When does c need a cedilla in French?

Only before a, o or u, where a plain c would sound like k: ça, garçon, reçu. Before e and i the c is already soft, so ç never appears there, which is why commencer gains one in nous commençons but not in vous commencez. Portuguese follows the same rule, as in açúcar.

Can I write c instead of ç?

In French and Portuguese, c is the accepted fallback where only ASCII fits, like URLs and email addresses, even though francais then reads with a k sound. In Turkish it changes the word: ç is a separate letter pronounced like ch, so çam means pine and cam means glass. There, keep the letter wherever the system allows it.

Why does psql store français as franÃ§ais on Windows?

psql decodes your script with its client encoding, which on Windows usually isn't UTF-8, so ç's two bytes are taken as two separate characters and saved that way. Put `SET client_encoding TO 'UTF8';` at the top of the script, or set PGCLIENTENCODING=UTF8 before running psql -f. Rows inserted before the fix stay double-encoded and need repairing separately.

Why is it Barça and not Barca?

Barça is the Catalan spelling of the nickname, and without the cedilla it would be read as Bar-ka. The club's own branding and the Catalan press write it with ç, so a search or hashtag using Barca misses part of what's out there. Unlike French, Catalan also puts ç at the end of words, as in feliç.

ASCII 231

Lowercase c with cedillaGreek small tau

ASCII 231 is ç on Windows and τ in old DOS.

ç (Small C with Cedilla) is byte 231 in Windows-1252, Unicode U+00E7. French, Portuguese, Catalan, Turkish and Albanian all use it. File downloads are a good place to lose it. HTTP header values are safest as ASCII, so a server that puts français.pdf straight into Content-Disposition can hand the browser raw UTF-8 bytes, and the file saves as franÃ§ais.pdf or with the letter dropped. Send the name the standard way instead, filename*=UTF-8''fran%C3%A7ais.pdf, with a plain ASCII filename= next to it as a fallback. Old DOS had ç at 135. The UTF-8 bytes C3 A7 misread as Ã§. The HTML entity is ç.

τ (Tau) is byte 231 in code page 437, the original IBM PC character set, and maps to Unicode U+03C4. Physics uses it for torque and for the time constant of an RC circuit, and some mathematicians use it for 2π, one full turn. Python adopted that last meaning: math.tau is 6.283185307179586. JavaScript has no Math.TAU, so code ported from Python needs 2 * Math.PI there. Code page 850 put the Icelandic þ on this byte. In UTF-8, τ is CF 84, and the HTML entity is τ.

Latin LetterGreek Letter

Technical Details

Code Example

</>

#include <stdio.h>

int main(void) {
    /* On Windows (code page 1252) byte 231 is ç.
       Modern terminals expect UTF-8, so print the code point, not the byte. */
    printf("\u00E7\n");   /* UTF-8: C3 A7 */

    unsigned char b = 231;         /* the raw Windows-1252 byte */
    printf("%d 0x%02X\n", b, b);    /* 231 0xE7 */
    return 0;
}

Code Example

</>

#include <stdio.h>

int main(void) {
    /* Byte 231 is τ only on a console using code page 437
       (chcp 437 on Windows, or DOSBox). On a UTF-8 terminal the
       lone byte 0xE7 is invalid and prints as garbage, often �. */
    putchar(231);

    /* char is signed on x86, so a plain char holding 0xE7 is -25.
       Use unsigned char when you compare or index by byte value. */
    char c = (char)231;
    unsigned char u = 231;
    printf("\n%d %d\n", c, u);   /* -25 231 */

    /* Portable: print the Unicode character as UTF-8 instead. */
    printf("\u03C4\n");          /* CF 84 */
    return 0;
}

How to Type It

Windows: Hold Alt and type 0231 on the numeric keypad, with Num Lock on. The leading zero picks the Windows code page. Alt+135, without the zero, types it too. No numeric keypad? In Word, type 00E7 and press Alt+X.
Mac: Press Option+C.
Linux: In GNOME, or any desktop running IBus, press Ctrl+Shift+U, type e7, then Space. With a Compose key set up, press Compose, comma, c.
Phone: Long-press c and pick ç.

How to Type It

Windows: Hold Alt and type 231 on the numeric keypad, with no leading zero. US PCs insert τ, but PCs set to code page 850, common in Western Europe, insert þ. No numeric keypad? In Word, type 03C4 and press Alt+X.
Mac: Press Control+Command+Space to open the Character Viewer, then search for "greek small letter tau".
Linux: In GNOME, or any desktop running IBus, press Ctrl+Shift+U, type 3c4, then Space.
Phone: Add the Greek keyboard in your phone's keyboard settings and type τ there, or tap the character at the top of this page to copy it.

Frequently Asked Questions

Only before a, o or u, where a plain c would sound like k: ça, garçon, reçu. Before e and i the c is already soft, so ç never appears there, which is why commencer gains one in nous commençons but not in vous commencez. Portuguese follows the same rule, as in açúcar.
In French and Portuguese, c is the accepted fallback where only ASCII fits, like URLs and email addresses, even though francais then reads with a k sound. In Turkish it changes the word: ç is a separate letter pronounced like ch, so çam means pine and cam means glass. There, keep the letter wherever the system allows it.
psql decodes your script with its client encoding, which on Windows usually isn't UTF-8, so ç's two bytes are taken as two separate characters and saved that way. Put `SET client_encoding TO 'UTF8';` at the top of the script, or set PGCLIENTENCODING=UTF8 before running psql -f. Rows inserted before the fix stay double-encoded and need repairing separately.
Barça is the Catalan spelling of the nickname, and without the cedilla it would be read as Bar-ka. The club's own branding and the Catalan press write it with ç, so a search or hashtag using Barca misses part of what's out there. Unlike French, Catalan also puts ç at the end of words, as in feliç.

ASCII 231

Lowercase c with cedillaGreek small tau

ASCII 231 is ç on Windows and τ in old DOS.

Latin Letter

Greek Letter

#include <stdio.h>

int main(void) {

/* On Windows (code page 1252) byte 231 is ç.

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

printf("\u00E7\n"); /* UTF-8: C3 A7 */

unsigned char b = 231; /* the raw Windows-1252 byte */

printf("%d 0x%02X\n", b, b); /* 231 0xE7 */

return 0;

}

#include <stdio.h>

int main(void) {

/* Byte 231 is τ only on a console using code page 437

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

lone byte 0xE7 is invalid and prints as garbage, often �. */

putchar(231);

/* char is signed on x86, so a plain char holding 0xE7 is -25.

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

char c = (char)231;

unsigned char u = 231;

printf("\n%d %d\n", c, u); /* -25 231 */

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

printf("\u03C4\n"); /* CF 84 */

return 0;

}