iPython:
%load_ext rpy2.ipython
import fwdpy11 as fp11
import numpy as np
import pandas as pd
from tqdm import tqdm_notebook
import msprime as msp
import libsequence as ls
Often, even if I get an environment somehow configured that works with the above, there's issues with dynamic libraries that crop up with (again from iPython):
%%R
library(tidyverse)
library(dbplyr)
Conda Info:
conda info
Current conda install:
platform : linux-64
conda version : 4.3.23
conda is private : False
conda-env version : 4.3.23
conda-build version : not installed
python version : 3.6.1.final.0
requests version : 2.14.2
root environment : /home/vinceb/anaconda3 (writable)
default environment : /home/vinceb/anaconda3/envs/fwpdy11
envs directories : /home/vinceb/anaconda3/envs
/home/vinceb/.conda/envs
package cache : /home/vinceb/anaconda3/pkgs
/home/vinceb/.conda/pkgs
channel URLs : https://conda.anaconda.org/bioconda/linux-64
https://conda.anaconda.org/bioconda/noarch
https://conda.anaconda.org/conda-forge/linux-64
https://conda.anaconda.org/conda-forge/noarch
https://repo.continuum.io/pkgs/free/linux-64
https://repo.continuum.io/pkgs/free/noarch
https://repo.continuum.io/pkgs/r/linux-64
https://repo.continuum.io/pkgs/r/noarch
https://repo.continuum.io/pkgs/pro/linux-64
https://repo.continuum.io/pkgs/pro/noarch
config file : /home/vinceb/.condarc
netrc file : None
offline mode : False
user-agent : conda/4.3.23 requests/2.14.2 CPython/3.6.1 Linux/3.10.0-514.26.2.el7.x86_64 CentOS Linux/7.3.1611 glibc/2.17
UID:GID : 1002:1002
@vsbuffalo, the most useful information here for us to help diagnose this is:
conda list --show-channel-urls
rpy2
is not available on conda-forge
for r=3.4.1
. It's available only for r=3.3.2
@vsbuffalo you have 3 channels there and I have a slide (#8) for you :wink:
I am re-rendering now to get rpy2
for r=3.4.1
in the channel. But we would still need a PR for r-dbplyr
.
@ocefpaf heh, nice slides. If we get r-dbplyr
into conda-forge
, that might solve it. It's a dependency to r-tidyverse
whenever database backends are used. I'll take a look at how this is done, maybe it's something I could help with.
@mingwandroid, sure here that is:
# packages in environment at /home/vinceb/anaconda3/envs/fwpdy11:
#
blas 1.1 openblas conda-forge
bleach 1.4.2 py36_0 bioconda
bzip2 1.0.6 1 conda-forge
ca-certificates 2017.7.27.1 0 conda-forge
cairo 1.14.6 4 conda-forge
certifi 2017.7.27.1 py36_0 conda-forge
curl 7.54.1 0 conda-forge
dbus 1.10.22 0 conda-forge
decorator 4.1.2 py36_0 conda-forge
entrypoints 0.2.3 py36_1 conda-forge
expat 2.2.1 0 conda-forge
fontconfig 2.12.1 4 conda-forge
freetype 2.7 1 conda-forge
fwdpy11 0.1.3a1 py36_gsl1.16_0 bioconda
gettext 0.19.7 1 conda-forge
glib 2.51.4 0 conda-forge
