mirror of
https://github.com/Zenithsiz/ftmemsim-valgrind.git
synced 2026-02-06 19:54:18 +00:00
696a45f8c9
Explicitly check that the symbol starts with "_Z" to only demangle modern style (gnu_v3) mangled C++ (and Rust) symbols. git-svn-id: svn://svn.valgrind.org/valgrind/trunk@16118
705 lines
22 KiB
C
705 lines
22 KiB
C
|
|
/*--------------------------------------------------------------------*/
|
|
/*--- Demangling of C++ mangled names. demangle.c ---*/
|
|
/*--------------------------------------------------------------------*/
|
|
|
|
/*
|
|
This file is part of Valgrind, a dynamic binary instrumentation
|
|
framework.
|
|
|
|
Copyright (C) 2000-2015 Julian Seward
|
|
jseward@acm.org
|
|
|
|
Rust demangler components are
|
|
Copyright (C) 2016-2016 David Tolnay
|
|
dtolnay@gmail.com
|
|
|
|
This program is free software; you can redistribute it and/or
|
|
modify it under the terms of the GNU General Public License as
|
|
published by the Free Software Foundation; either version 2 of the
|
|
License, or (at your option) any later version.
|
|
|
|
This program is distributed in the hope that it will be useful, but
|
|
WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
|
General Public License for more details.
|
|
|
|
You should have received a copy of the GNU General Public License
|
|
along with this program; if not, write to the Free Software
|
|
Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
|
|
02111-1307, USA.
|
|
|
|
The GNU General Public License is contained in the file COPYING.
|
|
*/
|
|
|
|
#include "pub_core_basics.h"
|
|
#include "pub_core_demangle.h"
|
|
#include "pub_core_libcassert.h"
|
|
#include "pub_core_libcbase.h"
|
|
#include "pub_core_libcprint.h"
|
|
#include "pub_core_mallocfree.h"
|
|
#include "pub_core_options.h"
|
|
|
|
#include "vg_libciface.h"
|
|
#include "demangle.h"
|
|
|
|
/* fwds */
|
|
static Bool rust_is_mangled ( const HChar* );
|
|
static void rust_demangle_sym ( HChar* );
|
|
|
|
|
|
/*------------------------------------------------------------*/
|
|
/*--- ---*/
|
|
/*------------------------------------------------------------*/
|
|
|
|
/* The demangler's job is to take a raw symbol name and turn it into
|
|
something a Human Bean can understand. Our mangling model
|
|
comprises a three stage pipeline. Mangling pushes names forward
|
|
through the pipeline (0, then 1, then 2) and demangling is
|
|
obviously the reverse. In practice it is highly unlikely that a
|
|
name would require all stages, but it is not impossible either.
|
|
|
|
0. If we're working with Rust, Rust names are lightly mangled by
|
|
the Rust front end.
|
|
|
|
1. Then the name is subject to standard C++ mangling.
|
|
|
|
2. Optionally, in relatively rare cases, the resulting name is then
|
|
itself encoded using Z-escaping (see pub_core_redir.h) so as to
|
|
become part of a redirect-specification.
|
|
|
|
Therefore, VG_(demangle) first tries to undo (2). If successful,
|
|
the soname part is discarded (humans don't want to see that).
|
|
Then, it tries to undo (1) (using demangling code from GNU/FSF) and
|
|
finally it tries to undo (0).
|
|
|
|
Finally, it changes the name of all symbols which are known to be
|
|
functions below main() to "(below main)". This helps reduce
|
|
variability of stack traces, something which has been a problem for
|
|
the testsuite for a long time.
|
|
|
|
--------
|
|
If do_cxx_demangle == True, it does all the above stages:
|
|
- undo (2) [Z-encoding]
|
|
- undo (1) [C++ mangling]
|
|
- if (1) succeeds, undo (0) [Rust mangling]
|
|
- do the below-main hack
|
|
|
|
Rust demangling (0) is only done if C++ demangling (1) succeeds
|
|
because Rust demangling is performed in-place, and it is difficult
|
|
to prove that we "own" the storage -- hence, that the in-place
|
|
operation is safe -- unless it is clear that it has come from the
|
|
C++ demangler, which returns its output in a heap-allocated buffer
|
|
which we can be sure we own. In practice (Nov 2016) this does not
|
|
seem to be a problem, since the Rust compiler appears to apply C++
|
|
mangling after Rust mangling, so we never encounter symbols that
|
|
require Rust demangling but not C++ demangling.
