Commit | Line | Data |
f40a6c71 |
1 | =head1 NAME |
2 | |
3 | perlclib - Internal replacements for standard C library functions |
4 | |
5 | =head1 DESCRIPTION |
6 | |
7 | One thing Perl porters should note is that F<perl> doesn't tend to use that |
8 | much of the C standard library internally; you'll see very little use of, |
9 | for example, the F<ctype.h> functions in there. This is because Perl |
10 | tends to reimplement or abstract standard library functions, so that we |
11 | know exactly how they're going to operate. |
12 | |
13 | This is a reference card for people who are familiar with the C library |
56d22bd2 |
14 | and who want to do things the Perl way; to tell them which functions |
f40a6c71 |
15 | they ought to use instead of the more normal C functions. |
16 | |
17 | =head2 Conventions |
18 | |
19 | In the following tables: |
20 | |
21 | =over 3 |
22 | |
23 | =item C<t> |
24 | |
25 | is a type. |
26 | |
27 | =item C<p> |
28 | |
29 | is a pointer. |
30 | |
31 | =item C<n> |
32 | |
33 | is a number. |
34 | |
35 | =item C<s> |
36 | |
37 | is a string. |
38 | |
39 | =back |
40 | |
41 | C<sv>, C<av>, C<hv>, etc. represent variables of their respective types. |
42 | |
43 | =head2 File Operations |
44 | |
45 | Instead of the F<stdio.h> functions, you should use the Perl abstraction |
46 | layer. Instead of C<FILE*> types, you need to be handling C<PerlIO*> |
56d22bd2 |
47 | types. Don't forget that with the new PerlIO layered I/O abstraction |
f40a6c71 |
48 | C<FILE*> types may not even be available. See also the C<perlapio> |
49 | documentation for more information about the following functions: |
50 | |
51 | Instead Of: Use: |
56d22bd2 |
52 | |
f40a6c71 |
53 | stdin PerlIO_stdin() |
54 | stdout PerlIO_stdout() |
55 | stderr PerlIO_stderr() |
56 | |
57 | fopen(fn, mode) PerlIO_open(fn, mode) |
58 | freopen(fn, mode, stream) PerlIO_reopen(fn, mode, perlio) (Deprecated) |
59 | fflush(stream) PerlIO_flush(perlio) |
60 | fclose(stream) PerlIO_close(perlio) |
61 | |
62 | =head2 File Input and Output |
63 | |
64 | Instead Of: Use: |
65 | |
66 | fprintf(stream, fmt, ...) PerlIO_printf(perlio, fmt, ...) |
67 | |
68 | [f]getc(stream) PerlIO_getc(perlio) |
69 | [f]putc(stream, n) PerlIO_putc(perlio, n) |
70 | ungetc(n, stream) PerlIO_ungetc(perlio, n) |
71 | |
72 | Note that the PerlIO equivalents of C<fread> and C<fwrite> are slightly |
73 | different from their C library counterparts: |
74 | |
75 | fread(p, size, n, stream) PerlIO_read(perlio, buf, numbytes) |
76 | fwrite(p, size, n, stream) PerlIO_write(perlio, buf, numbytes) |
77 | |
78 | fputs(s, stream) PerlIO_puts(perlio, s) |
79 | |
80 | There is no equivalent to C<fgets>; one should use C<sv_gets> instead: |
81 | |
82 | fgets(s, n, stream) sv_gets(sv, perlio, append) |
83 | |
84 | =head2 File Positioning |
85 | |
86 | Instead Of: Use: |
87 | |
88 | feof(stream) PerlIO_eof(perlio) |
89 | fseek(stream, n, whence) PerlIO_seek(perlio, n, whence) |
90 | rewind(stream) PerlIO_rewind(perlio) |
91 | |
92 | fgetpos(stream, p) PerlIO_getpos(perlio, sv) |
93 | fsetpos(stream, p) PerlIO_setpos(perlio, sv) |
94 | |
95 | ferror(stream) PerlIO_error(perlio) |
96 | clearerr(stream) PerlIO_clearerr(perlio) |
97 | |
98 | =head2 Memory Management and String Handling |
99 | |
702eb6d0 |
100 | Instead Of: Use: |
101 | |
102 | t* p = malloc(n) New(id, p, n, t) |
103 | t* p = calloc(n, s) Newz(id, p, n, t) |
104 | p = realloc(p, n) Renew(p, n, t) |
105 | memcpy(dst, src, n) Copy(src, dst, n, t) |
106 | memmove(dst, src, n) Move(src, dst, n, t) |
107 | memcpy/*(struct foo *) StructCopy(src, dst, t) |
108 | memset(dst, 0, n * sizeof(t)) Zero(dst, n, t) |
109 | memzero(dst, 0) Zero(dst, n, char) |
110 | free(p) Safefree(p) |
f40a6c71 |
111 | |
112 | strdup(p) savepv(p) |
113 | strndup(p, n) savepvn(p, n) (Hey, strndup doesn't exist!) |
114 | |
115 | strstr(big, little) instr(big, little) |
