What You Will Learn

• What the C character set contains and why the standard calls it the basic source character set.
• How letters, digits and special characters are used in real C syntax.
• The difference between the source character set and the execution character set.
• What whitespace really does, and why int x is not the same as intx.
• The job of each delimiter: colon, semicolon, parentheses, brackets, braces, hash and comma.
• How to classify characters with ctype.h and convert between characters and numbers.
• Typical beginner errors, a quiz, and interview questions asked on this topic.

Prerequisites

• A working compiler. On Ubuntu or Debian run sudo apt install gcc; on Windows use WSL or MinGW.
• Ability to save a .c file and compile it from a terminal. Chapter 1 of this free embedded C course covered setup.
• No earlier C knowledge beyond printf is required.

Why the C Character Set Matters

A C source file is just text. Before the compiler can understand a single idea in it, it performs lexical analysis: it reads characters one by one and groups them into the smallest meaningful units, called tokens. A token can be a keyword like while, a name like total, a number like 42, a string, an operator like +, or a punctuator like ;. The character set decides which characters are allowed to take part in this process.

Because the very first stage is character-driven, a single wrong character, for example a curly quote pasted from a web page, stops the build immediately. Understanding the set saves hours of confusing compiler messages.

The Three Building Blocks of the C Character Set

The basic character set is divided into letters, digits and special characters. Whitespace is added on top of these as a fourth group that controls layout.

Group	Members	Count	Typical use
Letters	A to Z and a to z	52	Keywords, variable and function names
Digits	0 to 9	10	Numeric constants, parts of names
Graphic symbols	Punctuation and operator characters	29	Operators, delimiters, quotes, escapes
Whitespace	Space, horizontal tab, vertical tab, form feed, newline	5	Separating tokens, formatting

Adding 52 letters, 10 digits and 29 graphic symbols gives the 91 graphic characters of the basic set. Notice that the underscore is counted among the 29 symbols, yet the language treats it like a letter when you build names.

Alphabets: Case Matters

C distinguishes uppercase from lowercase everywhere. Speed, speed and SPEED are three unrelated names, and While is not the keyword while. Behind the scenes every letter is stored as a small integer. In ASCII, 'A' is 65 and 'a' is 97. Uppercase and lowercase letters of the same pair differ by exactly 32, which is a single bit (bit 5). You will use this fact in the second example.

Digits: Characters Versus Numbers

The character '7' and the number 7 are different things. The character has a code (55 in ASCII); the number is a value. The C standard guarantees that the digit characters are consecutive, so '7' - '0' always gives 7 on any platform. That guarantee does not exist for letters on every encoding, so digit arithmetic is portable while letter arithmetic is only safe on ASCII-compatible systems.

Special Characters and What They Do in C

Instead of memorising a flat list of names, group the symbols by the role they play in the language.

Role	Characters	Where you meet them
Arithmetic operators	+ - * / %	total = a + b;
Relational and logical	< > = ! & |	if (x >= 10 && y != 0)
Bitwise	& | ^ ~	mask = flags & 0x0F;
Grouping delimiters	( ) [ ] { }	Calls, arrays, blocks
Separators	, ; : .	Argument lists, statement ends, labels, struct members
Conditional operator	? :	max = (a > b) ? a : b;
Quoting	' "	'A' and "hello"
Escape introducer		"linen" (next lecture)
Preprocessor	#	#include, #define
Name builder	_	sensor_value

Whitespace: The Invisible Separator

Space, horizontal tab, vertical tab, form feed and newline are collectively called whitespace. Their main job is to separate tokens. The compiler needs int total with a gap, because without it the lexer would see one long name, inttotal. Beyond that gap, extra whitespace is ignored, which is why you can indent code freely.

The lexer uses a rule called maximal munch: it always builds the longest token it can. That is why a+++b is read as a++ + b and not a + ++b. Whitespace is your tool for steering the lexer when that matters.

Source Character Set Versus Execution Character Set

There are two character sets in play, and mixing them up is a classic interview trap.

Aspect	Source character set	Execution character set
Where it applies	The text of your .c file	Characters inside the running program
Minimum content	The 91 graphic characters plus whitespace	The same, plus control characters like newline and backspace
Example	Writing the two characters and n	One newline character with code 10
Chosen by	Compiler, editor encoding	Compiler and target platform

On modern Linux both are usually ASCII-compatible UTF-8, so the difference feels invisible. On a microcontroller toolchain or a mainframe, it is real. This is also why the next lecture introduces escape sequences: they are the bridge from characters you can type to characters the program can use.

C Delimiters (Punctuators) in Detail

Delimiters give a program its structure. The standard calls them punctuators, but most books, and many interviewers, use the word delimiter.