|
|
|
|
If do_cxx_demangle == False, the C++ and Rust stags are skipped:
|
|
- undo (2) [Z-encoding]
|
|
- do the below-main hack
|
|
*/
|
|
|
|
/* Note that the C++ demangler is from GNU libiberty and is almost
|
|
completely unmodified. We use vg_libciface.h as a way to
|
|
impedance-match the libiberty code into our own framework.
|
|
|
|
The libiberty code included here was taken from the GCC repository
|
|
and is released under the LGPL 2.1 license, which AFAICT is compatible
|
|
with "GPL 2 or later" and so is OK for inclusion in Valgrind.
|
|
|
|
To update to a newer libiberty, use the "update-demangler" script
|
|
which is included in the valgrind repository. */
|
|
|
|
/* This is the main, standard demangler entry point. */
|
|
|
|
/* Upon return, *RESULT will point to the demangled name.
|
|
The memory buffer that holds the demangled name is allocated on the
|
|
heap and will be deallocated in the next invocation. Conceptually,
|
|
that buffer is owned by VG_(demangle). That means two things:
|
|
(1) Users of VG_(demangle) must not free that buffer.
|
|
(2) If the demangled name needs to be stashed away for later use,
|
|
the contents of the buffer need to be copied. It is not sufficient
|
|
to just store the pointer as it will point to deallocated memory
|
|
after the next VG_(demangle) invocation. */
|
|
void VG_(demangle) ( Bool do_cxx_demangling, Bool do_z_demangling,
|
|
/* IN */ const HChar *orig,
|
|
/* OUT */ const HChar **result )
|
|
{
|
|
/* Possibly undo (2) */
|
|
/* Z-Demangling was requested.
|
|
The fastest way to see if it's a Z-mangled name is just to attempt
|
|
to Z-demangle it (with NULL for the soname buffer, since we're not
|
|
interested in that). */
|
|
if (do_z_demangling) {
|
|
const HChar *z_demangled;
|
|
|
|
if (VG_(maybe_Z_demangle)( orig, NULL, /*soname*/
|
|
&z_demangled, NULL, NULL, NULL )) {
|
|
orig = z_demangled;
|
|
}
|
|
}
|
|
|
|
/* Possibly undo (1) */
|
|
if (do_cxx_demangling && VG_(clo_demangle)
|
|
&& orig != NULL && orig[0] == '_' && orig[1] == 'Z') {
|
|
/* !!! vvv STATIC vvv !!! */
|
|
static HChar* demangled = NULL;
|
|
/* !!! ^^^ STATIC ^^^ !!! */
|
|
|
|
/* Free up previously demangled name */
|
|
if (demangled) {
|
|
VG_(arena_free) (VG_AR_DEMANGLE, demangled);
|
|
demangled = NULL;
|
|
}
|
|
demangled = ML_(cplus_demangle) ( orig, DMGL_ANSI | DMGL_PARAMS );
|
|
|
|
*result = (demangled == NULL) ? orig : demangled;
|
|
|
|
if (demangled) {
|
|
/* Possibly undo (0). This is the only place where it is
|
|
safe, from a storage management perspective, to
|
|
Rust-demangle the symbol. That's because Rust demangling
|
|
happens in place, so we need to be sure that the storage
|
|
it is happening in is actually owned by us, and non-const.
|
|
In this case, the value returned by ML_(cplus_demangle)
|
|
does have that property. */
|
|
if (rust_is_mangled(demangled)) {
|
|
rust_demangle_sym(demangled);
|
|
}
|
|
*result = demangled;
|
|
} else {
|
|
*result = orig;
|
|
}
|
|
|
|
} else {
|
|
*result = orig;
|
|
}
|
|
|
|
// 13 Mar 2005: We used to check here that the demangler wasn't leaking
|
|
// by calling the (now-removed) function VG_(is_empty_arena)(). But,
|
|
// very rarely (ie. I've heard of it twice in 3 years), the demangler
|
|
// does leak. But, we can't do much about it, and it's not a disaster,
|
|
// so we just let it slide without aborting or telling the user.