116 | strcmp(s1, s2) strLE(s1, s2) / strEQ(s1, s2) / strGT(s1,s2) |
117 | strncmp(s1, s2, n) strnNE(s1, s2, n) / strnEQ(s1, s2, n) |
118 | |
119 | Notice the different order of arguments to C<Copy> and C<Move> than used |
120 | in C<memcpy> and C<memmove>. |
121 | |
122 | Most of the time, though, you'll want to be dealing with SVs internally |
123 | instead of raw C<char *> strings: |
124 | |
125 | strlen(s) sv_len(sv) |
126 | strcpy(dt, src) sv_setpv(sv, s) |
127 | strncpy(dt, src, n) sv_setpvn(sv, s, n) |
128 | strcat(dt, src) sv_catpv(sv, s) |
129 | strncat(dt, src) sv_catpvn(sv, s) |
130 | sprintf(s, fmt, ...) sv_setpvf(sv, fmt, ...) |
131 | |
328bf373 |
132 | Note also the existence of C<sv_catpvf> and C<sv_vcatpvfn>, combining |
f40a6c71 |
133 | concatenation with formatting. |
134 | |
9965345d |
135 | Sometimes instead of zeroing the allocated heap by using Newz() you |
136 | should consider "poisoning" the data. This means writing a bit |
137 | pattern into it that should be illegal as pointers (and floating point |
138 | numbers), and also hopefully surprising enough as integers, so that |
139 | any code attempting to use the data without forethought will break |
140 | sooner rather than later. Poisoning can be done using the Poison() |
141 | macro, which has similar arguments as Zero(): |
142 | |
143 | Poison(dst, n, t) |
144 | |
f40a6c71 |
145 | =head2 Character Class Tests |
146 | |
147 | There are two types of character class tests that Perl implements: one |
148 | type deals in C<char>s and are thus B<not> Unicode aware (and hence |
149 | deprecated unless you B<know> you should use them) and the other type |
150 | deal in C<UV>s and know about Unicode properties. In the following |
151 | table, C<c> is a C<char>, and C<u> is a Unicode codepoint. |
152 | |
153 | Instead Of: Use: But better use: |
154 | |
155 | isalnum(c) isALNUM(c) isALNUM_uni(u) |
156 | isalpha(c) isALPHA(c) isALPHA_uni(u) |
157 | iscntrl(c) isCNTRL(c) isCNTRL_uni(u) |
158 | isdigit(c) isDIGIT(c) isDIGIT_uni(u) |
159 | isgraph(c) isGRAPH(c) isGRAPH_uni(u) |
160 | islower(c) isLOWER(c) isLOWER_uni(u) |
161 | isprint(c) isPRINT(c) isPRINT_uni(u) |
162 | ispunct(c) isPUNCT(c) isPUNCT_uni(u) |
163 | isspace(c) isSPACE(c) isSPACE_uni(u) |
164 | isupper(c) isUPPER(c) isUPPER_uni(u) |
165 | isxdigit(c) isXDIGIT(c) isXDIGIT_uni(u) |
166 | |
167 | tolower(c) toLOWER(c) toLOWER_uni(u) |
168 | toupper(c) toUPPER(c) toUPPER_uni(u) |
169 | |
170 | =head2 F<stdlib.h> functions |
171 | |
172 | Instead Of: Use: |
173 | |
174 | atof(s) Atof(s) |
175 | atol(s) Atol(s) |
56d22bd2 |
176 | strtod(s, *p) Nothing. Just don't use it. |
f40a6c71 |
177 | strtol(s, *p, n) Strtol(s, *p, n) |
178 | strtoul(s, *p, n) Strtoul(s, *p, n) |
179 | |
53305cf1 |
180 | Notice also the C<grok_bin>, C<grok_hex>, and C<grok_oct> functions in |
2826e23d |
181 | F<numeric.c> for converting strings representing numbers in the respective |
f40a6c71 |
182 | bases into C<NV>s. |
183 | |
184 | In theory C<Strtol> and C<Strtoul> may not be defined if the machine perl is |
185 | built on doesn't actually have strtol and strtoul. But as those 2 |
186 | functions are part of the 1989 ANSI C spec we suspect you'll find them |
187 | everywhere by now. |
188 | |
189 | int rand() double Drand01() |
190 | srand(n) { seedDrand01((Rand_seed_t)n); |
191 | PL_srand_called = TRUE; } |
56d22bd2 |
192 | |
f40a6c71 |
193 | exit(n) my_exit(n) |
194 | system(s) Don't. Look at pp_system or use my_popen |
195 | |
196 | getenv(s) PerlEnv_getenv(s) |
197 | setenv(s, val) my_putenv(s, val) |
198 | |
199 | =head2 Miscellaneous functions |
200 | |
201 | You should not even B<want> to use F<setjmp.h> functions, but if you |
202 | think you do, use the C<JMPENV> stack in F<scope.h> instead. |
203 | |
204 | For C<signal>/C<sigaction>, use C<rsignal(signo, handler)>. |
205 | |
206 | =head1 SEE ALSO |
207 | |
208 | C<perlapi>, C<perlapio>, C<perlguts> |
209 | |