A newline is not necessarily a file byte
In C, fputs("hello\n", stdout) writes a text line. The program uses '\n' for its newline, but text output can convert that character to a platform-dependent external representation. Counting characters in the string therefore does not necessarily tell you how many bytes reach a file or console. section 8.3.3
Encoding determines how characters are represented. A C program can encounter different encodings in its source, during execution, and in external text. The C library handles text-I/O conversions where it can; portable string processing should avoid assuming a particular numeric code for a character. section 8.3.3 section 8.4
Three practical rules follow:
- Use character constants such as 'a' and '\n', and string literals such as "hello:\tx", rather than substituting presumed numeric character codes. An integer character constant has type int, not char. section 8.4
- Use <ctype.h> classifiers and case conversions for character classes instead of assuming every alphabetic character belongs to a fixed numeric range. Some classes may include platform-specific characters. section 8.4
- Treat numeric escapes carefully: a hexadecimal escape consumes subsequent hexadecimal digits. Separating adjacent string literals can mark its end. A numeric escape compiling on multiple platforms does not guarantee that its character code exists on each one. section 8.4
TAKEAWAY: A text character, a numeric character code, and a byte of external text are not interchangeable ideas. The limited basic character set cannot express many languages and scripts. Multibyte text and wide characters provide a route to handling extended characters, but operations on them must account for conversion rather than simply treating each char as one character. section 14.3 section 8.3.3
43 #define VBAR "\u2502" /**< a vertical bar character */
44 #define HBAR "\u2500" /**< a horizontal bar character */
45 #define TOPLEFT "\u250c" /**< topleft corner character */
46 #define TOPRIGHT "\u2510" /**< topright corner character */
!@%STYLE%@!
{"css":"{\"css\": \"font-weight: bold;\"}","target":"[[{\"line\":0,\"ch\":0},{\"line\":0,\"ch\":2}],[{\"line\":0,\"ch\":5},{\"line\":0,\"ch\":12}],[{\"line\":1,\"ch\":5},{\"line\":1,\"ch\":12}],[{\"line\":2,\"ch\":5},{\"line\":2,\"ch\":12}],[{\"line\":3,\"ch\":5},{\"line\":3,\"ch\":12}],[{\"line\":1,\"ch\":0},{\"line\":1,\"ch\":2}],[{\"line\":0,\"ch\":5},{\"line\":0,\"ch\":12}],[{\"line\":1,\"ch\":5},{\"line\":1,\"ch\":12}],[{\"line\":2,\"ch\":5},{\"line\":2,\"ch\":12}],[{\"line\":3,\"ch\":5},{\"line\":3,\"ch\":12}],[{\"line\":2,\"ch\":0},{\"line\":2,\"ch\":2}],[{\"line\":0,\"ch\":5},{\"line\":0,\"ch\":12}],[{\"line\":1,\"ch\":5},{\"line\":1,\"ch\":12}],[{\"line\":2,\"ch\":5},{\"line\":2,\"ch\":12}],[{\"line\":3,\"ch\":5},{\"line\":3,\"ch\":12}],[{\"line\":3,\"ch\":0},{\"line\":3,\"ch\":2}],[{\"line\":0,\"ch\":5},{\"line\":0,\"ch\":12}],[{\"line\":1,\"ch\":5},{\"line\":1,\"ch\":12}],[{\"line\":2,\"ch\":5},{\"line\":2,\"ch\":12}],[{\"line\":3,\"ch\":5},{\"line\":3,\"ch\":12}]]"}
!@%STYLE%@!
These string macros use \u escapes to identify box-drawing characters without writing their source-file glyphs directly. section 14.3
Measure characters after conversion
A separator-printing routine passes multibyte strings to fputs but uses mbsrlen to determine their character lengths before filling the remaining space with horizontal bars. Its helper calls mbsrtowcs to obtain a wide-character length. The example separates the bytes used for output from the characters counted for layout. section 14.3
Before printing the example’s multibyte characters, setlocale(LC_ALL, "") selects the environment’s locale. This is a concrete reminder that successful interpretation can depend on the execution environment, not just on the characters visible in a C source file. section 14.3
