diff --git a/dataFilter.py b/dataFilter.py
index 7f4a9d2c90eae390e326beee972c087a83db7ec1..8bda05b28190995996f0278b814f0eef78b4ddf0 100755
--- a/dataFilter.py
+++ b/dataFilter.py
@@ -3,7 +3,7 @@
# Todo:
# > use global vars for Col-indices. each function that changes em updates em
# > compat /w ecoTrack ? 1nt3gr4t3.
-# > think about putting blox-config from dataFilter rand dataSync in project-globl conf-file!
+# > think about putting blox-config from dataFilter and dataSync in project-global conf-file!
#
import sys, os
diff --git a/dataSync.py b/dataSync.py
index f6b7220d6de0edb3876eb9c6956e51268e683d10..093e18189ce1818a9ae6b86960436b6075a185d1 100755
--- a/dataSync.py
+++ b/dataSync.py
@@ -1,22 +1,20 @@
#!/usr/bin/python3
# clean, sort, merge raw data
-#
+# ecolux and ecotrack Projects
# output one huge table with all necessary fields for further processing/vizualization
#
-# files are 'skyglow.log','tags' and trackfiles (containing 'unit-x' in name, where x is a number)
-#
############################
# TODO
############################
+# > rewrite getXData to initXData
+# > add block name to blox. > add it to update-tags.sh
+#
# > parse moonmap and skyglowdict from skyglow Project sources
# -> maybe integrate skyglow project as submodule
# > integrate with ST
# > make Skyglow optional -> TEST
# > ST-Filenames should include 'unit-X'!
-#
# > how to process skylog error in merge()
-# > ECOTRACK: include temperature data. Also humidity??
-#
# > Speed Up! (takes 2min per 1mio lines)
# 1. Use Threads for
# https://realpython.com/python-concurrency/#multiprocessing-version
@@ -40,21 +38,29 @@ blox=[
("ecolux","./block-I/01-raw","./block-I/02-inspect","2020-07-21 00:00:00","2020-08-18 23:59:00"), # Block1: 21.7 - 18.8
("ecolux","./block-II/01-raw","./block-II/02-inspect","2020-09-15 00:00:00","2020-10-13 23:59:00"), # Block2: 15.9 - 13.10
("ecolux","./block-all/01-raw","./block-all/02-inspect","2020-07-21 00:00:00","2020-11-11 23:59:00"), # Blox1&2 + 4Weex
-("ecotrack","/opt/virtualBoxVMs/share/block-I/01-raw","/opt/virtualBoxVMs/share/block-I/02-inspect","2019-10-14 00:00:00","2019-10-16 23:59:00"),
-("ecolux","./test","./test","2019-09-14 00:00:00","2021-09-16 23:59:00"),
+("ecotrack","/opt/virtualBoxVMs/share/block-I/01-raw","/opt/virtualBoxVMs/share/block-I/02-inspect","2019-09-26 00:00:00","2019-10-21 23:59:00"),
+#temp 26.9 - 21.10
+("ecotrack","./test","./test","2019-09-26 00:00:00","2019-10-21 23:59:00"),
+#
+#NOTE: ecotrack time is in CEST (both track and temp) => UTC+2
+#("ecotrack",in,out,"2019-06-29 00:00:00", "2019-07-18 00:00:00") #block1
+#("ecotrack",in,out,"2019-08-05 00:00:00", "2019-08-16 00:00:00") #block2
+#("ecotrack",in,out,"2019-09-27 00:00:00","2019-10-11 00:00:00") #block3
+#
## -- add here -- ##
]
-TIME_FMT='%Y-%m-%d %H:%M:00'
+# TIME_FMT='%Y-%m-%d %H:%M:00'
+TIME_FMT='%Y-%m-%d %H:%M:%S'
noTime = dt.fromtimestamp(0)
TAG_LEN=len("04B94A7F7288588022")
RULER=f"{'#'*100}\n"
rootPath = os.path.abspath( os.path.dirname( __file__ ) ) #uses cwd(), which is set to scriptdir at start
+N_UNITS=12
def initFileStructure():
- # trackfile
- ##########
+ ## trackFiles
global T_DELIM, TIMESTAMP_LEN, T_COL_MS, T_COL_X, T_COL_Y, T_COL_TAG, T_MINLEN, T_MAXLEN, T_NCOLS
T_DELIM=';'
TIMESTAMP_LEN=10
@@ -64,18 +70,19 @@ def initFileStructure():
T_COL_Y=3
T_COL_TAG=4
#t_col_signalstrength=5
- if PROJ == "ecolux":
+ if PROJ == "ecolux": #track+light
T_MINLEN = 38 # without newline.
