UPDATE: 2022-10-28 20:15:39
ここでは{stringr}を使い、正規表現40本ノックを行う。100本の予定だったけど、やる元気がなかった。新しいパターンが実務で発生したら追加していく。
参考にするときはブラウザの検索機能を利用する。
下記は参考サイトおよび正規表現チェックツール。javascriptの正規表現マッチなので、Rでエスケープする場合とは異なるので注意。
下記はメタキャラクタと呼ばれるもの。これらを文字として使う場合エスケープする必要がある。
. ^ $ * ? { } [ ] \ | ( )| 正規表現 | パターン |
|---|---|
\ |
メタキャラクタのエスケープ。 |
. |
任意の1文字 に一致します。 |
+ |
直前の文字が1回以上 繰り返す場合に一致します。最長一致。 |
* |
直前の文字が0回以上 繰り返す場合に一致します。最長一致。 |
? |
直前の文字が0個か１個 の場合に一致します。最長一致。 |
+? |
直前の文字が1回以上 繰り返す場合に一致します。最短一致。 |
*? |
直前の文字が0回以上 繰り返す場合に一致します。最短一致。 |
?? |
直前の文字が0個か1個の場合に一致します。最短一致。 |
\| |
Or条件。 |
[...] |
文字リスト。括弧に含まれるいずれか１文字に一致します。 |
[^...] |
反転文字リスト。括弧に含まれるいずれか文字以外に一致します。 |
(...) |
正規表現のグループ化。 |
{n} |
直前の文字のn回の繰り返しに一致します。 |
{n,} |
少なくともn回の繰り返しに一致します。 |
{n,m} |
n回~m回の繰り返しに一致します。 |
^ |
直後の文字が行の先頭にある場合に一致します。 |
$ |
直前の文字が行の末尾にある場合に一致します。 |
\t |
タブに一致します。 |
\\r |
改行に一致します。CR(Carriage Return : 0x0D) |
\\n |
改行に一致します。LF(Line Feed : 0x0A) |
\\d |
すべての数字に一致します。[0-9] |
\\D |
すべての数字以外の文字に一致します。[^0-9] |
\\s |
垂直タブ以外のすべての空白文字に一致します。[ ] |
\\S |
すべての非空白文字に一致します。[^ ] |
\\w |
アルファベット、アンダーバー、数字に一致します。[a-zA-Z_0-9] |
\\W |
アルファベット、アンダーバー、数字以外の文字に一致します。[^a-zA-Z_0-9] |
\\< |
単語の先頭に一致します。 |
\\> |
単語の末尾に一致します。 |
\\b |
単語の両端の空文字列に一致します。 |
\\B |
単語の端にない場合、空の文字列に一致します。 |
(?=) |
先読み。X(?=Y)は、Xを探すがYが続く場合にだけ一致します。 |
(?!) |
否定先読み。X(?!Y)は、Xを探すがYが続かない場合にだけ一致します。 |
(?<=) |
後読み。(?<=Y)Xは、Xの前にYがある場合にのみ一致します。 |
(?<!) |
否定後読み。(?<!Y)Xは、Xの前にYがない場合にのみ一致します。 |
[:digit:] |
[0-9]に一致します。 |
[:lower:] |
小文字[a-z]に一致します。 |
[:upper:] |
大文字[A-Z]に一致します。 |
[:alpha:] |
[A-z]に一致します。 |
[:alnum:] |
英数字[A-z0-9]に一致します。 |
[:xdigit:] |
16進数[0-9A-Fa-f]に一致します。 |
[:blank:] |
スペースとタブに一致します。 |
[:space:] |
タブ、改行、垂直タブ、フォームフィード、CR、スペースに一致します。 |
[:punct:] |
句読文字に一致します。「!“#$%&’()*+,-./:;<=>?@[]^_{|}~」。| |[:cntrl:]|制御文字など\nや\rに一致します。[\x00-\x1F\x7F]。| |[0-9]| 数字に一致します。| |[a-z]| アルファベット小文字に一致します。| |[A-z]|アルファベットに一致します。 |[A-Z]| アルファベット大文字に一致します。| |[ぁ-ん]| ひらがなに一致します。| |[ァ-ヶ]| カタカナに一致します。| |[ｦ-ﾟ]` |
strings <- str_c("This is a ",
c("apple", "orange", "Banana", "cherry", "plum", "strawberry", "persimmon"),
".")
regs <- "This is a ......\\."
