def getItems(config):
configMeta = config["metadata"]
inputFile = configMeta["src"]
fieldnames, items = io.readCsv(inputFile, parseNumbers=False)
if "query" in configMeta:
items = lu.filterByQueryString(items, configMeta["query"])
print("%s items after filtering" % len(items))
# Add columns
sets, items = addColumnsToItems(items, config)
# Sort so that index corresponds to ID
items = sorted(items, key=lambda item: item["id"])
items = lu.addIndices(items)
return (sets, items)
help="File generated by html_to_csv.py")
parser.add_argument('-field',
dest="FIELD",
default="Locale",
help="Field to output")
parser.add_argument(
'-out',
dest="OUTPUT_FILE",
default="data/processed/MexicoAndCentralAmerica_locales.csv",
help="Output csv file")
a = parser.parse_args()
# Make sure output dirs exist
io.makeDirectories([a.OUTPUT_FILE])
fieldNames, items = io.readCsv(a.INPUT_FILE)
itemCount = len(items)
values = [item[a.FIELD] for item in items]
counter = collections.Counter(values)
counts = counter.most_common()
rows = []
for value, count in counts:
if len(str(value).strip()) < 1:
continue
row = {}
row[a.FIELD] = value
row["Count"] = count
rows.append(row)
from pprint import pprint
import random
import sys
import lib.io_utils as io
import lib.math_utils as mu
# input
parser = argparse.ArgumentParser()
parser.add_argument('-in', dest="INPUT_FILE", default="data/report_data.csv", help="File for input")
parser.add_argument('-start', dest="START_YEAR", default=1869, type=int, help="Start year")
parser.add_argument('-end', dest="END_YEAR", default=2018, type=int, help="End year")
parser.add_argument('-out', dest="OUTPUT_FILE", default="data/collections.json", help="File for output")
a = parser.parse_args()
_, data = io.readCsv(a.INPUT_FILE)
data = sorted(data, key=lambda k: k["year"])
latest = data[-1]
divisions = ["invertebrate zoology", "paleontology", "vertibrate zoology", "anthropology", "physical sciences"]
# entry = mu.roundInt(latest[]/1000.0)
# drawDots("img/dots_circle_invertebrate.png", totalDots, dotW, highlightCount=entry)
# runningTotal = entry
#
# entry = mu.roundInt(latest["paleontology"]/1000.0)
# drawDots("img/dots_circle_paleontology.png", totalDots, dotW, highlightCount=entry, highlightColor=[217,64,107], highlightOffset=runningTotal)
# runningTotal += entry
#
# entry = mu.roundInt(latest["vertibrate zoology"]/1000.0)
# drawDots("img/dots_circle_vertibrate.png", totalDots, dotW, highlightCount=entry, highlightColor=[226,169,17], highlightOffset=runningTotal)
dest="FIELD_LIST",
default="",
help="Comma-separated list of fields to output; leave blank for everything"
)
a = parser.parse_args()
# Make sure output dirs exist
io.makeDirectories([a.OUTPUT_FILE])
items = []
fieldNames = []
if "*" in a.INPUT_FILE:
files = glob.glob(a.INPUT_FILE)
for fn in files:
fFieldNames, fItems = io.readCsv(fn, parseNumbers=False)
# Infer region from filename
for j, fitem in enumerate(fItems):
fItems[j]["Region"] = re.sub(r'(?<!^)(?=[A-Z])', ' ',
os.path.basename(fn).split(".")[0])
fieldNames += fFieldNames
items += fItems
fieldNames = lu.unique(fieldNames)
else:
fieldNames, items = io.readCsv(a.INPUT_FILE, parseNumbers=False)
itemCount = len(items)
# this is where the normalized data will go
cleanedItems = [{
FIELDS_LISTS = a.FIELDS_LISTS.strip().split(",")
FIELDS_MERGE = io.parseQueryString(a.FIELDS_MERGE.strip())
# Make sure output dirs exist
io.makeDirectories([a.OUTPUT_FILE])
if len(a.DETAILED_OUTPUT_FILE) > 0:
io.makeDirectories([a.DETAILED_OUTPUT_FILE])
items = []
fieldNames = []
if "*" in a.INPUT_FILE:
files = glob.glob(a.INPUT_FILE)
for fn in files:
fFieldNames, fItems = io.readCsv(fn)
fieldNames += fFieldNames
items += fItems
fieldNames = lu.unique(fieldNames)
else:
fieldNames, items = io.readCsv(a.INPUT_FILE)
# make unique based on id
items = list({item['Catalog No']: item for item in items}.values())
itemCount = len(items)
# Parse lists
for i, item in enumerate(items):
for field in FIELDS_LISTS:
help="Max number of subjects (includes 'other')")
parser.add_argument('-out',
dest="OUTPUT_FILE",
default="data/photographic_images.json",
help="File for output")
a = parser.parse_args()
YEAR_RANGE = [1600, 2020]
gridW, gridH = tuple([int(t) for t in a.GRID_SIZE.split("x")])
# Make sure output dirs exist
io.makeDirectories(a.OUTPUT_FILE)
# retrieve data
fieldNames, data = io.readCsv(a.INPUT_FILE)
dataCount = len(data)
_, subjectData = io.readCsv(a.SUBJECTS_FILE)
grid = np.loadtxt(a.GRID_FILE, delimiter=",")
imageFiles = glob.glob(a.IMAGE_FILES)
imageFiles = sorted(imageFiles)
fileCount = len(imageFiles)
print("Loaded %s files" % fileCount)
# process subject data
subjectData = groupList(subjectData, "subject", sort=True)
subjectCount = len(subjectData)
mainSubjects = subjectData[:a.