- T_MAXLEN = 40 # ms column varies 1-3. Will be padded later for uniform len
+ T_MAXLEN = 40 # ms are variable in length (0-999). Will be padded later for uniform len
T_NCOLS = 6
- elif PROJ == "ecotrack": #ecotron temperature project 2019
+ elif PROJ == "ecotrack": #track+temp 2019
T_MINLEN = 36
T_MAXLEN = 38
T_NCOLS = 5
+ # elif PROJ == "schrebatron":
+ #SchrebaTron: ms is padded with zeros
- # lightfile
- ##########
if hasLight():
+ ## lightFile
global L_DELIM, L_COL_MOON_REAL, L_COL_MOON_DMX, L_COL_SKYGLOW, L_MINLEN, L_MAXLEN, L_NCOLS
L_DELIM='\t'
L_COL_MOON_REAL=1
@@ -84,9 +91,21 @@ def initFileStructure():
L_MINLEN = 61 #2020-07-21 15:15:00 0.0 0 [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
L_MAXLEN = 76 #2020-07-21 15:09:00 0.0 0 [15, 54, 4, 19, 0, 1, 5, 0, 40, 12, 123, 123]
L_NCOLS = 4
-
- # animalfile
- ###########
+
+ if hasTemp():
+ ## tempFile (C for Celsius)
+ global C_TIME_FMT, C_COL_TIME, C_COL_UNIT, C_COL_HABITAT, C_COL_TEMP, C_MIN_LEN, C_MAX_LEN, C_DELIM, C_NCOLS
+ C_TIME_FMT = "%Y-%m-%d %H:%M:%S.000"
+ C_COL_TIME = 0
+ C_COL_UNIT = 1
+ C_COL_HABITAT = 2 #??? qu'est-ce que c'est
+ C_COL_TEMP = 3
+ C_MIN_LEN = 30 #coarse approximation
+ C_MAX_LEN = 40
+ C_DELIM = ';'
+ C_NCOLS = 4
+
+ ## animalFile
global A_DELIM, A_MINLEN, A_MAXLEN, A_NCOLS, A_COL_SPECIES, A_COL_SPECIES_IND, A_COL_WEIGHT_WO_TAG_MG, A_COL_TAG
if PROJ == "ecolux":
A_DELIM='\t'
@@ -99,8 +118,8 @@ def initFileStructure():
A_COL_TAG=3
elif PROJ == "ecotrack":
A_DELIM=';'
- A_MINLEN = 39
- A_MAXLEN = 44
+ A_MINLEN = 30
+ A_MAXLEN = 45
A_NCOLS = 7 #/w comments
A_COL_SPECIES=1
A_COL_SPECIES_IND=0
@@ -110,10 +129,11 @@ def initFileStructure():
# out file
###########
- global OUT_FMT, DELIM, COL_UID, COL_TAG
+ global OUT_FMT, DELIM, COL_UID, COL_HID, COL_TAG
OUT_FMT='%Y%m%d'
DELIM=','
COL_UID=4
+ COL_HID=5
COL_TAG=8
@@ -200,6 +220,9 @@ skyglowDict = {
def hasLight():
return PROJ == "ecolux"
+def hasTemp():
+ return PROJ == "ecotrack"
+
def getUID(fileName):
""" extract and return unit-id from fileName """
f=fileName.split('-')
@@ -209,8 +232,7 @@ def getUID(fileName):
class Data:
- def __init__(self,animalFile, skyglowFile, trackFiles):
-
+ def __init__(self, animalFile, skyglowFile, tempFile, trackFiles):
self.animalData = self.getAnimalData(animalFile)
self.trackData = []
@@ -227,6 +249,9 @@ class Data:
self.startLight = self.getLightTime( self.lightData[0] )
self.endLight = self.getLightTime( self.lightData[-1] )
+ if hasTemp():
+ self.initTempData(tempFile)
+
global OUT_FILE
OUT_FILE = f"{PROJ}-data-{self.startTrack.strftime(OUT_FMT)}-{self.endTrack.strftime(OUT_FMT)}"
@@ -264,16 +289,7 @@ class Data:
x = int(cols[T_COL_X])
y = int(cols[T_COL_Y])
- if x in (1,2) and y in (1,2):
- habitat=1
- elif x in (5,6) and y in (1,2):
- habitat=2
- elif x in (1,2) and y in (5,6):
- habitat=3
- elif x in (5,6) and y in (5,6):
