def convertLidarDataToCsv(lidarFolder):
'''Make sure all lidar data is available in a readable text format'''
logger = logging.getLogger(__name__)
logger.info('Converting LIDAR files...')
# Loop through all lidar files in the folder
lidarFiles = os.listdir(lidarFolder)
for f in lidarFiles:
extension = icebridge_common.fileExtension(f)
# Only interested in a few file types
if (extension != '.qi') and (extension != '.hdf5') and (extension !=
'.h5'):
continue
# Handle paths
fullPath = os.path.join(lidarFolder, f)
outputPath = os.path.join(lidarFolder, os.path.splitext(f)[0] + '.csv')
if os.path.exists(outputPath):
continue
# Call the conversion
extract_icebridge_ATM_points.main([fullPath])
if not os.path.exists(outputPath):
raise Exception('Failed to parse LIDAR file: ' + fullPath)
def doFetch(options, outputFolder):
'''The main fetch function.
Returns the number of failures.'''
# Verify that required files exist
home = os.path.expanduser("~")
if not (os.path.exists(home + '/.netrc')
and os.path.exists(home + '/.urs_cookies')):
logger.error(
'Missing a required authentication file! See instructions here:\n'
+
' https://nsidc.org/support/faq/what-options-are-available-bulk-'
+ 'downloading-data-https-earthdata-login-enabled')
return -1
curlPath = asp_system_utils.which("curl")
curlOpts = ' -n -L '
cookiePaths = ' -b ~/.urs_cookies -c ~/.urs_cookies '
baseCurlCmd = curlPath + curlOpts + cookiePaths
logger.info('Creating output folder: ' + outputFolder)
os.system('mkdir -p ' + outputFolder)
isSouth = (options.site == 'AN')
if options.type == 'nav': # Nav fetching is much less complicated
return fetchNavData(options, outputFolder)
parsedIndexPath = fetchAndParseIndexFile(options, isSouth, baseCurlCmd,
outputFolder)
if not icebridge_common.fileNonEmpty(parsedIndexPath):
# Some dirs are weird, both images, fireball dems, and ortho.
# Just accept whatever there is, but with a warning.
logger.info('Warning: Missing index file: ' + parsedIndexPath)
# Store file information in a dictionary
# - Keep track of the earliest and latest frame
logger.info('Reading file list from ' + parsedIndexPath)
try:
(frameDict, urlDict) = icebridge_common.readIndexFile(parsedIndexPath)
except:
# We probably ran into old format index file. Must refetch.
logger.info('Could not read index file. Try again.')
options.refetchIndex = True
parsedIndexPath = fetchAndParseIndexFile(options, isSouth, baseCurlCmd,
outputFolder)
(frameDict, urlDict) = icebridge_common.readIndexFile(parsedIndexPath)
if options.stopAfterIndexFetch:
return 0
isLidar = (options.type in LIDAR_TYPES)
allFrames = sorted(frameDict.keys())
if not isLidar:
# The lidar frames use a totally different numbering than the image/ortho/dem frames
firstFrame = icebridge_common.getLargestFrame() # start big
lastFrame = icebridge_common.getSmallestFrame() # start small
for frameNumber in allFrames:
if frameNumber < firstFrame:
firstFrame = frameNumber
if frameNumber > lastFrame:
lastFrame = frameNumber
if options.allFrames:
options.startFrame = firstFrame
options.stopFrame = lastFrame
if isLidar:
# Based on image frames, determine which lidar frames to fetch.
if options.ignoreMissingLidar and len(frameDict.keys()) == 0:
# Nothing we can do if this run has no lidar and we are told to continue
logger.info("Warning: missing lidar, but continuing.")
lidarsToFetch = set()
else:
lidarsToFetch = lidarFilesInRange(frameDict, outputFolder,
options.startFrame,
options.stopFrame)
# There is always a chance that not all requested frames are available.
# That is particularly true for Fireball DEMs. Instead of failing,
# just download what is present and give a warning.
if options.startFrame not in frameDict and not isLidar:
logger.info("Warning: Frame " + str(options.startFrame) +
" is not found in this flight.")
if options.stopFrame and (options.stopFrame
not in frameDict) and not isLidar:
logger.info("Warning: Frame " + str(options.stopFrame) +
" is not found in this flight.")
allFilesToFetch = [
] # Files that we will fetch, relative to the current dir.
allUrlsToFetch = [] # Full url of each file.
# Loop through all found frames within the provided range
currentFileCount = 0
lastFrame = ""
if len(allFrames) > 0:
lastFrame = allFrames[len(allFrames) - 1]
hasTfw = (options.type == 'fireball')
hasXml = (isLidar or (options.type == 'ortho') or hasTfw)
numFetched = 0
skipCount = 0
for frame in allFrames:
# Skip frame outside of range
if isLidar:
if frameDict[frame] not in lidarsToFetch:
continue
else:
if ((frame < options.startFrame) or (frame > options.stopFrame)):
continue
# Handle the frame skip option
if options.frameSkip > 0:
if skipCount < options.frameSkip:
skipCount += 1
continue
skipCount = 0
filename = frameDict[frame]
