An R Interface to the Onigmo Regular Expression Library


Keywords
r, regex, regular-expressions, text-analysis
License
MirOS

Documentation

CRAN version r-universe version CI codecov Dependencies

Oniguruma Regular Expressions (for R)

NB: If you are looking for Oracle R Enterprise, please note you're in the wrong place!

Welcome to the ore package for R. This package provides an alternative to R's standard functions for manipulating strings with regular expressions, based on the Oniguruma regular expression library (rather than PCRE, as in base). Although the regex features of the two libraries are quite similar, the R interface provided by ore has some notable advantages:

  • Regular expressions are themselves first-class objects (of class ore), stored with attributes containing information such as the number of parenthesised groups present within them. This means that it is not necessary to compile a particular regex more than once.
  • Search results focus around the matched substrings (including parenthesised groups), rather than the locations of matches. This saves extra work with substr to extract the matches themselves.
  • Performance is substantially better, especially when matching against long strings.
  • Text can be easily drawn from a URL or other "connection", and matches early in large files can be found efficiently with incremental search.
  • Substitutions can be functions, as well as literal or back-referenced strings, and different replacements can be used for different matches.
  • There is wider support for character encodings.
  • Common subexpressions can be easily stored in and retrieved from a dictionary.
  • Matches can be efficiently obtained over only part of a text.
  • The core functions have more consistent names.

Oniguruma (or rather, the Onigmo fork of it) is the regular expression library used by the Ruby programming language, and ore is somewhat inspired by Ruby's regular expression features; although it is implemented in what aims to be a natural way for R users, including full vectorisation.

This README covers the package's R interface only, and assumes that the reader is already familiar with regular expressions. Please see the official reference document for details of supported regular expression syntax.

If you prefer the more verbose but also more friendly approach to creating regular expressions provided by Kevin Ushey and Jim Hester's excellent rex package, you can still use ore for performing the actual matching, and working with the results. None of the syntax generated by rex is known to be incompatible with Oniguruma.

Contents

Installation

The package is available from CRAN. The latest development version can be installed from r-universe or directly from GitHub using the remotes package.

# install.packages("remotes")
remotes::install_github("jonclayden/ore")

As of package version 1.7.1, two notable features are missing from the CRAN release, relative to the development version: the "I can eat glass" dataset, and connection support. These are both due to specific restrictions on what CRAN will accept, and neither currently has any known workaround.

Function mapping

The table below gives the approximate equivalence between the package's core functions and base R.

Effect ore syntax Base R syntax
Create a regex object regex <- ore(regex_string) (no equivalent)
Is there a match? ore_ismatch(regex, text) or text %~% regex grepl(regex, text, perl=TRUE)
Find the first match ore_search(regex, text) regexpr(regex, text, perl=TRUE)
Find match after character 10 ore_search(regex, text, start=10) (no equivalent)
Find all matches ore_search(regex, text, all=TRUE) gregexpr(regex, text, perl=TRUE)
Replace first match ore_subst(regex, replace, text) sub(regex, replace, text, perl=TRUE)
Replace all matches ore_subst(regex, replace, text, all=TRUE) gsub(regex, replace, text, perl=TRUE)
Split at matches ore_split(regex, text) strsplit(text, regex, perl=TRUE)

Basic usage

Let's consider a very simple example: a regular expression for matching a single decimal integer, either positive or negative. We create this regex as follows:

library(ore)
re <- ore("-?\\d+")

This syntax matches an optional minus sign, followed by one or more digits. Here we immediately introduce one of the differences between the regular expression capabilities of base R and the ore package: in the latter, regular expressions have class ore, rather than just being standard strings (although plain strings are also accepted by package functions). We can find the class of the regex object, and print it:

class(re)
## [1] "ore"
re
## Oniguruma regular expression: /-?\d+/
##  - 0 groups
##  - UTF-8 encoding
##  - ruby syntax

The ore() function compiles the regex string, retaining the compiled version for later use. The number of groups in the string is obtained definitively, because the string is parsed by the full Oniguruma parser.

