# to run python CIV_CIII_SiOIV_modelling_general.py line_to_fit path_to_data/ file_location_with_list_of_spectra_name # remember to after 'path_to_data' to include the symbol '/'

import matplotlib

matplotlib.use('Agg')

#matplotlib.use('Agg')

from matplotlib import pylab

import astropy.units as un

import numpy as np

import pyspeckit

import os

import sys

import numpy as np

from scipy.interpolate import InterpolatedUnivariateSpline as interpolation

from scipy.interpolate import UnivariateSpline as Interp

from scipy.ndimage.filters import gaussian_filter as gauss_conv

from astropy.cosmology import FlatLambdaCDM

from fitcode.line import *

import json

from matplotlib import rcParams

import time

plot_best_fits=1 # 1 if you want to plot the best fits. to run in cluster must be set to 0

treshold=0.05 #this is the percentage of most negative pixels that are going to be exclude in the fit after the continuum removal

# this procedure is done iteratively for 3 times.

#For low-z quasars with little absortion treshold=0.03 is ok (z<3). For high-z quasars treshold=0.05 would be better

use_exclude_files=0 # 0 if you want the code to fit the lines witouth using manually selected continuum windwos and manually selected

#absorption features. The files should be in the folder ./excludes/ with the following filenames:

# exclude_cont_CIV.txt and exclude_CIV.txt. Replace CIV for the name of other lines

# you want to fit e.g. CIII, SiOIV, Lya.

rcParams['text.usetex'] = True #Very important to force python to recognize Latex code

#------Executing the configuration file that includes the requiered parameters---#

execfile("./constraints_single.cfg")

#--- If you want to change width limits, line center limits, add gaussian components, this is the file to configure---#

if len(sys.argv)<4:

data_dir='./data/spectra/' #dir related with the spectra that will be fit

fn='./data/filenames.txt' #file with the list of name of the spectra located in data_dir

else:

data_dir=sys.argv[2]

fn=sys.argv[3]

filenames=np.genfromtxt(fn,unpack=1,dtype='str')

if len(sys.argv)<2:

line_to_fit='CIV'

else:

line_to_fit=sys.argv[1]

print line_to_fit

dataout_path=data_dir+'fit/'

plots_path= data_dir+'plots/'

xmin=1100

xmax=9000

n_spec=1

if not os.path.exists(dataout_path):

os.mkdir(dataout_path)

if not os.path.exists(plots_path):

os.mkdir(plots_path)

#pylab.ioff()

sample_dictionary={}

try:

if os.path.isfile( dataout_path + "sample_dictionary.json"):

with open(dataout_path + 'sample_dictionary.json', 'rb') as fp:

sample_dictionary=json.load(fp)

except:

pass

if line_to_fit in ['CIV','CIII','SiOIV','Lya','C']:

spec_division=['FUV']

if line_to_fit=='CIV':

xminc=1450

xmaxc=1723

if use_exclude_files:

exclude_file = "./excludes/exclude_cont_CIV.txt"

exclude_conti=np.loadtxt(exclude_file,skiprows=2)

exclude_cont=[1000,1445,1465,1695,1725,10000]

xminc1=exclude_cont[1]

xmaxc1=exclude_cont[-2]

if line_to_fit=='CIII':

xminc=1678

xmaxc=2017

exclude_cont=[1000,1695,1725,1960,2020,10000]

if use_exclude_files:

exclude_file = "./excludes/exclude_cont_CIII.txt"

exclude_conti=np.loadtxt(exclude_file,skiprows=2)

xminc1=exclude_cont[1]

xmaxc1=exclude_cont[-2]

if line_to_fit=='SiOIV':

xminc=1320

xmaxc=1480

if use_exclude_files:

exclude_file = "./excludes/exclude_cont_SiOIV.txt"

exclude_conti=np.loadtxt(exclude_file,skiprows=2)

exclude_cont=[1000,1340,1360,1430,1460,10000]

xminc1=exclude_cont[1]

xmaxc1=exclude_cont[-2]

if line_to_fit=='Lya':

print 'fitting Ly-alpha'

xminc=1100

xmaxc=1480

if use_exclude_files:

exclude_file = "./excludes/exclude_cont_Lya.txt"

exclude_conti=np.loadtxt(exclude_file,skiprows=2)

exclude_cont=[1000,1270,1280,1420,1460,10000]

xminc1=exclude_cont[1]

xmaxc1=exclude_cont[-2]

if line_to_fit=='C':

xminc=1450

xmaxc=2017

if use_exclude_files:

exclude_file = "./excludes/exclude_cont_C.txt"

exclude_conti=np.loadtxt(exclude_file,skiprows=2)