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 7 x 3
strings match1 match2
<chr> <lgl> <chr>
1 This is a apple. FALSE NA
2 This is a orange. TRUE This is a orange.
3 This is a Banana. TRUE This is a Banana.
4 This is a cherry. TRUE This is a cherry.
5 This is a plum. FALSE NA
6 This is a strawberry. FALSE NA
7 This is a persimmon. FALSE NAstrings <- c("Goooood morning", "Good morning", "God morning", "Gooddood morning")
regs <- "Gooo*d morning"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 4 x 3
strings match1 match2
<chr> <lgl> <chr>
1 Goooood morning TRUE Goooood morning
2 Good morning TRUE Good morning
3 God morning FALSE NA
4 Gooddood morning FALSE NAstrings <- c("Goooood morning", "Good morning", "God morning", "Gooddood morning")
regs <- "Gooo+d morning"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 4 x 3
strings match1 match2
<chr> <lgl> <chr>
1 Goooood morning TRUE Goooood morning
2 Good morning FALSE NA
3 God morning FALSE NA
4 Gooddood morning FALSE NA
strings <- c("Goooood morning", "Gooooood morning", "God morning", "Gooddood morning")
regs <- "Go{5}d morning"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 4 x 3
strings match1 match2
<chr> <lgl> <chr>
1 Goooood morning TRUE Goooood morning
2 Gooooood morning FALSE NA
3 God morning FALSE NA
4 Gooddood morning FALSE NAstrings <- c("Goooood morning", "Gooooood morning", "God morning", "Gooddood morning")
regs <- "Go{5,}d morning"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 4 x 3
strings match1 match2
<chr> <lgl> <chr>
1 Goooood morning TRUE Goooood morning
2 Gooooood morning TRUE Gooooood morning
3 God morning FALSE NA
4 Gooddood morning FALSE NAstrings <- c("Goooood morning", "Gooooood morning", "God morning", "Goooooooood morning", "Gooddood morning")
regs <- "Go{5,6}d morning"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 5 x 3
strings match1 match2
<chr> <lgl> <chr>
1 Goooood morning TRUE Goooood morning
2 Gooooood morning TRUE Gooooood morning
3 God morning FALSE NA
4 Goooooooood morning FALSE NA
5 Gooddood morning FALSE NAregs <- "Good-?morning"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 5 x 3
strings match1 match2
<chr> <lgl> <chr>
1 Good morning FALSE NA
2 Goodmorning TRUE Goodmorning
3 Good-morning TRUE Good-morning
4 Good/morning FALSE NA
5 Good--morning FALSE NAstrings <- c("Language R", "R", "R Language", "R script",
"Language Python", "Python", "Python Language", "R script")
regs <- "^R.*"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 8 x 3
strings match1 match2
<chr> <lgl> <chr>
1 Language R FALSE NA
2 R TRUE R
3 R Language TRUE R Language
4 R script TRUE R script
5 Language Python FALSE NA
6 Python FALSE NA
7 Python Language FALSE NA
8 R script TRUE R scriptstrings <- c("Language R", "R", "R Language", "R script",
"Language Python", "Python", "Python Language", "R script")
regs <- "script$"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 8 x 3
strings match1 match2
<chr> <lgl> <chr>
1 Language R FALSE NA
2 R FALSE NA
3 R Language FALSE NA
4 R script TRUE script
5 Language Python FALSE NA
6 Python FALSE NA
7 Python Language FALSE NA
8 R script TRUE scriptstrings <- c("I wish you every happiness.",
"He did not die happily.",
"happify our Social system.",
"His letter made me happy.")
regs <- ".*\\bhappy\\b.*"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 4 x 3
strings match1 match2
<chr> <lgl> <chr>
1 I wish you every happiness. FALSE NA
2 He did not die happily. FALSE NA
3 happify our Social system. FALSE NA
4 His letter made me happy. TRUE His letter made me happy.strings <- c("I like R.", "I like C.", "I like Python.", "I like C++.", "I like なでしこ.")
regs <- "I like [R|P].*"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 5 x 3
strings match1 match2
<chr> <lgl> <chr>
1 I like R. TRUE I like R.
2 I like C. FALSE NA
3 I like Python. TRUE I like Python.
4 I like C++. FALSE NA
5 I like なでしこ. FALSE NAこのようにもかけるが、I like Pyが意図したようにマッチしていないので、指定したいずれかの1文字とマッチすることに注意。
strings <- c("I like R.", "I like C.", "I like Python.", "I like C++.", "I like なでしこ.")
regs <- "I like [R|Python]."