We can now search another string for matches:

match <- ore_search(re, "I have 2 dogs, 3 cats and 4 hamsters")
class(match)
## [1] "orematch"
match
##   match:        2                            
## context: I have   dogs, 3 cats and 4 hamsters

The result of the search is an object of class orematch. This contains elements giving the offsets, lengths and content of matches, as well as those of any parenthesised groups. When printed, the object shows the original text with the matched substring extracted onto the line above (or coloured, if the crayon package is installed and a colour terminal is being used). This can be useful to check that the regular expression is capturing the text expected.

The start parameter to ore_search() can be used to indicate where in the text the search should begin. All matches (after the starting point) will be returned with all=TRUE:

ore_search(re, "I have 2 dogs, 3 cats and 4 hamsters", start=10)
##   match:                3                    
## context: I have 2 dogs,   cats and 4 hamsters
ore_search(re, "I have 2 dogs, 3 cats and 4 hamsters", all=TRUE)
##   match:        2       3          4         
## context: I have   dogs,   cats and   hamsters
##  number:        1       2          3

The text to be searched for matches can be a vector, in which case the return value will be a list of orematch objects:

ore_search(re, c("2 dogs","3 cats","4 hamsters"))
## <3 matches in 3 strings>
## 
## [[1]]
##   match: 2     
## context:   dogs
## 
## [[2]]
##   match: 3     
## context:   cats
## 
## [[3]]
##   match: 4         
## context:   hamsters

If there is no match the return value will be NULL, or a list with NULL for elements with no match.

Encodings

Both R and Oniguruma support alternative character encodings for strings, and this can affect matches. Consider the regular expression \b\w{4}\b, which matches words of exactly four letters. It behaves differently depending on the encoding that it is declared with:

re1 <- ore("\\b\\w{4}\\b", encoding="ASCII")
re2 <- ore("\\b\\w{4}\\b", encoding="UTF-8")
text <- enc2utf8("I'll have a piña colada")
ore_search(re1, text, all=TRUE)
##   match:      have              
## context: I'll      a piña colada
ore_search(re2, text, all=TRUE)
##   match:      have   piña       
## context: I'll      a      colada
##  number:      1===   2===

Note that, in a basic ASCII encoding, only ASCII word characters are matched to the \w character class. Since "ñ" is not directly representable in ASCII, the word "piña" is not considered a match.

Notice that base R's regular expression functions will not find the second match:

gregexpr("\\b\\w{4}\\b", text, perl=TRUE)
## [[1]]
## [1] 6
## attr(,"match.length")
## [1] 4

Alternative syntaxes

By default, Oniguruma and ore use Ruby's regular expression syntax, which is very similar to Perl's (and hence that of base R with perl=TRUE). However, the library does support alternative syntaxes, and ore currently also allows for literal string matching, which is equivalent to fixed=TRUE in base R.

Notice the difference in interpretation of a period in the following example:

ore_search(ore("."), "1.7")
##   match: 1  
## context:  .7
ore_search(ore(".",syntax="fixed"), "1.7")
##   match:  . 
## context: 1 7

In the first case the period has the usual regular expression interpretation of "any character", so it matches the first available character, the 1. In the second case the period has no special meaning, and it only matches a literal period in the search string.

