curl/src/var.c
Viktor Szakats 193cb00ce9
build: stop overriding standard memory allocation functions
Before this patch curl used the C preprocessor to override standard
memory allocation symbols: malloc, calloc, strdup, realloc, free.
The goal of these is to replace them with curl's debug wrappers in
`CURLDEBUG` builds, another was to replace them with the wrappers
calling user-defined allocators in libcurl. This solution needed a bunch
of workarounds to avoid breaking external headers: it relied on include
order to do the overriding last. For "unity" builds it needed to reset
overrides before external includes. Also in test apps, which are always
built as single source files. It also needed the `(symbol)` trick
to avoid overrides in some places. This would still not fix cases where
the standard symbols were macros. It was also fragile and difficult
to figure out which was the actual function behind an alloc or free call
in a specific piece of code. This in turn caused bugs where the wrong
allocator was accidentally called.

To avoid these problems, this patch replaces this solution with
`curlx_`-prefixed allocator macros, and mapping them _once_ to either
the libcurl wrappers, the debug wrappers or the standard ones, matching
the rest of the code in libtests.

This concludes the long journey to avoid redefining standard functions
in the curl codebase.

Note: I did not update `packages/OS400/*.c` sources. They did not
`#include` `curl_setup.h`, `curl_memory.h` or `memdebug.h`, meaning
the overrides were never applied to them. This may or may not have been
correct. For now I suppressed the direct use of standard allocators
via a local `.checksrc`. Probably they (except for `curlcl.c`) should be
updated to include `curl_setup.h` and use the `curlx_` macros.

This patch changes mappings in two places:
- `lib/curl_threads.c` in libtests: Before this patch it mapped to
  libcurl allocators. After, it maps to standard allocators, like
  the rest of libtests code.
- `units`: before this patch it mapped to standard allocators. After, it
  maps to libcurl allocators.

Also:
- drop all position-dependent `curl_memory.h` and `memdebug.h` includes,
  and delete the now unnecessary headers.
- rename `Curl_tcsdup` macro to `curlx_tcsdup` and define like the other
  allocators.
- map `curlx_strdup()` to `_strdup()` on Windows (was: `strdup()`).
  To fix warnings silenced via `_CRT_NONSTDC_NO_DEPRECATE`.
- multibyte: map `curlx_convert_*()` to `_strdup()` on Windows
  (was: `strdup()`).
- src: do not reuse the `strdup` name for the local replacement.
- lib509: call `_strdup()` on Windows (was: `strdup()`).
- test1132: delete test obsoleted by this patch.
- CHECKSRC.md: update text for `SNPRINTF`.
- checksrc: ban standard allocator symbols.

Follow-up to b12da22db1 #18866
Follow-up to db98daab05 #18844
Follow-up to 4deea9396b #18814
Follow-up to 9678ff5b1b #18776
Follow-up to 10bac43b87 #18774
Follow-up to 20142f5d06 #18634
Follow-up to bf7375ecc5 #18503
Follow-up to 9863599d69 #18502
Follow-up to 3bb5e58c10 #17827