Delimiter	Name	Purpose	Example
:	Colon	Ends a label, separates the parts of ? :, follows case and default	again:
;	Semicolon	Terminates a statement or declaration	total = 0;
( )	Parentheses	Group expressions, enclose call arguments and conditions	if (x > 0)
[ ]	Square brackets	Declare and index arrays	marks[2]
{ }	Curly braces	Form a block, enclose initialisers and function bodies	{ 70, 85 }
#	Hash	Begins a preprocessor directive	#define N 3
,	Comma	Separates declarators and arguments; also an operator	int a, b;

Hands-On Examples: Practise the C Character Set

Every example below was compiled with gcc -Wall -Wextra -std=c17 and the output shown is real. Type them in yourself rather than pasting; the muscle memory is part of the learning.

Example 1: Classify Every Character

Goal: walk through a string and count letters, digits, spaces and everything else. This shows the groups of the character set in action and introduces ctype.h.

#include <stdio.h>
#include <ctype.h>

int main(void)
{
    const char *text = "Sensor#7 reads 42.5C @ 09:30!";
    int letters = 0, digits = 0, spaces = 0, others = 0;

    for (const char *p = text; *p != ''; p++) {
        unsigned char ch = (unsigned char)*p;
        if (isalpha(ch))
            letters++;
        else if (isdigit(ch))
            digits++;
        else if (isspace(ch))
            spaces++;
        else
            others++;
    }

    printf("Text    : %sn", text);
    printf("Letters : %dn", letters);
    printf("Digits  : %dn", digits);
    printf("Spaces  : %dn", spaces);
    printf("Others  : %dn", others);
    return 0;
}

Build and run:

gcc -Wall -Wextra -std=c17 classify.c -o classify
./classify

Expected output:

Text    : Sensor#7 reads 42.5C @ 09:30!
Letters : 12
Digits  : 8
Spaces  : 4
Others  : 5

• The cast to unsigned char before calling isalpha is deliberate. The ctype functions accept values in the unsigned range, and passing a negative char is undefined behaviour.
• The string contains #, ., @, : and !, which makes the five “others”.

Try it yourself: change the string to your own name and a phone number

Replace the text with something like "Asha 98480-12345". You should see 4 letters, 10 digits, 1 space and 1 other (the hyphen). If your counts differ, check whether you accidentally typed a non-ASCII dash.

Example 2: Characters Are Small Numbers

Goal: prove that a character is an integer in disguise, and use that fact to change case and convert a digit to a value.

#include <stdio.h>

int main(void)
{
    char upper = 'G';
    char lower = upper ^ 0x20;   /* flip bit 5 */
    char digit = '7';

    printf("'%c' has code %dn", upper, upper);
    printf("'%c' has code %dn", lower, lower);
    printf("Gap between 'a' and 'A' : %dn", 'a' - 'A');
    printf("Digit '%c' is the number %dn", digit, digit - '0');

    for (char c = 'A'; c <= 'E'; c++)
        printf("%c=%d ", c, c);
    printf("n");
    return 0;
}

'G' has code 71
'g' has code 103
Gap between 'a' and 'A' : 32
Digit '7' is the number 7
A=65 B=66 C=67 D=68 E=69 

• upper ^ 0x20 flips bit 5, turning 'G' (71) into 'g' (103). The same trick converts back.
• digit - '0' is the portable way to turn a digit character into its value.
• The loop for (char c = 'A'; c <= 'E'; c++) works because characters support arithmetic and comparison like any integer.

Try it yourself: convert ‘z’ to uppercase without using toupper

Use 'z' - 32 or 'z' ^ 0x20; both give 90, which is 'Z'. Remember this only works for letters, because applying the XOR to a digit or symbol produces a different, unrelated character.

Example 3: Special Characters as Operators

Goal: see the arithmetic, bitwise, logical and pointer roles of special characters in one run.

#include <stdio.h>

int main(void)
{
    int a = 12, b = 5;
    int *p = &a;

    printf("a + b = %d, a - b = %dn", a + b, a - b);
    printf("a * b = %d, a / b = %d, a %% b = %dn", a * b, a / b, a % b);
    printf("a & b = %d, a | b = %d, a ^ b = %d, ~a = %dn", a & b, a | b, a ^ b, ~a);
    printf("!a = %d, a > b = %d, a == b = %dn", !a, a > b, a == b);
    printf("value through *p = %dn", *p);
    printf("100%% donen");
    return 0;
}

a + b = 17, a - b = 7
a * b = 60, a / b = 2, a % b = 2
a & b = 4, a | b = 13, a ^ b = 9, ~a = -13
!a = 0, a > b = 1, a == b = 0
value through *p = 12
100% done

• The percent sign has two jobs. In a % b it computes the remainder. In a printf format string you must write %% to print a literal percent.
• With a = 12 (binary 1100) and b = 5 (binary 0101), AND gives 0100 = 4, OR gives 1101 = 13, XOR gives 1001 = 9.
• The asterisk is multiplication in a * b but dereference in *p. The ampersand is bitwise AND between two values but address-of in &a. Context decides, which is why the lexer alone cannot understand a program.