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 5 x 3
strings match1 match2
<chr> <lgl> <chr>
1 I like R. TRUE I like R.
2 I like C. FALSE NA
3 I like Python. TRUE I like Py
4 I like C++. FALSE NA
5 I like なでしこ. FALSE NAstrings <- c("I like R.", "I like C.", "I like Python.", "I like C++.", "I like なでしこ.")
regs <- "I like [^R|^P].*"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 5 x 3
strings match1 match2
<chr> <lgl> <chr>
1 I like R. FALSE NA
2 I like C. TRUE I like C.
3 I like Python. FALSE NA
4 I like C++. TRUE I like C++.
5 I like なでしこ. TRUE I like なでしこ.strings <- c("I like R.", "I like S.", "I like TeX.", "I like C", "I like Python.")
# [R|S]でも同じ
regs <- "I like [R-S]."
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 5 x 3
strings match1 match2
<chr> <lgl> <chr>
1 I like R. TRUE I like R.
2 I like S. TRUE I like S.
3 I like TeX. FALSE NA
4 I like C FALSE NA
5 I like Python. FALSE NAアルファベットの範囲は注意が必要。
m <- c(LETTERS, letters, "[", "\\", "]", "^", "_", "`")
m[str_detect(m, "[A-Z]")]
[1] "A" "B" "C" "D" "E" "F" "G" "H" "I" "J" "K" "L" "M" "N" "O" "P" "Q" "R" "S" "T" "U" "V" "W" "X" "Y" "Z"
m[str_detect(m, "[a-z]")]
[1] "a" "b" "c" "d" "e" "f" "g" "h" "i" "j" "k" "l" "m" "n" "o" "p" "q" "r" "s" "t" "u" "v" "w" "x" "y" "z"
m[str_detect(m, "[A-z]")]
[1] "A" "B" "C" "D" "E" "F" "G" "H" "I" "J" "K" "L" "M" "N" "O" "P" "Q" "R" "S" "T" "U" "V" "W" "X" "Y" "Z"
[27] "a" "b" "c" "d" "e" "f" "g" "h" "i" "j" "k" "l" "m" "n" "o" "p" "q" "r" "s" "t" "u" "v" "w" "x" "y" "z"
[53] "[" "\\" "]" "^" "_" "`"strings <- c("I like R.", "I like S.", "I like TeX.", "I like C", "I like Python.")
regs <- "I like [^R-S].*"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 5 x 3
strings match1 match2
<chr> <lgl> <chr>
1 I like R. FALSE NA
2 I like S. FALSE NA
3 I like TeX. TRUE I like TeX.
4 I like C TRUE I like C
5 I like Python. TRUE I like Python.strings <- c("R version 11, '10",
"R version 01, 2020",
"11 R version 2020",
"R version 2020 11 10",
"R version 2020 11 ten",
"R version 11, '10")
regs <- "R version \\d\\d\\d\\d\\s\\d\\d\\s\\d\\d"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 6 x 3
strings match1 match2
<chr> <lgl> <chr>
1 R version 11, '10 FALSE NA
2 R version 01, 2020 FALSE NA
3 11 R version 2020 FALSE NA
4 R version 2020 11 10 TRUE R version 2020 11 10
5 R version 2020 11 ten FALSE NA
6 R version 11, '10 FALSE NAstrings <- c("R version 11, '10",
"R version 01, 2020",
"11 R version 2020",
"R version 2020 11 10",
"R version 2020 11, ten",
"R version 11, '10")
regs <- "R version \\d\\d\\d\\d\\s\\d\\d,\\s\\D\\D\\D"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 6 x 3
strings match1 match2
<chr> <lgl> <chr>
1 R version 11, '10 FALSE NA
2 R version 01, 2020 FALSE NA
3 11 R version 2020 FALSE NA
4 R version 2020 11 10 FALSE NA
5 R version 2020 11, ten TRUE R version 2020 11, ten
6 R version 11, '10 FALSE NAregs <- "R \\d{4,} version"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 5 x 3
strings match1 match2
<chr> <lgl> <chr>
1 R 01 version FALSE NA
2 R 012 version FALSE NA
3 R 0123 version TRUE R 0123 version
4 R 01234 version TRUE R 01234 version
5 R 012345 version TRUE R 012345 version
regs <- "R \\d{3,4} version"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 5 x 3