Alternatively, the ore_escape() function can be used to help escape substrings that would otherwise have special meaning in the default syntax:

ore_search(ore_escape("."), "1.7")
##   match:  . 
## context: 1 7

Substitutions

The ore_subst() function can be used to substitute regex matches with new text. Matched subgroups may be referred to using numerical or named back-references: \1, \2, etc.

re <- ore("\\b(\\w)(\\w)(\\w)(\\w)\\b")
text <- enc2utf8("I'll have a piña colada")
ore_subst(re, "\\3\\1\\2\\4", text, all=TRUE)
## [1] "I'll vhae a ñpia colada"

re <- ore("\\b(?<first>\\w)(?<second>\\w)(?<third>\\w)(?<fourth>\\w)\\b")
ore_subst(re, "\\k<third>\\k<first>\\k<second>\\k<fourth>", text, all=TRUE)
## [1] "I'll vhae a ñpia colada"

A function may also be provided, which will be used to generate replacement strings. For example, we could make all four-letter words uppercase:

re <- ore("\\b\\w{4}\\b")
text <- "I have 2 dogs, 3 cats and 4 hamsters"
ore_subst(re, toupper, text, all=TRUE)
## [1] "I HAVE 2 DOGS, 3 CATS and 4 hamsters"

There is also a variant called ore_repl(), which will replicate the source text to use multiple different replacements if needed. This is in turn used by es(), which does expression substitution (a.k.a. string interpolation): evaluating R code (within each "#{}" construct) and inserting the results into a string.

es("Test #{1:2}")
## [1] "Test 1" "Test 2"

Splitting

Strings can be split into parts using the ore_split() function.

ore_split("-?\\d+", "I have 2 dogs, 3 cats and 4 hamsters")
## [1] "I have "    " dogs, "    " cats and " " hamsters"

This finds all matches to the pattern, discards them, and then returns the remaining pieces of the original string.

String multiplexing

Sometimes we may want to classify the elements of a vector according to whether they match one or more regular expressions, or extract some information that may be in one of a number of formats. The ore_switch() function is designed for this purpose, taking any number of strings as arguments, named for the regular expressions used to select them. It works a little like the base R function ifelse().

ore_switch(c("2 dogs","some dogs","no dogs"), "-?\\d+"="number", "no number")
## [1] "number"    "no number" "no number"

Notice that the text to be matched comes first in this case. The second argument is named with the regular expression that selects it, and the third is unnamed and so will catch all strings that haven't already been matched, unconditionally.

This can also be used to achieve similar results to the alternation trick, to exclude more specific matches and retain less specific ones. For example, we can identify four-letter words, unless they are quoted:

strings <- c('Is it good?', 'Is it bad?', 'Is it "ugly"?')
ore_switch(strings, "\"\\b\\w{4}\\b\""=NA, "\\b\\w{4}\\b"="\\0")
## [1] "good" NA     NA

Here there is no catch-all case, so strings with no four-letter words in them will map to NA by default.

The pattern dictionary

It's not unusual to reuse parts of a regular expression many times. Perhaps, once you have an expression that captures certain common elements of your text, you might want to store it for regular use. Or maybe you want to make your regexes more readable by breaking them down into manageable chunks. The ore package's pattern dictionary can help.

To take a simple example, let's just consider a pattern for digits. We can add it to the dictionary using the ore_dict() function.

ore_dict(digits="\\d+")
## digits 
## "\\d+"

Now, we can create a regex using this pattern by naming it in a call to ore().

ore(digits)
## Oniguruma regular expression: /(\d+)/
##  - 1 group
##  - UTF-8 encoding
##  - ruby syntax

Notice the lack of quotation marks around the name, which distinguishes it from a normal pattern string. We can also reuse it multiple times, and add other regex syntax around it. Say, for example, that we want to find two sets of digits separated by word characters and/or space.

re <- ore(digits, "[\\w\\s]+", digits)
re
## Oniguruma regular expression: /(\d+)[\w\s]+(\d+)/
##  - 2 groups
##  - UTF-8 encoding
##  - ruby syntax

Notice that ore() constructs a full regex from the parts, wrapping each dictionary element in parentheses to make it a group. Now we can match it against our text.

ore_search(re, "I have 2 dogs, 3 cats and 4 hamsters")
##   match:                3 cats and 4         
## context: I have 2 dogs,              hamsters

The package comes with a small dictionary of fairly robust regexes for matching common elements like numbers or email addresses. These can be used "out of the box". For example,

ore_search(ore(number), "Numbers in various formats: -23, 0xbead5, .409 and 1.4e-5", all=TRUE)
##   match:                             -23  0xbead5  .409     1.4e-5
## context: Numbers in various formats:    ,        ,      and       
##  number:                             1==  2======  3===     4=====

Notice that, when using the dictionary, the ore() function must be called explicitly.