exclude_cont=[1000,1430,1460,1960,2020,10000]

xminc1=exclude_cont[1]

xmaxc1=exclude_cont[-2]

if line_to_fit in ['Halpha','Hbeta']:

spec_division=['OP']

if line_to_fit=='Halpha':

print 'Halpha'

xminc=6000.0

xmaxc=7000.0

if use_exclude_files:

exclude_file = "./excludes/exclude_cont_Halpha.txt"

exclude_conti=np.loadtxt(exclude_file,skiprows=2)

exclude_cont=[1000,6198,6215,6880,6920,10000]

xminc1=exclude_cont[1]

xmaxc1=exclude_cont[-2]

try:

sp_to_run=range(len(filenames))

except:

filenames=np.array([str(filenames)])

print 'only one file'

sp_to_run=[0]

#sp_to_run=[0]

for i in sp_to_run:

t0=time.time()

fileroot= filenames[i]

spectrum_file=file=data_dir+fileroot

print 'fitting ',line_to_fit,' in ', fileroot,'\n\n\n\n'

object_dictionary={}

object_dictionary1={}

plot_objpath=plots_path

if not os.path.exists(plot_objpath):

os.mkdir(plot_objpath)

sp = pyspeckit.Spectrum(spectrum_file)

if use_exclude_files:

exclude_cont=exclude_conti[i,:][:]

xminc1=exclude_cont[1]

xmaxc1=exclude_cont[-2]

print xminc1, xmaxc1, 'cont limits!!!!'

if line_to_fit=='Lya':

xminc1=1100

xmaxc1=1460

argxmin=np.argmin(np.abs(sp.xarr.value-xminc1))

argxmax=np.argmin(np.abs(sp.xarr.value-xmaxc1))

try:

sp.crop(argxmin,argxmax)

except:

continue

continuous=0.0*sp.data

CIV_fit=0.0*sp.data

CIII_fit=0.0*sp.data

C_fit=0.0*sp.data

SiOIV_fit=0.0*sp.data

Lya_fit=0.0*sp.data

Halpha_fit=0.0*sp.data

#------------opening spectrum i---------------------#

# -----------set up units properly------------------#

try:

mag_order=np.int((1)*np.round(np.log10(np.mean(sp.data))))

except:

continue

sp.xarr.unit==un.Angstrom

#sp.xarr.units='angstrom'

sp.xarr.xtype = 'wavelength'

sp.unit = r'$10^{'+str(mag_order)+'}$ erg s$^{-1}$ $\AA^{-1}$'

sp.unit = r'$10^{'+str(mag_order)+'}$ erg s$^{-1}$ $\AA^{-1}$'

sp.data = 10**(-1.0*mag_order)*np.array(sp.data.tolist())

sp.error= 10**(-1.0*mag_order)*np.array(sp.error.tolist())

#-------------- set up unit properly------------#

copy=sp.copy()

if copy.error.mean()==0:

copy.error=copy.data.mean()*0.1*copy.data

#-------------continuous fitting-------------------#

wlmin=sp.xarr[0]

wlmax=sp.xarr[-1]

wlimin_FUV=xminc

#exclude_cont=np.loadtxt(exclude_file,skiprows=2)

#exclude_cont=exclude_cont[i,:][:]

backup=sp.copy()

#backup.plotter(xmin=xminc1,xmax=xmaxc1)

backup.baseline.powerlaw=False

backup.baseline(xmin=xminc1, xmax=xmaxc1, exclude=exclude_cont, subtract=False, reset_selection=False, highlight_fitregion=False,powerlaw=False,quiet=True,LoudDebug=False,annotate=False)

continuous=backup.baseline.basespec

continuous_FUV=continuous

cut=np.percentile(continuous/sp.data,99.5)

ratio=continuous/sp.data

pylab.rcParams["figure.figsize"]=16,8

#sp.data[:index]=sp.data[:index] - balmer_template.data[:index]

#continuous,wlmin_FUV,L_model=continuous_substraction( i, sp, mag_order,FUV_limits,w)