Closes #19626
2025-11-28 10:44:26 +01:00

501 lines
14 KiB
C

/***************************************************************************
* _ _ ____ _
* Project ___| | | | _ \| |
* / __| | | | |_) | |
* | (__| |_| | _ <| |___
* \___|\___/|_| \_\_____|
*
* Copyright (C) Daniel Stenberg, <daniel@haxx.se>, et al.
*
* This software is licensed as described in the file COPYING, which
* you should have received as part of this distribution. The terms
* are also available at https://curl.se/docs/copyright.html.
*
* You may opt to use, copy, modify, merge, publish, distribute and/or sell
* copies of the Software, and permit persons to whom the Software is
* furnished to do so, under the terms of the COPYING file.
*
* This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
* KIND, either express or implied.
*
* SPDX-License-Identifier: curl
*
***************************************************************************/
#include "tool_setup.h"
#include "tool_cfgable.h"
#include "tool_getparam.h"
#include "tool_helpers.h"
#include "tool_findfile.h"
#include "tool_msgs.h"
#include "tool_parsecfg.h"
#include "tool_paramhlp.h"
#include "tool_writeout_json.h"
#include "tool_strdup.h"
#include "var.h"
#define MAX_EXPAND_CONTENT 10000000
#define MAX_VAR_LEN 128 /* max length of a name */
/* free everything */
void varcleanup(void)
{
struct tool_var *list = global->variables;
while(list) {
struct tool_var *t = list;
list = list->next;
curlx_free(CURL_UNCONST(t->content));
curlx_free(t);
}
}
static const struct tool_var *varcontent(const char *name, size_t nlen)
{
struct tool_var *list = global->variables;
while(list) {
if((strlen(list->name) == nlen) &&
!strncmp(name, list->name, nlen)) {
return list;
}
list = list->next;
}
return NULL;
}
#define ENDOFFUNC(x) (((x) == '}') || ((x) == ':'))
#define FUNCMATCH(ptr,name,len) \
(!strncmp(ptr, name, len) && ENDOFFUNC(ptr[len]))
#define FUNC_TRIM "trim"
#define FUNC_TRIM_LEN (sizeof(FUNC_TRIM) - 1)
#define FUNC_JSON "json"
#define FUNC_JSON_LEN (sizeof(FUNC_JSON) - 1)
#define FUNC_URL "url"
#define FUNC_URL_LEN (sizeof(FUNC_URL) - 1)
#define FUNC_B64 "b64"
#define FUNC_B64_LEN (sizeof(FUNC_B64) - 1)
#define FUNC_64DEC "64dec" /* base64 decode */
#define FUNC_64DEC_LEN (sizeof(FUNC_64DEC) - 1)
static ParameterError varfunc(char *c, /* content */
size_t clen, /* content length */
char *f, /* functions */
size_t flen, /* function string length */
struct dynbuf *out)
{
bool alloc = FALSE;
ParameterError err = PARAM_OK;
const char *finput = f;
/* The functions are independent and runs left to right */
while(*f && !err) {
if(*f == '}')
/* end of functions */
break;
/* On entry, this is known to be a colon already. In subsequent laps, it
is also known to be a colon since that is part of the FUNCMATCH()
checks */
f++;
if(FUNCMATCH(f, FUNC_TRIM, FUNC_TRIM_LEN)) {
size_t len = clen;
f += FUNC_TRIM_LEN;
if(clen) {
/* skip leading white space, including CRLF */
while(ISSPACE(*c)) {
c++;
len--;
}
while(len && ISSPACE(c[len-1]))
len--;
}
/* put it in the output */
curlx_dyn_reset(out);
if(curlx_dyn_addn(out, c, len)) {
err = PARAM_NO_MEM;
break;
}
}
else if(FUNCMATCH(f, FUNC_JSON, FUNC_JSON_LEN)) {
f += FUNC_JSON_LEN;
curlx_dyn_reset(out);
if(clen) {
if(jsonquoted(c, clen, out, FALSE)) {
err = PARAM_NO_MEM;
break;
}
}
}
else if(FUNCMATCH(f, FUNC_URL, FUNC_URL_LEN)) {
f += FUNC_URL_LEN;
curlx_dyn_reset(out);
if(clen) {
char *enc = curl_easy_escape(NULL, c, (int)clen);
if(!enc) {
err = PARAM_NO_MEM;
break;
}
/* put it in the output */
if(curlx_dyn_add(out, enc))
err = PARAM_NO_MEM;
curl_free(enc);
if(err)
break;
}
}
else if(FUNCMATCH(f, FUNC_B64, FUNC_B64_LEN)) {
f += FUNC_B64_LEN;
curlx_dyn_reset(out);
if(clen) {
