Skip to content

Implement String literal concatenation #988

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

Merged
merged 2 commits into from
Aug 18, 2022
Merged

Implement String literal concatenation #988

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
c382f48
Implement String literal concatenation
Thirumalai-Shaktivel Aug 18, 2022
720023d
Add tests and update the refs.
Thirumalai-Shaktivel Aug 18, 2022
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
1 change: 1 addition & 0 deletions src/lpython/parser/parser.yy
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -895,6 +895,7 @@ subscript

string
: string TK_STRING { $$ = STRING2($1, $2, @$); } // TODO
| string id TK_STRING { $$ = STRING4($1, STRING3($2, $3, @$), @$); }
| TK_STRING { $$ = STRING1($1, @$); }
| id TK_STRING { $$ = STRING3($1, $2, @$); }
;
Expand Down
79 changes: 75 additions & 4 deletions src/lpython/parser/semantics.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,6 +14,7 @@

#include <lpython/python_ast.h>
#include <libasr/string_utils.h>
#include <lpython/parser/parser_exception.h>

// This is only used in parser.tab.cc, nowhere else, so we simply include
// everything from LFortran::AST to save typing:
Expand Down Expand Up @@ -661,9 +662,78 @@ static inline ast_t* BOOLOP_01(Allocator &al, Location &loc,
#define COMPARE(x, op, y, l) make_Compare_t(p.m_a, l, \
EXPR(x), cmpopType::op, EXPRS(A2LIST(p.m_a, y)), 1)

char* concat_string(Allocator &al, ast_t *a, char *b) {
char *s = down_cast2<ConstantStr_t>(a)->m_value;
return LFortran::s2c(al, std::string(s) + std::string(b));
static inline ast_t* concat_string(Allocator &al, Location &l,
expr_t *string, std::string str, expr_t *string_literal) {
std::string str1 = "";
ast_t* tmp = nullptr;
Vec<expr_t *> exprs;
exprs.reserve(al, 4);

// TODO: Merge two concurrent ConstantStr's into one in the JoinedStr
if (string_literal) {
if (is_a<ConstantStr_t>(*string)
&& is_a<ConstantStr_t>(*string_literal)) {
str1 = std::string(down_cast<ConstantStr_t>(string)->m_value);
str = std::string(down_cast<ConstantStr_t>(string_literal)->m_value);
str1 = str1 + str;
tmp = make_ConstantStr_t(al, l, LFortran::s2c(al, str1), nullptr);
} else if (is_a<JoinedStr_t>(*string)
&& is_a<JoinedStr_t>(*string_literal)) {
JoinedStr_t *t = down_cast<JoinedStr_t>(string);
for (size_t i = 0; i < t->n_values; i++) {
exprs.push_back(al, t->m_values[i]);
}
t = down_cast<JoinedStr_t>(string_literal);
for (size_t i = 0; i < t->n_values; i++) {
exprs.push_back(al, t->m_values[i]);
}
tmp = make_JoinedStr_t(al, l, exprs.p, exprs.size());
} else if (is_a<JoinedStr_t>(*string)
&& is_a<ConstantStr_t>(*string_literal)) {
JoinedStr_t *t = down_cast<JoinedStr_t>(string);
for (size_t i = 0; i < t->n_values; i++) {
exprs.push_back(al, t->m_values[i]);
}
exprs.push_back(al, string_literal);
tmp = make_JoinedStr_t(al, l, exprs.p, exprs.size());
} else if (is_a<ConstantStr_t>(*string)
&& is_a<JoinedStr_t>(*string_literal)) {
exprs.push_back(al, string);
JoinedStr_t *t = down_cast<JoinedStr_t>(string_literal);
for (size_t i = 0; i < t->n_values; i++) {
exprs.push_back(al, t->m_values[i]);
}
tmp = make_JoinedStr_t(al, l, exprs.p, exprs.size());
} else if (is_a<ConstantBytes_t>(*string)
&& is_a<ConstantBytes_t>(*string_literal)) {
str1 = std::string(down_cast<ConstantBytes_t>(string)->m_value);
str1 = str1.substr(0, str1.size() - 1);
str = std::string(down_cast<ConstantBytes_t>(string_literal)->m_value);
str = str.substr(2, str.size());
str1 = str1 + str;
tmp = make_ConstantBytes_t(al, l, LFortran::s2c(al, str1), nullptr);
} else {
throw LFortran::parser_local::ParserError(
"The byte and non-byte literals can not be combined", l);
}
} else {
if (is_a<ConstantStr_t>(*string)) {
str1 = std::string(down_cast<ConstantStr_t>(string)->m_value);
str1 = str1 + str;
tmp = make_ConstantStr_t(al, l, LFortran::s2c(al, str1), nullptr);
} else if (is_a<JoinedStr_t>(*string)) {
JoinedStr_t *t = down_cast<JoinedStr_t>(string);
for (size_t i = 0; i < t->n_values; i++) {
exprs.push_back(al, t->m_values[i]);
}
exprs.push_back(al, (expr_t *)make_ConstantStr_t(al, l,
LFortran::s2c(al, str), nullptr));
tmp = make_JoinedStr_t(al, l, exprs.p, exprs.size());
} else {
LFORTRAN_ASSERT(false);
}
}
return tmp;
}