- habitat=4
- else:
- habitat=0
+ habitat = self.calcHabitat(x,y)
# add columns (date,time,uid,habitat)
line = line[:idx_after_ms] + DELIM + date + DELIM + time + DELIM + str(uid) + DELIM + str(habitat) + line[idx_after_ms:]
@@ -286,6 +302,32 @@ class Data:
assert (lines), "trackData: Not valid. Check times / formatting."
return lines
+ def calcHabitat(self,x,y):
+ if PROJ == "ecolux":
+ if x in (1,2) and y in (1,2):
+ return 1
+ elif x in (5,6) and y in (1,2):
+ return 2
+ elif x in (1,2) and y in (5,6):
+ return 3
+ elif x in (5,6) and y in (5,6):
+ return 4
+ else:
+ return 0
+ elif PROJ == "ecotrack":
+ if x in (1,2) and y in (1,2,5,6):
+ return 2
+ elif x in (1,2) and y in (3,4,7,8):
+ return 3
+ elif x in (3,4) and y in (1,2,5,6):
+ return 3
+ elif x in (3,4) and y in (3,4,7,8):
+ return 4
+ else:
+ return "NA"
+ else:
+ return "NA"
+
def getLightData(self,file):
with open(file) as f:
@@ -297,27 +339,54 @@ class Data:
assert (lines), "skyglow: Not valid. Check times / formatting."
return lines
+ def initTempData(self,file):
+ """ get temp data and save in a list of list of tuples >> tempData[unit][habitat] returns (time,temp)"""
+ pp(f"processing: {file}")
+ with open(file) as f:
+ lines = f.readlines()
+
+ lines = self.clean(lines, C_MIN_LEN, C_MAX_LEN, C_NCOLS, C_DELIM, self.getTempTime)
+ lines.sort()
+
+ self.startTemp = self.getTempTime( lines[0] )
+ self.endTemp = self.getTempTime( lines[-1] )
+ self.lenTemp = len(lines)
+
+ self.tempData = [[[],[],[],[]] for _ in range(N_UNITS) ] #create list of empty lists containing empty lists
+ for l in lines:
+ cols = l.split(DELIM)
+
+ uid = int( cols[C_COL_UNIT] ) - 1
+ hid = int( cols[C_COL_HABITAT] ) - 1
+ time = dt.strptime( cols[C_COL_TIME],C_TIME_FMT )
+ temp = float( cols[C_COL_TEMP] )
+
+ self.tempData[uid][hid].append( (time,temp) )
+
def getAnimalData(self,file):
with open(file) as f:
lines = f.readlines()
lines = self.clean(lines, A_MINLEN, A_MAXLEN, A_NCOLS, A_DELIM)
- #TODO: check for duplicate tags and print warning / ignore?
+ #TODO: check for duplicate tags and print warning / ignore?
assert (lines), "tags: Not valid. Check times / formatting."
+
return lines
def clean(self, lines, minLen, maxLen, nCols, sep, timeFunc=None):
""" remove trailing newline, empty, comments and all-digits and remove lines before startTime or after endTime"""
lines = [l.strip() for l in lines]
- lines = [l for l in lines if minLen <= len(l) <= maxLen] # ecolux: ms are variable in length (0-999), SchrebaTron: ms is padded with zeros
+ lines = [l for l in lines if minLen <= len(l) <= maxLen]
lines = list(filter(lambda q: q and q[0] != '#', lines))
lines = [l for l in lines if len(l.split(sep)) == nCols]
lines = [l.replace(sep,DELIM) for l in lines]
+ lines = [l.replace('\ufeff','') for l in lines] #cutout BOM (ByteOrderMark). Showed up in ecotrack temp-csv.