#-------------continuous subtraction-------------------#

argmax=np.argmax(sp.data)

posmax=sp.xarr.value[argmax]

fluxmax=sp.data[argmax]

dv=10000 #km/s maximum negative velocity to look for BAL features

dv0=0 #km/s miminum negative velocity to look for BAL features

dv1=20000

dv2=2000

dv3=1000

dv4=-1000

dv5=3000

dv6=-3000

dv7=20000

dv8=10000

dx=dv*posmax/3e5

dx0=dv0*posmax/3e5

dx1=dv1*posmax/3e5

dx2=dv2*posmax/3e5

dx3=dv3*posmax/3e5

dx4=dv4*posmax/3e5

dx5=dv5*posmax/3e5

dx6=dv6*posmax/3e5

dx7=dv7*posmax/3e5

dx8=dv8*posmax/3e5

poslim=posmax-dx

poslim0=posmax-dx0

poslim1=posmax-dx1

poslim2=posmax-dx2

poslim3=posmax-dx3

poslim4=posmax-dx4

poslim5=posmax-dx5

poslim6=posmax-dx6

poslim7=posmax-dx7

poslim8=posmax-dx8

arglim=np.argmin(np.abs(sp.xarr.value-poslim))

arglim0=np.argmin(np.abs(sp.xarr.value-poslim0))

arglim1=np.argmin(np.abs(sp.xarr.value-poslim1))

arglim2=np.argmin(np.abs(sp.xarr.value-poslim2))

arglim3=np.argmin(np.abs(sp.xarr.value-poslim3))

arglim4=np.argmin(np.abs(sp.xarr.value-poslim4))

arglim5=np.argmin(np.abs(sp.xarr.value-poslim5))

arglim6=np.argmin(np.abs(sp.xarr.value-poslim6))

arglim7=np.argmin(np.abs(sp.xarr.value-poslim7))

arglim8=np.argmin(np.abs(sp.xarr.value-poslim8))

sp.data=sp.data - continuous_FUV

SN=np.median(copy.data/sp.error)

print 'signal to noise ratio =',SN

if line_to_fit=='CIV':

limits_CIV[3]=(0,sp.data.max()) #Defining CIV ampplitud limits

limits_CIV[0]=(0,sp.data.max()/3.0) #Defining CIV ampplitud limits

limited_CIV[0]=(True,True) #Defining CIV ampplitud limits

guesses_CIV[3]=sp.data.max()

guesses_CIV[6]=sp.data.max()/3.0

guesses_CIV[9]=sp.data.max()

guesses_CIV[12]=sp.data.max()/3.0

if use_exclude_files:

exclude_cont=exclude_conti[i,:][:]

#exclude=[1489.2,1495.6,1497.73,1518.9,1521.0,1532.3]

#CIV_fit,object_dictionary['CIV_complex']=line_fitter(sp, "CIV", i,guesses_CIV, limits_CIV, limited_CIV, tied_CIV, xminc, xmaxc,magorder=mag_order,plot_path=plot_objpath, linenames=lines_dict['CIV_complex'],exclude=exclude, excluding=use_exclude_files, do_fit=fit_CIV)

CIV_fit1,object_dictionary1['CIV_complex']=line_fitter(sp, "CIV", i,guesses_CIV, limits_CIV, limited_CIV, tied_CIV, xminc1, xmaxc1,magorder=mag_order,plot_path=plot_objpath, linenames=lines_dict['CIV_complex'], excluding=use_exclude_files, do_fit=fit_CIV)

chi2_1=sp.specfit.chi2/sp.specfit.dof

dof_1=sp.specfit.dof

f=copy.data/continuous

dAIt=(1.0-f)

dAIc=(1.0-f)[arglim:arglim0]

dAIw=(1.0-f)[arglim1:arglim2]

dAIww=(1.0-f)[arglim7:arglim8]

threshold1=np.percentile(dAIt,97.0)

wAIc=dAIc>threshold1

wAIw=dAIw>threshold1

wAIww=dAIww>threshold1

nabs=1.0*len(dAIc[wAIc])

nblue=1.0*len(dAIc)

try:

rat_flux=nabs/nblue # fraction of points in the lower percentile of the relative flux (normalized wr to the continuum) between 0 and -10000km/s w/r to the line peak

except:

rat_flux=0

nabs=1.0*len(dAIw[wAIw])

nblue=1.0*len(dAIw)

try:

rat_fluxw=nabs/nblue # fraction of points in the lower percentile of the relative flux (normalized wr to the continuum) between -2000 and -20000km/s w/r to the line peak

except:

rat_fluxw=0

nabs=1.0*len(dAIww[wAIww])

nblue=1.0*len(dAIww)

try:

rat_fluxww=nabs/nblue # fraction of points in the lower percentile of the relative flux (normalized wr to the continuum) between -2000 and -20000km/s w/r to the line peak

except:

rat_fluxww=0

print 'flux fraction core=',rat_flux,'nobject=',i

print 'fux fraction wings=',rat_fluxw

#BI=

chi2=chi2_1

dof=dof_1

CIV_fit=CIV_fit1

object_dictionary['CIV_complex']=object_dictionary1['CIV_complex']

print 'chi2=',chi2_1

#if chi2>2.0:

# continue

sp=copy.copy()

sp1=copy.copy()

for j in range(1,4):