char *enc;
size_t elen;
CURLcode result = curlx_base64_encode((uint8_t *)c, clen, &enc, &elen);
if(result) {
err = PARAM_NO_MEM;
break;
}
/* put it in the output */
if(curlx_dyn_addn(out, enc, elen))
err = PARAM_NO_MEM;
curl_free(enc);
if(err)
break;
}
}
else if(FUNCMATCH(f, FUNC_64DEC, FUNC_64DEC_LEN)) {
f += FUNC_64DEC_LEN;
curlx_dyn_reset(out);
if(clen) {
unsigned char *enc;
size_t elen;
CURLcode result = curlx_base64_decode(c, &enc, &elen);
/* put it in the output */
if(result) {
if(curlx_dyn_add(out, "[64dec-fail]"))
err = PARAM_NO_MEM;
}
else {
if(curlx_dyn_addn(out, enc, elen))
err = PARAM_NO_MEM;
curl_free(enc);
}
if(err)
break;
}
}
else {
/* unsupported function */
errorf("unknown variable function in '%.*s'", (int)flen, finput);
err = PARAM_EXPAND_ERROR;
break;
}
if(alloc)
curlx_free(c);
clen = curlx_dyn_len(out);
c = memdup0(curlx_dyn_ptr(out), clen);
if(!c) {
err = PARAM_NO_MEM;
break;
}
alloc = TRUE;
}
if(alloc)
curlx_free(c);
if(err)
curlx_dyn_free(out);
return err;
}
ParameterError varexpand(const char *line, struct dynbuf *out,
bool *replaced)
{
CURLcode result;
char *envp;
bool added = FALSE;
const char *input = line;
*replaced = FALSE;
curlx_dyn_init(out, MAX_EXPAND_CONTENT);
do {
envp = strstr(line, "{{");
if((envp > line) && envp[-1] == '\\') {
/* preceding backslash, we want this verbatim */
/* insert the text up to this point, minus the backslash */
result = curlx_dyn_addn(out, line, envp - line - 1);
if(result)
return PARAM_NO_MEM;
/* output '{{' then continue from here */
result = curlx_dyn_addn(out, "{{", 2);
if(result)
return PARAM_NO_MEM;
line = &envp[2];
}
else if(envp) {
char name[MAX_VAR_LEN];
size_t nlen;
size_t i;
char *funcp;
char *clp = strstr(envp, "}}");
size_t prefix;
if(!clp) {
/* uneven braces */
warnf("missing close '}}' in '%s'", input);
break;
}
prefix = 2;
envp += 2; /* move over the {{ */
/* if there is a function, it ends the name with a colon */
funcp = memchr(envp, ':', clp - envp);
if(funcp)
nlen = funcp - envp;
else
nlen = clp - envp;
if(!nlen || (nlen >= sizeof(name))) {
warnf("bad variable name length '%s'", input);
/* insert the text as-is since this is not an env variable */
result = curlx_dyn_addn(out, line, clp - line + prefix);
if(result)
return PARAM_NO_MEM;
}
else {
/* insert the text up to this point */
result = curlx_dyn_addn(out, line, envp - prefix - line);
if(result)
return PARAM_NO_MEM;
/* copy the name to separate buffer */
memcpy(name, envp, nlen);
name[nlen] = 0;
/* verify that the name looks sensible */
for(i = 0; (i < nlen) &&
(ISALNUM(name[i]) || (name[i] == '_')); i++);
if(i != nlen) {
warnf("bad variable name: %s", name);
/* insert the text as-is since this is not an env variable */
result = curlx_dyn_addn(out, envp - prefix,
clp - envp + prefix + 2);
if(result)
return PARAM_NO_MEM;
}
else {
char *value;
size_t vlen = 0;
struct dynbuf buf;
const struct tool_var *v = varcontent(name, nlen);
if(v) {
value = (char *)CURL_UNCONST(v->content);
vlen = v->clen;
}
else
value = NULL;
curlx_dyn_init(&buf, MAX_EXPAND_CONTENT);
if(funcp) {
/* apply the list of functions on the value */
size_t flen = clp - funcp;
ParameterError err = varfunc(value, vlen, funcp, flen, &buf);
if(err)
return err;
value = curlx_dyn_ptr(&buf);
vlen = curlx_dyn_len(&buf);
}
if(value && vlen > 0) {
/* A variable might contain null bytes. Such bytes cannot be shown
using normal means, this is an error. */
char *nb = memchr(value, '\0', vlen);
if(nb) {
errorf("variable contains null byte");
return PARAM_EXPAND_ERROR;
}
}
/* insert the value */
result = curlx_dyn_addn(out, value, vlen);
curlx_dyn_free(&buf);
if(result)
return PARAM_NO_MEM;
added = true;
}
}
line = &clp[2];
}
} while(envp);
