Strings

A Char value represents a single character: it is just a 32-bit primitive type with a special literal representation and appropriate arithmetic behaviors, and which can be converted to a numeric value representing a Unicode code point. (Julia packages may define other subtypes of AbstractChar, e.g. to optimize operations for other text encodings.) Here is how Char values are input and shown:

You can easily convert a Char to its integer value, i.e. code point:

On 32-bit architectures, typeof(c) will be Int32. You can convert an integer value back to a Char just as easily:

Not all integer values are valid Unicode code points, but for performance, the Char conversion does not check that every character value is valid. If you want to check that each converted value is a valid code point, use the isvalid function:

As of this writing, the valid Unicode code points are U+0000 through U+D7FF and U+E000 through U+10FFFF. These have not all been assigned intelligible meanings yet, nor are they necessarily interpretable by applications, but all of these values are considered to be valid Unicode characters.

You can input any Unicode character in single quotes using \u followed by up to four hexadecimal digits or \U followed by up to eight hexadecimal digits (the longest valid value only requires six):

Julia uses your system’s locale and language settings to determine which characters can be printed as-is and which must be output using the generic, escaped \u or \U input forms. In addition to these Unicode escape forms, all of C’s traditional escaped input forms can also be used:

You can do comparisons and a limited amount of arithmetic with Char values:

String literals are delimited by double quotes or triple double quotes:

Long lines in strings can be broken up by preceding the newline with a backslash (\):

If you want to extract a character from a string, you index into it:

Many Julia objects, including strings, can be indexed with integers. The index of the first element (the first character of a string) is returned by firstindex(str), and the index of the last element (character) with lastindex(str). The keywords begin and end can be used inside an indexing operation as shorthand for the first and last indices, respectively, along the given dimension. String indexing, like most indexing in Julia, is 1-based: firstindex always returns 1 for any AbstractString. As we will see below, however, lastindex(str) is not in general the same as length(str) for a string, because some Unicode characters can occupy multiple "code units".

You can perform arithmetic and other operations with end, just like a normal value:

Using an index less than begin (1) or greater than end raises an error:

You can also extract a substring using range indexing:

Notice that the expressions str[k] and str[k:k] do not give the same result:

The former is a single character value of type Char, while the latter is a string value that happens to contain only a single character. In Julia these are very different things.

Range indexing makes a copy of the selected part of the original string. Alternatively, it is possible to create a view into a string using the type SubString. More simply, using the @views macro on a block of code converts all string slices into substrings. For example:

Several standard functions like chop, chomp or strip return a SubString.

Julia fully supports Unicode characters and strings. As discussed above, in character literals, Unicode code points can be represented using Unicode \u and \U escape sequences, as well as all the standard C escape sequences. These can likewise be used to write string literals:

Whether these Unicode characters are displayed as escapes or shown as special characters depends on your terminal’s locale settings and its support for Unicode. String literals are encoded using the UTF-8 encoding. UTF-8 is a variable-width encoding, meaning that not all characters are encoded in the same number of bytes ("code units"). In UTF-8, ASCII characters — i.e. those with code points less than 0x80 (128) — are encoded as they are in ASCII, using a single byte, while code points 0x80 and above are encoded using multiple bytes — up to four per character.

String indices in Julia refer to code units (= bytes for UTF-8), the fixed-width building blocks that are used to encode arbitrary characters (code points). This means that not every index into a String is necessarily a valid index for a character. If you index into a string at such an invalid byte index, an error is thrown:

In this case, the character ∀ is a three-byte character, so the indices 2 and 3 are invalid and the next character’s index is 4; this next valid index can be computed by nextind(s,1), and the next index after that by nextind(s,4) and so on.

Since end is always the last valid index into a collection, end-1 references an invalid byte index if the second-to-last character is multibyte.

The first case works, because the last character y and the space are one-byte characters, whereas end-2 indexes into the middle of the ∃ multibyte representation. The correct way for this case is using prevind(s, lastindex(s), 2) or, if you’re using that value to index into s you can write s[prevind(s, end, 2)] and end expands to lastindex(s).