#if use_exclude_files:

# continue

sp1=copy.copy()

residuals=sp1.data-(continuous+CIV_fit)

delta=treshold

d=(1.0-delta)**(j)

threshold=np.percentile(residuals,(1-d)*100)

wlow=residuals<threshold #finding the most negative residuals. 3 iterations each time removing (1-0.97^j)*100 percent of the points

sp1.error[wlow]=np.inf

#-----------------testing lines-------------------------#

#backup=sp1.copy()

#backup.plotter(xmin=xminc1,xmax=xmaxc1)

#backup.baseline.powerlaw=False

#backup.baseline(xmin=xminc1, xmax=xmaxc1, exclude=exclude_cont, subtract=False, reset_selection=False, highlight_fitregion=False,powerlaw=False,quiet=True,LoudDebug=False,annotate=False)

#continuous=backup.baseline.basespec

continuous_FUV=continuous

#-----------------testing lines-------------------------#

sp1.data=sp1.data - continuous

#sp1 = pyspeckit.Spectrum(xarr=sp.xarr[wlow],data=sp.data[wlow],error=sp.error[wlow])

CIV_fit1,object_dictionary1['CIV_complex']=line_fitter(sp1, "CIV", i,guesses_CIV, limits_CIV, limited_CIV, tied_CIV, xminc1, xmaxc1,magorder=mag_order,plot_path=plot_objpath, linenames=lines_dict['CIV_complex'], excluding=use_exclude_files, do_fit=fit_CIV)

chi2=sp1.specfit.chi2/sp1.specfit.dof

dof=sp1.specfit.dof

if chi2_1<chi2:

chi2=chi2_1

dof=dof_1

break

chi2_1=chi2

dof_1=dof

CIV_fit=CIV_fit1

object_dictionary['CIV_complex']=object_dictionary1['CIV_complex']

print 'chi2=',chi2

sp=copy.copy()

residuals=sp.data-(continuous+CIV_fit)

resibal=residuals[arglim:arglim0]

resicore=residuals[arglim3:arglim4]

#resicore1=residuals[arglim5:arglim6]

ratios=sp.data/(continuous+CIV_fit)

ratcore=ratios[arglim5:arglim6]

threshold1=np.percentile(residuals,3.0)

threshold2=np.percentile(residuals,97.0)

threshold3=np.percentile(ratios,3.0)

wlowbal=resibal<threshold1

wlowcore=ratcore<threshold3

wupcore=resicore>threshold2

nabs=1.0*len(resibal[wlowbal])

nabscore=1.0*len(ratcore[wlowcore])

nupcore=1.0*len(resicore[wupcore])

ncore=1.0*len(resicore)

nblue=1.0*len(resibal)

ncore1=1.0*len(ratios)

try:

rat_absw=nabs/nblue # fraction of points in the lowest 3 percentile of the residuals (most negative residualts) in the blue wing between 0 and -10000km/s w/r to the line peak

except:

rat_absw=0

try:

rat_res=nupcore/ncore # fraction of points in the higheest 3 percentile of the residuals(most positive residuals) in the line core between -1000km and -1000km/s w/r to the line peak

except:

rat_res=0

try:

rat_absc=nabscore/ncore1

except:

rat_absc=0

print 'abs fraction=',rat_absw,'nobject=',i

print 'res fraction=',rat_res

print 'abs core fraction=',rat_absc,'nobject=',i

#if use_exclude_files==0:

sp1=copy.copy()

sp1.error[wlow]=np.inf

#-----------------testing lines-------------------------#

backup=sp1.copy()

#backup.plotter(xmin=xminc1,xmax=xmaxc1)

#backup.baseline.powerlaw=False

#backup.baseline(xmin=xminc1, xmax=xmaxc1, exclude=exclude_cont, subtract=False, reset_selection=False, highlight_fitregion=False,powerlaw=False,quiet=True,LoudDebug=False,annotate=False)

#continuous=backup.baseline.basespec

continuous_FUV=continuous

#-----------------testing lines-------------------------#

sp1.data=sp.data - continuous

CIV_fit1,object_dictionary1['CIV_complex']=line_fitter(sp1, "CIV", i,guesses_CIV, limits_CIV, limited_CIV, tied_CIV, xminc1, xmaxc1,magorder=mag_order,plot_path=plot_objpath, linenames=lines_dict['CIV_complex'], excluding=use_exclude_files, do_fit=fit_CIV)

chi2_1=sp1.specfit.chi2/sp1.specfit.dof

dof_1=sp1.specfit.dof

if chi2_1<chi2:

chi2=chi2_1

dof=dof_1

CIV_fit=CIV_fit1

object_dictionary['CIV_complex']=object_dictionary1['CIV_complex']

if line_to_fit=='CIII':

if use_exclude_files:

exclude_cont=exclude_conti[i,:][:]