strings match1 match2
<chr> <lgl> <chr>
1 R 01 version FALSE NA
2 R 012 version TRUE R 012 version
3 R 0123 version TRUE R 0123 version
4 R 01234 version FALSE NA
5 R 012345 version FALSE NA
regs <- "R \\d{5} version"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 5 x 3
strings match1 match2
<chr> <lgl> <chr>
1 R 01 version FALSE NA
2 R 012 version FALSE NA
3 R 0123 version FALSE NA
4 R 01234 version TRUE R 01234 version
5 R 012345 version FALSE NAstrings <- c("2020-01-01","2020/01/01","2020 01 01","2020.01.01","01-01-2020","01 2020 01")
regs <- "\\d\\d\\d\\d[/|\\-|\\.]\\d\\d[/|\\-|\\.]\\d\\d"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 6 x 3
strings match1 match2
<chr> <lgl> <chr>
1 2020-01-01 TRUE 2020-01-01
2 2020/01/01 TRUE 2020/01/01
3 2020 01 01 FALSE NA
4 2020.01.01 TRUE 2020.01.01
5 01-01-2020 FALSE NA
6 01 2020 01 FALSE NAstrings <- c("おはようございます",
"084531ます",
"こんにちは",
"オハヨウゴザイマス",
"おハよウごザいマす",
"コンバンワ",
"Good morning")
regs <- "[ぁ-んァ-ヶ]{6,}"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 7 x 3
strings match1 match2
<chr> <lgl> <chr>
1 おはようございます TRUE おはようございます
2 084531ます FALSE NA
3 こんにちは FALSE NA
4 オハヨウゴザイマス TRUE オハヨウゴザイマス
5 おハよウごザいマす TRUE おハよウごザいマす
6 コンバンワ FALSE NA
7 Good morning FALSE NAstrings <- c("good",
"GOOD",
"GoOd",
"gOoD",
"GOod",
"goOd")
regs <- "[Gg][Oo][Oo]d"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 6 x 3
strings match1 match2
<chr> <lgl> <chr>
1 good TRUE good
2 GOOD FALSE NA
3 GoOd TRUE GoOd
4 gOoD FALSE NA
5 GOod TRUE GOod
6 goOd TRUE goOd
strings <- c(stringr::sentences[1:5])
# dから始まる単語
regs <- "\\bd\\w*\\b"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match_all(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 6 x 3
strings match1 match2[,1]
<chr> <lgl> <chr>
1 The birch canoe slid on the smooth planks. FALSE NA
2 Glue the sheet to the dark blue background. TRUE dark
3 It's easy to tell the depth of a well. TRUE depth
4 These days a chicken leg is a rare dish. TRUE days
5 These days a chicken leg is a rare dish. TRUE dish
6 Rice is often served in round bowls. FALSE NAstrings <- c("'りんご'は'リンゴ'であり'赤い果物'でもあり'Apple'でもある。",
"An 'Apple' is 'APPLE', 'Red fruit', 'リンゴ'.")
regs <- "'.*?'"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match_all(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 8 x 3
strings match1 match2[,1]
<chr> <lgl> <chr>
1 'りんご'は'リンゴ'であり'赤い果物'でもあり'Apple'でもある。 TRUE 'りんご'
2 'りんご'は'リンゴ'であり'赤い果物'でもあり'Apple'でもある。 TRUE 'リンゴ'
3 'りんご'は'リンゴ'であり'赤い果物'でもあり'Apple'でもある。 TRUE '赤い果物'
4 'りんご'は'リンゴ'であり'赤い果物'でもあり'Apple'でもある。 TRUE 'Apple'
5 An 'Apple' is 'APPLE', 'Red fruit', 'リンゴ'. TRUE 'Apple'
6 An 'Apple' is 'APPLE', 'Red fruit', 'リンゴ'. TRUE 'APPLE'
7 An 'Apple' is 'APPLE', 'Red fruit', 'リンゴ'. TRUE 'Red fruit'
8 An 'Apple' is 'APPLE', 'Red fruit', 'リンゴ'. TRUE 'リンゴ'strings <- c("<p><a>apple.com</a></p>",
"<h1><a>apple.com</a></h1>",
"<p>apple.com</p>",
"<p><a>apple.com</p></a>")
regs <- "<\\D+><\\w+>.*</\\w+></\\w+>"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 4 x 3
strings match1 match2
<chr> <lgl> <chr>
1 <p><a>apple.com</a></p> TRUE <p><a>apple.com</a></p>
2 <h1><a>apple.com</a></h1> FALSE NA
3 <p>apple.com</p> FALSE NA
4 <p><a>apple.com</p></a> TRUE <p><a>apple.com</p></a>最長マッチしている例。
strings <- c("This is a 'pen' and 'PEN'.")
