Should I use ¼ or 1/4 with CSS fraction styling?

Use plain 1/4 with `font-variant-numeric: diagonal-fractions` if your font has the OpenType frac feature, since it works for any fraction, like 3/16, while ready-made characters only exist for a few common ones. ¼ is still the safer choice in plain text or anywhere you can't control the font, because without that feature the styled version falls back to an ordinary 1/4.

Why doesn't a search for 1/4 find ¼, even after NFKC?

Because NFKC turns ¼ into 1⁄4 with the fraction slash U+2044, not the / on your keyboard, so the two still don't match. Normalize, then swap that slash: in Python, `unicodedata.normalize('NFKC', s).replace('\u2044', '/')` makes ¼ and 1/4 compare equal.

How do I parse 1¼ cups into 1.25?

Split the whole number from the fraction character and add them. In Python, `m = re.match(r'(\d*)\s*([¼½¾])', s)` captures both parts, and `int(m[1] or 0) + unicodedata.numeric(m[2])` gives 1.25 for 1¼ and 0.25 for ¼ alone. Extend the character class if your data uses ⅓ or ⅛ as well.

Why does Excel's SUM skip ¼?

Type 1/4 into Excel and it becomes a date (4 January or 1 April, depending on your locale), so ¼ seems like the fix. But Excel treats ¼ as text, and SUM skips it. To store a quarter as a number, type 0 1/4, which Excel reads as 0.25 in fraction format.

Why does printing ╝ in the bottom-right corner scroll the console?

Writing to the last cell of the Windows console moves the cursor past the end of the screen, and the console scrolls up a line to make room, dragging your whole frame with it. Write that corner without moving the cursor: WriteConsoleOutputCharacterW puts characters at given coordinates and leaves the cursor alone. On ANSI terminals, a newline after the bottom row does the same damage, so leave it out.

Why does Alt+188 give ╝ instead of ¼?

Without the leading zero, Windows looks 188 up in the DOS code page, and there it's the double-line corner ╝. Alt+0188 goes through Windows-1252 and gives ¼. In the DOS table the quarter sits at 172, right after ½ at 171, so Alt+172 is the zero-free way to type it.

ASCII 188

One quarterDouble box bottom-right corner

ASCII 188 is ¼ on Windows and ╝ in old DOS.

¼ (One Quarter) is byte 188 in Windows-1252, Unicode U+00BC. Recipes and measurements use it, as in ¼ cup or ¼ inch. Windows-1252 has only three fractions, ¼, ½ and ¾, so anything else, like 3/8, needs another approach. On the web you can type fractions as plain digits and let the font do the work: CSS font-variant-numeric: diagonal-fractions draws 1/4 or 3/8 as proper fractions in fonts that support it, and the text underneath stays searchable as 3/8. Old DOS had ¼ too, at 172. In UTF-8 it's C2 BC, which misreads as Â¼. The HTML entity is ¼.

╝ (Double Up and Left) is byte 188 in code page 437, the original IBM PC character set, and maps to Unicode U+255D. It's the bottom-right corner of an all-double frame, where ═ (205) comes in from the left and ║ (186) comes down from above; ╚ at 200 is its left-hand partner. In code page 437 every box piece was one byte, but in UTF-8 ╝ takes three (E2 95 9D). C code that lines up frames with strlen() or printf padding like %-20s counts bytes, not characters, so the right edge of a UTF-8 frame drifts. Pad by character count instead. The HTML entity is &boxUL;.

NumberBox Drawing

Technical Details

Code Example

</>

#include <stdio.h>

int main(void) {
    /* On Windows (code page 1252) byte 188 is ¼.
       Modern terminals expect UTF-8, so print the code point, not the byte. */
    printf("\u00BC\n");   /* UTF-8: C2 BC */

    unsigned char b = 188;         /* the raw Windows-1252 byte */
    printf("%d 0x%02X\n", b, b);    /* 188 0xBC */
    return 0;
}

Code Example

</>

#include <stdio.h>

int main(void) {
    /* Byte 188 is ╝ only on a console using code page 437
       (chcp 437 on Windows, or DOSBox). On a UTF-8 terminal the
       lone byte 0xBC is invalid and prints as garbage, often �. */
    putchar(188);