# Some files have an associated xml file. Fireball DEMs also have a tfw file.
currFilesToFetch = [filename]
if hasXml:
currFilesToFetch.append(icebridge_common.xmlFile(filename))
if hasTfw:
currFilesToFetch.append(icebridge_common.tfwFile(filename))
for filename in currFilesToFetch:
url = os.path.join(urlDict[frame], filename)
outputPath = os.path.join(outputFolder, filename)
allFilesToFetch.append(outputPath)
allUrlsToFetch.append(url)
# Restrict lidar fetch amount according to the parameter
if (isLidar and options.maxNumLidarToFetch > 0
and len(allFilesToFetch) > options.maxNumLidarToFetch):
# Ensure an even number, to fetch both the lidar file and its xml
if options.maxNumLidarToFetch % 2 == 1:
options.maxNumLidarToFetch += 1
allFilesToFetch = allFilesToFetch[0:options.maxNumLidarToFetch]
allUrlsToFetch = allUrlsToFetch[0:options.maxNumLidarToFetch]
icebridge_common.fetchFilesInBatches(baseCurlCmd, MAX_IN_ONE_CALL,
options.dryRun, outputFolder,
allFilesToFetch, allUrlsToFetch,
logger)
# Fetch from disk the set of already validated files, if any
validFilesList = icebridge_common.validFilesList(
os.path.dirname(outputFolder), options.startFrame, options.stopFrame)
validFilesSet = set()
validFilesSet = icebridge_common.updateValidFilesListFromDisk(
validFilesList, validFilesSet)
numInitialValidFiles = len(validFilesSet)
# Verify that all files were fetched and are in good shape
failedFiles = []
for outputPath in allFilesToFetch:
if options.skipValidate:
continue
if not icebridge_common.fileNonEmpty(outputPath):
logger.info('Missing file: ' + outputPath)
failedFiles.append(outputPath)
continue
if icebridge_common.hasImageExtension(outputPath):
if False:
# This check is just so slow. Turn it off for now.
# This will impact only the validation of jpegs,
# as the other files can be validated via the checksum.
# Jpegs will be validated when converting them to 1 band images
if outputPath in validFilesSet and os.path.exists(outputPath):
#logger.info('Previously validated: ' + outputPath) # verbose
continue
else:
if not icebridge_common.isValidImage(outputPath):
logger.info('Found an invalid image. Will wipe it: ' +
outputPath)
if os.path.exists(outputPath): os.remove(outputPath)
failedFiles.append(outputPath)
continue
else:
logger.info('Valid image: ' + outputPath)
validFilesSet.add(outputPath) # mark it as validated
# Sanity check: XML files must have the right latitude.
if icebridge_common.fileExtension(outputPath) == '.xml':
if outputPath in validFilesSet and os.path.exists(outputPath):
#logger.info('Previously validated: ' + outputPath) #verbose
continue
else:
if os.path.exists(outputPath):
try:
latitude = icebridge_common.parseLatitude(outputPath)
logger.info('Valid file: ' + outputPath)
validFilesSet.add(outputPath) # mark it as validated
except:
# Corrupted file
logger.info("Failed to parse latitude, will wipe: " +
outputPath)
if os.path.exists(outputPath): os.remove(outputPath)
failedFiles.append(outputPath)
# On a second thought, don't wipe files with wrong latitude, as
# next time we run fetch we will have to fetch them again.
# Hopefully they will be ignored.
#isGood = hasGoodLat(latitude, isSouth)
#if not isGood:
# logger.info("Wiping XML file " + outputPath + " with bad latitude " + \
# str(latitude))
# os.remove(outputPath)
# imageFile = icebridge_common.xmlToImage(outputPath)
# if os.path.exists(imageFile):
# logger.info("Wiping TIF file " + imageFile + " with bad latitude " + \
# str(latitude))
# os.remove(imageFile)
# Verify the chcksum
if hasXml and len(outputPath) >= 4 and outputPath[-4:] != '.xml' \
and outputPath[-4:] != '.tfw':
if outputPath in validFilesSet and os.path.exists(outputPath):
#logger.info('Previously validated: ' + outputPath) # verbose
continue
else:
isGood = icebridge_common.hasValidChkSum(outputPath, logger)
if not isGood:
xmlFile = icebridge_common.xmlFile(outputPath)
logger.info('Found invalid data. Will wipe: ' +
outputPath + ' ' + xmlFile)
if os.path.exists(outputPath): os.remove(outputPath)
if os.path.exists(xmlFile): os.remove(xmlFile)
failedFiles.append(outputPath)
failedFiles.append(xmlFile)
continue
else:
logger.info('Valid file: ' + outputPath)
validFilesSet.add(outputPath)
if hasTfw and icebridge_common.fileExtension(outputPath) == '.tfw':
if outputPath in validFilesSet and os.path.exists(outputPath):
#logger.info('Previously validated: ' + outputPath)
continue
else:
isGood = icebridge_common.isValidTfw(outputPath, logger)
if not isGood:
xmlFile = icebridge_common.xmlFile(outputPath)
logger.info('Found invalid tfw. Will wipe: ' + outputPath +
' ' + xmlFile)
if os.path.exists(outputPath): os.remove(outputPath)
if os.path.exists(xmlFile): os.remove(xmlFile)
failedFiles.append(outputPath)
failedFiles.append(xmlFile)
continue
else:
logger.info('Valid tfw file: ' + outputPath)
validFilesSet.add(outputPath)
# Write to disk the list of validated files, but only if new
# validations happened. First re-read that list, in case a
# different process modified it in the meantime, such as if two
# managers are running at the same time.
numFinalValidFiles = len(validFilesSet)
if numInitialValidFiles != numFinalValidFiles:
validFilesSet = \
icebridge_common.updateValidFilesListFromDisk(validFilesList, validFilesSet)
icebridge_common.writeValidFilesList(validFilesList, validFilesSet)
numFailed = len(failedFiles)
if numFailed > 0:
logger.info("Number of files that could not be processed: " +
str(numFailed))
return numFailed
def pairLidarFiles(lidarFolder, skipValidate, logger):
'''For each pair of lidar files generate a double size point cloud.
