How do I repair ESPAÃ‘A text that's already been saved?

Reverse the wrong decode: in Python, `s.encode('cp1252').decode('utf-8')` turns ESPAÃ‘A back into ESPAÑA. Run it before anything that straightens curly quotes, since the ‘ in the garble is really the second byte of Ñ, and once it becomes a plain apostrophe that byte is lost. Try it on a few rows first, because clean strings with accents will raise an error instead of converting.

Why does ESPAÑA show up as ESPAÃ‘A?

Check the database connection first: if it isn't set to UTF-8, text stored correctly comes back read as Windows-1252. With mysqli, call `mysqli_set_charset($link, 'utf8mb4')` right after connecting, and serve the page with charset=utf-8. If Ã‘ is already stored in the table, the connection fix only stops new damage, so repair the saved rows as well.

Why does uppercasing Muñoz give MUñOZ?

C's toupper() in the default C locale only changes ASCII letters, so a program that capitalizes names byte by byte turns Muñoz into MUñOZ. Use a Unicode-aware uppercase such as Python's str.upper() instead.

When do I use ╤ instead of ╦ in a table?

Use ╤ where a single column line drops from a double top border, the look of a double outer frame around single inner columns: ╔═╤═╗. ╦ is only for a double column line. That style also needs ╧ (207) on the bottom edge and ╪ (216) where rows cross, and code page 850 swapped all three for other characters, so it only draws right under 437.

What's the Alt code for Ñ, since Alt+209 gives ╤?

Alt+165, the old DOS code, which gives Ñ on both US and Western European PCs, and its lowercase ñ is Alt+164. Alt+0209 works too. Plain Alt+209 reads the DOS table, where 209 is the box piece ╤ on a US machine and Ð on a Western European one.

ASCII 209

Uppercase N with tildeBox T-junction pointing down, double horizontal

ASCII 209 is Ñ on Windows and ╤ in old DOS.

Ñ (Capital N with Tilde) is byte 209 in Windows-1252, Unicode U+00D1. Spanish capitals need it (AÑO, ESPAÑA), and it sits in surnames like Ibáñez and Núñez. Email is where those names get stuck. An address like muñoz@example.com is valid under the internationalized email standard (RFC 6531), but every mail server along the route has to support it, and a signup form with an ASCII-only pattern rejects it before it gets that far. Old DOS had Ñ at 165. Its UTF-8 bytes C3 91 misread as Ã‘. The HTML entity is Ñ.

╤ (Single Down, Double Horizontal) is byte 209 in code page 437, the original IBM PC character set, and maps to Unicode U+2564. It's a top-edge tee where a double line ═ (205) runs across and a single column line │ (179) drops down. Python's tabulate library uses it in its fancy_grid format, whose tables start with ╒═╤═╕. Printing one works in the Windows console, but redirect the output to a file (python report.py > out.txt) and Python can fall back to the ANSI code page, which has no ╤, and stop with UnicodeEncodeError. Run it with python -X utf8 or set PYTHONIOENCODING=utf-8. In UTF-8 it's E2 95 A4, and the HTML entity is &boxHd;.

Latin LetterBox Drawing

Technical Details

Code Example

</>

#include <stdio.h>

int main(void) {
    /* On Windows (code page 1252) byte 209 is Ñ.
       Modern terminals expect UTF-8, so print the code point, not the byte. */
    printf("\u00D1\n");   /* UTF-8: C3 91 */

    unsigned char b = 209;         /* the raw Windows-1252 byte */
    printf("%d 0x%02X\n", b, b);    /* 209 0xD1 */
    return 0;
}

Code Example

</>

#include <stdio.h>

int main(void) {
    /* Byte 209 is ╤ only on a console using code page 437
       (chcp 437 on Windows, or DOSBox). On a UTF-8 terminal the
       lone byte 0xD1 is invalid and prints as garbage, often �. */
    putchar(209);

