csaybar
commented
4 years ago Unfortunately, I've never been able to replicate this bug but I'm sure it happens in the last mile of ee_monitoring. Could you run the following lines to help me to understand better why this error occurs?.
library(rgee)
ee_Initialize(drive = TRUE, gcs = FALSE)
ee_monitoring_test <- function(task, eeTaskList = FALSE, quiet = FALSE) {
if (missing(task)) {
task <- ee$batch$Task$list()[[1]]
}
while (task$active() & task$state != "CANCEL_REQUESTED") {
if (!quiet) {
cat(sprintf("Polling for task (id: %s).\n", task$id))
}
Sys.sleep(5)
}
if (!quiet) {
cat(sprintf("State: %s\n", task$status()$state))
}
message(class(task$status()$state))
message(class(task$id))
if (task$status()$state != "COMPLETED") {
message(
"ERROR in Earth Engine servers: ",
task$status()$error_message
)
stop("ee_monitoring was forced to stop before getting results")
}
}
geometry = ee$Geometry$Polygon(
coords = list(
c(45.90741759102099, 41.79760334740121),
c(45.90741759102099, 41.68411789893928),
c(46.05847960273974, 41.68411789893928),
c(46.05847960273974, 41.79760334740121)
)
)
maskcloud1 <- function(image) {
QA60 <- image$select(list("QA60"))
return(image$updateMask(QA60$lt(1)))
}
S2 <- ee$ImageCollection("COPERNICUS/S2")$
filterDate("2015-01-01", "2016-10-31")$
filter(ee$Filter$lte("CLOUDY_PIXEL_PERCENTAGE", 20))$
filterBounds(geometry)$
map(maskcloud1)$
select(c('B4', 'B3', 'B2'))
ExpColl <- S2 $sort("system:time_start", FALSE)$limit(5)
# # Loop to output each image
count <- ExpColl$size()$getInfo()
ExpColl_list <- ExpColl$toList(count)
index <- 1
image = ee$Image(ExpColl_list$get(index-1))
name = image$get('system:index')$getInfo()
print(name)
task <- ee$batch$Export$image$toDrive(
image,
name,
scale = 10,
maxPixels = 1e9,
folder = "rgee_backup",
region = geometry
)
task$start()
ee_monitoring_test() 
andreatitolo
At submit an issue, please attached the following information of your
rgeesession:[X] You have the Python API installed (from terminal):
[X] You can find the credentials file on your system:
[X] You can run a simple EE command from R:
Attach your Python (reticulate) configuration:
Earthengine api, python and credentionals
Description
Hi Cesar, thank you so much for this package, I just discovered it recently, and is truly amazing. I am running into an issue: I was trying to export an image collection using
ee$batch$Export$image$toDrivefollowing the example provided in the export_ImageCollection.R example, using Method 01. If, as per the example, I useee_monitoring()without any variable inside, I get the "Python int too large to convert to C long" error and even if the task is stated as "COMPLETED", it exports only one image instead of all the images in the collection.However, if I use
ee_monitoring(quiet = TRUE)or if i remove ee_monitoring() entirely, the images are all exported correctly, as expected. I can do everything with rgee correctly and the installation seems fine to me, am I doing something wrong or this an expected behaviour?This issue might be similar to #47, however, if in that case it was negligible as the task succeeded anyway, my task is said to be completed even if it exported only one image.
What I Did
The result is:
with
ee_monitoring(quiet = TRUE)or when removing entirely ee_monitoring() the result is:and all the images are exported correctly.