We can use these later since they do not have any gaps between adjacent files.'''
logger.info('Generating lidar pairs...')
# Create the output folder
pairedFolder = icebridge_common.getPairedLidarFolder(lidarFolder)
os.system('mkdir -p ' + pairedFolder)
convLidarFile = icebridge_common.getConvertedLidarIndexFile(lidarFolder)
if not os.path.exists(convLidarFile):
raise Exception("Missing file: " + convLidarFile)
(lidarDict, dummyUrlDict) = icebridge_common.readIndexFile(convLidarFile)
lidarExt = ''
for frame in lidarDict:
lidarExt = icebridge_common.fileExtension(lidarDict[frame])
numLidarFiles = len(lidarDict.keys())
pairedDict = {}
# Loop through all pairs of csv files in the folder
badFiles = False
lidarKeys = sorted(lidarDict.keys())
for i in range(len(lidarKeys) - 1):
thisFile = lidarDict[lidarKeys[i]]
nextFile = lidarDict[lidarKeys[i + 1]]
date2, time2 = icebridge_common.parseTimeStamps(nextFile)
# Record the name with the second file
# - More useful because the time for the second file represents the middle of the file.
outputName = icebridge_common.lidar_pair_prefix(
) + date2 + '_' + time2 + lidarExt
pairedDict[lidarKeys[i]] = outputName
# Handle paths
path1 = os.path.join(lidarFolder, thisFile)
path2 = os.path.join(lidarFolder, nextFile)
outputPath = os.path.join(pairedFolder, outputName)
if not os.path.exists(path1) or not os.path.exists(path2):
logger.info("Cannot create " + outputPath +
" as we are missing its inputs")
# If the inputs are missing, but the output is there, most likely it is corrupt.
# Wipe it. Hopefully a subsequent fetch and convert step will bring it back.
if os.path.exists(outputPath):
logger.info("Wiping: " + outputPath)
os.system('rm -f ' + outputPath) # will not throw
badFiles = True
continue
# Skip existing valid files
if skipValidate:
if os.path.exists(outputPath):
logger.info("File exists, skipping: " + outputPath)
continue
else:
if icebridge_common.isValidLidarCSV(outputPath):
#logger.info("File exists and is valid, skipping: " + outputPath)
continue
# Concatenate the two files
cmd1 = 'cat ' + path1 + ' > ' + outputPath
cmd2 = 'tail -n +2 -q ' + path2 + ' >> ' + outputPath
logger.info(cmd1)
p = subprocess.Popen(cmd1, stdout=subprocess.PIPE, shell=True)
out, err = p.communicate()
logger.info(cmd2)
p = subprocess.Popen(cmd2, stdout=subprocess.PIPE, shell=True)
out, err = p.communicate()
if not icebridge_common.isValidLidarCSV(outputPath):
logger.error('Failed to generate merged LIDAR file, will wipe: ' +
outputPath)
os.system('rm -f ' + outputPath) # will not throw
badFiles = True
pairedLidarFile = icebridge_common.getPairedIndexFile(pairedFolder)
willWritePairedFile = False
if not os.path.exists(pairedLidarFile):
willWritePairedFile = True
else:
# Bugfix: Sometimes the written converted file has the wrong size, maybe
# something got interrupted earlier.
(lidarDictIn,
dummyUrlDict) = icebridge_common.readIndexFile(pairedLidarFile)
if lidarDictIn != pairedDict:
willWritePairedFile = True
if willWritePairedFile:
logger.info("Writing: " + pairedLidarFile)
icebridge_common.writeIndexFile(pairedLidarFile, pairedDict, {})
return (not badFiles)
def convertLidarDataToCsv(lidarFolder, startFrame, stopFrame, skipValidate,
logger):
'''Make sure all lidar data is available in a readable text format.
Returns false if any files failed to convert.'''
logger.info('Converting LIDAR files...')
lidarIndexPath = icebridge_common.csvIndexFile(lidarFolder)
(frameDict, urlDict) = icebridge_common.readIndexFile(lidarIndexPath)
if not skipValidate:
validFilesList = icebridge_common.validFilesList(
os.path.dirname(lidarFolder), startFrame, stopFrame)
validFilesSet = set()
validFilesSet = icebridge_common.updateValidFilesListFromDisk(
validFilesList, validFilesSet)
numInitialValidFiles = len(validFilesSet)
convDict = {}
# Loop through all files in the folder
badFiles = False
for frame in sorted(frameDict.keys()):
f = frameDict[frame]
extension = icebridge_common.fileExtension(f)
# Only interested in a few file types
if (extension != '.qi') and (extension != '.hdf5') and (extension !=
'.h5'):
convDict[frame] = f # these are already in plain text
continue
convDict[frame] = os.path.splitext(f)[0] + '.csv'
outputPath = os.path.join(lidarFolder, convDict[frame])
# Handle paths
fullPath = os.path.join(lidarFolder, f)
if not os.path.exists(fullPath):
logger.info("Cannot convert missing file: " + fullPath)
