From 727b04d5b9503ab205437657fe570abcd066f618 Mon Sep 17 00:00:00 2001
From: Marko Mecina <marko.mecina@univie.ac.at>
Date: Mon, 11 Dec 2023 18:13:03 +0100
Subject: [PATCH] flip CCD4 nodes in merged view (ED and FT)
---
.../dataprocessing/smile_raw_ce_converter.py | 52 ++++++++++---------
1 file changed, 27 insertions(+), 25 deletions(-)
diff --git a/Ccs/tools/dataprocessing/smile_raw_ce_converter.py b/Ccs/tools/dataprocessing/smile_raw_ce_converter.py
index 4930f74..5300ed6 100755
--- a/Ccs/tools/dataprocessing/smile_raw_ce_converter.py
+++ b/Ccs/tools/dataprocessing/smile_raw_ce_converter.py
@@ -1,4 +1,4 @@
-#!/usr/bin/env python
+#!/usr/bin/env python3
"""
Convert unprocessed CE raw data to FITS files. Product type is determined (guessed) based on CE size.
@@ -12,8 +12,12 @@ import numpy as np
from astropy.io import fits
# expected CE sizes in bytes
-SIZE_FF = 4511 * 4608 * 2
-SIZE_FT = 639 * 384 * 2 # 1 node
+NROWS_FF = 4511
+NCOLS_FF = 4608
+NROWS_FT = 639
+NCOLS_FT = 384
+SIZE_FF = NROWS_FF * NCOLS_FF * 2
+SIZE_FT = NROWS_FT * NCOLS_FT * 2 # 1 node
SIZE_ED = 64 # 1 event
FILE_PREFIX = 'SMILE_SXI_L1'
@@ -49,7 +53,7 @@ def convert_ce(cefile, fitsfile=None):
def mk_ff(data):
# create uint16 array from raw data and reshape
- arr = np.frombuffer(data, dtype='>H').reshape(4511, -1)
+ arr = np.frombuffer(data, dtype='>H').reshape(NROWS_FF, NCOLS_FF)
fnode = arr[:, ::2]
enode = arr[:, 1::2][:, ::-1]
ff = np.concatenate((fnode, enode), axis=1)
@@ -62,23 +66,16 @@ def mk_ff(data):
def mk_ft(data):
- arr = np.frombuffer(data, dtype='>H').reshape(-1, 639, 384)
+ arr = np.frombuffer(data, dtype='>H').reshape(-1, NROWS_FT, NCOLS_FT)
hdl = _mk_hdl('FT')
for n in range(arr.shape[0]):
hdl.append(fits.ImageHDU(data=arr[n, :, :], name='FT_CCD_NODE_{}'.format(n)))
- # arrange all nodes to full CCD view
+ # arrange all nodes to full CCD
if arr.shape[0] == 4:
- n00 = arr[0, :, :] # CCD2 F-side
- n01 = arr[1, :, ::-1] # CCD2 E-side
- n10 = arr[2, ::-1, :] # CCD4 F-side
- n11 = arr[3, ::-1, ::-1] # CCD4 E-side
- n0 = np.concatenate((n00, n01), 1)
- n1 = np.concatenate((n10, n11), 1)
-
- nn = np.concatenate((n0, n1), 0) # [::-1, :]
+ nn = _assemble_ft_frames_to_fp_view(arr)
hdl.append(fits.ImageHDU(data=nn, name='FULLCCD'))
@@ -87,14 +84,14 @@ def mk_ft(data):
def mk_ed(data):
# reshape into array of evt packets
- arr = np.frombuffer(data, dtype='>H').reshape(-1, 32)
+ arr = np.frombuffer(data, dtype='>H').reshape(-1, SIZE_ED // 2)
hdl = _mk_hdl('FT')
ts = int(hdl['PRIMARY'].header['OBS_ID'])
bindata = np.array([_mk_bin_entry(evt, ts) for evt in arr], dtype=ED_BIN_DTYPE)
# also add an HDU with event map
- nodes = np.zeros((2, 2, 639, 384))
+ nodes = np.zeros((2, 2, NROWS_FT, NCOLS_FT))
nodes[:] = np.nan
for _, _, ccd, col, row, node, fx in bindata:
@@ -103,15 +100,7 @@ def mk_ed(data):
except:
print(col, row, 'FAILED')
- n00 = nodes[0, 0, :, :] # CCD2 F-side
- n01 = nodes[0, 1, :, :][:, ::-1] # CCD2 E-side
- n10 = nodes[1, 0, :, :][::-1, :] # CCD4 F-side
- n11 = nodes[1, 1, :, :][::-1, ::-1] # CCD4 E-side
-
- n0 = np.concatenate((n00, n01), 1)
- n1 = np.concatenate((n10, n11), 1)
-
- ed_img = np.concatenate((n0, n1), 0) # [::-1, :]
+ ed_img = _assemble_ft_frames_to_fp_view(nodes)
hdl.append(fits.ImageHDU(data=ed_img, name='EVTMAP'))
@@ -129,6 +118,19 @@ def _mk_bin_entry(data, timestamp):
return timestamp, fc, ccdnr, col, row, node, evts
+def _assemble_ft_frames_to_fp_view(arr3d):
+
+ n00 = arr3d[0, :, :] # CCD2 F-side (lower left in FP view)
+ n01 = arr3d[1, :, ::-1] # CCD2 E-side (lower right in FP view)
+ n10 = arr3d[2, ::-1, ::-1] # CCD4 F-side (upper right in FP view)
+ n11 = arr3d[3, ::-1, :] # CCD4 E-side (upper left in FP view)
+
+ n0 = np.concatenate((n00, n01), axis=1) # CCD2
+ n1 = np.concatenate((n11, n10), axis=1) # CCD4
+
+ return np.concatenate((n0, n1), axis=0)
+
+
def _mk_hdl(dmode):
hdl = fits.HDUList()
phdu = fits.PrimaryHDU()
--
GitLab