+
if timeFunc:
lines = list(filter(lambda q: startTime <= timeFunc(q) < endTime, lines)) #only between start and endTime
-
+
return lines
def parseLightLine(self, line, uid):
@@ -339,11 +408,13 @@ class Data:
def merge(self):
- self.merge_animals()
+ self.mergeAnimals()
if hasLight():
- self.merge_light()
+ self.mergeLight()
+ if hasTemp():
+ self.mergeTemp()
- def merge_animals(self):
+ def mergeAnimals(self):
""" merge AnimalData into TrackData on TagID """
pp("merging animal data into track data. This might take a while. Time to move around. Or grab a Coffee.")
@@ -371,7 +442,7 @@ class Data:
pp(f"Didn't find Tags for {notFound} tracking events!")
- def merge_light(self):
+ def mergeLight(self):
"""merge LightData into TrackData on Time"""
pp("merging light data into track data")
@@ -380,15 +451,15 @@ class Data:
lightTime = self.startLight
trackTime = self.startTrack
- while( trackTime < lightTime ):
+ while trackTime < lightTime:
self.trackData[track_idx] += f'{DELIM}NA{DELIM}NA{DELIM}NA'
track_idx+=1
- if( track_idx >= len(self.trackData) ):
+ if track_idx >= len(self.trackData):
pp("skyglow.log starts after track.log...")
return
trackTime = self.getTrackTime(self.trackData[track_idx])
- pp(f"start merge @ {trackTime} -- idx {track_idx}")
+ pp(f"start light merge @ {trackTime} -- idx {track_idx}")
for trackLine in self.trackData[track_idx:]:
trackTime = self.getTrackTime(trackLine)
@@ -406,6 +477,54 @@ class Data:
self.trackData[track_idx] = trackLine
track_idx+=1
+ def mergeTemp(self):
+ """merge tempData into TrackData on Time"""
+ pp("merging temperature data into track data")
+
+ track_idx = 0
+ trackTime = self.startTrack
+ temp_idxs = [[0,0,0,0] for _ in range(N_UNITS) ]
+
+ #fill NA's for tracking events starting before temperature
+ while trackTime < self.startTemp:
+ self.trackData[track_idx] += f'{DELIM}NA'
+ track_idx+=1
+ if track_idx >= len(self.trackData):
+ pp("temperature starts after trackdata...")
+ return
+ trackTime = self.getTrackTime(self.trackData[track_idx])
+
+
+ pp(f"start temperature merge @ {trackTime} -- idx {track_idx}")
+
+ for trackLine in self.trackData[track_idx:]:
+
+ cols = trackLine.split(DELIM)
+ hid = int(cols[COL_HID]) - 1
+ uid = int(cols[COL_UID]) - 1
+ trackTime = self.getTrackTime(trackLine)
+
+# earliestTime = min(self.tempData[uid][1],self.tempData[uid][2],self.tempData[uid][3],self.tempData[uid][4])
+
+ # get temp idx for temp within 2:30min
+ delta = datetime.timedelta(minutes=2, seconds=30)
+ i=temp_idxs[uid][hid]
+ while self.tempData[uid][hid][i][0] < trackTime-delta:
+ # pp(f"{self.tempData[uid][hid][i][0]} < {trackTime} - {delta} ?? {self.tempData[uid][hid][i][0] < trackTime-delta}")
+ if i >= len(self.tempData[uid][hid]):
+ break;
+
+ i+=1
+
+ temp_idxs[uid][hid] = i
+ tempTime, temp = self.tempData[uid][hid][i]
+
+ # add temp col
+ trackLine += DELIM + str(temp)
+
+ self.trackData[track_idx] = trackLine
+ track_idx+=1
+
def getLightTime(self,line):
if not hasLight():
return
@@ -421,7 +540,7 @@ class Data:
winterTime = dt.strptime("2020-10-25 03:00:00",TIME_FMT)
