garrick
/
gathertweet
Sledovat 1
Oblíbit 0
Rozštěpit 0
Zdrojový kód Úkoly 1 Požadavky na natažení 0 Vydání 0 Wiki Aktivita
Procházet zdrojové kódy

Refactor gathertweet script into action functions

master
Garrick Aden-Buie před 7 roky
rodič
b81867358d
revize
ac7b015934
7 změnil soubory, kde provedl 331 přidání a 212 odebrání
Jednotný pohled Ukázat statistiku rozdílových dat
  1. +4
    -0
      NAMESPACE
  2. +152
    -0
      R/gathertweet_actions.R
  3. +2
    -0
      R/logging.R
  4. +1
    -1
      README.Rmd
  5. +61
    -56
      README.md
  6. +97
    -155
      inst/gathertweet.R
  7. +14
    -0
      man/gathertweet_search.Rd

+ 4
- 0
NAMESPACE Zobrazit soubor



export("%>%") export("%>%")
export(backup_tweets) export(backup_tweets)
export(gathertweet_search)
export(gathertweet_simplify)
export(gathertweet_timeline)
export(gathertweet_update)
export(get_user_info) export(get_user_info)
export(install_gathertweet) export(install_gathertweet)
export(last_seen_tweet) export(last_seen_tweet)

+ 152
- 0
R/gathertweet_actions.R Zobrazit soubor

#' @title gathertweet actions
#' @export
gathertweet_search <- function(
terms,
file = "tweets.rds",
n = 18000,
max_id = NULL,
since_id = "last",
type = "recent",
include_rts = FALSE,
geocode = NULL,
`no-parse` = FALSE,
token = NULL,
retryonratelimit = FALSE,
quiet = FALSE,
...
) {
log_info("Searching for \"{paste0(terms, collapse = '\", \"')}\"")

since_id <- if (is.null(max_id)) {
if (since_id == "last") {
last_seen_tweet(file = file)
} else if (since_id == "none") {
NULL
} else since_id
}
if (!is.null(since_id)) log_info("Tweets from {since_id}")
if (!is.null(max_id)) log_info("Tweets up to {max_id}")

tweets <- lapply(
terms,
function(term) rtweet::search_tweets(
q = term,
n = as.integer(n),
type = type,
include_rts = include_rts,
geocode = geocode,
max_id = max_id,
parse = isFALSE(`no-parse`),
token = token,
retryonratelimit = retryonratelimit,
verbose = isFALSE(quiet),
since_id = since_id
)
)



if (isTRUE(`no-parse`)) {
log_info("Saving un-parsed tweets in {file}")
saveRDS(tweets, file)
} else {
tweets <- dplyr::bind_rows(tweets)

if (nrow(tweets) == 0) {
log_info("No new tweets.")
exit()
}

tweets <- tweets[!duplicated(tweets$status_id), ]
tweets <- tweets[order(tweets$status_id), ]

log_info("Gathered {nrow(tweets)} tweets")
tweets <- save_tweets(tweets, file)

log_info("Total of {nrow(tweets)} tweets in {file}")
}

tweets
}

#' @export
gathertweet_update <- function(file = "tweets.rds", `no-parse` = FALSE, token = NULL, ...) {
logger("Updating tweets in {file}")
if (!file.exists(file)) {
log_fatal("`{file}` does not exist")
}
tweets <- update_tweets(
file = file,
# passed to rtweet::lookup_statuses()
parse = isFALSE(`no-parse`),
token = token
)
log_debug("Status lookup returned {nrow(tweets)} tweets")
tweets <- save_tweets(tweets, file)
log_debug("Total of {nrow(tweets)} tweets in {file}")
tweets
}