regs <- "'.*'"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 1 x 3
strings match1 match2
<chr> <lgl> <chr>
1 This is a 'pen' and 'PEN'. TRUE 'pen' and 'PEN'最短マッチしている例。
strings <- c("This is a 'pen' and 'PEN'.")
regs <- "'.*?'"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match_all(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 2 x 3
strings match1 match2[,1]
<chr> <lgl> <chr>
1 This is a 'pen' and 'PEN'. TRUE 'pen'
2 This is a 'pen' and 'PEN'. TRUE 'PEN'先読み：X(?=Y)で、Xを探すけどYが続く場合にだけ一致。
strings <- c("Mike is born in Heisei1 and 30 years old.",
"Tomy is born in Reiwa4 and 5 years old.")
regs <- "\\d+(?=\\syears)"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match_all(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 2 x 3
strings match1 match2[,1]
<chr> <lgl> <chr>
1 Mike is born in Heisei1 and 30 years old. TRUE 30
2 Tomy is born in Reiwa4 and 5 years old. TRUE 5否定-先読み：X(?!Y)は、Xを探すがYが続かない場合にだけ一致。
strings <- c("Mike is born in Heisei1 and 30 years old.",
"Tomy is born in Reiwa4 and 5 years old.")
regs <- "\\d+(?!\\syears|\\d+)"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match_all(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 2 x 3
strings match1 match2[,1]
<chr> <lgl> <chr>
1 Mike is born in Heisei1 and 30 years old. TRUE 1
2 Tomy is born in Reiwa4 and 5 years old. TRUE 4後読み：(?<=Y)Xは、Xの前にYがある場合にのみ一致。
strings <- c("Year2018 is Heisei30.",
"Year2020 is Reiwa1.")
regs <- "(?<=Heisei|Reiwa)\\d+"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match_all(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 2 x 3
strings match1 match2[,1]
<chr> <lgl> <chr>
1 Year2018 is Heisei30. TRUE 30
2 Year2020 is Reiwa1. TRUE 1否定-後読み：(?<!Y)Xは、Xの前にYがない場合にのみ一致。
strings <- c("Year2018 is Heisei30.",
"Year2020 is Reiwa1.")
regs <- "(?<!Heisei|Reiwa|\\d)\\d+"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match_all(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 2 x 3
strings match1 match2[,1]
<chr> <lgl> <chr>
1 Year2018 is Heisei30. TRUE 2018
2 Year2020 is Reiwa1. TRUE 2020strings <- c("123-4567","1234567","12-34567","1234-567","1-234567","12345678")
regs <- "^[0-9]{3}-?[0-9]{4}$"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 6 x 3
strings match1 match2
<chr> <lgl> <chr>
1 123-4567 TRUE 123-4567
2 1234567 TRUE 1234567
3 12-34567 FALSE NA
4 1234-567 FALSE NA
5 1-234567 FALSE NA
6 12345678 FALSE NA単語の境界にマッチする表現は\bで表現できる。\bは単語の境界にマッチする。
strings <- c("080-4567-1234","08012345678","01012345678","070-1234-5678","1-234567","12345678901")
regs <- "^0[7-9]0-?[0-9]{4}-?[0-9]{4}$"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 6 x 3
strings match1 match2
<chr> <lgl> <chr>
1 080-4567-1234 TRUE 080-4567-1234
2 08012345678 TRUE 08012345678
3 01012345678 FALSE NA
4 070-1234-5678 TRUE 070-1234-5678
5 1-234567 FALSE NA
6 12345678901 FALSE NAstrings <- c("01:10","12:34","28:10","01:82","04:10","21-10","15/10","5:1")
regs <- "(?:[01][0-9]|2[0-3]):[0-5][0-9]"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match_all(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 8 x 3
strings match1 match2[,1]
<chr> <lgl> <chr>
1 01:10 TRUE 01:10
2 12:34 TRUE 12:34
3 28:10 FALSE NA
4 01:82 FALSE NA
5 04:10 TRUE 04:10
6 21-10 FALSE NA
7 15/10 FALSE NA
8 5:1 FALSE NAstrings <- c("1920/11/01","1020-11-01","1999.12.01","20341101",