Extraction of a substring using range indexing also expects valid byte indices or an error is thrown:

Because of variable-length encodings, the number of characters in a string (given by length(s)) is not always the same as the last index. If you iterate through the indices 1 through lastindex(s) and index into s, the sequence of characters returned when errors aren’t thrown is the sequence of characters comprising the string s. Thus length(s) <= lastindex(s), since each character in a string must have its own index. The following is an inefficient and verbose way to iterate through the characters of s:

The blank lines actually have spaces on them. Fortunately, the above awkward idiom is unnecessary for iterating through the characters in a string, since you can just use the string as an iterable object, no exception handling required:

If you need to obtain valid indices for a string, you can use the nextind and prevind functions to increment/decrement to the next/previous valid index, as mentioned above. You can also use the eachindex function to iterate over the valid character indices:

To access the raw code units (bytes for UTF-8) of the encoding, you can use the codeunit(s,i) function, where the index i runs consecutively from 1 to ncodeunits(s). The codeunits(s) function returns an AbstractVector{UInt8} wrapper that lets you access these raw codeunits (bytes) as an array.

Strings in Julia can contain invalid UTF-8 code unit sequences. This convention allows to treat any byte sequence as a String. In such situations a rule is that when parsing a sequence of code units from left to right characters are formed by the longest sequence of 8-bit code units that matches the start of one of the following bit patterns (each x can be 0 or 1):

In particular this means that overlong and too-high code unit sequences and prefixes thereof are treated as a single invalid character rather than multiple invalid characters. This rule may be best explained with an example:

We can see that the first two code units in the string s form an overlong encoding of space character. It is invalid, but is accepted in a string as a single character. The next two code units form a valid start of a three-byte UTF-8 sequence. However, the fifth code unit \xe2 is not its valid continuation. Therefore code units 3 and 4 are also interpreted as malformed characters in this string. Similarly code unit 5 forms a malformed character because | is not a valid continuation to it. Finally the string s2 contains one too high code point.

Julia uses the UTF-8 encoding by default, and support for new encodings can be added by packages. For example, the LegacyStrings.jl package implements UTF16String and UTF32String types. Additional discussion of other encodings and how to implement support for them is beyond the scope of this document for the time being. For further discussion of UTF-8 encoding issues, see the section below on byte array literals. The transcode function is provided to convert data between the various UTF-xx encodings, primarily for working with external data and libraries.

One of the most common and useful string operations is concatenation:

It’s important to be aware of potentially dangerous situations such as concatenation of invalid UTF-8 strings. The resulting string may contain different characters than the input strings, and its number of characters may be lower than sum of numbers of characters of the concatenated strings, e.g.:

This situation can happen only for invalid UTF-8 strings. For valid UTF-8 strings concatenation preserves all characters in strings and additivity of string lengths.

Julia also provides * for string concatenation:

While * may seem like a surprising choice to users of languages that provide + for string concatenation, this use of * has precedent in mathematics, particularly in abstract algebra.

In mathematics, + usually denotes a commutative operation, where the order of the operands does not matter. An example of this is matrix addition, where A + B == B + A for any matrices A and B that have the same shape. In contrast, * typically denotes a noncommutative operation, where the order of the operands does matter. An example of this is matrix multiplication, where in general A * B != B * A. As with matrix multiplication, string concatenation is noncommutative: greet * whom != whom * greet. As such, * is a more natural choice for an infix string concatenation operator, consistent with common mathematical use.

More precisely, the set of all finite-length strings S together with the string concatenation operator * forms a free monoid (S, *). The identity element of this set is the empty string, "". Whenever a free monoid is not commutative, the operation is typically represented as \cdot, *, or a similar symbol, rather than +, which as stated usually implies commutativity.

Constructing strings using concatenation can become a bit cumbersome, however. To reduce the need for these verbose calls to string or repeated multiplications, Julia allows interpolation into string literals using $, as in Perl:

This is more readable and convenient and equivalent to the above string concatenation — the system rewrites this apparent single string literal into the call string(greet, ", ", whom, ".\n").

The shortest complete expression after the $ is taken as the expression whose value is to be interpolated into the string. Thus, you can interpolate any expression into a string using parentheses:

Both concatenation and string interpolation call string to convert objects into string form. However, string actually just returns the output of print, so new types should add methods to print or show instead of string.

Most non-AbstractString objects are converted to strings closely corresponding to how they are entered as literal expressions:

string is the identity for AbstractString and AbstractChar values, so these are interpolated into strings as themselves, unquoted and unescaped:

To include a literal $ in a string literal, escape it with a backslash:

When strings are created using triple-quotes ("""...""") they have some special behavior that can be useful for creating longer blocks of text.

First, triple-quoted strings are also dedented to the level of the least-indented line. This is useful for defining strings within code that is indented. For example:

In this case the final (empty) line before the closing """ sets the indentation level.

The dedentation level is determined as the longest common starting sequence of spaces or tabs in all lines, excluding the line following the opening """ and lines containing only spaces or tabs (the line containing the closing """ is always included). Then for all lines, excluding the text following the opening """, the common starting sequence is removed (including lines containing only spaces and tabs if they start with this sequence), e.g.:

Next, if the opening """ is followed by a newline, the newline is stripped from the resulting string.

is equivalent to

but

will contain a literal newline at the beginning.