    /* char is signed on x86, so a plain char holding 0xBC is -68.
       Use unsigned char when you compare or index by byte value. */
    char c = (char)188;
    unsigned char u = 188;
    printf("\n%d %d\n", c, u);   /* -68 188 */

    /* Portable: print the Unicode character as UTF-8 instead. */
    printf("\u255D\n");          /* E2 95 9D */
    return 0;
}

How to Type It

Windows: Hold Alt and type 0188 on the numeric keypad, with Num Lock on. The leading zero picks the Windows code page. Alt+172, without the zero, types it too. No numeric keypad? In Word, type 00BC and press Alt+X.
Mac: Press Control+Command+Space to open the Character Viewer, then search for "vulgar fraction one quarter".
Linux: In GNOME, or any desktop running IBus, press Ctrl+Shift+U, type bc, then Space. With a Compose key set up, press Compose, 1, 4.
Phone: Tap the character at the top of this page to copy it.

How to Type It

Windows: Hold Alt and type 188 on the numeric keypad, with no leading zero. Windows reads it as a DOS code and inserts ╝. No numeric keypad? In Word, type 255D and press Alt+X.
Mac: Press Control+Command+Space to open the Character Viewer, then search for "box drawings double up and left".
Linux: In GNOME, or any desktop running IBus, press Ctrl+Shift+U, type 255d, then Space.
Phone: Phone keyboards have no key for it. Tap the character at the top of this page to copy it.

Frequently Asked Questions

Use plain 1/4 with `font-variant-numeric: diagonal-fractions` if your font has the OpenType frac feature, since it works for any fraction, like 3/16, while ready-made characters only exist for a few common ones. ¼ is still the safer choice in plain text or anywhere you can't control the font, because without that feature the styled version falls back to an ordinary 1/4.
Because NFKC turns ¼ into 1⁄4 with the fraction slash U+2044, not the / on your keyboard, so the two still don't match. Normalize, then swap that slash: in Python, `unicodedata.normalize('NFKC', s).replace('\u2044', '/')` makes ¼ and 1/4 compare equal.
Split the whole number from the fraction character and add them. In Python, `m = re.match(r'(\d*)\s*([¼½¾])', s)` captures both parts, and `int(m[1] or 0) + unicodedata.numeric(m[2])` gives 1.25 for 1¼ and 0.25 for ¼ alone. Extend the character class if your data uses ⅓ or ⅛ as well.
Type 1/4 into Excel and it becomes a date (4 January or 1 April, depending on your locale), so ¼ seems like the fix. But Excel treats ¼ as text, and SUM skips it. To store a quarter as a number, type 0 1/4, which Excel reads as 0.25 in fraction format.
Without the leading zero, Windows looks 188 up in the DOS code page, and there it's the double-line corner ╝. Alt+0188 goes through Windows-1252 and gives ¼. In the DOS table the quarter sits at 172, right after ½ at 171, so Alt+172 is the zero-free way to type it.

Frequently Asked Questions

Writing to the last cell of the Windows console moves the cursor past the end of the screen, and the console scrolls up a line to make room, dragging your whole frame with it. Write that corner without moving the cursor: WriteConsoleOutputCharacterW puts characters at given coordinates and leaves the cursor alone. On ANSI terminals, a newline after the bottom row does the same damage, so leave it out.
Without the leading zero, Windows looks 188 up in the DOS code page, and there it's the double-line corner ╝. Alt+0188 goes through Windows-1252 and gives ¼. In the DOS table the quarter sits at 172, right after ½ at 171, so Alt+172 is the zero-free way to type it.

ASCII 188

One quarterDouble box bottom-right corner

ASCII 188 is ¼ on Windows and ╝ in old DOS.

Number

Box Drawing

#include <stdio.h>

int main(void) {

/* On Windows (code page 1252) byte 188 is ¼.

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

printf("\u00BC\n"); /* UTF-8: C2 BC */

unsigned char b = 188; /* the raw Windows-1252 byte */

printf("%d 0x%02X\n", b, b); /* 188 0xBC */

return 0;

}

#include <stdio.h>

int main(void) {

/* Byte 188 is ╝ only on a console using code page 437

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

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

putchar(188);

/* char is signed on x86, so a plain char holding 0xBC is -68.

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

char c = (char)188;

unsigned char u = 188;

printf("\n%d %d\n", c, u); /* -68 188 */

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

printf("\u255D\n"); /* E2 95 9D */

return 0;

}