We are working on a newer version of Phaser. The new SConscript file for
Phaser imports features from boost_python. Therefore it is currently not
straightforward to build it since an empty build directory does not yet
contain the boost_python module to be imported by SConscript (or
python). So SCons crashes when it reads the new Phaser SCsonscript
stating that boost_python cannot be imported. Currently our workaround
is to first build CCTBX alone and then subsequently to build Phaser.
My question is if it is possible to specify dependencies in
cctbx_project/boost_adaptbx/SConscript to force building boost_python
early on. On Windows boost_thread and a few other libraries are built
before other SConscript files are read. So I have put some
Depends(boost_thread, boost_python) andDepends(boost_thread,
boost_python_meta_ext)
statements into cctbx_project/boost_adaptbx/SConscript. These do seem to
force building the boost_python.lib and .dll files. But building the
boost_python_meta_ext python module fails with the error:
boost_adaptbx\meta_ext.cpp(14): fatal error C1083: Cannot open include
file: 'boost_adaptbx/type_id_eq.h'
It's possible that I'm not using SCons as it's supposed to be used. Let
me know if anyone has a suggestion on how to force building boost_python
before other SConscript are read.
Many thanks,
Rob
--
Robert Oeffner, Ph.D.
Research Associate, The Read Group
Department of Haematology,
Cambridge Institute for Medical Research
University of Cambridge
Cambridge Biomedical Campus
Wellcome Trust/MRC Building
Hills Road
Cambridge CB2 0XY
www.cimr.cam.ac.uk/investigators/read/index.html
tel: +44(0)1223 763234
Perhaps someone can reply Wen..
Pavel
-------- Forwarded Message --------
Subject: Re: indexing diffraction pattern
Date: Sat, 10 Nov 2018 09:30:53 -0500
From: Wen Jiang <jiang12(a)purdue.edu>
To: Pavel Afonine <pafonine(a)lbl.gov>
Pavel,
Sorry for insufficient details. The scenario is that I have a numpy 2D
array of float values that represents an image of "diffraction pattern"
(generated by an image processing operation, not actual X-ray
diffraction). I need to determine the peak positions and then find a 2D
lattice (two 2D vectors) going through these peaks. I have my own simple
functions that can detect the peaks and find a 2-D lattice to cross the
peaks but it often fails when the "diffraction pattern" is not very
clean. Thus, I want to find a more robust method that works well when
the peaks are relatively weak, not very sharp, and when there are some
spurious peaks off the lattice. I think these tasks are exactly what
cctbx has solved for indexing X-ray diffraction images.
Hope that these explanations help clarify the problem.
Thanks,
Wen
Prof. Wen Jiang
Department of Biological Sciences
Department of Chemistry (Courtesy)
Scientific Director, Purdue Cryo-EM Facility
Purdue University
Hockmeyer Hall of Structural Biology
240 S. Martin Jischke Drive
West Lafayette, IN 47907
http://jiang.bio.purdue.edu
On Sat, Nov 10, 2018 at 8:03 AM Pavel Afonine <pafonine(a)lbl.gov
<mailto:[email protected]>> wrote:
Hi Wen,
I'm not sure I fully understand your question... You say a function
that takes a numpy array.. Array of what?
Then, "a list of diffraction spots" -- these are measured in
diffraction experiment.
If you could provide more details as to what you are trying to do as
well as what exactly the inputs are then I will do my best to see
how to do this in CCTBX or forward your questions to CCTBX mailing
list so that others can comment.
All the best,
Pavel
On 11/10/18 03:56, Wen Jiang wrote:
> Hi Pavel,
> I am looking for a function that takes a numpy array and returns a
> list of diffraction spots and lattice vectors. I think
> cctbx should have such a function but I am not familiar with it.
> Can you point me in the right direction?
> Thanks,
>
> Wen
>
> Prof. Wen Jiang
> Department of Biological Sciences
> Department of Chemistry (Courtesy)
> Scientific Director, Purdue Cryo-EM Facility
> Purdue University
> Hockmeyer Hall of Structural Biology
> 240 S. Martin Jischke Drive
> West Lafayette, IN 47907
> http://jiang.bio.purdue.edu
>
Hi,
Trying to install cctbx on Windows 10,
1. install latest python 2 from Python.org
2. python2 bootstrap.py —builder=cctbxlite —with-python=\path-to-python-from-step-1
It fails because of missing module “builtins”. So I installed pip, then used it to install “future”. Then I had to install “six” with pip. Am I being stupid or cctbx does not install out of the box anymore?
Best wishes,
Luc J Bourhis
Hi folks,
It's great to see the momentum building behind the project to migrate to
Python 3. From the roadmap
<https://github.com/cctbx/cctbx_project/wiki/Python-3-roadmap>, there is
clearly still some way to go, but current activity makes me wonder if the
proposed deadline of complete migration by 2020 might be beaten easily.
What is not clear from the roadmap is whether there are any plans for a
post-transition grace period, in which we have switched from writing
forwards compatible Python 2 code to writing backwards compatible Python 3
code. Has this been thought about? Or will it be the case that the first
commit of Py3-only code seals the deal?
I ask because CCP4 has to support a large number of Python projects and
these have not all considered the Python 3 switch nearly as seriously as
cctbx yet. What I would like to request is a reasonable period of backwards
compatibility is ensured to allow other CCP4 projects to catch up, and that
this protected period is enshrined within the roadmap or developer
guidelines. What do you think?
Best wishes
-- David