"H32/11/01","2oo0/11/01","12020/11/01","899/11/01")
regs <- "^\\d{4}[-./]?\\d{1,2}[-./]?\\d{1,2}$"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match_all(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 8 x 3
strings match1 match2[,1]
<chr> <lgl> <chr>
1 1920/11/01 TRUE 1920/11/01
2 1020-11-01 TRUE 1020-11-01
3 1999.12.01 TRUE 1999.12.01
4 20341101 TRUE 20341101
5 H32/11/01 FALSE NA
6 2oo0/11/01 FALSE NA
7 12020/11/01 FALSE NA
8 899/11/01 FALSE NA単語の境界にマッチする表現は\bで表現できる。\bは単語の境界にマッチする。
strings <- c("正規表現(せいきひょうげん)は文字集合(Set)を1つの文字列(モジレツ)で表現(Expression)する方法の一つである。")
regs <- "\\(.+?\\)"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match_all(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 4 x 3
strings match1 match2[,1]
<chr> <lgl> <chr>
1 正規表現(せいきひょうげん)は文字集合(Set)を1つの文字列(モジレツ)で表現(Expression)する方法の一つである。 TRUE (せいきひょうげん)
2 正規表現(せいきひょうげん)は文字集合(Set)を1つの文字列(モジレツ)で表現(Expression)する方法の一つである。 TRUE (Set)
3 正規表現(せいきひょうげん)は文字集合(Set)を1つの文字列(モジレツ)で表現(Expression)する方法の一つである。 TRUE (モジレツ)
4 正規表現(せいきひょうげん)は文字集合(Set)を1つの文字列(モジレツ)で表現(Expression)する方法の一つである。 TRUE (Expression)strings <- c("FirstName = 'Taro'",
"LastName = 'Sato'",
"Age = 20",
"BOD = '1989/06/23",
"Hobby = 'A,B,C'",
"Memo")
regs <- "\\w+\\s*=\\s*.*"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match_all(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 6 x 3
strings match1 match2[,1]
<chr> <lgl> <chr>
1 FirstName = 'Taro' TRUE FirstName = 'Taro'
2 LastName = 'Sato' TRUE LastName = 'Sato'
3 Age = 20 TRUE Age = 20
4 BOD = '1989/06/23 TRUE BOD = '1989/06/23
5 Hobby = 'A,B,C' TRUE Hobby = 'A,B,C'
6 Memo FALSE NAstrings <- c("/usr/home/desktop/analysis.R",
"/usr/home/document/",
"../logs/error.log",
"~work/manuals/usermanual.Rmd",
"./workmanuals/user manual.html",
"/file.csv")
regs <- "(?<=/)(?!.*/).+"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 6 x 3
strings match1 match2
<chr> <lgl> <chr>
1 /usr/home/desktop/analysis.R TRUE analysis.R
2 /usr/home/document/ FALSE NA
3 ../logs/error.log TRUE error.log
4 ~work/manuals/usermanual.Rmd TRUE usermanual.Rmd
5 ./workmanuals/user manual.html TRUE user manual.html
6 /file.csv TRUE file.csvstrings <- c("/usr/home/desktop/analysis.R",
"/usr/home/document/",
"../logs/error.log",
"~work/manuals/usermanual.Rmd",
"./workmanuals/user manual.html",
"/file.csv")
regs <- ".*/"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 6 x 3
strings match1 match2
<chr> <lgl> <chr>
1 /usr/home/desktop/analysis.R TRUE /usr/home/desktop/
2 /usr/home/document/ TRUE /usr/home/document/
3 ../logs/error.log TRUE ../logs/
4 ~work/manuals/usermanual.Rmd TRUE ~work/manuals/
5 ./workmanuals/user manual.html TRUE ./workmanuals/
6 /file.csv TRUE /strings <- c("https://www.regs.com/sample.csv",
"https://www.regs.com/sample.csv#id=10",
"https://www.regs.com/sample.png",
"https://www.regs.com/sample.csv?param=1",
"https://www.regs.com/sample.csv?param=1?param=2",
"https://www.regs.com/",
"https://www.regs.com/sample.json")
regs <- "(?<=/)(?!.*/)[a-zA-Z0-9-_\\.]+"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match(string = strings, pattern = regs)) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 7 x 3
strings match1 match2
<chr> <lgl> <chr>
1 https://www.regs.com/sample.csv TRUE sample.csv
2 https://www.regs.com/sample.csv#id=10 TRUE sample.csv
3 https://www.regs.com/sample.png TRUE sample.png