|
|
}
|
|
|
|
|
|
/*------------------------------------------------------------*/
|
|
/*--- DEMANGLE Z-ENCODED NAMES ---*/
|
|
/*------------------------------------------------------------*/
|
|
|
|
/* Demangle a Z-encoded name as described in pub_tool_redir.h.
|
|
Z-encoded names are used by Valgrind for doing function
|
|
interception/wrapping.
|
|
|
|
Demangle 'sym' into its soname and fnname parts, putting them in
|
|
the specified buffers. Returns a Bool indicating whether the
|
|
demangled failed or not. A failure can occur because the prefix
|
|
isn't recognised, the internal Z-escaping is wrong, or because one
|
|
or the other (or both) of the output buffers becomes full. Passing
|
|
'so' as NULL is acceptable if the caller is only interested in the
|
|
function name part. */
|
|
|
|
Bool VG_(maybe_Z_demangle) ( const HChar* sym,
|
|
/*OUT*/const HChar** so,
|
|
/*OUT*/const HChar** fn,
|
|
/*OUT*/Bool* isWrap,
|
|
/*OUT*/Int* eclassTag,
|
|
/*OUT*/Int* eclassPrio )
|
|
{
|
|
static HChar *sobuf;
|
|
static HChar *fnbuf;
|
|
static SizeT buf_len = 0;
|
|
|
|
/* The length of the name after undoing Z-encoding is always smaller
|
|
than the mangled name. Making the soname and fnname buffers as large
|
|
as the demangled name is therefore always safe and overflow can never
|
|
occur. */
|
|
SizeT len = VG_(strlen)(sym) + 1;
|
|
|
|
if (buf_len < len) {
|
|
sobuf = VG_(arena_realloc)(VG_AR_DEMANGLE, "Z-demangle", sobuf, len);
|
|
fnbuf = VG_(arena_realloc)(VG_AR_DEMANGLE, "Z-demangle", fnbuf, len);
|
|
buf_len = len;
|
|
}
|
|
sobuf[0] = fnbuf[0] = '\0';
|
|
|
|
if (so)
|
|
*so = sobuf;
|
|
*fn = fnbuf;
|
|
|
|
# define EMITSO(ch) \
|
|
do { \
|
|
if (so) { \
|
|
sobuf[soi++] = ch; sobuf[soi] = 0; \
|
|
} \
|
|
} while (0)
|
|
# define EMITFN(ch) \
|
|
do { \
|
|
fnbuf[fni++] = ch; fnbuf[fni] = 0; \
|
|
} while (0)
|
|
|
|
Bool error, valid, fn_is_encoded, is_VG_Z_prefixed;
|
|
Int soi, fni, i;
|
|
|
|
error = False;
|
|
soi = 0;
|
|
fni = 0;
|
|
|
|
valid = sym[0] == '_'
|
|
&& sym[1] == 'v'
|
|
&& sym[2] == 'g'
|
|
&& (sym[3] == 'r' || sym[3] == 'w')
|
|
&& VG_(isdigit)(sym[4])
|
|
&& VG_(isdigit)(sym[5])
|
|
&& VG_(isdigit)(sym[6])
|
|
&& VG_(isdigit)(sym[7])
|
|
&& VG_(isdigit)(sym[8])
|
|
&& sym[9] == 'Z'
|
|
&& (sym[10] == 'Z' || sym[10] == 'U')
|
|
&& sym[11] == '_';
|
|
|
|
if (valid
|
|
&& sym[4] == '0' && sym[5] == '0' && sym[6] == '0' && sym[7] == '0'
|
|
&& sym[8] != '0') {
|
|
/* If the eclass tag is 0000 (meaning "no eclass"), the priority
|
|
must be 0 too. */
|
|
valid = False;
|
|
}
|
|
|
|
if (!valid)
|
|
return False;
|
|
|
|
fn_is_encoded = sym[10] == 'Z';
|
|
|
|
if (isWrap)
|
|
*isWrap = sym[3] == 'w';
|
|
|
|
if (eclassTag) {
|
|
*eclassTag = 1000 * ((Int)sym[4] - '0')
|
|