MAX_SUBJECTS] if subjectCount > a.MAX_SUBJECTS else subjectData
subjectLabels = [s["subject"] for s in mainSubjects]
parser.add_argument('-out', dest="OUTPUT_FILE", default="data/metadata.json", help="File for output")
a = parser.parse_args()
tileW, tileH = tuple([int(t) for t in a.TILE_SIZE.split("x")])
gridAssignment = None
with open(a.INPUT_GRID_FILE, "rb") as f:
gridAssignment = pickle.load(f)
grid, gridShape = gridAssignment
gridW, gridH = gridShape
# Make sure output dirs exist
io.makeDirectories(a.OUTPUT_FILE)
# retrieve data
fieldNames, data = io.readCsv(a.INPUT_FILE)
dataCount = len(data)
# retrieve images
imageFiles = glob.glob(a.IMAGE_FILES)
imageFiles = sorted(imageFiles)
fileCount = len(imageFiles)
print("Loaded %s files" % fileCount)
if fileCount != len(grid):
print("File count (%s) != grid count (%s)" % (fileCount, len(grid)))
sys.exit()
# initialize grid
rows = []
for i in range(gridH):
import sys
import lib.io_utils as io
import lib.list_utils as lu
import lib.math_utils as mu
# input
parser = argparse.ArgumentParser()
parser.add_argument('-in', dest="INPUT_FILE", default="data/MexicoAndCentralAmerica.csv", help="File generated by html_to_csv.py")
parser.add_argument('-out', dest="OUTPUT_FILE", default="data/processed/MexicoAndCentralAmerica_cleaned.csv", help="Output csv file")
a = parser.parse_args()
# Make sure output dirs exist
io.makeDirectories([a.OUTPUT_FILE])
fieldNames, items = io.readCsv(a.INPUT_FILE)
itemCount = len(items)
# this is where the clean data will go
cleanedItems = [{
"Id": item["Catalog No"],
"Acquisition Year": "",
"Acquisition Type": "",
"Latitude": 0,
"Longitude": 0,
"Country": "",
"Locale": "",
"Category": "",
"Hall": ""
} for item in items]
def getItems(config):
inputFile = config["metadataFile"]
idCol = config["identifierColumn"] if "identifierColumn" in config else None
fieldnames, items = io.readCsv(inputFile, parseNumbers=False)
if "metadataFilterQuery" in config:
items = lu.filterByQueryString(items, config["metadataFilterQuery"])
print("%s items after filtering" % len(items))
# map year, lat/lon, and category
columnMap = [("dateColumn", "year"), ("latitudeColumn", "lat"),
("longitudeColumn", "lon"), ("countryColumn", "country"),
("groupByColumn", "category")]
minimumYear = config["minimumYear"] if "minimumYear" in config else None
maximumYear = config["maximumYear"] if "maximumYear" in config else None
validItems = []
for i, item in enumerate(items):
validatedItem = item.copy()
isValid = True
for configKey, toColumn in columnMap:
if configKey not in config:
continue
value = item[config[configKey]]
if toColumn == "year":
value = su.validateYear(value, minimumYear, maximumYear)
elif toColumn == "lat":
value = su.validateLat(value)
elif toColumn == "lon":
value = su.validateLon(value)
if value is None:
isValid = False
break
validatedItem[toColumn] = value
if isValid:
validItems.append(validatedItem)
diff = len(items) - len(validItems)
print(f'Found {diff} invalid items.')
# Sort so that index corresponds to ID
if idCol is not None:
for i, item in enumerate(validItems):
validItems[i]["_id"] = str(item[idCol])
validItems = sorted(validItems, key=lambda item: item["_id"])
validItems = lu.addIndices(validItems)
if idCol is None:
for i, item in enumerate(validItems):
validItems[i]["_id"] = str(i)
# Retrieve categories
categories = []
itemsByCategory = lu.groupList(validItems,
"category",
sort=True,
desc=True)
if "groupLimit" in config and len(itemsByCategory) > config["groupLimit"]:
limit = config["groupLimit"] - 1
otherItems = itemsByCategory[limit:]
otherLabel = config["otherLabel"] if "otherLabel" in config else "Other"
otherCount = 0
for group in otherItems:
for item in group["items"]:
validItems[item["index"]]["category"] = otherLabel
otherCount += 1
itemsByCategory = itemsByCategory[:limit] + [{
"category": otherLabel,
"count": otherCount
}]
categoryColors = config["groupColors"]
colorCount = len(categoryColors)
for i, category in enumerate(itemsByCategory):
color = categoryColors[i % colorCount]
categories.append({
"text": category["category"],
"color": color,
"count": category["count"]
})
return (validItems, categories)
dest="OUTPUT_FILE",
default="data/processed/MexicoAndCentralAmerica_geocoded.csv",
help="Output csv file")
parser.add_argument('-wait',
dest="WAIT_SECONDS",
default=5,
type=int,
help="Seconds to wait before each request")
a = parser.parse_args()
LOCALE_FIELD = "Locale"
# Make sure output dirs exist
io.makeDirectories([a.OUTPUT_FILE, a.CACHE_FILE])
fieldNames, items = io.readCsv(a.INPUT_FILE)
itemCount = len(items)
for i, item in enumerate(items):
items[i]["LookupString"] = ""
# only lookup items that have country and locale with values
if len(item["Country"]) < 1 or len(item[LOCALE_FIELD]) < 1:
continue
items[i]["LookupString"] = item[LOCALE_FIELD] + ", " + item["Country"]
locales = []
localeLookup = {}
if os.path.isfile(a.CACHE_FILE):
_, locales = io.readCsv(a.CACHE_FILE)
