iagapov / ocelot Goto Github PK

add helical undulator parameter

or at least up to second order

function definition does not conform to standards

1. must return bool (True/False)
2. comments must be in more or less correct english
3. change name to something like check_edges()

rename cpbd.optics.track into step and move to cpbd.tracking
rename cpbd.tracking to cpbd.track

Traceback (most recent call last):
File "C:/Users/tomins/Documents/Dropbox/DESY/repository/ocelot/demos/optics/ex6.py", line 92, in
s2 = fwhm(x,np.abs(of.mesh.points[:,of.ny/2]))
File "C:\Users\tomins\Documents\Dropbox\DESY\repository\ocelot\common\math_op.py", line 126, in fwhm
return x[downs[-1]] - x[ups[0]]
IndexError: list index out of range

Traceback (most recent call last):
File "/home/ilya/workspace/desy/petra3/petra_twiss.py", line 40, in
eb.integrals_id()
File "/home/ilya/workspace/ocelot/cpbd/e_beam_params.py", line 125, in integrals_id
B = K2field(elem.Kx, lu = elem.lperiod)
NameError: global name 'K2field' is not defined

for zero length sextupoles will be better to use multipole.
taking into account multipoles in chromaticity calculation

from ocelot import *
from ocelot.gui.accelerator import *

d2 = Drift(l=1.0)
sf1 = Sextupole(l=0.25, k2= 260)

beam = Beam()
beam.E = 5.0
beam.sigma_E = 0.002
beam.beta_x = 1
beam.beta_y = 1

tw0 = Twiss(beam)

lat = MagneticLattice( (d2,sf1,d2))

tws=twiss(lat, Twiss(beam), nPoints = 1000)

plot_opt_func(lat, tws)

plt.show()

remove ocelot.rad
remove ocelot.lib and move ocelot.lib.genera -> ocelot.genera
consider renaming genera -> sr to improve understandability, but not necessarily

charge = 1e-09
Traceback (most recent call last):
File "track_flash2.py", line 40, in
p_array, charge_array = astra_beam2particleArray(filename='elegant_files/flash_out_200000.ast')
File "/data/netapp/xfel/products/test/ocelot/adaptors/astra2ocelot.py", line 50, in astra_beam2particleArray
xxstg = exact_xp_2_xxstg(xp, gamref)
File "/data/netapp/xfel/products/test/ocelot/adaptors/astra2ocelot.py", line 13, in exact_xp_2_xxstg
u = u/np.linalg.norm(u, 2, 1).reshape((N, 1))
TypeError: norm() takes at most 2 arguments (3 given)
[iagapov@it-hpc-wgs01]/data/netapp/xfel/products/desy/demos/ebeam/flash%

beam file interpolation on transform

starting from get_envelope in
https://github.com/iagapov/desy/blob/master/flash/track_flash.py

put together a library of similar processing functions, including

-> slice emittance and twiss parameters
-> current profile

remove one of this possibility for clarity

rename module common -> math.py
on mac some scripts imports this module instead of standard math module

C:\Users\iagapov\workspace\ocelot/lib/genera/build/genera_libs/undulator.so
platform = win32 os.name = nt
Traceback (most recent call last):
File "C:\Users\iagapov\workspace\ocelot\demos\sr\spatial.py", line 5, in
from ocelot.lib.genera.src.python.radiation import generaSR
File "C:\Users\iagapov\workspace\ocelot\lib\genera\src\python\radiation\generaSR.py", line 6, in
from ocelot.lib.genera.src.python.trajectory.lat_trajectory import trace4radiation
File "C:\Users\iagapov\workspace\ocelot\lib\genera\src\python\trajectory\lat_trajectory.py", line 2, in
from undulator import und_trace
File "C:\Users\iagapov\workspace\ocelot\lib\genera\src\python\trajectory\undulator.py", line 40, in
exec(open(path_to_ocelot + "/lib/genera/src/cpp/compile.py"))
File "C:\Users\iagapov\workspace\ocelot/lib/genera/src/cpp/compile.py", line 73, in
compile(dir_path)
File "C:\Users\iagapov\workspace\ocelot/lib/genera/src/cpp/compile.py", line 33, in compile
p = subprocess.Popen(args, stdout=subprocess.PIPE, stderr=subprocess.STDOUT)
File "C:\Python27\lib\subprocess.py", line 709, in init
errread, errwrite)
File "C:\Python27\lib\subprocess.py", line 957, in _execute_child
startupinfo)
WindowsError: [Error 2] The system cannot find the file specified

scipy.weave is unsupported now. We need to choose standard tool for speeding up some calculations. Maybe cython?
Replace scipy.weave in code.