Stripping of the newline is performed after the dedentation. For example:

If the newline is removed using a backslash, dedentation will be respected as well:

Trailing whitespace is left unaltered.

Triple-quoted string literals can contain " characters without escaping.

Note that line breaks in literal strings, whether single- or triple-quoted, result in a newline (LF) character \n in the string, even if your editor uses a carriage return \r (CR) or CRLF combination to end lines. To include a CR in a string, use an explicit escape \r; for example, you can enter the literal string "a CRLF line ending\r\n".

You can lexicographically compare strings using the standard comparison operators:

You can search for the index of a particular character using the findfirst and findlast functions:

You can start the search for a character at a given offset by using the functions findnext and findprev:

You can use the occursin function to check if a substring is found within a string:

The last example shows that occursin can also look for a character literal.

Two other handy string functions are repeat and join:

Some other useful functions include:

There are situations when you want to construct a string or use string semantics, but the behavior of the standard string construct is not quite what is needed. For these kinds of situations, Julia provides non-standard string literals. A non-standard string literal looks like a regular double-quoted string literal, but is immediately prefixed by an identifier, and may behave differently from a normal string literal.

Regular expressions, byte array literals, and version number literals, as described below, are some examples of non-standard string literals. Users and packages may also define new non-standard string literals. Further documentation is given in the Metaprogramming section.

Sometimes you are not looking for an exact string, but a particular pattern. For example, suppose you are trying to extract a single date from a large text file. You don’t know what that date is (that’s why you are searching for it), but you do know it will look something like YYYY-MM-DD. Regular expressions allow you to specify these patterns and search for them.

Julia uses version 2 of Perl-compatible regular expressions (regexes), as provided by the PCRE library (see the PCRE2 syntax description for more details). Regular expressions are related to strings in two ways: the obvious connection is that regular expressions are used to find regular patterns in strings; the other connection is that regular expressions are themselves input as strings, which are parsed into a state machine that can be used to efficiently search for patterns in strings. In Julia, regular expressions are input using non-standard string literals prefixed with various identifiers beginning with r. The most basic regular expression literal without any options turned on just uses r"...":

To check if a regex matches a string, use occursin:

As one can see here, occursin simply returns true or false, indicating whether a match for the given regex occurs in the string. Commonly, however, one wants to know not just whether a string matched, but also how it matched. To capture this information about a match, use the match function instead:

If the regular expression does not match the given string, match returns nothing — a special value that does not print anything at the interactive prompt. Other than not printing, it is a completely normal value and you can test for it programmatically:

If a regular expression does match, the value returned by match is a RegexMatch object. These objects record how the expression matches, including the substring that the pattern matches and any captured substrings, if there are any. This example only captures the portion of the substring that matches, but perhaps we want to capture any non-blank text after the comment character. We could do the following:

When calling match, you have the option to specify an index at which to start the search. For example:

You can extract the following info from a RegexMatch object:

For when a capture doesn’t match, instead of a substring, m.captures contains nothing in that position, and m.offsets has a zero offset (recall that indices in Julia are 1-based, so a zero offset into a string is invalid). Here is a pair of somewhat contrived examples:

It is convenient to have captures returned as an array so that one can use destructuring syntax to bind them to local variables. As a convenience, the RegexMatch object implements iterator methods that pass through to the captures field, so you can destructure the match object directly:

Captures can also be accessed by indexing the RegexMatch object with the number or name of the capture group:

Captures can be referenced in a substitution string when using replace by using \n to refer to the nth capture group and prefixing the substitution string with s. Capture group 0 refers to the entire match object. Named capture groups can be referenced in the substitution with \g<groupname>. For example:

Numbered capture groups can also be referenced as \g<n> for disambiguation, as in:

You can modify the behavior of regular expressions by some combination of the flags i, m, s, and x after the closing double quote mark. These flags have the same meaning as they do in Perl, as explained in this excerpt from the perlre manpage:

For example, the following regex has all three flags turned on:

The r"..." literal is constructed without interpolation and unescaping (except for quotation mark " which still has to be escaped). Here is an example showing the difference from standard string literals:

Triple-quoted regex strings, of the form r"""...""", are also supported (and may be convenient for regular expressions containing quotation marks or newlines).