CIII_fit,object_dictionary['CIII_complex']=line_fitter(sp, "CIII", i,guesses_CIII, limits_CIII, limited_CIII, tied_CIII, xminc1, xmaxc1,magorder=mag_order,plot_path=plot_objpath, linenames=lines_dict['CIII_complex'], excluding=use_exclude_files, do_fit=fit_CIII)

chi2=sp.specfit.chi2/sp.specfit.dof

dof=sp.specfit.dof

print 'chi2=',chi2

print 'nobject=',i

if line_to_fit=='SiOIV':

if use_exclude_files:

exclude_cont=exclude_conti[i,:][:]

SiOIV_fit1,object_dictionary1['SiOIV_complex']=line_fitter(sp, "SiOIV1", i,guesses_SiOIV1, limits_SiOIV1, limited_SiOIV1, tied_SiOIV1, xminc1, xmaxc1,magorder=mag_order,plot_path=plot_objpath, linenames=lines_dict['SiOIV_complex'], excluding=use_exclude_files, do_fit=fit_Si)

chi2_1=sp.specfit.chi2/sp.specfit.dof

dof_1=sp.specfit.dof

f=copy.data/continuous

dAIt=(1.0-f)

dAIc=(1.0-f)[arglim:arglim0]

dAIw=(1.0-f)[arglim1:arglim2]

threshold1=np.percentile(dAIt,97.0)

wAIc=dAIc>threshold1

wAIw=dAIw>threshold1

nabs=1.0*len(dAIc[wAIc])

nblue=1.0*len(dAIc)

try:

rat_flux=nabs/nblue # fraction of points in the lower percentile of the relative flux (normalized wr to the continuum) between 0 and -10000km/s w/r to the line peak

except:

rat_flux=0

nabs=1.0*len(dAIw[wAIw])

nblue=1.0*len(dAIw)

try:

rat_fluxw=nabs/nblue # fraction of points in the lower percentile of the relative flux (normalized wr to the continuum) between -2000 and -20000km/s w/r to the line peak

except:

rat_fluxw=0

print 'flux fraction core=',rat_flux,'nobject=',i

print 'fux fraction wings=',rat_fluxw

#BI=

chi2=chi2_1

dof=dof_1

SiOIV_fit=SiOIV_fit1

object_dictionary['SiOIV_complex']=object_dictionary1['SiOIV_complex']

print 'chi2=',chi2_1

#if chi2>2.0:

# continue

sp=copy.copy()

sp1=copy.copy()

for j in range(1,4):

sp1=copy.copy()

residuals=sp1.data-(continuous+SiOIV_fit)

delta=0.03

d=(1.0-delta)**(j)

threshold=np.percentile(residuals,(1-d)*100)

wlow=residuals<threshold #finding the most negative residuals. 3 iterations each time removing (1-0.97^j)*100 percent of the points

sp1.data=sp1.data - continuous

sp1.error[wlow]=np.inf

#sp1 = pyspeckit.Spectrum(xarr=sp.xarr[wlow],data=sp.data[wlow],error=sp.error[wlow])

SiOIV_fit1,object_dictionary1['SiOIV_complex']=line_fitter(sp1,"SiOIV1", i,guesses_SiOIV1, limits_SiOIV1, limited_SiOIV1, tied_SiOIV1, xminc1, xmaxc1, magorder=mag_order,plot_path=plot_objpath, linenames=lines_dict['SiOIV_complex'], do_fit=fit_Si)

chi2=sp1.specfit.chi2/sp1.specfit.dof

dof=sp1.specfit.dof

if chi2_1<chi2:

chi2=chi2_1

dof=dof_1

break

chi2_1=chi2

dof_1=dof

SiOIV_fit=SiOIV_fit1

object_dictionary['SiOIV_complex']=object_dictionary1['SiOIV_complex']

print 'chi2=',chi2

sp=copy.copy()

residuals=sp.data-(continuous+SiOIV_fit)

resibal=residuals[arglim:arglim0]

resicore=residuals[arglim3:arglim4]

threshold1=np.percentile(residuals,3.0)

threshold2=np.percentile(residuals,97.0)

wlowbal=resibal<threshold1

wupcore=resicore>threshold2

nabs=1.0*len(resibal[wlowbal])

nupcore=1.0*len(resicore[wupcore])

ncore=1.0*len(resicore)

nblue=1.0*len(resibal)

try:

rat_absw=nabs/nblue # fraction of points in the lowest 3 percentile of the residuals (most negative residualts) in the blue wing between 0 and -10000km/s w/r to the line peak