continue
# If the input is invalid, wipe both it, its xml, and the output
# Hopefully there will be a subsquent fetch step where it will get
# refetched.
if not icebridge_common.hasValidChkSum(fullPath, logger):
logger.info("Will wipe invalid file: " + fullPath)
xmlFile = icebridge_common.xmlFile(fullPath)
os.system('rm -f ' + fullPath) # will not throw
os.system('rm -f ' + xmlFile) # will not throw
os.system('rm -f ' + outputPath) # will not throw
badFiles = True
continue
# Skip existing valid files
if skipValidate:
if os.path.exists(outputPath):
logger.info("File exists, skipping: " + outputPath)
continue
else:
if outputPath in validFilesSet and os.path.exists(outputPath):
#logger.info('Previously validated: ' + outputPath) # verbose
continue
if icebridge_common.isValidLidarCSV(outputPath):
#logger.info("File exists and is valid, skipping: " + outputPath)
continue
# Call the conversion
logger.info("Process " + fullPath)
extract_icebridge_ATM_points.main([fullPath])
# Check the result
if not icebridge_common.isValidLidarCSV(outputPath):
logger.error('Failed to parse LIDAR file, will wipe: ' +
outputPath)
os.system('rm -f ' + outputPath) # will not throw
badFiles = True
else:
if not skipValidate:
validFilesSet.add(outputPath) # mark it as validated
convLidarFile = icebridge_common.getConvertedLidarIndexFile(lidarFolder)
willWriteConvFile = False
if not os.path.exists(convLidarFile):
willWriteConvFile = True
else:
# Bugfix: Sometimes the written converted file has the wrong size, maybe
# something got interrupted earlier.
(lidarDictIn,
dummyUrlDict) = icebridge_common.readIndexFile(convLidarFile)
if lidarDictIn != convDict:
willWriteConvFile = True
if willWriteConvFile:
logger.info("Writing: " + convLidarFile)
icebridge_common.writeIndexFile(convLidarFile, convDict, {})
if not skipValidate:
# Write to disk the list of validated files, but only if new
# validations happened. First re-read that list, in case a
# different process modified it in the meantime, such as if two
# managers are running at the same time.
numFinalValidFiles = len(validFilesSet)
if numInitialValidFiles != numFinalValidFiles:
validFilesSet = icebridge_common.updateValidFilesListFromDisk(
validFilesList, validFilesSet)
icebridge_common.writeValidFilesList(validFilesList, validFilesSet)
return (not badFiles)
def doFetch(options, outputFolder):
'''The main fetch function.
Returns the number of failures.'''
# Verify that required files exist
home = os.path.expanduser("~")
if not (os.path.exists(home+'/.netrc') and os.path.exists(home+'/.urs_cookies')):
logger.error('Missing a required authentication file! See instructions here:\n' +
' https://nsidc.org/support/faq/what-options-are-available-bulk-' +
'downloading-data-https-earthdata-login-enabled')
return -1
curlPath = asp_system_utils.which("curl")
curlOpts = ' -n -L '
cookiePaths = ' -b ~/.urs_cookies -c ~/.urs_cookies '
baseCurlCmd = curlPath + curlOpts + cookiePaths
logger.info('Creating output folder: ' + outputFolder)
os.system('mkdir -p ' + outputFolder)
isSouth = (options.site == 'AN')
if options.type == 'nav': # Nav fetching is much less complicated
return fetchNavData(options, outputFolder)
parsedIndexPath = fetchAndParseIndexFile(options, isSouth, baseCurlCmd, outputFolder)
if not icebridge_common.fileNonEmpty(parsedIndexPath):
# Some dirs are weird, both images, fireball dems, and ortho.
# Just accept whatever there is, but with a warning.
logger.info('Warning: Missing index file: ' + parsedIndexPath)
# Store file information in a dictionary
# - Keep track of the earliest and latest frame
logger.info('Reading file list from ' + parsedIndexPath)
try:
(frameDict, urlDict) = icebridge_common.readIndexFile(parsedIndexPath)
except:
# We probably ran into old format index file. Must refetch.
logger.info('Could not read index file. Try again.')
options.refetchIndex = True
parsedIndexPath = fetchAndParseIndexFile(options, isSouth, baseCurlCmd, outputFolder)
(frameDict, urlDict) = icebridge_common.readIndexFile(parsedIndexPath)
if options.stopAfterIndexFetch:
return 0
isLidar = (options.type in LIDAR_TYPES)
allFrames = sorted(frameDict.keys())
if not isLidar:
# The lidar frames use a totally different numbering than the image/ortho/dem frames
firstFrame = icebridge_common.getLargestFrame() # start big
lastFrame = icebridge_common.getSmallestFrame() # start small
for frameNumber in allFrames:
if frameNumber < firstFrame:
firstFrame = frameNumber
if frameNumber > lastFrame:
lastFrame = frameNumber
if options.allFrames:
options.startFrame = firstFrame
options.stopFrame = lastFrame
if isLidar:
# Based on image frames, determine which lidar frames to fetch.
if options.ignoreMissingLidar and len(frameDict.keys()) == 0:
# Nothing we can do if this run has no lidar and we are told to continue
logger.info("Warning: missing lidar, but continuing.")
lidarsToFetch = set()
else:
lidarsToFetch = lidarFilesInRange(frameDict, outputFolder,
options.startFrame, options.stopFrame)
# There is always a chance that not all requested frames are available.
# That is particularly true for Fireball DEMs. Instead of failing,
# just download what is present and give a warning.
if options.startFrame not in frameDict and not isLidar:
logger.info("Warning: Frame " + str(options.startFrame) +
" is not found in this flight.")
if options.stopFrame and (options.stopFrame not in frameDict) and not isLidar:
logger.info("Warning: Frame " + str(options.stopFrame) +
" is not found in this flight.")
allFilesToFetch = [] # Files that we will fetch, relative to the current dir.
allUrlsToFetch = [] # Full url of each file.
