Why does café not equal café in my code?

One of them stores é as a single code point, U+00E9, and the other as a plain e followed by a combining acute accent, U+0301. They render identically, but they're different sequences, so comparisons and lengths disagree. File names from macOS often arrive in the split form. Normalize both sides to NFC before comparing: `s.normalize('NFC')` in JavaScript, or unicodedata.normalize('NFC', s) in Python.

How do I make a search for cafe find café?

Compare with the accents removed on both sides. In MySQL 8 the default collation, utf8mb4_0900_ai_ci, already does this, since ai stands for accent-insensitive. In PostgreSQL, enable the unaccent extension and compare `unaccent(name) = unaccent($1)`, ideally backed by an index on the unaccented value if the table is large.

Does résumé need both accents?

No. English dictionaries accept résumé, resumé and resume for the job document. The accents are the only thing separating it from the verb resume, though, so résumé or resumé reads more clearly in a heading or a job ad. Pick one form and stick with it throughout.

Is it événement or évènement in French?

Both are correct. Événement is the traditional spelling, and the 1990 spelling rectifications added évènement because the second vowel is pronounced as an open è. The same change covers future and conditional forms like céderai, which can now be written cèderai. Dictionaries list both, so choose one per document.

I typed Alt+233 for é and got Θ. What happened?

Add the leading zero: Alt+0233 gives é, and Alt+0201 gives the capital É. Without the zero, Windows looks the number up in the machine's DOS code page instead, which is why a US system shows Θ and a Western European one shows a different letter altogether.

ASCII 233

Lowercase e with acute accentGreek capital theta

ASCII 233 is é on Windows and Θ in old DOS.

é (Small E with Acute) is byte 233 in Windows-1252, Unicode U+00E9. French writes café, été and école with it, and it shows up in Spanish, Portuguese, Italian, Czech and Hungarian as well. If you're here because of Ã©, the quickest fix starts with finding which layer is wrong, and the raw bytes tell you. Dump the stored value with xxd, with HEX() in MySQL, or as bytes in your language. If you see C3 A9, the data is correct UTF-8 and only the reader is wrong, so fix the charset in the HTTP header, the meta tag or the database connection. If you see C3 83 C2 A9, the text was mangled once before it was saved, and you need to convert the stored data back rather than change a setting. A lone E9, the Windows-1252 form of é, means the reverse problem and shows up as � in a UTF-8 reader. Old DOS had é at 130. The HTML entity is é.

Θ (Capital Theta) is byte 233 in code page 437, the original IBM PC character set, and maps to Unicode U+0398. Programmers know it from complexity analysis. Θ(n log n) says an algorithm grows at exactly that rate, bounded above and below, while O(n log n) only promises it grows no faster and Ω(n log n), with the Ω at 234, only that it grows no slower. Saying O when you mean Θ is technically true but weaker than it sounds: bubble sort is O(n³) too. Code page 850 put Ú on this byte. In UTF-8, Θ is CE 98, and the HTML entity is Θ.

Latin LetterGreek Letter

Technical Details

Code Example

</>

#include <stdio.h>

int main(void) {
    /* On Windows (code page 1252) byte 233 is é.
       Modern terminals expect UTF-8, so print the code point, not the byte. */
    printf("\u00E9\n");   /* UTF-8: C3 A9 */

    unsigned char b = 233;         /* the raw Windows-1252 byte */
    printf("%d 0x%02X\n", b, b);    /* 233 0xE9 */
    return 0;
}

Code Example

</>

#include <stdio.h>

int main(void) {
    /* Byte 233 is Θ only on a console using code page 437
       (chcp 437 on Windows, or DOSBox). On a UTF-8 terminal the
       lone byte 0xE9 is invalid and prints as garbage, often �. */
    putchar(233);

    /* char is signed on x86, so a plain char holding 0xE9 is -23.
       Use unsigned char when you compare or index by byte value. */
    char c = (char)233;
    unsigned char u = 233;
    printf("\n%d %d\n", c, u);   /* -23 233 */

