Skip to main content

Converting from UTF8

The FromUTF8 class provides methods to convert UTF-8 text to various other formats and perform text transformations.

toHtmlEntities​

Convert UTF-8 text to ASCII, writing every other character as an HTML entity: a named one when it exists (&atilde;), otherwise a numeric one (&#1055;). &, < and > are left as they are, so the text may contain markup.

$str = \ByJG\Convert\FromUTF8::toHtmlEntities('João');
echo $str; // Jo&atilde;o

$str = \ByJG\Convert\FromUTF8::toHtmlEntities('Привет');
echo $str; // &#1055;&#1088;&#1080;&#1074;&#1077;&#1090;

Use Cases​

  • Generating HTML-safe output
  • Ensuring compatibility with systems that don't support UTF-8
  • Email templates that need entity encoding
  • Legacy system integration

toMimeEncodedWord​

Convert UTF-8 text to MIME encoded-word format according to RFC 2047. This is commonly used for encoding non-ASCII characters in email headers (like Subject, From, To).

$str = \ByJG\Convert\FromUTF8::toMimeEncodedWord('João');
echo $str; // =?UTF-8?Q?Jo=C3=A3o?=

$str = \ByJG\Convert\FromUTF8::toMimeEncodedWord('Hello World');
echo $str; // Hello World (unchanged, no encoding needed)

RFC 2047 Format​

The output format is: =?charset?encoding?encoded-text?=

  • charset: Always UTF-8
  • encoding: Always Q (Quoted-Printable)
  • encoded-text: The text with special characters encoded as =XX where XX is the hexadecimal byte value

Use Cases​

  • Email subject lines with special characters
  • Email header fields (From, To, CC, etc.)
  • Ensuring email compatibility across different mail servers
  • Supporting international characters in email metadata

Notes​

  • Plain ASCII text is returned unchanged
  • Only the words that need it are encoded: Test ũ becomes Test =?UTF-8?Q?=C5=A9?=
  • A long header is folded into several encoded words, one per line, as RFC 2047 requires
  • It is mb_encode_mimeheader($text, 'UTF-8', 'Q')

removeAccent​

Remove the accents and diacritical marks of the Latin letters, converting them to their ASCII equivalents.

$str = \ByJG\Convert\FromUTF8::removeAccent('João');
echo $str; // Joao

$str = \ByJG\Convert\FromUTF8::removeAccent('Café résumé');
echo $str; // Cafe resume

$str = \ByJG\Convert\FromUTF8::removeAccent('Zürich');
echo $str; // Zurich

$str = \ByJG\Convert\FromUTF8::removeAccent('Škoda Dvořák Việt Nam');
echo $str; // Skoda Dvorak Viet Nam

It works for a precomposed letter (é) and for a letter followed by a combining mark (e + U+0301), using the Unicode decomposition (Normalizer).

Character Conversions​

Some notable conversions:

  • À Á Â Ã Ä Å → A
  • È É Ê Ë → E
  • Ñ → N
  • Ç → C
  • Æ → AE, Œ → OE
  • ß → ss
  • Ø → O, Ł → L, Đ → D
  • Þ → TH, ð → d
  • © → (C)
  • ® → (R)
  • ° → o.
  • ½ → 1/2

What Is Left Alone​

  • Other scripts (Cyrillic, Greek, Japanese, Hebrew, …), where those marks are part of the letter
  • Symbols and punctuation outside Latin-1 (€, ✓, “, –) and emoji
  • A text that is not valid UTF-8, which is returned unchanged

Use onlyAscii() when the result has to be ASCII.

Use Cases​

  • Creating URL slugs
  • Normalizing search queries
  • Database comparisons
  • File naming (when special characters are not allowed)
  • Generating ASCII-only identifiers
  • Legacy system integration that doesn't support UTF-8

onlyAscii​

Convert UTF-8 text to printable ASCII (space to ~, plus tabs and line breaks). The emoji are removed, the accents are removed, and every character still outside that range — the other control characters included — is replaced with a default character.

$str = \ByJG\Convert\FromUTF8::onlyAscii('João');
echo $str; // Joao

$str = \ByJG\Convert\FromUTF8::onlyAscii('Hello 世界', '?');
echo $str; // Hello ?? (one ? per character)

$str = \ByJG\Convert\FromUTF8::onlyAscii('Café 世界', '');
echo $str; // Cafe  (accents removed, the rest dropped)

Parameters​

  • $text (string): The UTF-8 text to convert
  • $defaultChar (string): Replaces each character that is still not ASCII once the accents and emoji are removed (default: empty string, which removes them). Tabs and line breaks are kept.

Use Cases​

  • Strict ASCII-only output requirements
  • Systems that cannot handle any non-ASCII characters
  • Log files that need to be ASCII-only
  • Network protocols with ASCII-only constraints

Difference from removeAccent​

  • removeAccent(): Converts accented Latin letters to their base form (á → a) and leaves the rest of the text as it is
  • onlyAscii(): Does the same, then removes or replaces every character with no ASCII equivalent

removeEmoji​

Remove all emoji characters from UTF-8 text, including complex emoji sequences and flag emojis.

$str = \ByJG\Convert\FromUTF8::removeEmoji('Hello 👋 World 🌍');
echo $str; // Hello  World

$str = \ByJG\Convert\FromUTF8::removeEmoji('Great! 😀 I love it ❤️');
echo $str; // Great!  I love it

$str = \ByJG\Convert\FromUTF8::removeEmoji('Visit us in 🇺🇸 or 🇧🇷');
echo $str; // Visit us in  or

What Gets Removed​

  • Basic emoji (😀, 😊, ❤️, etc.)
  • Flag emoji (🇺🇸, 🇧🇷, 🇯🇵, etc.)
  • Complex emoji sequences (family emoji, skin tones, etc.)
  • Emoji with modifiers

Use Cases​

  • Data sanitization for systems that don't support emoji
  • Text analytics where emoji would interfere
  • Creating plain text versions of content
  • SMS systems that don't support emoji
  • Database fields with character set limitations
  • Search indexing where emoji should be ignored

Notes​

  • It matches the Unicode Extended_Pictographic property, so emoji added in newer Unicode versions are removed too, with no list to keep up to date
  • Spaces may remain where emoji were removed