if(added && *line) {
/* add the "suffix" as well */
result = curlx_dyn_add(out, line);
if(result)
return PARAM_NO_MEM;
}
*replaced = added;
if(!added)
curlx_dyn_free(out);
return PARAM_OK;
}
/*
* Created in a way that is not revealing how variables are actually stored so
* that we can improve this if we want better performance when managing many
* at a later point.
*/
static ParameterError addvariable(const char *name,
size_t nlen,
const char *content,
size_t clen,
bool contalloc)
{
struct tool_var *p;
const struct tool_var *check = varcontent(name, nlen);
DEBUGASSERT(nlen);
if(check)
notef("Overwriting variable '%s'", check->name);
p = curlx_calloc(1, sizeof(struct tool_var) + nlen);
if(p) {
memcpy(p->name, name, nlen);
/* the null termination byte is already present from above */
p->content = contalloc ? content : memdup0(content, clen);
if(p->content) {
p->clen = clen;
p->next = global->variables;
global->variables = p;
return PARAM_OK;
}
curlx_free(p);
}
return PARAM_NO_MEM;
}
#define MAX_FILENAME 10000
ParameterError setvariable(const char *input)
{
const char *name;
size_t nlen;
char *content = NULL;
size_t clen = 0;
bool contalloc = FALSE;
const char *line = input;
ParameterError err = PARAM_OK;
bool import = FALSE;
char *ge = NULL;
char buf[MAX_VAR_LEN];
curl_off_t startoffset = 0;
curl_off_t endoffset = CURL_OFF_T_MAX;
if(*input == '%') {
import = TRUE;
line++;
}
name = line;
while(*line && (ISALNUM(*line) || (*line == '_')))
line++;
nlen = line - name;
if(!nlen || (nlen >= MAX_VAR_LEN)) {
warnf("Bad variable name length (%zd), skipping", nlen);
return PARAM_OK;
}
if(import) {
/* this does not use curl_getenv() because we want "" support for blank
content */
if(*line) {
/* if there is a default action, we need to copy the name */
memcpy(buf, name, nlen);
buf[nlen] = 0;
name = buf;
}
ge = getenv(name);
if(!*line && !ge) {
/* no assign, no variable, fail */
errorf("Variable '%s' import fail, not set", name);
return PARAM_EXPAND_ERROR;
}
else if(ge) {
/* there is a value to use */
content = ge;
clen = strlen(ge);
}
}
if(*line == '[' && ISDIGIT(line[1])) {
/* is there a byte range specified? [num-num] */
line++;
if(curlx_str_number(&line, &startoffset, CURL_OFF_T_MAX) ||
curlx_str_single(&line, '-'))
return PARAM_VAR_SYNTAX;
if(curlx_str_single(&line, ']')) {
if(curlx_str_number(&line, &endoffset, CURL_OFF_T_MAX) ||
curlx_str_single(&line, ']'))
return PARAM_VAR_SYNTAX;
}
if(startoffset > endoffset)
return PARAM_VAR_SYNTAX;
}
if(content)
;
else if(*line == '@') {
/* read from file or stdin */
FILE *file;
bool use_stdin;
line++;
use_stdin = !strcmp(line, "-");
if(use_stdin)
file = stdin;
else {
file = curlx_fopen(line, "rb");
if(!file) {
char errbuf[STRERROR_LEN];
errorf("Failed to open %s: %s", line,
curlx_strerror(errno, errbuf, sizeof(errbuf)));
err = PARAM_READ_ERROR;
}
}
if(!err) {
err = file2memory_range(&content, &clen, file, startoffset, endoffset);
/* in case of out of memory, this should fail the entire operation */
if(clen)
contalloc = TRUE;
}
if(!use_stdin && file)
curlx_fclose(file);
if(err)
return err;
}
else if(*line == '=') {
line++;
clen = strlen(line);
/* this is the exact content */
content = (char *)CURL_UNCONST(line);
if(startoffset || (endoffset != CURL_OFF_T_MAX)) {
if(startoffset >= (curl_off_t)clen)
clen = 0;
else {
/* make the end offset no larger than the last byte */
if(endoffset >= (curl_off_t)clen)
endoffset = clen - 1;
clen = (size_t)(endoffset - startoffset) + 1;
content += startoffset;
}
}
}
else {
warnf("Bad --variable syntax, skipping: %s", input);
return PARAM_OK;
}
err = addvariable(name, nlen, content, clen, contalloc);
if(err) {
if(contalloc)
curlx_free(content);
}
return err;
}