使用 write.csv 和 write_csv {r} 导出到 csv 时出错
Error when exporting to csv using write.csv and write_csv {r}
我有一个类似于下面的数据框（只是更长）。当我尝试导出它时（使用不同的 methods/packages，我不断收到相同的错误：
Error in dimnames(X) <- list(dn[[1L]], unlist(collabs[nc > 0], use.names = FALSE)) :
length of 'dimnames' [2] not equal to array extent
我不确定错误的实际含义，谷歌搜索也没有太大帮助。
这是数据框的一个片段（下面可重现的数据是整个数据框的头部）。
> head(tweets2)
possibly_sensitive id author_id public_metrics.retweet_count public_metrics.reply_count
1 FALSE 1457730445089640453 4568748862 0 1
2 FALSE 1456653659811549193 4568748862 0 0
3 FALSE 1455688889889435650 4568748862 0 0
4 FALSE 1455571797919870980 4568748862 0 0
5 FALSE 1455557277369393160 4568748862 0 0
6 FALSE 1455288530159157249 4568748862 0 6
public_metrics.like_count public_metrics.quote_count
1 0 0
2 4 0
3 1 0
4 1 0
5 0 0
6 60 0
可重现的数据：
test2 <- structure(list(possibly_sensitive = c(FALSE, FALSE, FALSE, FALSE,
FALSE, FALSE), id = c("1457730445089640453", "1456653659811549193",
"1455688889889435650", "1455571797919870980", "1455557277369393160",
"1455288530159157249"), author_id = c("4568748862", "4568748862",
"4568748862", "4568748862", "4568748862", "4568748862"), public_metrics = structure(list(
retweet_count = c(0L, 0L, 0L, 0L, 0L, 0L), reply_count = c(1L,
0L, 0L, 0L, 0L, 6L), like_count = c(0L, 4L, 1L, 1L, 0L, 60L
), quote_count = c(0L, 0L, 0L, 0L, 0L, 0L)), row.names = c(NA,
6L), class = "data.frame"), entities = structure(list(urls = list(
structure(list(start = 237L, end = 260L, url = "X",
expanded_url = "X",
display_url = "X"), class = "data.frame", row.names = 1L),
structure(list(start = 241L, end = 264L, url = "X",
expanded_url = "X",
display_url = "X"), class = "data.frame", row.names = 1L),
structure(list(), .Names = character(0)), structure(list(
start = 262L, end = 285L, url = "X",
expanded_url = "https://twitter.com/nationalpost/status/1455243316866555908",
display_url = "twitter.com/nationalpost/s…"), class = "data.frame", row.names = 1L),
structure(list(), .Names = character(0)), structure(list(
start = c(218L, 218L), end = c(241L, 241L), url = c("X",
"X"), expanded_url = c("https://twitter.com/ziad_aboultaif/status/1455288530159157249/photo/1",
"https://twitter.com/ziad_aboultaif/status/1455288530159157249/photo/1"
), display_url = c("pic.twitter.com/SNxB53sDoE", "pic.twitter.com/SNxB53sDoE"
)), class = "data.frame", row.names = 1:2)), annotations = list(
structure(list(start = 0L, end = 5L, probability = 0.9922,
type = "Place", normalized_text = "Canada"), class = "data.frame", row.names = 1L),
structure(list(), .Names = character(0)), structure(list(), .Names = character(0)),
structure(list(start = 26L, end = 31L, probability = 0.9765,
type = "Place", normalized_text = "Canada"), class = "data.frame", row.names = 1L),
structure(list(), .Names = character(0)), structure(list(
start = c(43L, 81L), end = c(66L, 96L), probability = c(0.6856,
0.5389), type = c("Organization", "Place"), normalized_text = c("44th Canadian Parliament",
"Edmonton Manning")), class = "data.frame", row.names = 1:2)),
hashtags = list(structure(list(), .Names = character(0)),
structure(list(start = 8L, end = 21L, tag = "VeteransWeek"), class = "data.frame", row.names = 1L),
structure(list(), .Names = character(0)), structure(list(), .Names = character(0)),
structure(list(), .Names = character(0)), structure(list(), .Names = character(0))),
mentions = list(structure(list(), .Names = character(0)),
structure(list(), .Names = character(0)), structure(list(
start = 0L, end = 14L, username = "egyptincanada",
id = "1044603931496779777"), class = "data.frame", row.names = 1L),
structure(list(), .Names = character(0)), structure(list(
start = 0L, end = 11L, username = "Mukiza2021", id = "715376684501852160"), class = "data.frame", row.names = 1L),
structure(list(), .Names = character(0)))), row.names = c(NA,
6L), class = "data.frame"), conversation_id = c("1457730445089640453",