+ 100 * ((Int)sym[5] - '0')
|
|
+ 10 * ((Int)sym[6] - '0')
|
|
+ 1 * ((Int)sym[7] - '0');
|
|
vg_assert(*eclassTag >= 0 && *eclassTag <= 9999);
|
|
}
|
|
|
|
if (eclassPrio) {
|
|
*eclassPrio = ((Int)sym[8]) - '0';
|
|
vg_assert(*eclassPrio >= 0 && *eclassPrio <= 9);
|
|
}
|
|
|
|
/* Now check the soname prefix isn't "VG_Z_", as described in
|
|
pub_tool_redir.h. */
|
|
is_VG_Z_prefixed =
|
|
sym[12] == 'V' &&
|
|
sym[13] == 'G' &&
|
|
sym[14] == '_' &&
|
|
sym[15] == 'Z' &&
|
|
sym[16] == '_';
|
|
if (is_VG_Z_prefixed) {
|
|
vg_assert2(0, "symbol with a 'VG_Z_' prefix: %s.\n"
|
|
"see pub_tool_redir.h for an explanation.", sym);
|
|
}
|
|
|
|
/* Now scan the Z-encoded soname. */
|
|
i = 12;
|
|
while (True) {
|
|
|
|
if (sym[i] == '_')
|
|
/* Found the delimiter. Move on to the fnname loop. */
|
|
break;
|
|
|
|
if (sym[i] == 0) {
|
|
error = True;
|
|
goto out;
|
|
}
|
|
|
|
if (sym[i] != 'Z') {
|
|
EMITSO(sym[i]);
|
|
i++;
|
|
continue;
|
|
}
|
|
|
|
/* We've got a Z-escape. */
|
|
i++;
|
|
switch (sym[i]) {
|
|
case 'a': EMITSO('*'); break;
|
|
case 'c': EMITSO(':'); break;
|
|
case 'd': EMITSO('.'); break;
|
|
case 'h': EMITSO('-'); break;
|
|
case 'p': EMITSO('+'); break;
|
|
case 's': EMITSO(' '); break;
|
|
case 'u': EMITSO('_'); break;
|
|
case 'A': EMITSO('@'); break;
|
|
case 'D': EMITSO('$'); break;
|
|
case 'L': EMITSO('('); break;
|
|
case 'P': EMITSO('%'); break;
|
|
case 'R': EMITSO(')'); break;
|
|
case 'S': EMITSO('/'); break;
|
|
case 'Z': EMITSO('Z'); break;
|
|
default: error = True; goto out;
|
|
}
|
|
i++;
|
|
}
|
|
|
|
vg_assert(sym[i] == '_');
|
|
i++;
|
|
|
|
/* Now deal with the function name part. */
|
|
if (!fn_is_encoded) {
|
|
|
|
/* simple; just copy. */
|
|
while (True) {
|
|
if (sym[i] == 0)
|
|
break;
|
|
EMITFN(sym[i]);
|
|
i++;
|
|
}
|
|
goto out;
|
|
|
|
}
|
|
|
|
/* else use a Z-decoding loop like with soname */
|
|
while (True) {
|
|
|
|
if (sym[i] == 0)
|
|
break;
|
|
|
|
if (sym[i] != 'Z') {
|
|
EMITFN(sym[i]);
|
|
i++;
|
|
continue;
|
|
}
|
|
|
|
/* We've got a Z-escape. */
|
|
i++;
|
|
switch (sym[i]) {
|
|
case 'a': EMITFN('*'); break;
|
|
case 'c': EMITFN(':'); break;
|
|
case 'd': EMITFN('.'); break;
|
|
case 'h': EMITFN('-'); break;
|
|
case 'p': EMITFN('+'); break;
|
|
case 's': EMITFN(' '); break;
|
|
case 'u': EMITFN('_'); break;
|
|
case 'A': EMITFN('@'); break;
|
|
case 'D': EMITFN('$'); break;
|
|
case 'L': EMITFN('('); break;
|
|
case 'P': EMITFN('%'); break;
|
|
case 'R': EMITFN(')'); break;
|
|
case 'S': EMITFN('/'); break;
|
|
case 'Z': EMITFN('Z'); break;
|
|
default: error = True; goto out;
|
|
}
|
|
i++;
|
|
}
|
|
|
|
out:
|
|
EMITSO(0);
|
|
EMITFN(0);
|
|
|
|
if (error) {
|
|
/* Something's wrong. Give up. */
|
|
VG_(message)(Vg_UserMsg,
|
