Merge pull request #1 from flyconnectome/fix/cave-infrastructure

Switch from Zetta to cave infrastructure

.github/workflows/R-CMD-check.yaml

@@ -49,7 +49,7 @@ jobs:
         shell: Rscript {0}
 
       - name: writetoken
-        run: fafbseg::flywire_set_token(token=Sys.getenv("CHUNKEDGRAPH_SECRET"), domain='wclee.api.zetta.ai')
+        run: fafbseg::flywire_set_token(token=Sys.getenv("CHUNKEDGRAPH_SECRET"))
         shell: Rscript {0}
 
       - name: Check

R/cloudvolume.R

@@ -21,12 +21,16 @@ fanc_api_url <- function(endpoint="") {
 #' @return The path to the token file (invisibly)
 #' @export
 fanc_set_token <- function(token=NULL) {
-  # check we have the
-  fafbseg::flywire_set_token(token=token, domain='wclee.api.zetta.ai')
+  # path=fafbseg::flywire_set_token(token=token, domain='cave.fanc-fly.com')
+  path=fafbseg::flywire_set_token(token=token)
+  # clear the token cache so the new one is immediately available
+  fanc_token(cached=FALSE)
+  invisible(path)
 }
 
 fanc_token <- function(cached=TRUE) {
-  fafbseg::chunkedgraph_token(url='wclee.api.zetta.ai', cached = cached)
+  # fafbseg::chunkedgraph_token(url='cave.fanc-fly.com', cached = cached)
+  fafbseg::chunkedgraph_token(cached = cached)
 }
 
 fanc_token_available <- function() {
@@ -42,24 +46,24 @@ fanc_token_available <- function() {
 #' dr_fanc()
 #' }
 dr_fanc <- function() {
-  zetta_report()
+  fanc_api_report()
   cat("\n\n")
   res = fafbseg:::py_report()
   cat("\n")
   try(fafbseg:::check_cloudvolume_reticulate(min_version = "3.12"))
   invisible(res)
 }
 
-zetta_report <- function() {
-  message("Zetta Neuroglancer / API access\n----")
+fanc_api_report <- function() {
+  message("FANC Neuroglancer / CAVE API access\n----")
 
   token=try(fanc_token(cached = F), silent = FALSE)
   if(inherits(token, "try-error")) {
     FUN=if(requireNamespace('usethis', quietly = T)) usethis::ui_todo else message
-    FUN(paste('No valid Zetta token found. Set your token by doing:\n',
+    FUN(paste('No valid FANC API token found. Set your token by doing:\n',
                   "{ui_code('fanc_set_token()')}"))
   } else{
-    cat("Valid Zetta ChunkedGraph token is set!\n")
+    cat("Valid FANC API ChunkedGraph token is set!\n")
   }
   ff=dir(fafbseg:::cv_secretdir(), pattern = '-secret\\.json$')
   if(length(ff)){

R/ids.R

@@ -62,7 +62,7 @@ fanc_leaves <- function(x, integer64=TRUE, ...) {
 #' @export
 #' @importFrom nat xyzmatrix
 #' @examples
-#' # a point from neuroglancer, should map to 648518346479013777
+#' # a point from neuroglancer, should map to 648518346498932033
 #' fanc_xyz2id(cbind(34495, 82783, 1954), rawcoords=TRUE)
 fanc_xyz2id <- function(xyz, rawcoords=FALSE, voxdims=c(4.3, 4.3, 45),
                         root=TRUE, ...){

R/urls.R

@@ -13,7 +13,7 @@
 #' browseURL(fanc_scene())
 #' }
 fanc_scene <- function(ids=NULL) {
-  url="https://neuromancer-seung-import.appspot.com/?json_url=https://api.zetta.ai/json/325890137970388411"
+  url="https://neuromancer-seung-import.appspot.com/?json_url=https://global.daf-apis.com/nglstate/api/v1/5969075557629952"
   url=sub("?json_url=", "?", url, fixed = T)
   parts=unlist(strsplit(url, "?", fixed = T))
   json=try(flywire_fetch(parts[2], token=fanc_token(), return = 'text', cache = TRUE))

R/zetta-api.R

@@ -1,32 +1,28 @@
 #' Fetch change log information for one or more neurons
 #'
+#' @details As of August 2021 this is a simple wrapper of
+#'   \code{fabseg::\link{flywire_change_log}}. For now the old (and less
+#'   convenient format) available from the zetta API can be obtained with the
+#'   private \code{fancr:::fanc_change_log_zetta} function.
+#'
 #' @param x One or more fanc ids in any format understandable by
 #'   \code{\link[fafbseg]{ngl_segments}}
-#' @param ... Additional argument passed to \code{\link{pbsapply}} and then on
-#'   to \code{\link{flywire_fetch}} (via the private \code{fanc_fetch}).
-#'
-#' @return A list containing information on the changes applied to a given body
-#'
-#'   \itemize{
-#'
-#'   \item \code{n_splits},\code{n_mergers} The number of splits/mergers applied
-#'   to the neuron
-#'
-#'   \item \code{user_info} A data.frame containing user ids or emails and the
-#'   number of splits/mergers
-#'
-#'   \item \code{operations_ids} An integer vector of operation ids
-#'
-#'   \item \code{past_ids} Previous ids associated with this body
-#'
-#'   }
+#' @return a \code{data.frame} See \code{fabseg::\link{flywire_change_log}} for
+#'   details
 #' @export
 #' @importFrom pbapply pbsapply
+#' @inheritParams fafbseg::flywire_change_log
 #' @examples
 #' \donttest{
 #' fanc_change_log("648518346473954669")
 #' }
-fanc_change_log <- function(x, ...) {
+
+fanc_change_log <- function(x, tz="UTC", filtered=TRUE, OmitFailures=TRUE,
+                            ...) {
+  with_fanc(flywire_change_log(x=x, tz=tz, filtered = filtered, OmitFailures = OmitFailures, ...))
+}
+
+fanc_change_log_zetta <- function(x, ...) {
   baseurl=fanc_api_url(endpoint = "root/")
   x=fafbseg::ngl_segments(x, as_character = T)
   if(length(x)>1) {

tests/testthat/test-ids.R

@@ -1,6 +1,6 @@
 test_that("fanc_xyz2id works", {
   expect_equal(fanc_xyz2id(cbind(34495, 82783, 1954), rawcoords=TRUE),
-               "648518346479013777")
+               "648518346498932033")
 
   expect_equal(
     fanc_xyz2id(cbind(34495, 82783, 1954), rawcoords=TRUE, root=F),

tests/testthat/test-urls.R

@@ -1,3 +1,3 @@
 test_that("fanc_scene works", {
-  expect_type(sc <- fanc_scene("648518346479013777"), 'character')
+  expect_type(sc <- fanc_scene("648518346498932033"), 'character')
 })

tests/testthat/test-zetta-api.R

@@ -1,8 +1,8 @@
 test_that("changelog works", {
   skip_if_offline()
   skip_if_not(fanc_token_available(),
-              message="Unable to obtain a Zetta / FANC access token")
-  expect_type(res <- fanc_change_log("648518346473954669"), "list")
-  expect_named(res, c("n_splits", "n_mergers", "user_info", "operations_ids",
-                      "past_ids"))
+              message="Unable to obtain a FANC access token")
+  expect_s3_class(res <- fanc_change_log("648518346473954669"), "data.frame")
+  expect_named(res, c("operation_id", "timestamp", "user_id", "before_root_ids",
+                      "after_root_ids", "is_merge", "user_name"))
 })