search

bedatadriven / activityinfo-R

ActivityInfo R Language Client
https://www.activityinfo.org/support/docs/R/
18 stars 12 forks source link

List of common use cases to document #122

Open nickdickinson opened 5 months ago

nickdickinson commented 5 months ago

What are common use cases that we could document?

Let's use this issue to track these for the next release.

nickdickinson commented 1 month ago

Use cases from existing issues and features. Shall we include this in the work?:

nickdickinson commented 1 month ago

@jamiewhths In our meeting, let's talk about how to link up with the customer success team and see if you can give me access to any commonly used scripts. The prior comment is mainly to help organize related issues that we may be able to close at the same time.

jamiewhths commented 1 month ago

@nickdickinson sent you a file via email from CS which covers some of the most common use cases in an early form of documentation which could be expanded out.

jamiewhths commented 1 month ago

Order of priority:

Ryo-N7 commented 1 month ago

a few more examples:

search_resource_activityinfo <- function(resourceId) {

alldbsdf <- activityinfo::getDatabases() %>% dplyr::select(databaseId, label) alldb_ids <- alldbsdf$databaseId alldb_labels <- alldbsdf$label

alldbrecs <- purrr::map2(alldb_ids, alldb_labels, ~ getdbrecs_df(.x, .y))

alldbrecs_df <- purrr::reduce(alldbrecs, bind_rows)

if (is.na(resourceId)) { return(alldbrecs_df) } else { dbrec_filtered <- alldbrecs_df %>% dplyr::filter(resourceId == resourceId) return(dbrec_filtered) } }


- get all 'Inactive' roles from databases

```R
alldbs <- getDatabases()

get_all_inactive_ids <- function(databaseId) {
  ## search db schema for role info
  dbschm <- getDatabaseTree(databaseId)

  dbroles <- dbschm[["roles"]]

  ## the ID for "Inactive" role
  inactiveid <- dbroles %>% 
    keep(~ .x$label == "Inactive") %>% 
    map_chr(~ .x$id)

  inactdf <- data.frame("db_id" = databaseId,
                        "inactive_id" = inactiveid)

  return(inactdf)
}

safe_get_all_inactive_ids <- purrr::safely(get_all_inactive_ids)

allinactids_output <- map(alldbsclean$databaseId, 
                      ~ safe_get_all_inactive_ids(databaseId = .x))

allinactids_df <- allinactids_output %>% 
  map("result") %>% 
  discard(is.null) %>% 
  reduce(bind_rows)
PrList <- getDatabases()

get_public_visibles <- function(dblabel, dbid) {
  cat(paste0("\n\nDatabase: ", dblabel, "\n\n"))
  dbrecs <- getDatabaseResources(dbid)
  dbrecs <- dbrecs %>% 
    filter(visibility == "PUBLIC", type == "FORM") %>% 
    mutate(project = dblabel,
           url = AVDGconvenience::getActivityInfoURL(formId = id)) %>% 
    select(project, id, label, visibility, url)
  cat(paste0("\n\nDatabase: ", dblabel, "... Done!\n\n"))
  return(dbrecs)
}

allpubvisdf <- map2_dfr(PrList$label, PrList$databaseId, ~ get_public_visibles(dblabel = .x, dbid = .y))
# Generalized ActivityInfo database back-up script
# Run at the end of every month (or whatever other time interval)

# Packages ----
# Install the following packages:

# install.packages("devtools")
# devtools::install_github("bedatadriven/activityinfo-R")
# install.packages("here")

library(activityinfo)
library(here)

# Generate ActivityInfo API token ----
# Please see the following instructions below to generate your API token so you can interact with ActivityInfo through R:
#   
# 1. Click on your account name in the top right corner bar and click on "Account settings".
# 2. Click on "API Tokens" in the left side bar.
# 3. Then click "Add".
# 4. Set a label for your API token then choose "Read & Write" button in "Scope", finally click "Generate" to create the API token.
# 5. You'll be given a chance to copy-paste your full API token somewhere. Do so in a secure location as you won't have another opportunity to copy the full token after this time.
# 6. Now in your R session you can use the `activityInfoToken()` function to set everything up. Note, in general, __you should never include your API token in an R source file or check such sources with tokens into version control tools like git__.

# Set up file/directory paths ----
# Folders are pre-created in the back-up directory according to the date
# the `here()` function sets up the file-path to the current project root
path <- here("activityinfo_database_backup")

#  If directory/folder doesn't exist yet, create it.
if (!dir.exists(path)) {
  dir.create(path)
}

## Monthly back-ups:
yr <- format(Sys.Date(), "%Y")
yrmonth <- format(Sys.Date(), "%Y_%m")
fileLoc <- sprintf("%s/%s/%s", path, yr, yrmonth)

## Weekly back-ups
yr <- format(Sys.Date(), "%Y")
yrmonth <- format(Sys.Date(), "%Y_%m")
yrweek <- as.numeric(format(Sys.Date(), "%U"))  ## NOTE: week of the YEAR (not the month)
fileLoc <- sprintf("%s/%s/%s/%s", path, yr, yrmonth, yrweek)

#  If directory/folder doesn't exist yet, create it.
if (!dir.exists(fileLoc)) {
  dir.create(fileLoc, recursive = TRUE)
}

# Get all existing forms in database ----
# Resources are the forms and folders in ActivityInfo
database_id <- "c5h9mudkz3sqce52" 
formlist <- getDatabaseResources(database_id)

# We only want resources that are forms and subforms 
formListClean <- subset(formlist, type %in% c("FORM", "SUB_FORM")) 

# Grab data and save ----
# The queryTable function gets ALL the columns, including those of the referenced forms which can be extraneous
# So first use getFormSchema() to get the individual fields and use this to limit the fields obtained from queryTable()
# We will only get the fieldId and fieldLabel because we already obtained the formId, parentId, and formLabel earlier

for (n in 1:nrow(formListClean)) {
  message(sprintf("Form %s of %s", n, nrow(formListClean)))
  schema <- getFormSchema(as.character(formListClean[n, "id"]))
  ## erase `/`, square brackets, parentheses, etc. from labels as that would mess up the file names
  schema$label <- gsub(x = schema$label, pattern = "\\/", replacement = "-")
  schema$label <- gsub(x = schema$label, pattern = "\\[|\\]", replacement = "-")
  schema$label <- gsub(x = schema$label, pattern = "\\(|\\)", replacement = "-")

  if (length(schema$elements) == 0) { next } # Do not proceed if form has no fields
  # queryTable() function looks like this:
  # queryTable(formid,
  #    "fieldLabel" = "fieldId",
  #    "fieldLabel" = "fieldId")
  # So we need to store these individual columns as a string to pass it to queryTable()
  schemaElements <- c("_id")
  schemaHeaders <- c("ID")

  for (index in 1:length(schema$elements)) {
    id <- schema$elements[[index]]$id
    label <- schema$elements[[index]]$label
    schemaElements <- append(schemaElements, id)
    schemaHeaders <- append(schemaHeaders, label)
    toString(schemaElements)
    toString(schemaHeaders)
    cols <- c(setNames(schemaElements, schemaHeaders))
  }
  # Run queryTable() to grab the data
  data <- queryTable(schema$id, cols)
  # Write data to its designated folder
  message(sprintf("Writing data for %s\n", schema$label))
  write.csv(data, file = sprintf("%s/%s_%s.csv", fileLoc, schema$label, Sys.Date()))
}

message("End of script.")