except:

rat_absw=0

try:

rat_res=nupcore/ncore # fraction of points in the higheest 3 percentile of the residuals(most positive residuals) in the line core between -1000km and -1000km/s w/r to the line peak

except:

rat_res=0

print 'abs fraction=',rat_absw,'nobject=',i

print 'res fraction=',rat_res

sp1.data=sp.data - continuous

sp1.error[wlow]=np.inf

SiOIV_fit,object_dictionary['SiOIV_complex']=line_fitter(sp1,"SiOIV1", i,guesses_SiOIV1, limits_SiOIV1, limited_SiOIV1, tied_SiOIV1, xminc1, xmaxc1, magorder=mag_order,plot_path=plot_objpath, linenames=lines_dict['SiOIV_complex'], do_fit=fit_Si)

chi2_1=sp1.specfit.chi2/sp1.specfit.dof

dof_1=sp1.specfit.dof

if chi2_1<chi2:

chi2=chi2_1

dof=dof_1

SiOIV_fit=SiOIV_fit1

object_dictionary['SiOIV_complex']=object_dictionary1['SiOIV_complex']

"""

SiOIV_fit,object_dictionary['SiOIV_complex']=line_fitter(sp,"SiOIV1", i,guesses_SiOIV1, limits_SiOIV1, limited_SiOIV1, tied_SiOIV1, xminc, xmaxc, magorder=mag_order,plot_path=plot_objpath, linenames=lines_dict['SiOIV_complex'], do_fit=fit_Si)

chi2=sp.specfit.chi2/sp.specfit.dof

dof=sp.specfit.dof

print 'chi2=',chi2

if chi2>2.0:

continue

"""

if line_to_fit=='Lya':

if use_exclude_files:

exclude_cont=exclude_conti[i,:][:]

Lya_fit,object_dictionary['Lya_complex']=line_fitter(sp,"Lya", i,guesses_Lya, limits_Lya, limited_Lya, tied_Lya, xminc1, xmaxc1, magorder=mag_order,plot_path=plot_objpath, linenames=lines_dict['Lya_complex'], do_fit=fit_Si)

chi2=sp.specfit.chi2/sp.specfit.dof

dof=sp.specfit.dof

print 'chi2=',chi2

if line_to_fit=='C':

if use_exclude_files:

exclude_cont=exclude_conti[i,:][:]

C_fit,object_dictionary['C_complex']=line_fitter(sp, "C", i,guesses_C, limits_C, limited_C, tied_C, xminc1, xmaxc1,magorder=mag_order,plot_path=plot_objpath, linenames=lines_dict['C_complex'], excluding=use_exclude_files, do_fit=fit_C)

chi2=sp.specfit.chi2/sp.specfit.dof

dof=sp.specfit.dof

print 'chi2=',chi2

if line_to_fit=='Halpha':

if use_exclude_files:

exclude_cont=exclude_conti[i,:][:]

Halpha_fit,object_dictionary['Halpha_complex']=line_fitter(sp,"Halpha", i, guesses_Halpha, limits_Halpha, limited_Halpha, tied_Halpha,

xmin_Halpha, xmax_Halpha, magorder=mag_order,plot_path=plot_objpath,

linenames=lines_dict['Halpha_complex'], do_fit=fit_Halpha)

chi2=sp.specfit.chi2/sp.specfit.dof

dof=sp.specfit.dof

print 'chi2=',chi2

#del(number,mag,redshift)

if fileroot in sample_dictionary:

try:

sample_dictionary[fileroot]['model'][line_to_fit+'_complex']['lines']=object_dictionary[line_to_fit+'_complex']

print 'opened'

except:

sample_dictionary[fileroot]['model'][line_to_fit+'_complex']={}

sample_dictionary[fileroot]['model'][line_to_fit+'_complex']['lines']={}

sample_dictionary[fileroot]['model'][line_to_fit+'_complex']['lines']=object_dictionary[line_to_fit+'_complex']

print fileroot, i

else:

print fileroot, ' does not exists'

sample_dictionary[fileroot]={}

sample_dictionary[fileroot]['model']={}

sample_dictionary[fileroot]['model'][line_to_fit+'_complex']={}

sample_dictionary[fileroot]['model'][line_to_fit+'_complex']['lines']={}

sample_dictionary[fileroot]['model'][line_to_fit+'_complex']['lines']=object_dictionary[line_to_fit+'_complex']

sp = pyspeckit.Spectrum(spectrum_file)

sp.crop(argxmin,argxmax)

#sp.data=sp.data[~wlow]

#sp.error=sp.error[~wlow]

#sp.xarr=sp.xarr[~wlow]