4 https://www.regs.com/sample.csv?param=1 TRUE sample.csv
5 https://www.regs.com/sample.csv?param=1?param=2 TRUE sample.csv
6 https://www.regs.com/ FALSE NA
7 https://www.regs.com/sample.json TRUE sample.jsonFails to unnest list of matrices #788に関連して、スクリプトがワークアラウンドされている。
strings <- c("hoge@abc10.com",
"fuga@gmail.co.jp",
"piyo@yahoo.co.jp",
"30-example.com",
"dsvsfv@ezweb.ne.jp")
regs <- "^\\w+([-+.]\\w+)*@\\w+([-.]\\w+)*\\.\\w+([-.]\\w+)*$"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match_all(string = strings, pattern = regs)) %>%
mutate(open = map(match2, ~ .[,1])) %>%
unnest(open, keep_empty = TRUE) %>%
select(-match2)
# A tibble: 5 x 3
strings match1 open
<chr> <lgl> <chr>
1 hoge@abc10.com TRUE hoge@abc10.com
2 fuga@gmail.co.jp TRUE fuga@gmail.co.jp
3 piyo@yahoo.co.jp TRUE piyo@yahoo.co.jp
4 30-example.com FALSE NA
5 dsvsfv@ezweb.ne.jp TRUE dsvsfv@ezweb.ne.jp
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match_all(string = strings, pattern = regs)) %>%
mutate(match2 = map(match2, ~ as_tibble(., .name_repair = "unique"))) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 5 x 6
strings match1 ...1 ...2 ...3 ...4
<chr> <lgl> <chr> <chr> <chr> <chr>
1 hoge@abc10.com TRUE hoge@abc10.com NA NA NA
2 fuga@gmail.co.jp TRUE fuga@gmail.co.jp NA .co NA
3 piyo@yahoo.co.jp TRUE piyo@yahoo.co.jp NA .co NA
4 30-example.com FALSE NA NA NA NA
5 dsvsfv@ezweb.ne.jp TRUE dsvsfv@ezweb.ne.jp NA .ne NA 単語の境界にマッチする表現は\bで表現できる。\bは単語の境界にマッチする。
strings <- c(
"tmp-sample.csv", "sample.csv",
"sample2-csv-specs.csv", "sample2.csv2.specs.json",
"sample_cars.xlsx", "sample-houses.csv",
"sample_Trees.csv","sample-cars.R",
"sample-houses.r","sample-final.json","Sample-final2.json"
)
regs <- ".*\\.(csv|json)$"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match_all(string = strings, pattern = regs)) %>%
mutate(match2 = map(match2, ~ as_tibble(., .name_repair = "unique"))) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 11 x 4
strings match1 ...1 ...2
<chr> <lgl> <chr> <chr>
1 tmp-sample.csv TRUE tmp-sample.csv csv
2 sample.csv TRUE sample.csv csv
3 sample2-csv-specs.csv TRUE sample2-csv-specs.csv csv
4 sample2.csv2.specs.json TRUE sample2.csv2.specs.json json
5 sample_cars.xlsx FALSE NA NA
6 sample-houses.csv TRUE sample-houses.csv csv
7 sample_Trees.csv TRUE sample_Trees.csv csv
8 sample-cars.R FALSE NA NA
9 sample-houses.r FALSE NA NA
10 sample-final.json TRUE sample-final.json json
11 Sample-final2.json TRUE Sample-final2.json json strings <- c(
"tmp-sample.csv", "sample.csv",
"sample2-csv-specs.csv", "sample2.csv2.specs.json",
"sample_cars.xlsx", "sample-houses.csv",
"sample_Trees.csv","sample-cars.R",
"sample-houses.r","sample-final.json","Sample-final2.json"
)
regs <- "(S|s)ample(\\_|\\-)[a-zA-Z0-9]*\\.(csv|json)$"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match_all(string = strings, pattern = regs)) %>%
mutate(match2 = map(match2, ~ as_tibble(., .name_repair = "unique"))) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 11 x 6
strings match1 ...1 ...2 ...3 ...4
<chr> <lgl> <chr> <chr> <chr> <chr>
1 tmp-sample.csv FALSE NA NA NA NA
2 sample.csv FALSE NA NA NA NA
3 sample2-csv-specs.csv FALSE NA NA NA NA
4 sample2.csv2.specs.json FALSE NA NA NA NA
5 sample_cars.xlsx FALSE NA NA NA NA
6 sample-houses.csv TRUE sample-houses.csv s - csv