# Loop through all found frames within the provided range
currentFileCount = 0
lastFrame = ""
if len(allFrames) > 0:
lastFrame = allFrames[len(allFrames)-1]
hasTfw = (options.type == 'fireball')
hasXml = ( isLidar or (options.type == 'ortho') or hasTfw )
numFetched = 0
skipCount = 0
for frame in allFrames:
# Skip frame outside of range
if isLidar:
if frameDict[frame] not in lidarsToFetch:
continue
else:
if ((frame < options.startFrame) or (frame > options.stopFrame) ):
continue
# Handle the frame skip option
if options.frameSkip > 0:
if skipCount < options.frameSkip:
skipCount += 1
continue
skipCount = 0
filename = frameDict[frame]
# Some files have an associated xml file. Fireball DEMs also have a tfw file.
currFilesToFetch = [filename]
if hasXml:
currFilesToFetch.append(icebridge_common.xmlFile(filename))
if hasTfw:
currFilesToFetch.append(icebridge_common.tfwFile(filename))
for filename in currFilesToFetch:
url = os.path.join(urlDict[frame], filename)
outputPath = os.path.join(outputFolder, filename)
allFilesToFetch.append(outputPath)
allUrlsToFetch.append(url)
# Restrict lidar fetch amount according to the parameter
if (isLidar and options.maxNumLidarToFetch > 0 and
len(allFilesToFetch) > options.maxNumLidarToFetch):
# Ensure an even number, to fetch both the lidar file and its xml
if options.maxNumLidarToFetch % 2 == 1:
options.maxNumLidarToFetch += 1
allFilesToFetch = allFilesToFetch[0:options.maxNumLidarToFetch]
allUrlsToFetch = allUrlsToFetch [0:options.maxNumLidarToFetch]
icebridge_common.fetchFilesInBatches(baseCurlCmd, MAX_IN_ONE_CALL, options.dryRun,
outputFolder,
allFilesToFetch, allUrlsToFetch, logger)
# Fetch from disk the set of already validated files, if any
validFilesList = icebridge_common.validFilesList(os.path.dirname(outputFolder),
options.startFrame, options.stopFrame)
validFilesSet = set()
validFilesSet = icebridge_common.updateValidFilesListFromDisk(validFilesList, validFilesSet)
numInitialValidFiles = len(validFilesSet)
# Verify that all files were fetched and are in good shape
failedFiles = []
for outputPath in allFilesToFetch:
if options.skipValidate:
continue
if not icebridge_common.fileNonEmpty(outputPath):
logger.info('Missing file: ' + outputPath)
failedFiles.append(outputPath)
continue
if icebridge_common.hasImageExtension(outputPath):
if False:
# This check is just so slow. Turn it off for now.
# This will impact only the validation of jpegs,
# as the other files can be validated via the checksum.
# Jpegs will be validated when converting them to 1 band images
if outputPath in validFilesSet and os.path.exists(outputPath):
#logger.info('Previously validated: ' + outputPath) # verbose
continue
else:
if not icebridge_common.isValidImage(outputPath):
logger.info('Found an invalid image. Will wipe it: ' + outputPath)
if os.path.exists(outputPath): os.remove(outputPath)
failedFiles.append(outputPath)
continue
else:
logger.info('Valid image: ' + outputPath)
validFilesSet.add(outputPath) # mark it as validated
# Sanity check: XML files must have the right latitude.
if icebridge_common.fileExtension(outputPath) == '.xml':
if outputPath in validFilesSet and os.path.exists(outputPath):
#logger.info('Previously validated: ' + outputPath) #verbose
continue
else:
if os.path.exists(outputPath):
try:
latitude = icebridge_common.parseLatitude(outputPath)
logger.info('Valid file: ' + outputPath)
validFilesSet.add(outputPath) # mark it as validated
except:
# Corrupted file
logger.info("Failed to parse latitude, will wipe: " + outputPath)
if os.path.exists(outputPath): os.remove(outputPath)
failedFiles.append(outputPath)