#continuous=continuous[~wlow]

# -----------set up unit properly------------------#

mag_order=np.int((1)*np.round(np.log10(np.mean(sp.data))))

sp.xarr.units='angstroms'

sp.xarr.xtype = 'wavelength'

sp.unit = r'$10^{'+str(mag_order)+'}$ erg s$^{-1}$ $\AA^{-1}$'

#sp.units = r'$10^{'+str(mag_order)+'}$ erg s$^{-1}$ $\AA^{-1}$'

sp.data = 10**(-1.0*mag_order)*np.array(sp.data.tolist())

sp.error= 10**(-1.0*mag_order)*np.array(sp.error.tolist())

#-------------- set up unit properly------------#

#----Superposition plot --------#

total=(continuous + CIV_fit + C_fit +

+ SiOIV_fit +CIII_fit+Lya_fit+Halpha_fit)

total_lines=( CIV_fit + C_fit +

+ SiOIV_fit +CIII_fit+Lya_fit+Halpha_fit)

plot_file=plot_objpath + line_to_fit + "_"+ fileroot.split('.txt')[0] + ".png"

if plot_best_fits:

pylab.rcParams["figure.figsize"]=16,6

copy1=sp.copy()

argxmin=np.argmin(np.abs(sp.xarr.value-xminc))

argxmax=np.argmin(np.abs(sp.xarr.value-xmaxc))

copy1.crop(argxmin,argxmax)

pylab.figure()

pylab.ylim(ymin=0,ymax=1.1*copy1.data.max())

pylab.xlim(xmin=xminc-100.0,xmax=xmaxc+100.0)

#pylab.ylabel(r'$10^{'+str(mag_order-8)+'}$ erg s$^{-1}$ $\AA^{-1}$')

#pylab.xlabel(r'$\AA$')

pylab.yscale('linear')

try:

pylab.plot(sp.xarr,sp.data,'k',label=' fluxcenter rat='+str(np.round(rat_flux,2))

+' fluxwingfar rat='+str(np.round(rat_fluxww,2)) +' fluxwing rat='+str(np.round(rat_fluxw,2)) + ' SN='+str(np.round(SN,2))+'chi2='+str(np.round(chi2,2))+' abs rat='+str(np.round(rat_absw,2))+' exccess rat='+str(np.round(rat_res,2)) +' abs rat core='+str(np.round(rat_absc,2)))

except:

pylab.plot(sp.xarr,sp.data,'k')

pylab.plot(sp.xarr, total,'r')

pylab.plot(sp.xarr, total_lines,'r')

total=continuous

pylab.plot(sp.xarr, total,'gray')

#pylab.legend(fontsize=10)

#sp.plotter.figure.savefig(plot_file,format='pdf', dpi=600,bbox_inches='tight')

pylab.savefig(plot_file)

pylab.close('all')

#----Superposition plot --------#

obj_path=dataout_path+fileroot + '/'

if not os.path.exists(obj_path):

os.mkdir(obj_path)

np.savetxt(obj_path +'obs_spectrum.txt' ,np.transpose([sp.xarr,sp.data]))

if line_to_fit=='CIV':

np.savetxt(obj_path +'CIV_fit.txt' ,np.transpose([sp.xarr,CIV_fit]) )

np.savetxt(obj_path +'continuous_CIV.txt' ,np.transpose([sp.xarr,continuous]))

if line_to_fit=='CIII':

np.savetxt(obj_path +'CIII_fit.txt' ,np.transpose([sp.xarr,CIII_fit]) )

np.savetxt(obj_path +'continuous_CIII.txt' ,np.transpose([sp.xarr,continuous]))

if line_to_fit=='SiOIV':

np.savetxt(obj_path +'SiOIV_fit.txt' ,np.transpose([sp.xarr,SiOIV_fit]) )

np.savetxt(obj_path +'continuous_SiOIV.txt' ,np.transpose([sp.xarr,continuous]))

if line_to_fit=='Lya':

np.savetxt(obj_path +'Lya_fit.txt' ,np.transpose([sp.xarr,Lya_fit]) )

np.savetxt(obj_path +'continuous_Lya.txt' ,np.transpose([sp.xarr,continuous]))

if line_to_fit=='Halpha':

np.savetxt(obj_path +'Halpha_fit.txt' ,np.transpose([sp.xarr,Halpha_fit]) )

np.savetxt(obj_path +'continuous_Halpha.txt' ,np.transpose([sp.xarr,continuous]))