    /* Portable: print the Unicode character as UTF-8 instead. */
    printf("\u0398\n");          /* CE 98 */
    return 0;
}

How to Type It

Windows: Hold Alt and type 0233 on the numeric keypad, with Num Lock on. The leading zero picks the Windows code page. Alt+130, without the zero, types it too. No numeric keypad? In Word, type 00E9 and press Alt+X.
Mac: Press Option+E, then e.
Linux: In GNOME, or any desktop running IBus, press Ctrl+Shift+U, type e9, then Space. With a Compose key set up, press Compose, apostrophe, e.
Phone: Long-press e and pick é.

How to Type It

Windows: Hold Alt and type 233 on the numeric keypad, with no leading zero. US PCs insert Θ, but PCs set to code page 850, common in Western Europe, insert Ú. No numeric keypad? In Word, type 0398 and press Alt+X.
Mac: Press Control+Command+Space to open the Character Viewer, then search for "greek capital letter theta".
Linux: In GNOME, or any desktop running IBus, press Ctrl+Shift+U, type 398, then Space.
Phone: Add the Greek keyboard in your phone's keyboard settings and type Θ there, or tap the character at the top of this page to copy it.

Frequently Asked Questions

One of them stores é as a single code point, U+00E9, and the other as a plain e followed by a combining acute accent, U+0301. They render identically, but they're different sequences, so comparisons and lengths disagree. File names from macOS often arrive in the split form. Normalize both sides to NFC before comparing: `s.normalize('NFC')` in JavaScript, or unicodedata.normalize('NFC', s) in Python.
Compare with the accents removed on both sides. In MySQL 8 the default collation, utf8mb4_0900_ai_ci, already does this, since ai stands for accent-insensitive. In PostgreSQL, enable the unaccent extension and compare `unaccent(name) = unaccent($1)`, ideally backed by an index on the unaccented value if the table is large.
No. English dictionaries accept résumé, resumé and resume for the job document. The accents are the only thing separating it from the verb resume, though, so résumé or resumé reads more clearly in a heading or a job ad. Pick one form and stick with it throughout.
Both are correct. Événement is the traditional spelling, and the 1990 spelling rectifications added évènement because the second vowel is pronounced as an open è. The same change covers future and conditional forms like céderai, which can now be written cèderai. Dictionaries list both, so choose one per document.
Add the leading zero: Alt+0233 gives é, and Alt+0201 gives the capital É. Without the zero, Windows looks the number up in the machine's DOS code page instead, which is why a US system shows Θ and a Western European one shows a different letter altogether.

Frequently Asked Questions

Add the leading zero: Alt+0233 gives é, and Alt+0201 gives the capital É. Without the zero, Windows looks the number up in the machine's DOS code page instead, which is why a US system shows Θ and a Western European one shows a different letter altogether.

ASCII 233

Lowercase e with acute accentGreek capital theta

ASCII 233 is é on Windows and Θ in old DOS.

Latin Letter

Greek Letter

#include <stdio.h>

int main(void) {

/* On Windows (code page 1252) byte 233 is é.

Modern terminals expect UTF-8, so print the code point, not the byte. */

printf("\u00E9\n"); /* UTF-8: C3 A9 */

unsigned char b = 233; /* the raw Windows-1252 byte */

printf("%d 0x%02X\n", b, b); /* 233 0xE9 */

return 0;

}

#include <stdio.h>

int main(void) {

/* Byte 233 is Θ only on a console using code page 437

(chcp 437 on Windows, or DOSBox). On a UTF-8 terminal the

lone byte 0xE9 is invalid and prints as garbage, often �. */

putchar(233);

/* char is signed on x86, so a plain char holding 0xE9 is -23.

Use unsigned char when you compare or index by byte value. */

char c = (char)233;

unsigned char u = 233;

printf("\n%d %d\n", c, u); /* -23 233 */

/* Portable: print the Unicode character as UTF-8 instead. */

printf("\u0398\n"); /* CE 98 */

return 0;

}