#' @export
gathertweet_timeline <- function(
users,
file = "tweets.rds",
n = 3200,
max_id = NULL,
home = TRUE,
`no-parse` = FALSE,
token = NULL,
include_rts = FALSE,
...
) {
log_info("Gathering tweets by {collapse(users)}")

n <- as.integer(n)
if (n > 3200) {
log_warn("Twitter API for timelines returns a maximum of 3200 tweets per user")
}

tweets <- rtweet::get_timeline(
user = users,
n = n,
max_id = max_id,
home = isTRUE(home),
parse = isFALSE(`no-parse`),
check = TRUE,
token = token,
include_rts = isTRUE(include_rts)
)

tweets <- tweets[!duplicated(tweets$status_id), ]
tweets <- tweets[order(tweets$status_id), ]

log_info("Gathered {nrow(tweets)} tweets from {length(users)} users")
tweets <- save_tweets(tweets, file)

log_info("Total of {nrow(tweets)} tweets in {file}")
tweets
}

#' @export
gathertweet_simplify <- function(
file = "tweets.rds",
fields = NULL,
output = NULL,
...
) {
logger("Simplifying tweets in {file}")
if (!file.exists(file)) {
log_fatal("`{file}` does not exist")
}
tweets_simplified <- simplify_tweets(
tweets = NULL,
file = file,
.fields = fields
)
log_debug("Simplified {nrow(tweets_simplified)} tweets")
if (is.null(output)) {
output <- gathertweet:::path_add(file, append = "_simplified")
}
log_info("Saving simplified tweets to {output}")
save_tweets(tweets_simplified, output)
}

+ 2
- 0
R/logging.R Zobrazit soubor

futile.logger::flog.layout(gathertweet_layout, name = "gathertweet") futile.logger::flog.layout(gathertweet_layout, name = "gathertweet")
} }


collapse <- function(..., sep = ", ") paste(..., collapse = sep)

#' @title Logging functions #' @title Logging functions
#' @export #' @export
logger <- function(..., level = "info", envir = parent.frame()) { logger <- function(..., level = "info", envir = parent.frame()) {

+ 1
- 1
README.Rmd Zobrazit soubor

```{r setup, include = FALSE} ```{r setup, include = FALSE}
knitr::opts_chunk$set( knitr::opts_chunk$set(
collapse = TRUE, collapse = TRUE,
cache = TRUE,
cache = FALSE,
comment = "", comment = "",
prompt = TRUE, prompt = TRUE,
fig.path = "man/figures/README-", fig.path = "man/figures/README-",

+ 61
- 56
README.md Zobrazit soubor



``` bash ``` bash
> gathertweet search --n 100 --quiet "#rstats" > gathertweet search --n 100 --quiet "#rstats"
[2019-05-04 14:52:15] [INFO] ---- gathertweet search start ----
[2019-05-04 14:52:15] [INFO] Searching for "#rstats"
[2019-05-04 14:52:16] [INFO] Gathered 100 tweets
[2019-05-04 14:52:16] [INFO] Total of 100 tweets in tweets.rds
[2019-05-04 14:52:16] [INFO] ---- gathertweet search complete ----
INFO [2019-05-06 21:56:27] ---- gathertweet search start ----
INFO [2019-05-06 21:56:27] Searching for "#rstats"
INFO [2019-05-06 21:56:28] Gathered 98 tweets
INFO [2019-05-06 21:56:28] Total of 98 tweets in tweets.rds
INFO [2019-05-06 21:56:28] ---- gathertweet search complete ----
``` ```


Get more tweets, automatically starting from end of the last search Get more tweets, automatically starting from end of the last search


``` bash ``` bash
> gathertweet search --n 100 --quiet "#rstats" > gathertweet search --n 100 --quiet "#rstats"
[2019-05-04 14:53:17] [INFO] ---- gathertweet search start ----
[2019-05-04 14:53:17] [INFO] Searching for "#rstats"
[2019-05-04 14:53:17] [INFO] Tweets from 1124748486971359232
[2019-05-04 14:53:17] [INFO] Gathered 1 tweets
[2019-05-04 14:53:17] [INFO] Total of 100 tweets in tweets.rds
[2019-05-04 14:53:17] [INFO] ---- gathertweet search complete ----
INFO [2019-05-06 21:57:29] ---- gathertweet search start ----
INFO [2019-05-06 21:57:29] Searching for "#rstats"
INFO [2019-05-06 21:57:29] Tweets from 1125579895403352064
INFO [2019-05-06 21:57:29] No new tweets.
``` ```


Update the stored data about those \#rstats tweets Update the stored data about those \#rstats tweets