Additional convenience functions

The ore_ismatch function will return a logical vector indicating whether or not a match is present in each element of a character vector. The infix notation %~% is a shorthand way to achieve the same thing. Either way, the full match data can be obtained without repeating the search, using the ore_lastmatch() function.

if ("I have 2 dogs, 3 cats and 4 hamsters" %~% "-?\\d+")
  print(ore_lastmatch())
##   match:        2                            
## context: I have   dogs, 3 cats and 4 hamsters

The %~~% operator works likewise, except that all matches will be found (i.e. it sets all=TRUE when calling ore_search()). Finally, the %~|% operator filters a vector, returning just elements which match the regular expression.

Text matching the entire regex, or parenthesised groups, can be extracted using the matches() and groups() convenience functions, or even more concisely using indexing.

# An example from ?regexpr
re <- "^(([^:]+)://)?([^:/]+)(:([0-9]+))?(/.*)"
text <- "http://stat.umn.edu:80/xyz"
match <- ore_search(re, text)
matches(match)
## [1] "http://stat.umn.edu:80/xyz"
match[1]
## [1] "http://stat.umn.edu:80/xyz"
groups(match)
##      [,1]      [,2]   [,3]           [,4]  [,5] [,6]  
## [1,] "http://" "http" "stat.umn.edu" ":80" "80" "/xyz"
match[1,3]
## [1] "stat.umn.edu"

Searching in files

Since version 1.3.0 of the package, it has been possible to search directly within files, using their native encoding if it is supported by Onigmo (which supports many more encodings than R does internally). Binary files may also be searched, but in that case the regex is fixed to use ASCII encoding, and the file is examined byte-by-byte.

For example, using a test file provided with the package source, and if your local iconv supports the Shift JIS encoding, you can try

path <- system.file("tinytest", "sjis.txt", package="ore")
match <- ore_search("\\p{Katakana}+", ore_file(path,encoding="SHIFT_JIS"), all=TRUE)
matches(match)
## [1] "コ"       "ディング" "ファイル"

Note that if you read the file using R's readLines function, it will be re-encoded to UTF-8. The same matches will be found, but the byte offsets are different:

match <- ore_search("\\p{Katakana}+", ore_file(path,encoding="SHIFT_JIS"), all=TRUE)
match$byteOffsets
## [1] 18 22 44
match <- ore_search("\\p{Katakana}+", readLines(file(path,encoding="SHIFT_JIS")), all=TRUE)
match$byteOffsets
## [1] 22 28 61

Hence, if you want to know where in a file the match can be found, the first of these approaches will give the right answer, while the latter will not.

Version 1.7.0 of the package added support for R connections, which allows gzipped files, URLs and other sources to be used directly. For example, let's look for the first mention of an iDevice on Apple's home page:

ore_search("\\bi[A-Z]\\w+", url("https://www.apple.com"))
##   match:                                  iPhone                                 
## context: ... of Apple and shop everything       , iPad, Apple Watch, Mac, and ...

Related packages

As noted at the beginning of this README, base R provides some regular expression functions, although they are less varied, flexible and fast than those in the ore package. There are other related and alternative packages available:

  • rematch2 provides a convenient wrapper around base R's functions.
  • The stringi package provides an extensive set of string-processing facilities, wrapping the ICU library. stringr offers an alternative interface.
  • re2 provides an interface to RE2, another regular expression library.
  • The glue package provides string interpolation, similar to es().
  • Various packages aim to provide a friendlier interface to regular expressions or related functionality, including rex, nc, rebus and RVerbalExpressions.