# On a second thought, don't wipe files with wrong latitude, as
# next time we run fetch we will have to fetch them again.
# Hopefully they will be ignored.
#isGood = hasGoodLat(latitude, isSouth)
#if not isGood:
# logger.info("Wiping XML file " + outputPath + " with bad latitude " + \
# str(latitude))
# os.remove(outputPath)
# imageFile = icebridge_common.xmlToImage(outputPath)
# if os.path.exists(imageFile):
# logger.info("Wiping TIF file " + imageFile + " with bad latitude " + \
# str(latitude))
# os.remove(imageFile)
# Verify the chcksum
if hasXml and len(outputPath) >= 4 and outputPath[-4:] != '.xml' \
and outputPath[-4:] != '.tfw':
if outputPath in validFilesSet and os.path.exists(outputPath):
#logger.info('Previously validated: ' + outputPath) # verbose
continue
else:
isGood = icebridge_common.hasValidChkSum(outputPath, logger)
if not isGood:
xmlFile = icebridge_common.xmlFile(outputPath)
logger.info('Found invalid data. Will wipe: ' + outputPath + ' ' + xmlFile)
if os.path.exists(outputPath): os.remove(outputPath)
if os.path.exists(xmlFile): os.remove(xmlFile)
failedFiles.append(outputPath)
failedFiles.append(xmlFile)
continue
else:
logger.info('Valid file: ' + outputPath)
validFilesSet.add(outputPath)
if hasTfw and icebridge_common.fileExtension(outputPath) == '.tfw':
if outputPath in validFilesSet and os.path.exists(outputPath):
#logger.info('Previously validated: ' + outputPath)
continue
else:
isGood = icebridge_common.isValidTfw(outputPath, logger)
if not isGood:
xmlFile = icebridge_common.xmlFile(outputPath)
logger.info('Found invalid tfw. Will wipe: ' + outputPath + ' ' + xmlFile)
if os.path.exists(outputPath): os.remove(outputPath)
if os.path.exists(xmlFile): os.remove(xmlFile)
failedFiles.append(outputPath)
failedFiles.append(xmlFile)
continue
else:
logger.info('Valid tfw file: ' + outputPath)
validFilesSet.add(outputPath)
# Write to disk the list of validated files, but only if new
# validations happened. First re-read that list, in case a
# different process modified it in the meantime, such as if two
# managers are running at the same time.
numFinalValidFiles = len(validFilesSet)
if numInitialValidFiles != numFinalValidFiles:
validFilesSet = \
icebridge_common.updateValidFilesListFromDisk(validFilesList, validFilesSet)
icebridge_common.writeValidFilesList(validFilesList, validFilesSet)
numFailed = len(failedFiles)
if numFailed > 0:
logger.info("Number of files that could not be processed: " + str(numFailed))
return numFailed
def doFetch(options, outputFolder):
# Verify that required files exist
home = os.path.expanduser("~")
if not (os.path.exists(home + '/.netrc')
and os.path.exists(home + '/.urs_cookies')):
logger.error(
'Missing a required authentication file! See instructions here:\n'
+
' https://nsidc.org/support/faq/what-options-are-available-bulk-downloading-data-https-earthdata-login-enabled'
)
return -1
curlPath = asp_system_utils.which("curl")
curlOpts = ' -n -L '
cookiePaths = ' -b ~/.urs_cookies -c ~/.urs_cookies '
baseCurlCmd = curlPath + curlOpts + cookiePaths
logger.info('Creating output folder: ' + outputFolder)
os.system('mkdir -p ' + outputFolder)
isSouth = (options.site == 'AN')
parsedIndexPath = fetchAndParseIndexFile(options, isSouth, baseCurlCmd,
outputFolder)
if not icebridge_common.fileNonEmpty(parsedIndexPath):
# Some dirs are weird, both images, dems, and ortho.
# Just accept whatever there is, but with a warning.
logger.info('Warning: Missing index file: ' + parsedIndexPath)
# Store file information in a dictionary
# - Keep track of the earliest and latest frame
logger.info('Reading file list from ' + parsedIndexPath)
try:
(frameDict, urlDict) = readIndexFile(parsedIndexPath)
except:
# We probably ran into old format index file. Must refetch.
logger.info('Could not read index file. Try again.')
options.refetchIndex = True
parsedIndexPath = fetchAndParseIndexFile(options, isSouth, baseCurlCmd,
outputFolder)
(frameDict, urlDict) = readIndexFile(parsedIndexPath)
allFrames = sorted(frameDict.keys())
firstFrame = icebridge_common.getLargestFrame() # start big
lastFrame = icebridge_common.getSmallestFrame() # start small
for frameNumber in allFrames:
if frameNumber < firstFrame:
firstFrame = frameNumber
if frameNumber > lastFrame:
lastFrame = frameNumber
if options.allFrames:
options.startFrame = firstFrame
options.stopFrame = lastFrame
# There is always a chance that not all requested frames are available.
# That is particularly true for Fireball DEMs. Instead of failing,
# just download what is present and give a warning.
if options.startFrame not in frameDict:
logger.info("Warning: Frame " + str(options.startFrame) + \
" is not found in this flight.")
if options.stopFrame and (options.stopFrame not in frameDict):
logger.info("Warning: Frame " + str(options.stopFrame) + \
" is not found in this flight.")
allFilesToFetch = [
] # Files that we will fetch, relative to the current dir.
allUrlsToFetch = [] # Full url of each file.
# Loop through all found frames within the provided range
currentFileCount = 0
lastFrame = ""
if len(allFrames) > 0:
lastFrame = allFrames[len(allFrames) - 1]
hasTfw = (options.type == 'dem')
hasXml = ((options.type in LIDAR_TYPES) or (options.type == 'ortho')
or hasTfw)
numFetched = 0
for frame in allFrames:
if (frame >= options.startFrame) and (frame <= options.stopFrame):
filename = frameDict[frame]