|
"m_demangle: error Z-demangling: %s\n", sym);
|
|
return False;
|
|
}
|
|
|
|
return True;
|
|
}
|
|
|
|
|
|
/*------------------------------------------------------------*/
|
|
/*--- DEMANGLE RUST NAMES ---*/
|
|
/*------------------------------------------------------------*/
|
|
|
|
/*
|
|
* Mangled Rust symbols look like this:
|
|
*
|
|
* _$LT$std..sys..fd..FileDesc$u20$as$u20$core..ops..Drop$GT$::drop::hc68340e1baa4987a
|
|
*
|
|
* The original symbol is:
|
|
*
|
|
* <std::sys::fd::FileDesc as core::ops::Drop>::drop
|
|
*
|
|
* The last component of the path is a 64-bit hash in lowercase hex, prefixed
|
|
* with "h". Rust does not have a global namespace between crates, an illusion
|
|
* which Rust maintains by using the hash to distinguish things that would
|
|
* otherwise have the same symbol.
|
|
*
|
|
* Any path component not starting with a XID_Start character is prefixed with
|
|
* "_".
|
|
*
|
|
* The following escape sequences are used:
|
|
*
|
|
* "," => $C$
|
|
* "@" => $SP$
|
|
* "*" => $BP$
|
|
* "&" => $RF$
|
|
* "<" => $LT$
|
|
* ">" => $GT$
|
|
* "(" => $LP$
|
|
* ")" => $RP$
|
|
* " " => $u20$
|
|
* "\"" => $u22$
|
|
* "'" => $u27$
|
|
* "+" => $u2b$
|
|
* ";" => $u3b$
|
|
* "[" => $u5b$
|
|
* "]" => $u5d$
|
|
* "{" => $u7b$
|
|
* "}" => $u7d$
|
|
* "~" => $u7e$
|
|
*
|
|
* A double ".." means "::" and a single "." means "-".
|
|
*
|
|
* The only characters allowed in the mangled symbol are a-zA-Z0-9 and _.:$
|
|
*/
|
|
|
|
static const HChar *hash_prefix = "::h";
|
|
static const SizeT hash_prefix_len = 3;
|
|
static const SizeT hash_len = 16;
|
|
|
|
static Bool is_prefixed_hash(const HChar *start);
|
|
static Bool looks_like_rust(const HChar *sym, SizeT len);
|
|
static Bool unescape(const HChar **in, HChar **out,
|
|
const HChar *seq, HChar value);
|
|
|
|
/*
|
|
* INPUT:
|
|
* sym: symbol that has been through BFD-demangling
|
|
*
|
|
* This function looks for the following indicators:
|
|
*
|
|
* 1. The hash must consist of "h" followed by 16 lowercase hex digits.
|
|
*
|
|
* 2. As a sanity check, the hash must use between 5 and 15 of the 16 possible
|
|
* hex digits. This is true of 99.9998% of hashes so once in your life you
|
|
* may see a false negative. The point is to notice path components that
|
|
* could be Rust hashes but are probably not, like "haaaaaaaaaaaaaaaa". In
|
|
* this case a false positive (non-Rust symbol has an important path
|
|
* component removed because it looks like a Rust hash) is worse than a
|
|
* false negative (the rare Rust symbol is not demangled) so this sets the
|
|
* balance in favor of false negatives.
|
|
*
|
|
* 3. There must be no characters other than a-zA-Z0-9 and _.:$
|
|
*
|
|
* 4. There must be no unrecognized $-sign sequences.