``` bash ``` bash
> gathertweet update > gathertweet update
[2019-05-04 14:53:18] [INFO] ---- gathertweet update start ----
[2019-05-04 14:53:18] [INFO] Updating tweets in tweets.rds
[2019-05-04 14:53:18] [INFO] Getting 100 tweets
[2019-05-04 14:53:19] [INFO] ---- gathertweet update complete ----
INFO [2019-05-06 21:57:30] ---- gathertweet update start ----
INFO [2019-05-06 21:57:30] Updating tweets in tweets.rds
INFO [2019-05-06 21:57:30] Getting 98 tweets
INFO [2019-05-06 21:57:31] ---- gathertweet update complete ----
``` ```


``` bash ``` bash
> ls -lh > ls -lh
total 40K total 40K
-rw-rw-r-- 1 garrick garrick 39K May 4 14:53 tweets.rds
-rw-rw-r-- 1 garrick garrick 39K May 6 21:57 tweets.rds
``` ```


Gather user timelines Gather user timelines


``` bash ``` bash
> gathertweet timeline hadleywickham jennybryan dataandme > gathertweet timeline hadleywickham jennybryan dataandme
[2019-05-04 21:11:54] [INFO] ---- gathertweet timeline start ----
[2019-05-04 21:11:54] [INFO] Gathering tweets by hadleywickham, jennybryan, dataandme
[2019-05-04 21:12:23] [INFO] Gathered 7368 tweets from 3 users
[2019-05-04 21:12:23] [INFO] Total of 7368 tweets in tweets.rds
[2019-05-04 21:12:23] [INFO] ---- gathertweet timeline complete ----
INFO [2019-05-06 21:57:32] ---- gathertweet timeline start ----
INFO [2019-05-06 21:57:32] Gathering tweets by hadleywickham, jennybryan, dataandme
WARN [2019-05-06 21:57:32] Twitter API for timelines returns a maximum of 3200 tweets per user
INFO [2019-05-06 21:58:01] Gathered 7427 tweets from 3 users
INFO [2019-05-06 21:58:02] Total of 7524 tweets in tweets.rds
INFO [2019-05-06 21:58:02] ---- gathertweet timeline complete ----
``` ```


### Schedule tweet gathering using cron ### Schedule tweet gathering using cron
Usage: Usage:
gathertweet search [--file=<file>] [options] [--] <terms>... gathertweet search [--file=<file>] [options] [--] <terms>...
gathertweet timeline [options] [--] <users>... gathertweet timeline [options] [--] <users>...
gathertweet update [--file=<file> --token=<token> --backup --backup-dir=<dir> --polite --debug-args]
gathertweet update [--file=<file> --and-simplify --polite --debug-args --token=<token> --backup --backup-dir=<dir>]
gathertweet simplify [--file=<file> --output=<output> --debug-args --polite] [<fields>...] gathertweet simplify [--file=<file> --output=<output> --debug-args --polite] [<fields>...]
Arguments
<terms> Search terms. Individual search terms are queried separately,
but duplicated tweets are removed from the stored results.
Each search term counts against the 15 minute rate limit of 180
searches, which can be avoided by manually joining search terms
into a single query. WARNING: Wrap queries with spaces in
'single quotes': double quotes are allowed inside single quotes only.
<fields> Tweet fields that should be included. Default value will include
`status_id`, `created_at`, `user_id`, `screen_name`, `text`,
`favorite_count`, `retweet_count`, `is_quote`, `hashtags`,
`mentions_screen_name`, `profile_url`, `profile_image_url`,
`media_url`, `urls_url`, `urls_expanded_url`.
Options: Options:
-h --help Show this screen. -h --help Show this screen.
--file <file> Name of RDS file where tweets are stored [default: tweets.rds]
--file <file> Name of RDS file where tweets are stored
[default: tweets.rds]
--no-parse Disable parsing of the results --no-parse Disable parsing of the results
--token <token> See {rtweet} for more information --token <token> See {rtweet} for more information
--retryonratelimit Wait and retry when rate limited (only relevant when n exceeds 18000 tweets)
--quiet Disable printing of {rtweet} processing/retrieval messages
--retryonratelimit Wait and retry when rate limited (only relevant when n
exceeds 18000 tweets)
--quiet Disable printing of {rtweet} processing messages