# Some files have an associated xml file. DEMs also have a tfw file.
currFilesToFetch = [filename]
if hasXml:
currFilesToFetch.append(icebridge_common.xmlFile(filename))
if hasTfw:
currFilesToFetch.append(icebridge_common.tfwFile(filename))
for filename in currFilesToFetch:
url = os.path.join(urlDict[frame], filename)
outputPath = os.path.join(outputFolder, filename)
allFilesToFetch.append(outputPath)
allUrlsToFetch.append(url)
if options.maxNumToFetch > 0 and len(
allFilesToFetch) > options.maxNumToFetch:
allFilesToFetch = allFilesToFetch[0:options.maxNumToFetch]
allUrlsToFetch = allUrlsToFetch[0:options.maxNumToFetch]
icebridge_common.fetchFilesInBatches(baseCurlCmd, MAX_IN_ONE_CALL,
options.dryRun, outputFolder,
allFilesToFetch, allUrlsToFetch,
logger)
# Verify that all files were fetched and are in good shape
failedFiles = []
for outputPath in allFilesToFetch:
if options.skipValidate: continue
if not icebridge_common.fileNonEmpty(outputPath):
logger.info('Missing file: ' + outputPath)
failedFiles.append(outputPath)
continue
if icebridge_common.hasImageExtension(outputPath):
if not icebridge_common.isValidImage(outputPath):
logger.info('Found an invalid image. Will wipe it: ' +
outputPath)
if os.path.exists(outputPath): os.remove(outputPath)
failedFiles.append(outputPath)
continue
else:
logger.info('Valid image: ' + outputPath)
# Sanity check: XML files must have the right latitude.
if icebridge_common.fileExtension(outputPath) == '.xml':
if os.path.exists(outputPath):
latitude = icebridge_common.parseLatitude(outputPath)
isGood = hasGoodLat(latitude, isSouth)
if not isGood:
logger.info("Wiping XML file " + outputPath + " with bad latitude " + \
str(latitude))
os.remove(outputPath)
imageFile = icebridge_common.xmlToImage(outputPath)
if os.path.exists(imageFile):
logger.info("Wiping TIF file " + imageFile + " with bad latitude " + \
str(latitude))
os.remove(imageFile)
# Verify the chcksum
if hasXml and len(outputPath) >= 4 and outputPath[-4:] != '.xml' \
and outputPath[-4:] != '.tfw':
isGood = icebridge_common.hasValidChkSum(outputPath)
if not isGood:
xmlFile = icebridge_common.xmlFile(outputPath)
logger.info('Found invalid data. Will wipe it: ' + outputPath +
' ' + xmlFile)
if os.path.exists(outputPath): os.remove(outputPath)
if os.path.exists(xmlFile): os.remove(xmlFile)
failedFiles.append(outputPath)
failedFiles.append(xmlFile)
continue
else:
logger.info('Valid chksum: ' + outputPath)
if hasTfw and icebridge_common.fileExtension(outputPath) == '.tfw':
isGood = icebridge_common.isValidTfw(outputPath)
if not isGood:
xmlFile = icebridge_common.xmlFile(outputPath)
logger.info('Found invalid data. Will wipe it: ' + outputPath +
' ' + xmlFile)
if os.path.exists(outputPath): os.remove(outputPath)
if os.path.exists(xmlFile): os.remove(xmlFile)
failedFiles.append(outputPath)
failedFiles.append(xmlFile)
continue
else:
logger.info('Valid tfw file: ' + outputPath)
numFailed = len(failedFiles)
if numFailed > 0:
logger.info("Number of files that could not be processed: " +
str(numFailed))
return numFailed
def pairLidarFiles(lidarFolder, skipValidate, logger):
'''For each pair of lidar files generate a double size point cloud.
We can use these later since they do not have any gaps between adjacent files.'''
logger.info('Generating lidar pairs...')
# Create the output folder
pairedFolder = icebridge_common.getPairedLidarFolder(lidarFolder)
os.system('mkdir -p ' + pairedFolder)
convLidarFile = icebridge_common.getConvertedLidarIndexFile(lidarFolder)
if not os.path.exists(convLidarFile):
raise Exception("Missing file: " + convLidarFile)
(lidarDict, dummyUrlDict) = icebridge_common.readIndexFile(convLidarFile)
lidarExt = ''
for frame in lidarDict:
lidarExt = icebridge_common.fileExtension(lidarDict[frame])
numLidarFiles = len(lidarDict.keys())
pairedDict = {}
# Loop through all pairs of csv files in the folder
badFiles = False
lidarKeys = sorted(lidarDict.keys())
for i in range(len(lidarKeys)-1):
thisFile = lidarDict[lidarKeys[i ]]
nextFile = lidarDict[lidarKeys[i+1]]
date2, time2 = icebridge_common.parseTimeStamps(nextFile)
# Record the name with the second file
# - More useful because the time for the second file represents the middle of the file.
outputName = icebridge_common.lidar_pair_prefix() + date2 +'_'+ time2 + lidarExt
pairedDict[lidarKeys[i]] = outputName
# Handle paths
path1 = os.path.join(lidarFolder, thisFile)
path2 = os.path.join(lidarFolder, nextFile)
outputPath = os.path.join(pairedFolder, outputName)
if not os.path.exists(path1) or not os.path.exists(path2):
logger.info("Cannot create " + outputPath + " as we are missing its inputs")