hours= ( 2 if lightTime < winterTime else 1 )
- lightTime -= datetime.timedelta(hours)
+ lightTime -= datetime.timedelta(hours=hours) # -2h in summmer, -1h in winter
return lightTime
@@ -436,19 +555,32 @@ class Data:
return dt.fromtimestamp(timestamp)
+ def getTempTime(self,line):
+ cols = line.split(DELIM)
+ return dt.strptime(cols[C_COL_TIME],C_TIME_FMT)
+
def header(self):
h= f"# {PROJ} data\n" \
f"{RULER}" \
- f"# len(Track): {len(self.trackData)}\n" \
- f"# startTrack: {self.startTrack}\n" \
- f"# endTrack: {self.endTrack}\n" \
- f"# len(animal): {len(self.animalData)}\n"
+ f"# len(Track): {len(self.trackData)}\n" \
+ f"# startTrack: {self.startTrack}\n" \
+ f"# endTrack: {self.endTrack}\n" \
+ f"# len(animal): {len(self.animalData)}\n"
if hasLight():
- h += f"# len(Light): {len(self.lightData)}\n" \
- f"# startLight: {self.startLight}\n" \
- f"# endLight: {self.endLight}\n"
+ h += f"# len(Light): {len(self.lightData)}\n" \
+ f"# startLight: {self.startLight}\n" \
+ f"# endLight: {self.endLight}\n"
+ if hasTemp():
+ h += f"# len(Temperature): {self.lenTemp}\n" \
+ f"# startTemp: {self.startTemp}\n" \
+ f"# endTemp: {self.endTemp}\n"
h += RULER
- h += f"#timestamp, ms, date, time, unit, habitat, x, y, tag, species, speciesnumber, weight_without_tag[mg]{', moon_real[mLux], moon_eco[mLux], skyglow[Lux]' if hasLight() else ''}\n"
+ h += f"#timestamp, ms, date, time, unit, habitat, x, y, tag, species, speciesnumber, weight_without_tag[mg]"
+ if hasLight():
+ h+=', moon_real[mLux], moon_eco[mLux], skyglow[Lux]'
+ if hasTemp():
+ h+=', temp[C°]'
+ h +="\n"
h += RULER
return h
@@ -475,29 +607,41 @@ def xtract():
def getFileList():
- animalFile, skyglowFile, trackFiles = '','',[]
print( "Get files." )
+ animalFile, skyglowFile, tempFile, trackFiles = '','','',[]
+ animalFileName="tags"
+ skyglowFileName="skyglow.log"
+ trackFileSub="unit-"
+ tempFileSub="emperature"
+
for root, dirs, files in os.walk(INPUTDIR): #walk recursively
for file in files:
- if "unit" in file and not file.endswith(".tgz") and not file.endswith(".tar.gz"):
+ if file.endswith(".tgz") or file.endswith(".tar.gz"):
+ continue
+ elif trackFileSub in file:
trackFiles.append(file)
- elif file == "tags":
+ elif file == animalFileName:
animalFile = INPUTDIR+os.sep+file
print(f"Found animalFile: {animalFile}")
- elif hasLight() and file == "skyglow.log":
+ elif hasLight() and file == skyglowFileName:
skyglowFile = INPUTDIR+os.sep+file
print(f"Found logfile: {skyglowFile}")
+ elif hasTemp() and tempFileSub in file:
+ print(f"Found tempFile: {file}")
+ tempFile = INPUTDIR+os.sep+file
else:
pp(f"Ignoring ({file})")
+ if hasTemp() and not tempFile:
+ sys.exit(f"Couldn't find temperature data (filename containing '{tempFileSub}')")
if not trackFiles:
- sys.exit("No tracking data found. (filename containing unit-X, where X is a number)")
+ sys.exit(f"No tracking data found. (filenames containing '{trackFileSub}X', where X is a number)")
if hasLight() and not skyglowFile:
- sys.exit(f"Couldn't find 'skyglow.log'")