--polite Only allow one process (search|update) to run at a time --polite Only allow one process (search|update) to run at a time
--backup Create a backup of existing tweet file before writing any new files
--backup-dir <dir> Location for backups, use "" for current directory. [default: backups]
--debug-args Print values of the arguments only
--and-simplify Create additional simplified tweet set with default values.
--backup Create a backup of existing tweet file
--backup-dir <dir> Location for backups [default: backups]
--debug-args Debug input arguments
--and-simplify Create additional simplified tweet set.
Run `gathertweet simplify` manually for more control. Run `gathertweet simplify` manually for more control.
search:
<terms> Search terms. Individual search terms are queried separately,
but duplicated tweets are removed from the stored results.
Each search term counts against the 15 minute rate limit of 180
searches, which can be avoided by manually joining search terms
into a single query. NOTE: Wrap queries with spaces in
'single quotes': only use double quotes within single quotes.
--type <type> Type of search results: "recent", "mixed", or "popular"
[default: recent]
--geocode <geocode> Geographical limiter of the template
"latitude,longitude,radius"
--since_id <since_id> Return results with an ID greather than (newer than) or
equal to since_id, automatically extracted from the
existing tweets <file>, if it exists, and ignored when
<max_id> is set. Use "none" for all available tweets,
or "last" for the maximum seen status_id in existing
tweets. [default: last]
search and timeline: search and timeline:
-n, --n <n> Number of tweets to return [default: 18000]
--include_rts Logical indicating whether retweets should be included
--max_id <max_id> Return results with an ID less than (older than) or equal to max_id
search:
--type <type> Type of search results: "recent", "mixed", or "popular". [default: recent]
--geocode <geocode> Geographical limiter of the template "latitude,longitude,radius"
--since_id <since_id> Return results with an ID greather than (newer than) or equal to since_id,
automatically extracted from the existing tweets <file>, if it exists, and
ignored when <max_id> is set. Use "none" for all available tweets,
or "last" for the maximum seen status_id in existing tweets. [default: last]
-n, --n <n> Number of tweets to return [default: 18000]
--include_rts Logical indicating whether retweets should be included
(default is to exclude RTs)
--max_id <max_id> Return tweets with an ID less (older) than or equal to
timeline: timeline:
--home If included, returns home-timeline instead of user-timeline.
<users> A list of users as user names, IDs, or a mixture of both,
separated by spaces.
--home If included, returns home-timeline instead of user-timeline.
simplify: simplify:
--output <output> Output file, default is input file with `_simplified` appended to name.
<fields> Tweet fields that should be included. By default includes:
`status_id`, `created_at`, `user_id`, `screen_name`, `text`,
`favorite_count`, `retweet_count`, `is_quote`, `hashtags`,
`mentions_screen_name`, `profile_url`, `profile_image_url`,
`media_url`, `urls_url`, `urls_expanded_url`.
--output <output> Output file, default is input file with `_simplified`
appended to name.

+ 97
- 155
inst/gathertweet.R Zobrazit soubor

#! /usr/bin/env Rscript #! /usr/bin/env Rscript


# Usage -------------------------------------------------------------------
# Usage -----------------------------------------------------------------------
'Gather tweets from the command line 'Gather tweets from the command line


Usage: Usage:
gathertweet search [--file=<file>] [options] [--] <terms>... gathertweet search [--file=<file>] [options] [--] <terms>...