# If the inputs are missing, but the output is there, most likely it is corrupt.
# Wipe it. Hopefully a subsequent fetch and convert step will bring it back.
if os.path.exists(outputPath):
logger.info("Wiping: " + outputPath)
os.system('rm -f ' + outputPath) # will not throw
badFiles = True
continue
# Skip existing valid files
if skipValidate:
if os.path.exists(outputPath):
logger.info("File exists, skipping: " + outputPath)
continue
else:
if icebridge_common.isValidLidarCSV(outputPath):
#logger.info("File exists and is valid, skipping: " + outputPath)
continue
# Concatenate the two files
cmd1 = 'cat ' + path1 + ' > ' + outputPath
cmd2 = 'tail -n +2 -q ' + path2 + ' >> ' + outputPath
logger.info(cmd1)
p = subprocess.Popen(cmd1, stdout=subprocess.PIPE, shell=True,
universal_newlines=True)
out, err = p.communicate()
logger.info(cmd2)
p = subprocess.Popen(cmd2, stdout=subprocess.PIPE, shell=True,
universal_newlines=True)
out, err = p.communicate()
if not icebridge_common.isValidLidarCSV(outputPath):
logger.error('Failed to generate merged LIDAR file, will wipe: ' + outputPath)
os.system('rm -f ' + outputPath) # will not throw
badFiles = True
pairedLidarFile = icebridge_common.getPairedIndexFile(pairedFolder)
willWritePairedFile = False
if not os.path.exists(pairedLidarFile):
willWritePairedFile = True
else:
# Bugfix: Sometimes the written converted file has the wrong size, maybe
# something got interrupted earlier.
(lidarDictIn, dummyUrlDict) = icebridge_common.readIndexFile(pairedLidarFile)
if lidarDictIn != pairedDict:
willWritePairedFile = True
if willWritePairedFile:
logger.info("Writing: " + pairedLidarFile)
icebridge_common.writeIndexFile(pairedLidarFile, pairedDict, {})
return (not badFiles)
def convertLidarDataToCsv(lidarFolder, startFrame, stopFrame,
skipValidate, logger):
'''Make sure all lidar data is available in a readable text format.
Returns false if any files failed to convert.'''
logger.info('Converting LIDAR files...')
lidarIndexPath = icebridge_common.csvIndexFile(lidarFolder)
(frameDict, urlDict) = icebridge_common.readIndexFile(lidarIndexPath)
if not skipValidate:
validFilesList = icebridge_common.validFilesList(os.path.dirname(lidarFolder),
startFrame, stopFrame)
validFilesSet = set()
validFilesSet = icebridge_common.updateValidFilesListFromDisk(validFilesList, validFilesSet)
numInitialValidFiles = len(validFilesSet)
convDict = {}
# Loop through all files in the folder
badFiles = False
for frame in sorted(frameDict.keys()):
f = frameDict[frame]
extension = icebridge_common.fileExtension(f)
# Only interested in a few file types
if (extension != '.qi') and (extension != '.hdf5') and (extension != '.h5'):
convDict[frame] = f # these are already in plain text
continue
convDict[frame] = os.path.splitext(f)[0] + '.csv'
outputPath = os.path.join(lidarFolder, convDict[frame])
# Handle paths
fullPath = os.path.join(lidarFolder, f)
if not os.path.exists(fullPath):
logger.info("Cannot convert missing file: " + fullPath)
continue
# If the input is invalid, wipe both it, its xml, and the output
# Hopefully there will be a subsquent fetch step where it will get
# refetched.
if not icebridge_common.hasValidChkSum(fullPath, logger):
logger.info("Will wipe invalid file: " + fullPath)
xmlFile = icebridge_common.xmlFile(fullPath)
os.system('rm -f ' + fullPath) # will not throw
os.system('rm -f ' + xmlFile) # will not throw
os.system('rm -f ' + outputPath) # will not throw
badFiles = True
continue
# Skip existing valid files
if skipValidate:
if os.path.exists(outputPath):
logger.info("File exists, skipping: " + outputPath)
continue
else:
if outputPath in validFilesSet and os.path.exists(outputPath):
#logger.info('Previously validated: ' + outputPath) # verbose
continue
if icebridge_common.isValidLidarCSV(outputPath):
#logger.info("File exists and is valid, skipping: " + outputPath)
continue
# Call the conversion
logger.info("Process " + fullPath)
extract_icebridge_ATM_points.main([fullPath])
# Check the result
if not icebridge_common.isValidLidarCSV(outputPath):
logger.error('Failed to parse LIDAR file, will wipe: ' + outputPath)
os.system('rm -f ' + outputPath) # will not throw
badFiles = True
else:
if not skipValidate:
validFilesSet.add(outputPath) # mark it as validated
convLidarFile = icebridge_common.getConvertedLidarIndexFile(lidarFolder)
willWriteConvFile = False
if not os.path.exists(convLidarFile):
willWriteConvFile = True
else:
# Bugfix: Sometimes the written converted file has the wrong size, maybe
# something got interrupted earlier.
(lidarDictIn, dummyUrlDict) = icebridge_common.readIndexFile(convLidarFile)
if lidarDictIn != convDict:
willWriteConvFile = True
if willWriteConvFile:
logger.info("Writing: " + convLidarFile)
icebridge_common.writeIndexFile(convLidarFile, convDict, {})
if not skipValidate:
# Write to disk the list of validated files, but only if new
# validations happened. First re-read that list, in case a
# different process modified it in the meantime, such as if two
# managers are running at the same time.
numFinalValidFiles = len(validFilesSet)
if numInitialValidFiles != numFinalValidFiles:
validFilesSet = icebridge_common.updateValidFilesListFromDisk(validFilesList,
validFilesSet)
icebridge_common.writeValidFilesList(validFilesList, validFilesSet)
return (not badFiles)