char* unescape(Allocator &al, LFortran::Str &s) {
Expand All @@ -684,8 +754,9 @@ char* unescape(Allocator &al, LFortran::Str &s) {
// `x.int_n` is of type BigInt but we store the int64_t directly in AST
#define INTEGER(x, l) make_ConstantInt_t(p.m_a, l, x, nullptr)
#define STRING1(x, l) make_ConstantStr_t(p.m_a, l, unescape(p.m_a, x), nullptr)
#define STRING2(x, y, l) make_ConstantStr_t(p.m_a, l, concat_string(p.m_a, x, y.c_str(p.m_a)), nullptr)
#define STRING2(x, y, l) concat_string(p.m_a, l, EXPR(x), y.str(), nullptr)
#define STRING3(id, x, l) PREFIX_STRING(p.m_a, l, name2char(id), x.c_str(p.m_a))
#define STRING4(x, s, l) concat_string(p.m_a, l, EXPR(x), "", EXPR(s))
#define FLOAT(x, l) make_ConstantFloat_t(p.m_a, l, x, nullptr)
#define COMPLEX(x, l) make_ConstantComplex_t(p.m_a, l, 0, x, nullptr)
#define BOOL(x, l) make_ConstantBool_t(p.m_a, l, x, nullptr)
Expand Down
24 changes: 24 additions & 0 deletions tests/parser/string1.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -47,3 +47,27 @@
RB"Text"

rf'\N{AMPERSAND}'

(f"Text{a}, {b}"
f"Text {a}")

(f"Text{a}, {b}"
"Text")

("Text"
f"{b}, Text")

(f"Text {a}"
r"Text")

(r"Text"
r"Text")

(r"Text"
"Text")

(r"Text"
f"{a} Text")