+ sys.exit(f"Couldn't find {skyglowFileName}")
if not animalFile:
- sys.exit(f"Couldn't find 'tags'")
+ sys.exit(f"Couldn't find {animalFileName}")
- return animalFile, skyglowFile, trackFiles
+ return animalFile, skyglowFile, tempFile, trackFiles
def make_rel_abs_path(path):
@@ -509,10 +653,10 @@ def make_rel_abs_path(path):
def main():
initFileStructure()
- xtract()
+ # xtract()
- animalFile, skyglowFile, trackFiles = getFileList()
- data = Data(animalFile,skyglowFile,trackFiles)
+ animalFile, skyglowFile, tempFile, trackFiles = getFileList()
+ data = Data(animalFile,skyglowFile,tempFile,trackFiles)
data.merge()
data.write()
@@ -520,10 +664,10 @@ def main():
if( platform.system() == "Linux" ):
os.system(rootPath+"/updateGitIgnore.sh")
- # os.system(rootPath+"/unknown-tags.sh ") + BLOCKS[1] <--- I , II, Ia, all ... TODO
+ os.system(rootPath+"/unknown-tags.sh ") #+ BLOCKS[1] <--- I , II, Ia, all ... TODO
if __name__ == "__main__" :
- known_projects=["ecotrack","ecolux","schrebatron"]
+ known_projects=["ecotrack","ecolux"] #,"schrebatron"
print(f" Process raw data for which block?\n {RULER}")
for idx, b in enumerate(blox):
print(f" {idx+1}) Project {b[0]}")
@@ -543,7 +687,7 @@ if __name__ == "__main__" :
raise Exception(f"Unknown Project '{PROJ}'")
except Exception as e:
- print(f"Error: {e}\n")
+ print(f"Error: {e}\n")
print("You can add a configuration by adding a line to the 'blox' structure. It has following form:\n")
print(" (Project, inDir, outDir, start, end)\n" )
print(f"> Project must be in {known_projects}" )
@@ -554,4 +698,4 @@ if __name__ == "__main__" :
start = time.time()
main()
end = time.time()
- print(f"Took {end - start}")
+ print(f"Took {end - start} seconds")
diff --git a/dataViz.py b/dataViz.py
index 5e775faed65cadb93deb125d2fdb0cf8b089e738..4c7ac559a2ca1a09f05e934f8bf30332922262a3 100755
--- a/dataViz.py
+++ b/dataViz.py
@@ -2,7 +2,7 @@
#
# VIZ (ET/ST)
###################
-# sa: /home/kr69sugu/code/idiv/schrebatron3000Data/08_alpha_rep1_rep2/02_data_inspect/ana.py
+# sa: ../schrebatrondata/08_alpha_rep1_rep2/02_data_inspect/ana.py
# preq: should work on one huge table with all data!
# > amount PatchCrossingEvents - Time -> activity histogram per unit/rep
# > detection_duration_plot
diff --git a/unknown-tags.sh b/unknown-tags.sh
index 8686c5ea56c213be58a34149ccb071da7d6f829a..453f0611313d0ad39e943a744d187840ce591a49 100755
--- a/unknown-tags.sh
+++ b/unknown-tags.sh
@@ -10,7 +10,7 @@ if (( $# < 1 )); then
echo "2. $this n list"
echo " write list of unknown tags together with their occurrences to file unknowntags-block-{n}"
echo ""
- echo "Looks for data-file ( ${scriptdir}/block-{n}/02-inspect/ecolux-data-*-filtered )"
+ echo "Looks for data-file ( ${scriptdir}/block-{n}/02-inspect/eco*-data-*-filtered )"
exit
fi
@@ -25,7 +25,7 @@ COL_TAG=13
HEADER_SIZE=1
# fi
-in=`ls ${scriptdir}/block-${1}/02-inspect/ecolux-data-*-filtered | head -1`
+in=`ls ${scriptdir}/block-${1}/02-inspect/eco*-data-*-filtered | head -1`
LEN_W_HEADER=`wc -l "$in" | cut -d" " -f1`
LEN=$(( $LEN_W_HEADER - $HEADER_SIZE))