7 sample_Trees.csv TRUE sample_Trees.csv s _ csv
8 sample-cars.R FALSE NA NA NA NA
9 sample-houses.r FALSE NA NA NA NA
10 sample-final.json TRUE sample-final.json s - json
11 Sample-final2.json TRUE Sample-final2.json S - json 先読み、後読みを利用した正規表現を使う。先読みはX(?=Y)で、Xを探すけどYが続く場合にだけ一致。後読みは、(?<=Y)Xは、Xの前にYがある場合にのみ一致。
strings <- c(
"https://w.com/?utm_source=newsletter&utm_medium=email&utm_campaign=camp_a&utm_content=url01",
"https://w.com/?utm_source=newsletter&utm_medium=email&utm_campaign=camp_bb&utm_content=url03",
"https://w.com/?utm_source=newsletter&utm_medium=email&utm_campaign=camp_ca&utm_content=url02",
"https://w.com/?utm_source=newsletter&utm_medium=email&utm_campaign=camp_mjk&utm_content=url04")
regs <- "(?<=campaign=).*?(?=&utm_content)"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match_all(string = strings, pattern = regs)) %>%
mutate(match2 = map(match2, ~ as_tibble(., .name_repair = "unique"))) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 4 x 3
strings match1 ...1
<chr> <lgl> <chr>
1 https://w.com/?utm_source=newsletter&utm_medium=email&utm_campaign=camp_a&utm_content=url01 TRUE camp_a
2 https://w.com/?utm_source=newsletter&utm_medium=email&utm_campaign=camp_bb&utm_content=url03 TRUE camp_bb
3 https://w.com/?utm_source=newsletter&utm_medium=email&utm_campaign=camp_ca&utm_content=url02 TRUE camp_ca
4 https://w.com/?utm_source=newsletter&utm_medium=email&utm_campaign=camp_mjk&utm_content=url04 TRUE camp_mjk後読みは、(?<=Y)Xは、Xの前にYがある場合にのみ一致。
strings <- c(
"https://w.com/?utm_source=newsletter&utm_medium=email&utm_campaign=camp_a&utm_content=url01",
"https://w.com/?utm_source=newsletter&utm_medium=email&utm_campaign=camp_bb&utm_content=url03",
"https://w.com/?utm_source=newsletter&utm_medium=email&utm_campaign=camp_ca&utm_content=url02",
"https://w.com/?utm_source=newsletter&utm_medium=email&utm_campaign=camp_mjk&utm_content=url04")
regs <- "(?<=utm_content=).*$"
tibble(strings) %>%
mutate(match1 = str_detect(string = strings, pattern = regs),
match2 = str_match_all(string = strings, pattern = regs)) %>%
mutate(match2 = map(match2, ~ as_tibble(., .name_repair = "unique"))) %>%
unnest(match2, keep_empty = TRUE)
# A tibble: 4 x 3
strings match1 ...1
<chr> <lgl> <chr>
1 https://w.com/?utm_source=newsletter&utm_medium=email&utm_campaign=camp_a&utm_content=url01 TRUE url01
2 https://w.com/?utm_source=newsletter&utm_medium=email&utm_campaign=camp_bb&utm_content=url03 TRUE url03
3 https://w.com/?utm_source=newsletter&utm_medium=email&utm_campaign=camp_ca&utm_content=url02 TRUE url02
4 https://w.com/?utm_source=newsletter&utm_medium=email&utm_campaign=camp_mjk&utm_content=url04 TRUE url04アルファベットが続かなければマッチして、検索クエリに文字列を直す表現。
strings <- c(
"date\t20100701\tsort\t安い順\tregion\tハワイ\tcountry\tハワイ島",
"sort\t安い順",
"date\t20100701\tsort\t安い順\tregion\tハワイ\tcountry\tハワイ島"
)
strings
[1] "date\t20100701\tsort\t安い順\tregion\tハワイ\tcountry\tハワイ島"
[2] "sort\t安い順"
[3] "date\t20100701\tsort\t安い順\tregion\tハワイ\tcountry\tハワイ島"
tibble(strings) %>%
mutate(strings = str_replace_all(strings, "\\t(?=[^\x01-\x7E])", "="),
strings = str_replace_all(strings, "\\t(?=[0-9])", "="),
strings = str_replace_all(strings, "\\t", "&"),
strings = str_c("?", strings))
# A tibble: 3 x 1
strings
<chr>
1 ?date=20100701&sort=安い順®ion=ハワイ&country=ハワイ島
2 ?sort=安い順
3 ?date=20100701&sort=安い順®ion=ハワイ&country=ハワイ島