gathertweet timeline [options] [--] <users>... gathertweet timeline [options] [--] <users>...
gathertweet update [--file=<file> --token=<token> --backup --backup-dir=<dir> --polite --debug-args]
gathertweet update [--file=<file> --and-simplify --polite --debug-args --token=<token> --backup --backup-dir=<dir>]
gathertweet simplify [--file=<file> --output=<output> --debug-args --polite] [<fields>...] gathertweet simplify [--file=<file> --output=<output> --debug-args --polite] [<fields>...]


Arguments
<terms> Search terms. Individual search terms are queried separately,
but duplicated tweets are removed from the stored results.
Each search term counts against the 15 minute rate limit of 180
searches, which can be avoided by manually joining search terms
into a single query. WARNING: Wrap queries with spaces in
\'single quotes\': double quotes are allowed inside single quotes only.

<fields> Tweet fields that should be included. Default value will include
`status_id`, `created_at`, `user_id`, `screen_name`, `text`,
`favorite_count`, `retweet_count`, `is_quote`, `hashtags`,
`mentions_screen_name`, `profile_url`, `profile_image_url`,
`media_url`, `urls_url`, `urls_expanded_url`.

Options: Options:
-h --help Show this screen. -h --help Show this screen.
--file <file> Name of RDS file where tweets are stored [default: tweets.rds]
--file <file> Name of RDS file where tweets are stored
[default: tweets.rds]
--no-parse Disable parsing of the results --no-parse Disable parsing of the results
--token <token> See {rtweet} for more information --token <token> See {rtweet} for more information
--retryonratelimit Wait and retry when rate limited (only relevant when n exceeds 18000 tweets)
--quiet Disable printing of {rtweet} processing/retrieval messages
--retryonratelimit Wait and retry when rate limited (only relevant when n
exceeds 18000 tweets)
--quiet Disable printing of {rtweet} processing messages
--polite Only allow one process (search|update) to run at a time --polite Only allow one process (search|update) to run at a time
--backup Create a backup of existing tweet file before writing any new files
--backup-dir <dir> Location for backups, use "" for current directory. [default: backups]
--debug-args Print values of the arguments only
--and-simplify Create additional simplified tweet set with default values.
--backup Create a backup of existing tweet file
--backup-dir <dir> Location for backups [default: backups]
--debug-args Debug input arguments
--and-simplify Create additional simplified tweet set.
Run `gathertweet simplify` manually for more control. Run `gathertweet simplify` manually for more control.
search:
<terms> Search terms. Individual search terms are queried separately,
but duplicated tweets are removed from the stored results.
Each search term counts against the 15 minute rate limit of 180
searches, which can be avoided by manually joining search terms
into a single query. NOTE: Wrap queries with spaces in
\'single quotes\': only use double quotes within single quotes.
--type <type> Type of search results: "recent", "mixed", or "popular"
[default: recent]
--geocode <geocode> Geographical limiter of the template
"latitude,longitude,radius"
--since_id <since_id> Return results with an ID greather than (newer than) or
equal to since_id, automatically extracted from the
existing tweets <file>, if it exists, and ignored when
<max_id> is set. Use "none" for all available tweets,
or "last" for the maximum seen status_id in existing
tweets. [default: last]


search and timeline: search and timeline:
-n, --n <n> Number of tweets to return [default: 18000]
--include_rts Logical indicating whether retweets should be included
--max_id <max_id> Return results with an ID less than (older than) or equal to max_id

search:
--type <type> Type of search results: "recent", "mixed", or "popular". [default: recent]