"1456653659811549193", "1455288530159157249", "1455571797919870980",
"1455288530159157249", "1455288530159157249"), source = c("Twitter Web App",
"Twitter Web App", "Twitter for Android", "Twitter Web App",
"Twitter for Android", "Twitter Web App"), text = c("Canada can’t afford a cabinet that is bent on radical policies in the midst of an economic crisis. What we need right now is to give workers their paychecks back and ensure a recovery for all. \n\nThis is not the time for a “great reset”. X",
"On this #VeteransWeek, let us remember the valiant efforts our veterans have made to keep us safe. \n\nToday, there are still many who saw the horrors of the Second World War and beyond. \n\nWe are forever indebted to our veterans. \n\nThank you. X",
"@egyptincanada Thank you your excellency.", "Five months after placing Canada’s flag on half-mast, it is time for our symbol of unity to start flying again. \nCanadians have shown that we are united towards reconciliation and it’s time to continue this work while our Maple Leaf flies high, strong and free. X",
"@Mukiza2021 Thank you my friend.", "I am honoured to be sworn-in today for the 44th Canadian Parliament representing Edmonton Manning. \n\nI will continue the hard work I have been doing for the last six years and I look forward to serving on your behalf. X"
), referenced_tweets = list(structure(list(type = "quoted", id = "1456333073621012480"), class = "data.frame", row.names = 1L),
structure(list(), .Names = character(0)), structure(list(
type = "replied_to", id = "1455688445586718722"), class = "data.frame", row.names = 1L),
structure(list(type = "quoted", id = "1455243316866555908"), class = "data.frame", row.names = 1L),
structure(list(type = "replied_to", id = "1455552008111345664"), class = "data.frame", row.names = 1L),
structure(list(), .Names = character(0))), lang = c("en",
"en", "en", "en", "en", "en"), created_at = c("2021-11-08T15:23:16.000Z",
"2021-11-05T16:04:30.000Z", "2021-11-03T00:10:51.000Z", "2021-11-02T16:25:34.000Z",
"2021-11-02T15:27:52.000Z", "2021-11-01T21:39:58.000Z"), attachments = structure(list(
media_keys = list(structure(list(), .Names = character(0)),
"3_1456653607873523714", structure(list(), .Names = character(0)),
structure(list(), .Names = character(0)), structure(list(), .Names = character(0)),
c("3_1455288520818544641", "3_1455288520994697217"))), row.names = c(NA,
6L), class = "data.frame"), in_reply_to_user_id = c(NA, NA, "1044603931496779777",
NA, "715376684501852160", NA), geo = structure(list(place_id = c(NA_character_,
NA_character_, NA_character_, NA_character_, NA_character_, NA_character_
), coordinates = structure(list(coordinates = list(NULL, NULL,
NULL, NULL, NULL, NULL), type = c(NA_character_, NA_character_,
NA_character_, NA_character_, NA_character_, NA_character_)), row.names = c(NA,
6L), class = "data.frame")), row.names = c(NA, 6L), class = "data.frame")), row.names = c(NA,
6L), class = "data.frame")
如果您查看 data.frame 的结构，您会注意到它是一个复杂的对象 - data.frame 的 data.frame 和列表以及您拥有的东西.
> str(test2)
'data.frame': 6 obs. of 14 variables:
$ possibly_sensitive : logi FALSE FALSE FALSE FALSE FALSE FALSE
$ id : chr "1457730445089640453" "1456653659811549193" "1455688889889435650" "1455571797919870980" ...
$ author_id : chr "4568748862" "4568748862" "4568748862" "4568748862" ...
$ public_metrics :'data.frame': 6 obs. of 4 variables:
..$ retweet_count: int 0 0 0 0 0 0
..$ reply_count : int 1 0 0 0 0 6
..$ like_count : int 0 4 1 1 0 60
..$ quote_count : int 0 0 0 0 0 0
$ entities :'data.frame': 6 obs. of 4 variables:
..$ urls :List of 6
.. ..$ :'data.frame': 1 obs. of 5 variables:
.. .. ..$ start : int 237
.. .. ..$ end : int 260
.. .. ..$ url : chr "X"
.. .. ..$ expanded_url: chr "X"
.. .. ..$ display_url : chr "X"
.. ..$ :'data.frame': 1 obs. of 5 variables:
您需要先将 data.frame 设为“平面”，然后才能安全地将其保存为 tab/csv 文件。
或者，您可以将此复杂对象另存为 .RData 二进制文件。不过，这只能通过 R 供其他用户使用。
正如 Roman 所提到的，您的数据框不是“扁平”的，并且某些列中有数据框，例如public_metrics.