(b"Text"
b"Text")
4 changes: 2 additions & 2 deletions tests/reference/ast_new-string1-96b90b3.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,11 +2,11 @@
"basename": "ast_new-string1-96b90b3",
"cmd": "lpython --show-ast --new-parser --no-color {infile} -o {outfile}",
"infile": "tests/parser/string1.py",
"infile_hash": "e4090ab45efb09242e14f3dd494ca869a32e71de3617306d959d8721",
"infile_hash": "fd64c289d2ab5638fcc3093bda8f07fdde0e034798d4c04791fd441b",
"outfile": null,
"outfile_hash": null,
"stdout": "ast_new-string1-96b90b3.stdout",
"stdout_hash": "1666ca8ebba593a8bf35e7906efabf5726626b7c82b130324bd78a43",
"stdout_hash": "e23911e24aada1bf320ffde6cca77cc2a19700d09a2e9b01bbd37afa",
"stderr": null,
"stderr_hash": null,
"returncode": 0
Expand Down
2 changes: 1 addition & 1 deletion tests/reference/ast_new-string1-96b90b3.stdout
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,4 +5,4 @@ anotherline
Text
" ())) (Expr (ConstantStr "Text" ())) (Expr (ConstantStr "a\tb\nA\tB" ())) (Expr (ConstantStr "1,2,3 # comment" "u")) (Expr (ConstantStr "Text" "u")) (Expr (ConstantStr "Text" ())) (Expr (JoinedStr [(ConstantStr "Text " ()) (FormattedValue (Name id Load) -1 ())])) (Expr (JoinedStr [(ConstantStr "Text " ()) (FormattedValue (Name id Load) -1 ())])) (Expr (JoinedStr [(ConstantStr "Text " ()) (FormattedValue (Name id Load) -1 ())])) (Expr (JoinedStr [(ConstantStr "Text " ()) (FormattedValue (Name id Load) -1 ())])) (Expr (JoinedStr [(ConstantStr "Text " ()) (FormattedValue (Name id Load) -1 ())])) (Expr (JoinedStr [(ConstantStr "
Text " ()) (FormattedValue (Name id Load) -1 ()) (ConstantStr "
" ())])) (Expr (JoinedStr [(ConstantStr "Text " ()) (FormattedValue (Name id Load) -1 ())])) (Expr (JoinedStr [(ConstantStr "Text " ()) (FormattedValue (Name id Load) -1 ())])) (Expr (ConstantBytes "b'Text'" ())) (Expr (ConstantBytes "b'\nText\n'" ())) (Expr (ConstantBytes "b'Text'" ())) (Expr (ConstantBytes "b'Text'" ())) (Expr (ConstantBytes "b'Text'" ())) (Expr (ConstantBytes "b'\nText\n'" ())) (Expr (ConstantBytes "b'Text'" ())) (Expr (ConstantBytes "b'Text'" ())) (Expr (JoinedStr [(ConstantStr "\N" ()) (FormattedValue (Name AMPERSAND Load) -1 ())]))] [])
" ())])) (Expr (JoinedStr [(ConstantStr "Text " ()) (FormattedValue (Name id Load) -1 ())])) (Expr (JoinedStr [(ConstantStr "Text " ()) (FormattedValue (Name id Load) -1 ())])) (Expr (ConstantBytes "b'Text'" ())) (Expr (ConstantBytes "b'\nText\n'" ())) (Expr (ConstantBytes "b'Text'" ())) (Expr (ConstantBytes "b'Text'" ())) (Expr (ConstantBytes "b'Text'" ())) (Expr (ConstantBytes "b'\nText\n'" ())) (Expr (ConstantBytes "b'Text'" ())) (Expr (ConstantBytes "b'Text'" ())) (Expr (JoinedStr [(ConstantStr "\N" ()) (FormattedValue (Name AMPERSAND Load) -1 ())])) (Expr (JoinedStr [(ConstantStr "Text" ()) (FormattedValue (Name a Load) -1 ()) (ConstantStr ", " ()) (FormattedValue (Name b Load) -1 ()) (ConstantStr "Text " ()) (FormattedValue (Name a Load) -1 ())])) (Expr (JoinedStr [(ConstantStr "Text" ()) (FormattedValue (Name a Load) -1 ()) (ConstantStr ", " ()) (FormattedValue (Name b Load) -1 ()) (ConstantStr "Text" ())])) (Expr (JoinedStr [(ConstantStr "Text" ()) (FormattedValue (Name b Load) -1 ()) (ConstantStr ", Text" ())])) (Expr (JoinedStr [(ConstantStr "Text " ()) (FormattedValue (Name a Load) -1 ()) (ConstantStr "Text" ())])) (Expr (ConstantStr "TextText" ())) (Expr (ConstantStr "TextText" ())) (Expr (JoinedStr [(ConstantStr "Text" ()) (FormattedValue (Name a Load) -1 ()) (ConstantStr " Text" ())])) (Expr (ConstantBytes "b'TextText'" ()))] [])