--geocode <geocode> Geographical limiter of the template "latitude,longitude,radius"
--since_id <since_id> Return results with an ID greather than (newer than) or equal to since_id,
automatically extracted from the existing tweets <file>, if it exists, and
ignored when <max_id> is set. Use "none" for all available tweets,
or "last" for the maximum seen status_id in existing tweets. [default: last]
-n, --n <n> Number of tweets to return [default: 18000]
--include_rts Logical indicating whether retweets should be included
(default is to exclude RTs)
--max_id <max_id> Return tweets with an ID less (older) than or equal to


timeline: timeline:
--home If included, returns home-timeline instead of user-timeline.
<users> A list of users as user names, IDs, or a mixture of both,
separated by spaces.
--home If included, returns home-timeline instead of user-timeline.


simplify: simplify:
--output <output> Output file, default is input file with `_simplified` appended to name.
<fields> Tweet fields that should be included. By default includes:
`status_id`, `created_at`, `user_id`, `screen_name`, `text`,
`favorite_count`, `retweet_count`, `is_quote`, `hashtags`,
`mentions_screen_name`, `profile_url`, `profile_image_url`,
`media_url`, `urls_url`, `urls_expanded_url`.
--output <output> Output file, default is input file with `_simplified`
appended to name.
' -> doc ' -> doc


library(docopt) library(docopt)
exit() exit()
} }


library(gathertweet)
collapse <- function(..., sep = ", ") paste(..., collapse = sep)
do_gathertweet <- function() {
library(gathertweet)
collapse <- function(..., sep = ", ") paste(..., collapse = sep)


# Which action was called?
valid_actions <- c("search", "update", "simplify", "timeline")
action <- names(Filter(isTRUE, args[valid_actions]))
if (!length(action)) {
log_fatal("Please specify a valid action: {collapse(valid_actions)}")
}
# Which action was called?
valid_actions <- c("search", "update", "simplify", "timeline")
action <- names(Filter(isTRUE, args[valid_actions]))
if (!length(action)) {
log_fatal("Please specify a valid action: {collapse(valid_actions)}")
}


if (args$polite) {
lockfile <- paste0(".gathertweet_",
digest::digest(args[c("file", "search", "update", "simplify")]),
".lock")
lck <- filelock::lock(lockfile, exclusive = TRUE, timeout = 0)
gathertweet:::stopifnot_locked(lck, "Another gathertweet {action} process is currently running for {args$file}")
}
if (args$polite) {
lockfile <- paste0(
".gathertweet_",
digest::digest(args[c("file", "search", "update", "simplify")]),
".lock"
)
lck <- filelock::lock(lockfile, exclusive = TRUE, timeout = 0)
gathertweet:::stopifnot_locked(
lck,
"Another gathertweet {action} process is currently running for {args$file}"
)
on.exit({
filelock::unlock(lck)
unlink(lockfile)
})
}


log_info("---- gathertweet {action} start ----")
log_info("---- gathertweet {action} start ----")

if (isTRUE(args$backup)) {
backup_tweets(args$file, backup_dir = args[["backup-dir"]])
}


# Search ------------------------------------------------------------------
if (isTRUE(args$search)) {
# Also simplify if --and-simplify flag is called
if (args[["--and-simplify"]]) args$simplify <- TRUE if (args[["--and-simplify"]]) args$simplify <- TRUE


log_info("Searching for \"{paste0(args$terms, collapse = '\", \"')}\"")
tweets <-
# Search ----
if (isTRUE(args$search)) {


max_id <- args[["max_id"]]
since_id <- args[["since_id"]]
since_id <- if (is.null(max_id)) {
if (since_id == "last") {
last_seen_tweet(file = args$file)
} else if (since_id == "none") {
NULL
} else since_id
}
if (!is.null(since_id)) log_info("Tweets from {since_id}")
if (!is.null(max_id)) log_info("Tweets up to {max_id}")

tweets <- lapply(
args$term,
function(term) rtweet::search_tweets(
q = term,
n = as.integer(args$n),