update_transfer_maps does not seem to work correctly with cavities

so that each script needs only from ocelot import *

ksi_x = -25.994508998
ksi_y = -22.8071219474
Chromatism compensate: Before: {'SF': 1.5, 'SD': -1.5}
Chromatism compensate: After: {'SF': 5.8914775395193395, 'SD': -6.8036102026266407}
0
No module named msvccompiler in numpy.distutils; trying from distutils
creating c:\users\iagapov\appdata\local\temp\scipy-iagapov-sushtr\python27_intermediate\compiler_e3b0c44298fc1c149afbf4c8996fb924
Cannot build msvcr library: "msvcr90d.dll" not found
Cannot build msvcr library: "msvcr90d.dll" not found
No module named msvccompiler in numpy.distutils; trying from distutils
Cannot build msvcr library: "msvcr90d.dll" not found

using expressions
h_eV_s = const.codata.value('Planck constant in eV s')

in common.globals

is not recommended since the value might be installation dependant
and not always verifyable

use explicit values

Consider deleting the line

beam_new.z = beam_new.z[:-i_diff]

in the update_beam function in hxrss_common.py

The number of slices in the radiation file is larger than the original one in the beam file. i_diff is defined as such difference. If you redefine beam_new.z with beam_new.z = beam_new.z[:-i_diff]
you actually cut the original beamfile, ending up with less slices than in the original beamfile. Deleting this line you end up with a number of slices in the beam file equal to the original one.

thus optics for bend and sbend not the same

functions related to undulator parameters need to be moved to appropriate modules.

rename:

phys_ref.pdf --> electron_optics_ref.pdf

rename:

OCELOT Program
Physical Methods Manual

-->>
OCELOT: electron optics physics reference

demos -- to be distributed with code, simple clean examples, on version control
repository -- only related to working machines (xfel), clean, separate version control + xfel.eu website
scripts(work) -- work in progress scripts. Version control? Partial version control?

add temperature parameter to Fhkl lookup

optical functions with cavities on differ from elegant

e.g. something likephase advance e1 ->e2
constr[e2] = {'dmux', pi, e1}

Script enclosed. Bad matching in both planes.

SASE3_issue.txt

and a function to attach solvers according to z range or area.

undulator transfer map order 2 by default. Twiss does not work (standard sase_tdp.py script)

for r56 for drifts and cavities is 0. Should not be 0.

add test for mad-x translation to ocelot.test -- 1 example

in the screen object, after the call info like nx ny etc is removed. Better keep it.

RFcavity is already present as Cavity. Maybe add missing parameters to 'Cavity'.
Maybe calculate harmonic number from

fel.py line 38, check formulae
a = p.aw02 / (2*(2+p.aw02))
etc.
(make sure there's no confusion between k_rms and k_max)

cpbd.beam.get_envelope() should calculate twiss parameters with taking into account dispersion.

in the lower panel where the elements are shown:

1. Show element name and type when the mouse is focused on the element
2. Then the color-coding label (quad, bend, sext) can also be removed

with coupling twiss does not work with error

ocelot/cpbd/optics.py:170: RuntimeWarning: divide by zero encountered in double_scalars
d_mux = arctan(M[0, 1]/(M[0, 0]_m.beta_x - M[0, 1]_m.alpha_x))

this should be changed to some other warning/error indicating that in coupled case decoupled twiss is not correct

test case:

D0 = Drift (l = 5., id = "D0")
D1 = Drift (l = 1., id = "D1")
D2 = Drift (l = 2.0, id = "D2")

Q1 = Quadrupole (l = 0.1, k1 = -0.6, id = "Q1")
Q2 = Quadrupole (l = 0.1, k1 = 0.01, id = "Q2", tilt = pi/4)

sec = (D0, Q1, D1, Q2, D2)

beam = Beam()
beam.E = 6.0
beam.sigma_E = 0.002
beam.emit_xn = 0.4e-6
beam.emit_yn = 0.4e-6
beam.gamma_rel = beam.E / (0.511e-3)
beam.emit_x = beam.emit_xn / beam.gamma_rel
beam.emit_y = beam.emit_yn / beam.gamma_rel
beam.beta_x = 10.0
beam.beta_y = 3.0
beam.alpha_x = 2.0
beam.alpha_y = -2.0

lat = MagneticLattice(sec)
tw0 = Twiss(beam)
tws=twiss(lat, tw0, nPoints = 1000)

following code should work:

navi = Navigator()
while navi.z0 < lat.totalLen:
track(lat, plist, dz=dz, navi=navi, order=1)

navi.reset()
while navi.z0 < lat.totalLen:
track(lat, plist, dz=dz, navi=navi, order=1)

why such name? Maybe rename to 'Marker'?

in optics module

1. Allow for clean inclusion/exclusion of weave code in cpbd
2. check performance on cluster

remove ocelot.gui.file.txt from version control

please have a look if is possible to add fintx parameter to bends
http://frs.home.cern.ch/frs/Xdoc/Introduction/bend.html

This is used for lcls lattice

currently fint1 = fint2 = fint