gmp 6.1.2 0 conda-forge
graphite2 1.3.9 0 conda-forge
gsl 1.16 0 conda-forge
gst-plugins-base 1.8.0 0 conda-forge
gstreamer 1.8.0 2 conda-forge
h5py 2.7.0 np113py36_1 conda-forge
harfbuzz 1.4.3 0 conda-forge
hdf5 1.8.18 1 conda-forge
html5lib 0.999999999 py36_0 conda-forge
htslib 1.5 0 bioconda
icu 58.1 1 conda-forge
ipykernel 4.6.1 py36_0 conda-forge
ipython 6.1.0 py36_0 conda-forge
ipython_genutils 0.2.0 py36_0 conda-forge
ipywidgets 7.0.0 py36_0 conda-forge
jedi 0.10.2 py36_0 conda-forge
jinja2 2.9.6 py36_0 conda-forge
jpeg 9b 0 conda-forge
jsonschema 2.6.0 py36_0 conda-forge
jupyter 1.0.0 py36_0 conda-forge
jupyter_client 5.1.0 py36_0 conda-forge
jupyter_console 5.1.0 py36_0 conda-forge
jupyter_core 4.3.0 py36_0 conda-forge
krb5 1.14.2 0 conda-forge
libffi 3.2.1 3 conda-forge
libgcc 5.2.0 0 defaults
libgfortran 3.0.0 1 defaults
libiconv 1.14 4 conda-forge
libpng 1.6.28 1 conda-forge
libsequence 1.9.1 htslib1.5_1 bioconda
libsodium 1.0.10 0 conda-forge
libssh2 1.8.0 1 conda-forge
libtiff 4.0.6 7 conda-forge
libxcb 1.12 1 conda-forge
libxml2 2.9.4 4 conda-forge
markupsafe 1.0 py36_0 conda-forge
mistune 0.7.4 py36_0 conda-forge
msprime 0.4.0 py36_gsl1.16_2 bioconda
nbconvert 5.2.1 py36_1 conda-forge
nbformat 4.3.0 py36_0 conda-forge
ncurses 5.9 10 conda-forge
notebook 5.0.0 py36_0 conda-forge
numpy 1.13.1 py36_blas_openblas_200 [blas_openblas] conda-forge
openblas 0.2.19 2 conda-forge
openssl 1.0.2l 0 conda-forge
pandas 0.20.3 py36_1 conda-forge
pandoc 1.19.2 0 conda-forge
pandocfilters 1.4.1 py36_0 conda-forge
pango 1.40.4 0 conda-forge
pcre 8.39 0 conda-forge
pexpect 4.2.1 py36_0 conda-forge
pickleshare 0.7.4 py36_0 conda-forge
pip 9.0.1 py36_0 conda-forge
pixman 0.34.0 0 conda-forge
prompt_toolkit 1.0.15 py36_0 conda-forge
ptyprocess 0.5.2 py36_0 conda-forge
pygments 2.2.0 py36_0 conda-forge
pylibseq 0.1.9.post0 py36_0 bioconda
pyparsing 2.2.0 py36_0 conda-forge
pyqt 5.6.0 py36_4 conda-forge
python 3.6.2 0 conda-forge
python-dateutil 2.6.1 py36_0 conda-forge
pytz 2017.2 py36_0 conda-forge
pyzmq 16.0.2 py36_2 conda-forge
qt 5.6.2 3 conda-forge
qtconsole 4.3.1 py36_0 conda-forge
r-assertthat 0.1 r3.4.1_0 conda-forge
r-base 3.4.1 0 conda-forge
r-bh 1.62.0_1 r3.4.1_0 conda-forge
r-bindr 0.1 r3.4.1_0 defaults
r-bindrcpp 0.2 r3.4.1_0 defaults
r-broom 0.4.2 r3.4.1_0 defaults
r-cellranger 1.1.0 r3.4.1_0 conda-forge
r-colorspace 1.3_2 r3.4.1_0 conda-forge
r-curl 2.4 r3.4.1_0 conda-forge
r-dbi 0.6_1 r3.4.1_0 conda-forge
r-dbplyr 1.1.0 r3.4.1_0 defaults
r-devtools 1.13.2 r3.4.1_0 conda-forge
r-dichromat 2.0_0 r3.4.1_0 conda-forge
r-digest 0.6.10 r3.4.1_0 conda-forge
r-dplyr 0.7.0 r3.4.1_0 defaults
r-forcats 0.2.0 r3.4.1_0 conda-forge
r-foreign 0.8_67 r3.4.1_0 conda-forge
r-ggplot2 2.2.0 r3.4.1_0 conda-forge