type = args$type,
include_rts = args$include_rts,
geocode = args$geocode,
max_id = max_id,
parse = !args[["no-parse"]],
token = args$token,
retryonratelimit = args$retryonratelimit,
verbose = !args$quiet,
since_id = since_id
)
)
do.call("gathertweet_search", args)


tweets <- dplyr::bind_rows(tweets)
# Update ----
} else if (isTRUE(args$update)) {


if (nrow(tweets) == 0) {
log_info("No new tweets.")
exit()
}
do.call("gathertweet_update", args)


tweets <- tweets[!duplicated(tweets$status_id), ]
tweets <- tweets[order(tweets$status_id), ]

log_info("Gathered {nrow(tweets)} tweets")
if (args$backup) backup_tweets(args$file, backup_dir = args[["backup-dir"]])
tweets <- save_tweets(tweets, args$file)

log_info("Total of {nrow(tweets)} tweets in {args$file}")

# Update ------------------------------------------------------------------
} else if (isTRUE(args$update)) {
logger("Updating tweets in {args$file}")
tweets <- update_tweets(
file = args$file,
# passed to rtweet::lookup_statuses()
parse = !args[["no-parse"]],
token = args$token
)
log_debug("Status lookup returned {nrow(tweets)} tweets")
if (args$backup) backup_tweets(args$file, backup_dir = args[["backup-dir"]])
tweets <- save_tweets(tweets, args$file)
log_debug("Total of {nrow(tweets)} tweets in {args$file}")

} else if (isTRUE(args$timeline)) {
if (!length(args$users)) {
stop("Please provide a list of users as user names, user IDs, or a mixture of both.")
}
# Timeline ----
} else if (isTRUE(args$timeline)) {
if (!length(args$users)) {
stop("Please provide a list of users as user names, user IDs, ",
"or a mixture of both.")
}


log_info("Gathering tweets by {collapse(args$users)}")
if (args[["--and-simplify"]]) args$simplify <- TRUE
do.call("gathertweet_timeline", args)
}


tweets <- rtweet::get_timeline(
user = args[["users"]],
n = min(as.integer(args[["n"]]), 3200),
max_id = args[["max_id"]],
home = isTRUE(args[["home"]]),
parse = isFALSE(args[["no-parse"]]),
check = TRUE,
token = args$token,
include_rts = isTRUE(args[["include-rts"]])
)

tweets <- tweets[!duplicated(tweets$status_id), ]
tweets <- tweets[order(tweets$status_id), ]

log_info("Gathered {nrow(tweets)} tweets from {length(args$users)} users")
if (args$backup) backup_tweets(args$file, backup_dir = args[["backup-dir"]])
tweets <- save_tweets(tweets, args$file)

log_info("Total of {nrow(tweets)} tweets in {args$file}")
}


# Simplify ----------------------------------------------------------------
if (isTRUE(args$simplify)) {
do.call("gathertweet_simplify", args)
}


# Simplify ----------------------------------------------------------------
if (isTRUE(args$simplify)) {
logger("Simplifying tweets in {args$file}")
tweets_simplified <- simplify_tweets(
tweets = NULL,
file = args$file,
.fields = args$fields
)
log_debug("Simplified {nrow(tweets_simplified)} tweets")
if (is.null(args$output)) {
args$output <- gathertweet:::path_add(args$file, append = "_simplified")
if (args$polite) {
filelock::unlock(lck)
unlink(lockfile)
} }
log_info("Saving simplified tweets to {args$output}")
tweets_simplfied <- save_tweets(tweets_simplified, args$output)
}


if (args$polite) {
filelock::unlock(lck)
unlink(lockfile)
log_info("---- gathertweet {action} complete ----")
} }


log_info("---- gathertweet {action} complete ----")
do_gathertweet()

+ 14
- 0
man/gathertweet_search.Rd Zobrazit soubor

% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/gathertweet_actions.R
\name{gathertweet_search}
\alias{gathertweet_search}
\title{gathertweet actions}
\usage{
gathertweet_search(terms, file = "tweets.rds", n = 18000,
max_id = NULL, since_id = "last", type = "recent",
include_rts = FALSE, geocode = NULL, `no-parse` = FALSE,
token = NULL, retryonratelimit = FALSE, quiet = FALSE, ...)
}
\description{
gathertweet actions
}

Zapsat Náhled
Načítá se…
Zrušit
Uložit