The Regex() constructor may be used to create a valid regex string programmatically. This permits using the contents of string variables and other string operations when constructing the regex string. Any of the regex codes above can be used within the single string argument to Regex(). Here are some examples:

Note the use of the \Q...\E escape sequence. All characters between the \Q and the \E are interpreted as literal characters (after string interpolation). This escape sequence can be useful when interpolating, possibly malicious, user input.

Another useful non-standard string literal is the byte-array string literal: b"...". This form lets you use string notation to express read only literal byte arrays — i.e. arrays of UInt8 values. The type of those objects is CodeUnits{UInt8, String}. The rules for byte array literals are the following:

There is some overlap between these rules since the behavior of \x and octal escapes less than 0x80 (128) are covered by both of the first two rules, but here these rules agree. Together, these rules allow one to easily use ASCII characters, arbitrary byte values, and UTF-8 sequences to produce arrays of bytes. Here is an example using all three:

The ASCII string "DATA" corresponds to the bytes 68, 65, 84, 65. \xff produces the single byte 255. The Unicode escape \u2200 is encoded in UTF-8 as the three bytes 226, 136, 128. Note that the resulting byte array does not correspond to a valid UTF-8 string:

As it was mentioned CodeUnits{UInt8, String} type behaves like read only array of UInt8 and if you need a standard vector you can convert it using Vector{UInt8}:

Also observe the significant distinction between \xff and \uff: the former escape sequence encodes the byte 255, whereas the latter escape sequence represents the code point 255, which is encoded as two bytes in UTF-8:

Character literals use the same behavior.

For code points less than \u80, it happens that the UTF-8 encoding of each code point is just the single byte produced by the corresponding \x escape, so the distinction can safely be ignored. For the escapes \x80 through \xff as compared to \u80 through \uff, however, there is a major difference: the former escapes all encode single bytes, which — unless followed by very specific continuation bytes — do not form valid UTF-8 data, whereas the latter escapes all represent Unicode code points with two-byte encodings.

If this is all extremely confusing, try reading "The Absolute Minimum Every Software Developer Absolutely, Positively Must Know About Unicode and Character Sets". It’s an excellent introduction to Unicode and UTF-8, and may help alleviate some confusion regarding the matter.

Version numbers can easily be expressed with non-standard string literals of the form v"...". Version number literals create VersionNumber objects which follow the specifications of semantic versioning, and therefore are composed of major, minor and patch numeric values, followed by pre-release and build alpha-numeric annotations. For example, v"0.2.1-rc1+win64" is broken into major version 0, minor version 2, patch version 1, pre-release rc1 and build win64. When entering a version literal, everything except the major version number is optional, therefore e.g. v"0.2" is equivalent to v"0.2.0" (with empty pre-release/build annotations), v"2" is equivalent to v"2.0.0", and so on.

VersionNumber objects are mostly useful to easily and correctly compare two (or more) versions. For example, the constant VERSION holds Julia version number as a VersionNumber object, and therefore one can define some version-specific behavior using simple statements as:

Note that in the above example the non-standard version number v"0.3-" is used, with a trailing -: this notation is a Julia extension of the standard, and it’s used to indicate a version which is lower than any 0.3 release, including all of its pre-releases. So in the above example the code would only run with stable 0.2 versions, and exclude such versions as v"0.3.0-rc1". In order to also allow for unstable (i.e. pre-release) 0.2 versions, the lower bound check should be modified like this: v"0.2-" <= VERSION.

Another non-standard version specification extension allows one to use a trailing + to express an upper limit on build versions, e.g. VERSION > v"0.2-rc1+" can be used to mean any version above 0.2-rc1 and any of its builds: it will return false for version v"0.2-rc1+win64" and true for v"0.2-rc2".

It is good practice to use such special versions in comparisons (particularly, the trailing - should always be used on upper bounds unless there’s a good reason not to), but they must not be used as the actual version number of anything, as they are invalid in the semantic versioning scheme.

Besides being used for the VERSION constant, VersionNumber objects are widely used in the Pkg module, to specify packages versions and their dependencies.

Raw strings without interpolation or unescaping can be expressed with non-standard string literals of the form raw"...". Raw string literals create ordinary String objects which contain the enclosed contents exactly as entered with no interpolation or unescaping. This is useful for strings which contain code or markup in other languages which use $ or \ as special characters.

The exception is that quotation marks still must be escaped, e.g. raw"\"" is equivalent to "\"". To make it possible to express all strings, backslashes then also must be escaped, but only when appearing right before a quote character:

Notice that the first two backslashes appear verbatim in the output, since they do not precede a quote character. However, the next backslash character escapes the backslash that follows it, and the last backslash escapes a quote, since these backslashes appear before a quote.