|
|
|
@ -1793,8 +1793,9 @@ inline NodeRef<Key> Parse(Span<const char> in, const Ctx& ctx)
|
|
|
|
|
// Get threshold
|
|
|
|
|
int next_comma = FindNextChar(in, ',');
|
|
|
|
|
if (next_comma < 1) return false;
|
|
|
|
|
int64_t k;
|
|
|
|
|
if (!ParseInt64(std::string(in.begin(), in.begin() + next_comma), &k)) return false;
|
|
|
|
|
const auto k_to_integral{ToIntegral<int64_t>(std::string_view(in.begin(), next_comma))};
|
|
|
|
|
if (!k_to_integral.has_value()) return false;
|
|
|
|
|
const int64_t k{k_to_integral.value()};
|
|
|
|
|
in = in.subspan(next_comma + 1);
|
|
|
|
|
// Get keys. It is compatible for both compressed and x-only keys.
|
|
|
|
|
std::vector<Key> keys;
|
|
|
|
@ -1948,21 +1949,19 @@ inline NodeRef<Key> Parse(Span<const char> in, const Ctx& ctx)
|
|
|
|
|
} else if (Const("after(", in)) {
|
|
|
|
|
int arg_size = FindNextChar(in, ')');
|
|
|
|
|
if (arg_size < 1) return {};
|
|
|
|
|
int64_t num;
|
|
|
|
|
if (!ParseInt64(std::string(in.begin(), in.begin() + arg_size), &num)) return {};
|
|
|
|
|
if (num < 1 || num >= 0x80000000L) return {};
|
|
|
|
|
constructed.push_back(MakeNodeRef<Key>(internal::NoDupCheck{}, ctx.MsContext(), Fragment::AFTER, num));
|
|
|
|
|
const auto num{ToIntegral<int64_t>(std::string_view(in.begin(), arg_size))};
|
|
|
|
|
if (!num.has_value() || *num < 1 || *num >= 0x80000000L) return {};
|
|
|
|
|
constructed.push_back(MakeNodeRef<Key>(internal::NoDupCheck{}, ctx.MsContext(), Fragment::AFTER, *num));
|
|
|
|
|
in = in.subspan(arg_size + 1);
|
|
|
|
|
script_size += 1 + (num > 16) + (num > 0x7f) + (num > 0x7fff) + (num > 0x7fffff);
|
|
|
|
|
script_size += 1 + (*num > 16) + (*num > 0x7f) + (*num > 0x7fff) + (*num > 0x7fffff);
|
|
|
|
|
} else if (Const("older(", in)) {
|
|
|
|
|
int arg_size = FindNextChar(in, ')');
|
|
|
|
|
if (arg_size < 1) return {};
|
|
|
|
|
int64_t num;
|
|
|
|
|
if (!ParseInt64(std::string(in.begin(), in.begin() + arg_size), &num)) return {};
|
|
|
|
|
if (num < 1 || num >= 0x80000000L) return {};
|
|
|
|
|
constructed.push_back(MakeNodeRef<Key>(internal::NoDupCheck{}, ctx.MsContext(), Fragment::OLDER, num));
|
|
|
|
|
const auto num{ToIntegral<int64_t>(std::string_view(in.begin(), arg_size))};
|
|
|
|
|
if (!num.has_value() || *num < 1 || *num >= 0x80000000L) return {};
|
|
|
|
|
constructed.push_back(MakeNodeRef<Key>(internal::NoDupCheck{}, ctx.MsContext(), Fragment::OLDER, *num));
|
|
|
|
|
in = in.subspan(arg_size + 1);
|
|
|
|
|
script_size += 1 + (num > 16) + (num > 0x7f) + (num > 0x7fff) + (num > 0x7fffff);
|
|
|
|
|
script_size += 1 + (*num > 16) + (*num > 0x7f) + (*num > 0x7fff) + (*num > 0x7fffff);
|
|
|
|
|
} else if (Const("multi(", in)) {
|
|
|
|
|
if (!parse_multi_exp(in, /* is_multi_a = */false)) return {};
|
|
|
|
|
} else if (Const("multi_a(", in)) {
|
|
|
|
@ -1970,13 +1969,13 @@ inline NodeRef<Key> Parse(Span<const char> in, const Ctx& ctx)
|
|
|
|
|
} else if (Const("thresh(", in)) {
|
|
|
|
|
int next_comma = FindNextChar(in, ',');
|
|
|
|
|
if (next_comma < 1) return {};
|
|
|
|
|
if (!ParseInt64(std::string(in.begin(), in.begin() + next_comma), &k)) return {};
|
|
|
|
|
if (k < 1) return {};
|
|
|
|
|
const auto k{ToIntegral<int64_t>(std::string_view(in.begin(), next_comma))};
|
|
|
|
|
if (!k.has_value() || *k < 1) return {};
|
|
|
|
|
in = in.subspan(next_comma + 1);
|
|
|
|
|
// n = 1 here because we read the first WRAPPED_EXPR before reaching THRESH
|
|
|
|
|
to_parse.emplace_back(ParseContext::THRESH, 1, k);
|
|
|
|
|
to_parse.emplace_back(ParseContext::THRESH, 1, *k);
|
|
|
|
|
to_parse.emplace_back(ParseContext::WRAPPED_EXPR, -1, -1);
|
|
|
|
|
script_size += 2 + (k > 16) + (k > 0x7f) + (k > 0x7fff) + (k > 0x7fffff);
|
|
|
|
|
script_size += 2 + (*k > 16) + (*k > 0x7f) + (*k > 0x7fff) + (*k > 0x7fffff);
|
|
|
|
|
} else if (Const("andor(", in)) {
|
|
|
|
|
to_parse.emplace_back(ParseContext::ANDOR, -1, -1);
|
|
|
|
|
to_parse.emplace_back(ParseContext::CLOSE_BRACKET, -1, -1);
|
|
|
|
|