Try it yourself: predict ~b before running

For b = 5 the answer is -6. The tilde flips every bit, and on a two’s complement machine flipping all bits of n gives -(n + 1).

Example 4: Delimiters at Work

Goal: use a colon label, square brackets, braces, commas, a semicolon and a hash directive in one small program.

#include <stdio.h>
#define LIMIT 3

int main(void)
{
    int marks[LIMIT] = { 70, 85, 90 }, total = 0, i = 0;

again:
    total += marks[i];
    i++;
    if (i < LIMIT)
        goto again;

    {
        int average = total / LIMIT;
        printf("total=%d average=%dn", total, average);
    }
    return 0;
}

total=245 average=81

• #define LIMIT 3 starts with a hash, so the preprocessor handles it before compilation.
• again: is a label: a name followed by a colon. goto again; jumps back to it, forming a loop.
• The bare pair of braces creates an inner block whose variable average exists only inside it.
• 245 divided by 3 is 81 in integer arithmetic; the fractional part is discarded.

Try it yourself: remove one semicolon and read the error

Delete the semicolon after total += marks[i]. GCC reports an “expected ‘;'” style error that names the token which followed. The compiler only realises something is wrong when that next token arrives, so the real fault is at the end of the previous line. Learning to look one line up is a real-world debugging skill.

Common Mistakes with the C Character Set

Mistake	Symptom	Fix
Pasting curly quotes from a document	Error such as stray ‘342’ in program	Retype the quotes as plain " and '
Using 'abc' for a string	Warning about a multi-character constant	Single quotes are for one character, double quotes for strings
Writing = instead of == in a condition	Condition always true or false	Compile with -Wall, which flags it
Forgetting %% in printf	Garbage or missing output	Double the percent sign
Passing a plain char to isalpha	Undefined behaviour for negative values	Cast to unsigned char
Using @ or $ in code	Stray character error	Keep them inside strings and comments

Best Practices, Performance and Security Notes

• Save source files as UTF-8 without a byte-order mark and keep code itself pure ASCII; put non-ASCII text only in comments or in string data you intend to handle.
• Prefer isalpha, isdigit and isspace over hard-coded numeric ranges. They are clearer and respect the platform encoding.
• Always compile with -Wall -Wextra. Most character-level mistakes are caught as warnings.
• Performance: character classification is cheap, usually a table lookup, so do not hand-roll range checks for speed.
• Security: when you validate input, decide explicitly which characters are allowed (an allow-list) instead of trying to block bad ones. Unexpected control characters in input are a common source of injection and log-forging bugs in embedded Linux services.

Summary and Key Takeaways

• The C character set is 52 letters, 10 digits, 29 graphic symbols and the whitespace characters.
• The lexer groups characters into tokens using maximal munch; whitespace separates tokens.
• Characters are small integers; '7' - '0' is 7 and a case flip is a single XOR with 32 in ASCII.
• @, $ and the back-quote are not in the basic source set.
• Delimiters: colon for labels, semicolon to end statements, parentheses for grouping, brackets for arrays, braces for blocks, hash for the preprocessor, comma as a separator.
• Source and execution character sets are different ideas; escape sequences connect them.

Quiz: Test Your Understanding

Pick an answer first, then open the box to check.

Interview Questions on the C Character Set

Frequently Asked Questions

What is the C character set in simple words?

It is the list of characters you are allowed to type in a C program: letters, digits, a fixed group of symbols and whitespace.

Does C support Unicode characters in source code?

Modern compilers read UTF-8 files and allow non-ASCII text in strings and comments. Using non-ASCII letters in identifiers is allowed by the standard but poorly portable, so avoid it in embedded work.

Is the underscore a letter in C?

The standard lists it as a symbol, but for building identifiers it behaves like a letter, so it can start a name.

Why does my program fail with a stray character error?

Most often a smart quote, long dash or invisible non-breaking space was pasted from a document. Retype the line.

Are delimiters and operators the same thing?

No. Operators compute values; delimiters give structure. Some symbols, such as the comma and the colon, play both roles depending on context.

Is this free embedded systems course enough for interview preparation?

It is designed to be: each lecture pairs theory with runnable examples, a quiz and interview questions. Work through the examples by hand for the best result.

Conclusion

You have now seen the C character set from the compiler’s point of view: a stream of characters that is sliced into tokens, with whitespace as separators and delimiters as the structural glue. The programs you ran proved the key ideas: characters are numbers, the same symbol can have different jobs depending on context, and tiny character mistakes produce confusing errors. In the next lecture of this free embedded C programming course we move to escape sequences and trigraphs, which let a program use characters that cannot simply be typed.