#np.savetxt(obj_path +'H_fit.txt' ,np.transpose([sp.xarr,H_fit]) )

sample_dictionary[fileroot]['model']

if line_to_fit=='CIV':

sample_dictionary[fileroot]['model']['CIV_complex']['datafile']= obj_path +'CIV_fit.txt'

sample_dictionary[fileroot]['model']['CIV_complex']['continuous']={}

sample_dictionary[fileroot]['model']['CIV_complex']['continuous']['datafile']= obj_path + 'continuous_CIV.txt'

sample_dictionary[fileroot]['model']['CIV_complex']['chi2']= chi2

sample_dictionary[fileroot]['model']['CIV_complex']['dof']= dof

sample_dictionary[fileroot]['model']['CIV_complex']['abs_ratio']=rat_absw

sample_dictionary[fileroot]['model']['CIV_complex']['abs_core_ratio']=rat_absc

sample_dictionary[fileroot]['model']['CIV_complex']['res_ratio']=rat_res

sample_dictionary[fileroot]['model']['CIV_complex']['core_ratio']=rat_flux

sample_dictionary[fileroot]['model']['CIV_complex']['wing_ratio']=rat_fluxw

sample_dictionary[fileroot]['model']['CIV_complex']['wing_far_ratio']=rat_fluxww

sample_dictionary[fileroot]['model']['CIV_complex']['SN']= SN

if line_to_fit=='CIII':

sample_dictionary[fileroot]['model']['CIII_complex']['datafile']= obj_path +'CIII_fit.txt'

sample_dictionary[fileroot]['model']['CIII_complex']['continuous']={}

sample_dictionary[fileroot]['model']['CIII_complex']['continuous']['datafile']= obj_path + 'continuous_CIII.txt'

sample_dictionary[fileroot]['model']['CIII_complex']['continuous']['chi2']= chi2

sample_dictionary[fileroot]['model']['CIII_complex']['continuous']['dof']= dof

if line_to_fit=='SiOIV':

sample_dictionary[fileroot]['model']['SiOIV_complex']['datafile']= obj_path +'SiOIV_fit.txt'

sample_dictionary[fileroot]['model']['SiOIV_complex']['continuous']={}

sample_dictionary[fileroot]['model']['SiOIV_complex']['continuous']['datafile']= obj_path + 'continuous_SiOIV.txt'

sample_dictionary[fileroot]['model']['SiOIV_complex']['continuous']['chi2']= chi2

sample_dictionary[fileroot]['model']['SiOIV_complex']['continuous']['dof']= dof

sample_dictionary[fileroot]['model']['SiOIV_complex']['abs_ratio']=rat_absw

sample_dictionary[fileroot]['model']['SiOIV_complex']['res_ratio']=rat_res

sample_dictionary[fileroot]['model']['SiOIV_complex']['core_ratio']=rat_flux

sample_dictionary[fileroot]['model']['SiOIV_complex']['wing_ratio']=rat_fluxw

sample_dictionary[fileroot]['model']['SiOIV_complex']['SN']= SN

if line_to_fit=='Lya':

sample_dictionary[fileroot]['model']['Lya_complex']['datafile']= obj_path +'Lya_fit.txt'

sample_dictionary[fileroot]['model']['Lya_complex']['continuous']={}

sample_dictionary[fileroot]['model']['Lya_complex']['continuous']['datafile']= obj_path + 'continuous_Lya.txt'

sample_dictionary[fileroot]['model']['Lya_complex']['continuous']['chi2']= chi2

sample_dictionary[fileroot]['model']['Lya_complex']['continuous']['dof']= dof

if line_to_fit=='Halpha':

sample_dictionary[fileroot]['model']['Halpha_complex']['datafile']= obj_path +'Halpha_fit.txt'

sample_dictionary[fileroot]['model']['Halpha_complex']['continuous']={}

sample_dictionary[fileroot]['model']['Halpha_complex']['continuous']['datafile']= obj_path + 'continuous_Halpha.txt'

sample_dictionary[fileroot]['model']['Halpha_complex']['continuous']['chi2']= chi2

sample_dictionary[fileroot]['model']['Halpha_complex']['continuous']['dof']= dof

sample_dictionary[fileroot]['obs_spectrum']= obj_path +'obs_spectrum.txt'

with open(obj_path + line_to_fit +'_dictionary.json', 'wb') as fp:

#json.dump(object_dictionary[line_to_fit+'_complex'], fp)

json.dump(sample_dictionary[fileroot]['model'][line_to_fit+'_complex'], fp)

with open(dataout_path + 'sample_dictionary.json', 'wb') as fp:

json.dump(sample_dictionary, fp)

t1=time.time()

print 'elapsed time=', t1-t0

with open(dataout_path + 'sample_dictionary.json', 'wb') as fp:

json.dump(sample_dictionary, fp)

#with open('data.json', 'rb') as fp:

# data = json.load(fp)