r-git2r 0.19.0 r3.4.1_0 conda-forge
r-glue 1.1.1 r3.4.1_0 defaults
r-gtable 0.2.0 r3.4.1_0 conda-forge
r-haven 1.0.0 r3.4.1_0 conda-forge
r-hms 0.3 r3.4.1_0 conda-forge
r-httr 1.2.1 r3.4.1_0 conda-forge
r-jsonlite 1.4 r3.4.1_0 conda-forge
r-labeling 0.3 r3.4.1_0 conda-forge
r-lattice 0.20_34 r3.4.1_0 conda-forge
r-lazyeval 0.2.0 r3.4.1_0 conda-forge
r-lubridate 1.6.0 r3.4.1_0 conda-forge
r-magrittr 1.5 r3.4.1_0 conda-forge
r-mass 7.3_45 r3.4.1_0 conda-forge
r-memoise 1.1.0 r3.4.1_0 conda-forge
r-mime 0.5 r3.4.1_0 conda-forge
r-mnormt 1.5_5 r3.4.1_0 conda-forge
r-modelr 0.1.0 r3.4.1_0 conda-forge
r-munsell 0.4.3 r3.4.1_0 conda-forge
r-nlme 3.1_131 r3.4.1_0 conda-forge
r-openssl 0.9.6 r3.4.1_1 conda-forge
r-pkgconfig 2.0.1 r3.4.1_0 defaults
r-plogr 0.1_1 r3.4.1_0 conda-forge
r-plyr 1.8.4 r3.4.1_0 conda-forge
r-psych 1.7.3.21 r3.4.1_0 conda-forge
r-purrr 0.2.2 1 bioconda
r-r6 2.2.0 r3.4.1_0 conda-forge
r-rcolorbrewer 1.1_2 r3.4.1_0 conda-forge
r-rcpp 0.12.10 r3.4.1_0 conda-forge
r-readr 1.1.0 r3.4.1_0 conda-forge
r-readxl 1.0.0 r3.4.1_0 conda-forge
r-rematch 1.0.1 r3.4.1_0 conda-forge
r-reshape2 1.4.2 r3.4.1_0 conda-forge
r-rlang 0.1.2 r3.4.1_0 conda-forge
r-rstudioapi 0.6 r3.4.1_0 conda-forge
r-rvest 0.3.2 r3.4.1_0 conda-forge
r-scales 0.4.1 r3.4.1_0 conda-forge
r-selectr 0.3_1 r3.4.1_0 conda-forge
r-stringi 1.1.2 r3.4.1_0 conda-forge
r-stringr 1.2.0 r3.4.1_0 conda-forge
r-tibble 1.3.3 r3.4.1_0 defaults
r-tidyr 0.6.1 r3.4.1_0 conda-forge
r-tidyverse 1.1.1 r3.4.1_0 conda-forge
r-whisker 0.3_2 r3.4.1_0 conda-forge
r-withr 1.0.2 r3.4.1_0 conda-forge
r-xml2 1.1.1 r3.4.1_0 conda-forge
readline 6.2 0 conda-forge
rpy2 2.8.6 py36r3.4.1_2 defaults
scipy 0.19.1 py36_blas_openblas_202 [blas_openblas] conda-forge
setuptools 36.2.2 py36_0 conda-forge
simplegeneric 0.8.1 py36_0 conda-forge
singledispatch 3.4.0.3 py36_0 conda-forge
sip 4.18 py36_1 conda-forge
six 1.10.0 py36_1 conda-forge
sqlite 3.13.0 1 conda-forge
svgwrite 1.1.6 py36_0 bioconda
tbb 4.4_20150728 0 bioconda
terminado 0.6 py36_0 conda-forge
testpath 0.3.1 py36_0 conda-forge
tk 8.5.19 2 conda-forge
tornado 4.5.1 py36_0 conda-forge
tqdm 4.7.2 py36_0 bioconda
traitlets 4.3.2 py36_0 conda-forge
wcwidth 0.1.7 py36_0 conda-forge
webencodings 0.5 py36_0 conda-forge
wheel 0.29.0 py36_0 conda-forge
widgetsnbextension 3.0.2 py36_0 conda-forge
xorg-libxau 1.0.8 3 conda-forge
xorg-libxdmcp 1.1.2 3 conda-forge
xz 5.2.3 0 conda-forge
zeromq 4.2.1 1 conda-forge
zlib 1.2.8 3 conda-forge
@vsbuffalo, please, if you have time do help! Here is the recipe used to build the package on defaults
:
https://github.com/conda/conda-recipes/tree/master/r-packages/r-dbplyr
I added a PR for r-dbplyr with https://github.com/conda-forge/staged-recipes/pull/3861.