|
|
*
|
|
* 5. There must be no sequence of three or more dots in a row ("...").
|
|
*/
|
|
static Bool rust_is_mangled(const HChar *sym)
|
|
{
|
|
SizeT len, len_without_hash;
|
|
|
|
if (!sym)
|
|
return False;
|
|
|
|
len = VG_(strlen)(sym);
|
|
if (len <= hash_prefix_len + hash_len)
|
|
/* Not long enough to contain "::h" + hash + something else */
|
|
return False;
|
|
|
|
len_without_hash = len - (hash_prefix_len + hash_len);
|
|
if (!is_prefixed_hash(sym + len_without_hash))
|
|
return False;
|
|
|
|
return looks_like_rust(sym, len_without_hash);
|
|
}
|
|
|
|
/*
|
|
* A hash is the prefix "::h" followed by 16 lowercase hex digits. The hex
|
|
* digits must comprise between 5 and 15 (inclusive) distinct digits.
|
|
*/
|
|
static Bool is_prefixed_hash(const HChar *str)
|
|
{
|
|
const HChar *end;
|
|
Bool seen[16];
|
|
SizeT i;
|
|
Int count;
|
|
|
|
if (VG_(strncmp)(str, hash_prefix, hash_prefix_len))
|
|
return False;
|
|
str += hash_prefix_len;
|
|
|
|
VG_(memset)(seen, False, sizeof(seen));
|
|
for (end = str + hash_len; str < end; str++)
|
|
if (*str >= '0' && *str <= '9')
|
|
seen[*str - '0'] = True;
|
|
else if (*str >= 'a' && *str <= 'f')
|
|
seen[*str - 'a' + 10] = True;
|
|
else
|
|
return False;
|
|
|
|
/* Count how many distinct digits seen */
|
|
count = 0;
|
|
for (i = 0; i < 16; i++)
|
|
if (seen[i])
|
|
count++;
|
|
|
|
return count >= 5 && count <= 15;
|
|
}
|
|
|
|
static Bool looks_like_rust(const HChar *str, SizeT len)
|
|
{
|
|
const HChar *end = str + len;
|
|
|
|
while (str < end) {
|
|
switch (*str) {
|
|
case '$':
|
|
if (!VG_(strncmp)(str, "$C$", 3))
|
|
str += 3;
|
|
else if (!VG_(strncmp)(str, "$SP$", 4)
|
|
|| !VG_(strncmp)(str, "$BP$", 4)
|
|
|| !VG_(strncmp)(str, "$RF$", 4)
|
|
|| !VG_(strncmp)(str, "$LT$", 4)
|
|
|| !VG_(strncmp)(str, "$GT$", 4)
|
|
|| !VG_(strncmp)(str, "$LP$", 4)
|
|
|| !VG_(strncmp)(str, "$RP$", 4))
|
|
str += 4;
|
|
else if (!VG_(strncmp)(str, "$u20$", 5)
|
|
|| !VG_(strncmp)(str, "$u22$", 5)
|
|
|| !VG_(strncmp)(str, "$u27$", 5)
|
|
|| !VG_(strncmp)(str, "$u2b$", 5)
|
|
|| !VG_(strncmp)(str, "$u3b$", 5)
|
|
|| !VG_(strncmp)(str, "$u5b$", 5)
|
|
|| !VG_(strncmp)(str, "$u5d$", 5)
|
|
|| !VG_(strncmp)(str, "$u7b$", 5)
|
|
|| !VG_(strncmp)(str, "$u7d$", 5)
|
|
|| !VG_(strncmp)(str, "$u7e$", 5))
|
|
str += 5;
|
|
else
|
|
return False;
|
|
break;
|
|
case '.':
|
|
/* Do not allow three or more consecutive dots */
|
|
if (!VG_(strncmp)(str, "...", 3))
|
|
return False;
|
|
/* Fall through */
|
|
case 'a' ... 'z':
|
|
case 'A' ... 'Z':
|
|
case '0' ... '9':
|
|
case '_':
|
|
case ':':
|
|
str++;
|
|
break;
|
|
default:
|
|
return False;
|
|
}
|
|
}
|
|
|
|
return True;
|
|
}
|
|
|
|
/*
|
|
* INPUT:
|
|
* sym: symbol for which rust_is_mangled(sym) returns True
|
|
*
|
|
* The input is demangled in-place because the mangled name is always longer
|
|
* than the demangled one.