    /* char is signed on x86, so a plain char holding 0xD1 is -47.
       Use unsigned char when you compare or index by byte value. */
    char c = (char)209;
    unsigned char u = 209;
    printf("\n%d %d\n", c, u);   /* -47 209 */

    /* Portable: print the Unicode character as UTF-8 instead. */
    printf("\u2564\n");          /* E2 95 A4 */
    return 0;
}

How to Type It

Windows: Hold Alt and type 0209 on the numeric keypad, with Num Lock on. The leading zero picks the Windows code page. Alt+165, without the zero, types it too. No numeric keypad? In Word, type 00D1 and press Alt+X.
Mac: Press Option+N, then Shift+N.
Linux: In GNOME, or any desktop running IBus, press Ctrl+Shift+U, type d1, then Space. With a Compose key set up, press Compose, tilde, N.
Phone: Tap Shift, then long-press N and pick Ñ.

How to Type It

Windows: Hold Alt and type 209 on the numeric keypad, with no leading zero. US PCs insert ╤, but PCs set to code page 850, common in Western Europe, insert Ð. No numeric keypad? In Word, type 2564 and press Alt+X.
Mac: Press Control+Command+Space to open the Character Viewer, then search for "box drawings down single and horizontal double".
Linux: In GNOME, or any desktop running IBus, press Ctrl+Shift+U, type 2564, 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

Reverse the wrong decode: in Python, `s.encode('cp1252').decode('utf-8')` turns ESPAÃ‘A back into ESPAÑA. Run it before anything that straightens curly quotes, since the ‘ in the garble is really the second byte of Ñ, and once it becomes a plain apostrophe that byte is lost. Try it on a few rows first, because clean strings with accents will raise an error instead of converting.
Check the database connection first: if it isn't set to UTF-8, text stored correctly comes back read as Windows-1252. With mysqli, call `mysqli_set_charset($link, 'utf8mb4')` right after connecting, and serve the page with charset=utf-8. If Ã‘ is already stored in the table, the connection fix only stops new damage, so repair the saved rows as well.
C's toupper() in the default C locale only changes ASCII letters, so a program that capitalizes names byte by byte turns Muñoz into MUñOZ. Use a Unicode-aware uppercase such as Python's str.upper() instead.
Alt+165, the old DOS code, which gives Ñ on both US and Western European PCs, and its lowercase ñ is Alt+164. Alt+0209 works too. Plain Alt+209 reads the DOS table, where 209 is the box piece ╤ on a US machine and Ð on a Western European one.

Frequently Asked Questions

Use ╤ where a single column line drops from a double top border, the look of a double outer frame around single inner columns: ╔═╤═╗. ╦ is only for a double column line. That style also needs ╧ (207) on the bottom edge and ╪ (216) where rows cross, and code page 850 swapped all three for other characters, so it only draws right under 437.
Alt+165, the old DOS code, which gives Ñ on both US and Western European PCs, and its lowercase ñ is Alt+164. Alt+0209 works too. Plain Alt+209 reads the DOS table, where 209 is the box piece ╤ on a US machine and Ð on a Western European one.

ASCII 209

Uppercase N with tildeBox T-junction pointing down, double horizontal

ASCII 209 is Ñ on Windows and ╤ in old DOS.

Latin Letter

Box Drawing

#include <stdio.h>

int main(void) {

/* On Windows (code page 1252) byte 209 is Ñ.

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

printf("\u00D1\n"); /* UTF-8: C3 91 */

unsigned char b = 209; /* the raw Windows-1252 byte */

printf("%d 0x%02X\n", b, b); /* 209 0xD1 */

return 0;

}

#include <stdio.h>

int main(void) {

/* Byte 209 is ╤ only on a console using code page 437

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

lone byte 0xD1 is invalid and prints as garbage, often �. */

putchar(209);

/* char is signed on x86, so a plain char holding 0xD1 is -47.

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

char c = (char)209;

unsigned char u = 209;

printf("\n%d %d\n", c, u); /* -47 209 */

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

printf("\u2564\n"); /* E2 95 A4 */

return 0;

}