您可以使用 tidyr::unpack() 来展平这些列，然后将其写入 CSV。
test2 %>%
unpack(cols = c(public_metrics, entities, attachments, geo)) %>%
unpack(cols = coordinates) %>%
write_csv("test2.csv")
我有一个类似于下面的数据框（只是更长）。当我尝试导出它时（使用不同的 methods/packages，我不断收到相同的错误：
Error in dimnames(X) <- list(dn[[1L]], unlist(collabs[nc > 0], use.names = FALSE)) :
length of 'dimnames' [2] not equal to array extent
我不确定错误的实际含义，谷歌搜索也没有太大帮助。
这是数据框的一个片段（下面可重现的数据是整个数据框的头部）。
> head(tweets2)
possibly_sensitive id author_id public_metrics.retweet_count public_metrics.reply_count
1 FALSE 1457730445089640453 4568748862 0 1
2 FALSE 1456653659811549193 4568748862 0 0
3 FALSE 1455688889889435650 4568748862 0 0
4 FALSE 1455571797919870980 4568748862 0 0
5 FALSE 1455557277369393160 4568748862 0 0
6 FALSE 1455288530159157249 4568748862 0 6
public_metrics.like_count public_metrics.quote_count
1 0 0
2 4 0
3 1 0
4 1 0
5 0 0
6 60 0
可重现的数据：
test2 <- structure(list(possibly_sensitive = c(FALSE, FALSE, FALSE, FALSE,
FALSE, FALSE), id = c("1457730445089640453", "1456653659811549193",
"1455688889889435650", "1455571797919870980", "1455557277369393160",
"1455288530159157249"), author_id = c("4568748862", "4568748862",
"4568748862", "4568748862", "4568748862", "4568748862"), public_metrics = structure(list(
retweet_count = c(0L, 0L, 0L, 0L, 0L, 0L), reply_count = c(1L,
0L, 0L, 0L, 0L, 6L), like_count = c(0L, 4L, 1L, 1L, 0L, 60L
), quote_count = c(0L, 0L, 0L, 0L, 0L, 0L)), row.names = c(NA,
6L), class = "data.frame"), entities = structure(list(urls = list(
structure(list(start = 237L, end = 260L, url = "X",
expanded_url = "X",
display_url = "X"), class = "data.frame", row.names = 1L),
structure(list(start = 241L, end = 264L, url = "X",
expanded_url = "X",
display_url = "X"), class = "data.frame", row.names = 1L),
structure(list(), .Names = character(0)), structure(list(
start = 262L, end = 285L, url = "X",
expanded_url = "https://twitter.com/nationalpost/status/1455243316866555908",
display_url = "twitter.com/nationalpost/s…"), class = "data.frame", row.names = 1L),
structure(list(), .Names = character(0)), structure(list(
start = c(218L, 218L), end = c(241L, 241L), url = c("X",
"X"), expanded_url = c("https://twitter.com/ziad_aboultaif/status/1455288530159157249/photo/1",
"https://twitter.com/ziad_aboultaif/status/1455288530159157249/photo/1"
), display_url = c("pic.twitter.com/SNxB53sDoE", "pic.twitter.com/SNxB53sDoE"
)), class = "data.frame", row.names = 1:2)), annotations = list(
structure(list(start = 0L, end = 5L, probability = 0.9922,
type = "Place", normalized_text = "Canada"), class = "data.frame", row.names = 1L),
structure(list(), .Names = character(0)), structure(list(), .Names = character(0)),
structure(list(start = 26L, end = 31L, probability = 0.9765,
type = "Place", normalized_text = "Canada"), class = "data.frame", row.names = 1L),
structure(list(), .Names = character(0)), structure(list(
start = c(43L, 81L), end = c(66L, 96L), probability = c(0.6856,
0.5389), type = c("Organization", "Place"), normalized_text = c("44th Canadian Parliament",
"Edmonton Manning")), class = "data.frame", row.names = 1:2)),
hashtags = list(structure(list(), .Names = character(0)),
structure(list(start = 8L, end = 21L, tag = "VeteransWeek"), class = "data.frame", row.names = 1L),
structure(list(), .Names = character(0)), structure(list(), .Names = character(0)),
structure(list(), .Names = character(0)), structure(list(), .Names = character(0))),
mentions = list(structure(list(), .Names = character(0)),
structure(list(), .Names = character(0)), structure(list(
start = 0L, end = 14L, username = "egyptincanada",
id = "1044603931496779777"), class = "data.frame", row.names = 1L),
structure(list(), .Names = character(0)), structure(list(
start = 0L, end = 11L, username = "Mukiza2021", id = "715376684501852160"), class = "data.frame", row.names = 1L),
structure(list(), .Names = character(0)))), row.names = c(NA,
6L), class = "data.frame"), conversation_id = c("1457730445089640453",
"1456653659811549193", "1455288530159157249", "1455571797919870980",