|
|
*/
|
|
static void rust_demangle_sym(HChar *sym)
|
|
{
|
|
const HChar *in;
|
|
HChar *out;
|
|
const HChar *end;
|
|
|
|
if (!sym)
|
|
return;
|
|
|
|
const SizeT sym_len = VG_(strlen)(sym);
|
|
const HChar* never_after = sym + sym_len;
|
|
|
|
in = sym;
|
|
out = sym;
|
|
end = sym + sym_len - (hash_prefix_len + hash_len);
|
|
|
|
while (in < end) {
|
|
switch (*in) {
|
|
case '$':
|
|
if (!(unescape(&in, &out, "$C$", ',')
|
|
|| unescape(&in, &out, "$SP$", '@')
|
|
|| unescape(&in, &out, "$BP$", '*')
|
|
|| unescape(&in, &out, "$RF$", '&')
|
|
|| unescape(&in, &out, "$LT$", '<')
|
|
|| unescape(&in, &out, "$GT$", '>')
|
|
|| unescape(&in, &out, "$LP$", '(')
|
|
|| unescape(&in, &out, "$RP$", ')')
|
|
|| unescape(&in, &out, "$u20$", ' ')
|
|
|| unescape(&in, &out, "$u22$", '\"')
|
|
|| unescape(&in, &out, "$u27$", '\'')
|
|
|| unescape(&in, &out, "$u2b$", '+')
|
|
|| unescape(&in, &out, "$u3b$", ';')
|
|
|| unescape(&in, &out, "$u5b$", '[')
|
|
|| unescape(&in, &out, "$u5d$", ']')
|
|
|| unescape(&in, &out, "$u7b$", '{')
|
|
|| unescape(&in, &out, "$u7d$", '}')
|
|
|| unescape(&in, &out, "$u7e$", '~'))) {
|
|
goto fail;
|
|
}
|
|
break;
|
|
case '_':
|
|
/*
|
|
* If this is the start of a path component and the next
|
|
* character is an escape sequence, ignore the
|
|
* underscore. The mangler inserts an underscore to make
|
|
* sure the path component begins with a XID_Start
|
|
* character.
|
|
*/
|
|
if ((in == sym || in[-1] == ':') && in[1] == '$')
|
|
in++;
|
|
else
|
|
*out++ = *in++;
|
|
break;
|
|
case '.':
|
|
if (in[1] == '.') {
|
|
/* ".." becomes "::" */
|
|
*out++ = ':';
|
|
*out++ = ':';
|
|
in += 2;
|
|
} else {
|
|
/* "." becomes "-" */
|
|
*out++ = '-';
|
|
in++;
|
|
}
|
|
break;
|
|
case 'a' ... 'z':
|
|
case 'A' ... 'Z':
|
|
case '0' ... '9':
|
|
case ':':
|
|
*out++ = *in++;
|
|
break;
|
|
default:
|
|
goto fail;
|
|
}
|
|
}
|
|
goto done;
|
|
|
|
fail:
|
|
*out++ = '?'; /* This is pretty lame, but it's hard to do better. */
|
|
done:
|
|
*out++ = '\0';
|
|
|
|
vg_assert(out <= never_after);
|
|
}
|
|
|
|
static Bool unescape(const HChar **in, HChar **out,
|
|
const HChar *seq, HChar value)
|
|
{
|
|
SizeT len = VG_(strlen)(seq);
|
|
|
|
if (VG_(strncmp)(*in, seq, len))
|
|
return False;
|
|
|
|
**out = value;
|
|
|
|
*in += len;
|
|
*out += 1;
|
|
|
|
return True;
|
|
}
|
|
|
|
|
|
/*--------------------------------------------------------------------*/
|
|
/*--- end ---*/
|
|
/*--------------------------------------------------------------------*/
|