Skip to content

Instrument strchr to use the string refinement solver #5200

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 6 commits into
base: develop
Choose a base branch
from
Open

Instrument strchr to use the string refinement solver #5200

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
941eed4
Add c-specific refinement utils
hannes-steffenhagen-diffblue Feb 7, 2020
8d82639
Extend index-of for c
hannes-steffenhagen-diffblue Feb 7, 2020
3f76f13
Implement strchr in terms of index-of
petr-bauch Feb 7, 2020
80c88cc
Handle the c-specific cases in string refinement
petr-bauch Feb 7, 2020
ddf72d1
Add tests for strchr
petr-bauch Feb 7, 2020
0879ad0
Add case for byte-extracting arrays
petr-bauch Feb 7, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
79 changes: 76 additions & 3 deletions src/solvers/strings/string_refinement_util.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,19 +14,27 @@ Author: Diffblue Ltd.
#include <numeric>
#include <unordered_set>
#include <util/arith_tools.h>
#include <util/expr_cast.h>
#include <util/expr_iterator.h>
#include <util/expr_util.h>
#include <util/graph.h>
#include <util/magic.h>
#include <util/make_unique.h>
#include <util/ssa_expr.h>
#include <util/std_expr.h>
#include <util/string2int.h>
#include <util/string_constant.h>
#include <util/unicode.h>

bool is_char_type(const typet &type)
{
return type.id() == ID_unsignedbv && to_unsignedbv_type(type).get_width() <=
STRING_REFINEMENT_MAX_CHAR_WIDTH;
if(type.id() == ID_unsignedbv)
return to_unsignedbv_type(type).get_width() <=
STRING_REFINEMENT_MAX_CHAR_WIDTH;

if(type.id() == ID_signedbv)
return to_signedbv_type(type).get_width() <= 8;
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🚫 Magic number here should have some explanation

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

documentation of the function needs to be updated, and does it make sense to have signed characters?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@romainbrenguier char is allowed to be either signed or unsigned in C.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Turns out the magic is not needed at all.


return false;
}

bool is_char_array_type(const typet &type, const namespacet &ns)
Expand Down Expand Up @@ -192,3 +200,68 @@ array_exprt interval_sparse_arrayt::concretize(
size, it == entries.end() ? default_value : it->second);
return array;
}

exprt maybe_byte_extract_expr(const exprt &expr)
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

❓ What's "maybe" about this? Would returning an optional make sense here?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Well it always returns some expr. But it acts as identity for expressions that are not byte-extract.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

💡 This function is non-trivial, I'd consider writing a unit test for it to check it is doing everything you expect

{
if(!can_cast_expr<byte_extract_exprt>(expr))
return expr;

const auto &byte_extract_expr = to_byte_extract_expr(expr);
const auto &offset = byte_extract_expr.offset();
PRECONDITION(offset.id() == ID_constant);
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

❔ how come this is a precondition? Seems like should be documented at least

if(offset.id() != ID_constant)
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

❓ Unreachable? You've just asserted this isn't true

{
return expr;
}
const auto &constant_offset = to_constant_expr(offset);
const auto &op = byte_extract_expr.op();
auto numeric_offset =
string2optional_int(id2string(constant_offset.get_value()));
PRECONDITION(numeric_offset.has_value());
if(*numeric_offset == 0)
{
return op;
}

array_exprt::operandst offset_operands;
offset_operands.insert(
offset_operands.end(),
op.operands().begin() + *numeric_offset,
op.operands().end());
const auto extracted_array_suffix =
array_exprt{offset_operands, to_array_type(op.type())};

return extracted_array_suffix;
}

exprt maybe_remove_string_exprs(const exprt &expr)
{
return [&]() {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

❓ What's the purpose of this lambda...

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I am fairly sure there isn’t one. Probably happened by accident during a refactor.

if(
auto const &maybe_string_constant =
expr_try_dynamic_cast<string_constantt>(expr))
{
return static_cast<const exprt &>(maybe_string_constant->to_array_expr());
}
else
{
return expr;
}
}();
}

exprt massage_weird_arrays_into_non_weird_arrays(const exprt &expr)
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🚫 This of course needs documentation, or a much better name! And again, a unit test seems wise as a technical documentation of what qualifies as "weird" and "not weird" arrays

{
auto const byte_extracted = maybe_byte_extract_expr(expr);
auto const string_removed = maybe_remove_string_exprs(byte_extracted);
PRECONDITION(string_removed.type().id() == ID_array);
if(string_removed.id() == ID_if)
{
return if_exprt{
string_removed.op0(),
massage_weird_arrays_into_non_weird_arrays(string_removed.op1()),
massage_weird_arrays_into_non_weird_arrays(string_removed.op2()),
string_removed.type()};
}
return string_removed;
}
5 changes: 5 additions & 0 deletions src/solvers/strings/string_refinement_util.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -140,4 +140,9 @@ class interval_sparse_arrayt final : public sparse_arrayt
}
};

exprt maybe_remove_string_exprs(const exprt &expr);

exprt maybe_byte_extract_expr(const exprt &expr);

exprt massage_weird_arrays_into_non_weird_arrays(const exprt &expr);
#endif // CPROVER_SOLVERS_REFINEMENT_STRING_REFINEMENT_UTIL_H