So for completeness, here is an update. First, I can rework my code so r-dbplyr is no longer necessary (but then new issues crop up). Then, the following works:
$ conda create -n default-fwdpy11-2 python=3.6 r-base=3.4.1 r-tidyverse r-devtools r-rsqlite gsl=1.16 fwdpy11 pylibseq ipython jupyter numpy pandas tqdm msprime scipy
$ conda install -c conda-forge -n default-fwdpy11 rpy2
while trying to install it all in one go, e.g.:
$ conda create -n default-fwdpy11-2 python=3.6 r-base=3.4.1 r-tidyverse r-devtools r-rsqlite gsl=1.16 fwdpy11 pylibseq ipython jupyter numpy pandas tqdm msprime scipy rpy2
does not work.
Essentially, I can get a reproducible-ish environment through iteratively crafting with the command line interface. In general, issues are avoided and conda is quite useful if one minimizes the channels used (I think this is what @ocefpaf's slides suggest) as this creates dependency nightmares. As a conda-beginner, it seems avoiding default
and r
entirely would be useful, but not always feasible — this is where I got bitten. AFAIK, there's still not a way to install r-dplyr=0.7
from r
or default
without re-causing the original issue. I can try writing a recipe, as this is a rather important release (the behavior of collect()
changes drastically).
As someone who works on r
and defaults
(pedantically r
is a part of defaults
now) as well as on conda-forge
, I would like to know what is missing from r
\ defaults
that prevents you being able to use them.
I want all the options to be viable basically (and mixable in future).
it seems avoiding default and r entirely would be useful, but not always feasible
The long-term goal is to make conda-forge
and defaults
as compatible as possible to avoid such headaches. And you are correct, we need at least those two, but if we can avoid other channels we should.
r-base 3.4.1
was added recently to conda-forge
, right when we had the token issue, and we need to re-render the recipes to get the packages. rpy2
, for example, is stuck b/c of that (missing some dependencies). See https://github.com/conda-forge/rpy2-feedstock/pull/2
I'll see what I can next weekend, but I cannot promise anything (king of buried in work here).
If you want to help take a look at https://conda-forge.org/docs/conda_smithy.html and try to send re-rendering PRs to the feedtocks we need to be available for r 3.4.1
, like r-ggplot
. We do have a bot to do that but it goes really slow b/c it can only process a few feedstocks per day to avoid bringing our CIs down.
I want all the options to be viable basically (and mixable in future).
For mixing conda-forge and defaults, one issue is that defaults
is using a broken gettext version when compiling r-base
. gettext=0.19.8
accidentally bumped the soversion and this was reverted in gettext=0.19.8.1
.
conda-forge
is using 0.19.8.1
while defaults is using 0.19.8
@isuruf, thanks.
r-dbplyr is now available.
I was having problems that I couldn't diagnose, so I deleted all of miniconda/.jupyter/etc, and reinstalled.
The error
How I installed everything
These were the steps that I ran:
My setup
And