Sewer

This page has been automatically translated using the Google Translate API services. We are working on improving texts. Thank you for your understanding and patience.

Even the grandest palaces needed sewers. Tom Lehrer

Functions

Types and Constants

Sewer is the first library within the NAppGUI SDK (Figure 1). It declares the basic types, the Unicode support, assertions, pointers safe manipulation, elementary math functions, Standard I/O and dynamic memory allocation. It is also used as a "sink" to bury the unsightly preprocessor macros necessary to configure the compiler, CPU, platforms, etc. As dependencies only has a few headers of the C standard library:

The C standard library (cstdlib) is not part of the C language, but implements functions of great utility for the developer that solve typical programming problems. Any C programmer has used it more or less and its study is usually linked to learning the language itself (Figure 2).

This library is located halfway between the application and system calls and provides a portable API for file access, dynamic memory, I/O, time, etc (Figure 3). It also implements mathematical functions, conversion, search, string management, etc. In one way or another, NAppGUI integrates its functionality, so it's not necessary (or advisable) to use cstdlib directly in the application layer. The reasons that have motivated this design decision can be summarized in:

• Small differences: Unix-like systems do not support the secure cstdlib versions implemented by Microsoft (strcpy_s() and others). The use of classical functions (without the suffix _s) is insecure and will trigger annoying warnings in Visual Studio.
• Security: Related to the previous one, osbs avoids buffer overflow vulnerabilities in the processing of memory blocks and strings.
• Compiler support: Each compiler implements this library in its own way. Binary modules that include cstdlib functions may not be able to link with other versions of the same compiler, as with Visual Studio.

In the interest of portability, avoid using functions from the C Standard Library directly in applications. Always use a NAppGUI equivalent function.

• Duplicity: Much of the functionality of cstdlib is already implemented in osbs using direct system calls (files, dynamic memory, I/O, time, etc.)
• Completeness: The cstdlib functions related to files (fopen() and others) do not include support for directory management. Files and directories presents a complete API based on system calls.
• Performance: In certain cases, especially in mathematical functions and memory management, it may be interesting to change the implementation of cstdlib to an improved one. All applications will benefit from the change, without having to modify your code.
• Clarity: The behavior of some cstdlib functions is not entirely clear and can lead to confusion. For example, strtoul has a very particular functioning that we must remember every time we use it.

1 2 3 4 5 6 7

char *s1 = "-56"; char *s2 = "asCr"; char *s3 = "467Xd"; int v1, v2, v3; v1 = strtoul(s1, NULL, 10); // v1 = 4294967240, errno = OK v2 = strtoul(s2, NULL, 10); // v2 = 0, errno = OK v3 = strtoul(s3, NULL, 10); // v3 = 467, errno = OK

• Style: The use of osbs functions does not break the aesthetics of an application written with NAppGUI.

1 2 3 4 5 6 7

real32_t a1 = 1.43f; real64_t a2 = .38; real32_t c = (real32_t)cosf((float)a1); real64_t t = (real64_t)tan((double)a2); ... real32_t c = bmath_cosf(a1); real64_t t = bmath_tand(a2);

• Independence: NAppGUI internally uses a very small subset of cstdlib functions. It is possible that in the future we will make our own implementations and completely disconnect the support of the standard library.

int8_t

8-bit signed integer. It can represent a value between INT8_MIN and INT8_MAX.

int16_t

16-bit signed integer. It can represent a value between INT16_MIN and INT16_MAX.

int32_t

32-bit signed integer. It can represent a value between INT32_MIN and INT32_MAX.

int64_t

64-bit signed integer. It can represent a value between INT64_MIN and INT64_MAX.

uint8_t

8-bit unsigned integer. It can represent a value between 0 and UINT8_MAX.

uint16_t

16-bit unsigned integer. It can represent a value between 0 and UINT16_MAX.

uint32_t

32-bit unsigned integer. It can represent a value between 0 and UINT32_MAX.

uint64_t

64-bit unsigned integer. It can represent a value between 0 and UINT64_MAX.

char_t

8-bit character type (Unicode). A single character may need 1, 2, 3 or 4 elements (bytes), depending on UTF encodings.

byte_t

8-bit type to store generic memory blocks.

bool_t

8-bit boolean. Only two values are allowed TRUE (1) and FALSE (0).

real

32 or 64-bit floating point number.

real32_t

32-bit floating point number. The C float type.

real64_t

64-bit floating point number. The C double type.

TRUE

const bool_t TRUE = 1;

True.

FALSE

const bool_t FALSE = 0;

False.

NULL

const void* NULL = 0;

Null pointer.

INT8_MIN

const int8_t INT8_MIN = 0x80;

-128.

INT8_MAX

const int8_t INT8_MAX = 0x7F;

127.

INT16_MIN

const int16_t INT16_MIN = 0x8000;

-32.768.

INT16_MAX

const int16_t INT16_MAX = 0x7FFF;

32.767.

INT32_MIN

const int32_t INT32_MIN = 0x80000000;

-2.147.483.648.

INT32_MAX

const int32_t INT32_MAX = 0x7FFFFFFF;

2.147.483.647.

INT64_MIN

const int64_t INT64_MIN = 0x8000000000000000;

-9.223.372.036.854.775.808.

INT64_MAX

const int64_t INT64_MAX = 0x7FFFFFFFFFFFFFFF;

9.223.372.036.854.775.807.

UINT8_MAX

const uint8_t UINT8_MAX = 0xFF;

255.

UINT16_MAX

const uint16_t UINT16_MAX = 0xFFFF;

65.535.

UINT32_MAX

const uint32_t UINT32_MAX = 0xFFFFFFFF;

4.294.967.295.

UINT64_MAX

const uint64_t UINT64_MAX = 0xFFFFFFFFFFFFFFFF;

18.446.744.073.709.551.615.

enum unicode_t

Represents the UTF encodings.

enum unicode_t
{
    ekUTF8,
    ekUTF16,
    ekUTF32
};

struct REnv

Random numbers environment.

struct REnv;

FPtr_destroy

Destructor function prototype.

void
(*FPtr_destroy)(type **item);

FPtr_copy

Copy constructor function prototype.

type*
(*FPtr_copy)(const type *item);

Return

The new object that is an exact copy of the input.

FPtr_scopy

Unallocated memory copy constructor prototype.

void
(*FPtr_scopy)(type *dest,
              const type *src);

Remarks

In this copy operation, the memory required by the object has already been allocated. We must create dynamic memory for the fields of the object that require it, but not for the object itself. Usually used to copy arrays of objects (not pointers to objects).

FPtr_compare

Comparison function prototype.

int
(*FPtr_compare)(const type *item1,
                const type *item2);

Return

Comparison result.

FPtr_compare_ex

Similar to FPtr_compare, but receive an additional parameter that may influence the comparison.

int
(*FPtr_compare_ex)(const type *item1,
                   const type *item2,
                   const dtype *data);

Return

Comparison result.

FPtr_assert

Callback function prototype called when an assert occurs.

void
(*FPtr_assert)(type *item,
               const uint32_t group,
               const char_t *caption,
               const char_t *detail,
               const char_t *file,
               const uint32_t line);

unref ()

Mark the parameter as non-referenced, disabling the compiler's warnings.

void
unref(param);

1
2
3
4
5

static void i_OnClick(App *app, Event *e)
{
    unref(e);
    app_click_action(app);
}