"1455288530159157249", "1455288530159157249"), source = c("Twitter Web App",
"Twitter Web App", "Twitter for Android", "Twitter Web App",
"Twitter for Android", "Twitter Web App"), text = c("Canada can’t afford a cabinet that is bent on radical policies in the midst of an economic crisis. What we need right now is to give workers their paychecks back and ensure a recovery for all. \n\nThis is not the time for a “great reset”. X",
"On this #VeteransWeek, let us remember the valiant efforts our veterans have made to keep us safe. \n\nToday, there are still many who saw the horrors of the Second World War and beyond. \n\nWe are forever indebted to our veterans. \n\nThank you. X",
"@egyptincanada Thank you your excellency.", "Five months after placing Canada’s flag on half-mast, it is time for our symbol of unity to start flying again. \nCanadians have shown that we are united towards reconciliation and it’s time to continue this work while our Maple Leaf flies high, strong and free. X",
"@Mukiza2021 Thank you my friend.", "I am honoured to be sworn-in today for the 44th Canadian Parliament representing Edmonton Manning. \n\nI will continue the hard work I have been doing for the last six years and I look forward to serving on your behalf. X"
), referenced_tweets = list(structure(list(type = "quoted", id = "1456333073621012480"), class = "data.frame", row.names = 1L),
structure(list(), .Names = character(0)), structure(list(
type = "replied_to", id = "1455688445586718722"), class = "data.frame", row.names = 1L),
structure(list(type = "quoted", id = "1455243316866555908"), class = "data.frame", row.names = 1L),
structure(list(type = "replied_to", id = "1455552008111345664"), class = "data.frame", row.names = 1L),
structure(list(), .Names = character(0))), lang = c("en",
"en", "en", "en", "en", "en"), created_at = c("2021-11-08T15:23:16.000Z",
"2021-11-05T16:04:30.000Z", "2021-11-03T00:10:51.000Z", "2021-11-02T16:25:34.000Z",
"2021-11-02T15:27:52.000Z", "2021-11-01T21:39:58.000Z"), attachments = structure(list(
media_keys = list(structure(list(), .Names = character(0)),
"3_1456653607873523714", structure(list(), .Names = character(0)),
structure(list(), .Names = character(0)), structure(list(), .Names = character(0)),
c("3_1455288520818544641", "3_1455288520994697217"))), row.names = c(NA,
6L), class = "data.frame"), in_reply_to_user_id = c(NA, NA, "1044603931496779777",
NA, "715376684501852160", NA), geo = structure(list(place_id = c(NA_character_,
NA_character_, NA_character_, NA_character_, NA_character_, NA_character_
), coordinates = structure(list(coordinates = list(NULL, NULL,
NULL, NULL, NULL, NULL), type = c(NA_character_, NA_character_,
NA_character_, NA_character_, NA_character_, NA_character_)), row.names = c(NA,
6L), class = "data.frame")), row.names = c(NA, 6L), class = "data.frame")), row.names = c(NA,
6L), class = "data.frame")
如果您查看 data.frame 的结构，您会注意到它是一个复杂的对象 - data.frame 的 data.frame 和列表以及您拥有的东西.
> str(test2)
'data.frame': 6 obs. of 14 variables:
$ possibly_sensitive : logi FALSE FALSE FALSE FALSE FALSE FALSE
$ id : chr "1457730445089640453" "1456653659811549193" "1455688889889435650" "1455571797919870980" ...
$ author_id : chr "4568748862" "4568748862" "4568748862" "4568748862" ...
$ public_metrics :'data.frame': 6 obs. of 4 variables:
..$ retweet_count: int 0 0 0 0 0 0
..$ reply_count : int 1 0 0 0 0 6
..$ like_count : int 0 4 1 1 0 60
..$ quote_count : int 0 0 0 0 0 0
$ entities :'data.frame': 6 obs. of 4 variables:
..$ urls :List of 6
.. ..$ :'data.frame': 1 obs. of 5 variables:
.. .. ..$ start : int 237
.. .. ..$ end : int 260
.. .. ..$ url : chr "X"
.. .. ..$ expanded_url: chr "X"
.. .. ..$ display_url : chr "X"
.. ..$ :'data.frame': 1 obs. of 5 variables:
您需要先将 data.frame 设为“平面”，然后才能安全地将其保存为 tab/csv 文件。
或者，您可以将此复杂对象另存为 .RData 二进制文件。不过，这只能通过 R 供其他用户使用。
正如 Roman 所提到的，您的数据框不是“扁平”的，并且某些列中有数据框，例如public_metrics.
您可以使用 tidyr::unpack() 来展平这些列，然后将其写入 CSV。
test2 %>%
unpack(cols = c(public_metrics, entities, attachments, geo)) %>%
unpack(cols = coordinates) %>%
write_csv("test2.csv")