Skip to content

DO NOT MERGE -- Tag strings #103766

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

Closed
gvanrossum wants to merge 12 commits into from
Closed

DO NOT MERGE -- Tag strings #103766

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
12 commits
Select commit Hold shift + click to select a range
1b61a5e
Tagged String syntax -- RAISES SYNTAX ERROR
gvanrossum May 1, 2022
a17f336
Some progress -- translate foo'bar' into foo('bar')
gvanrossum May 2, 2022
a9922f4
Correctly unparse TagString node
gvanrossum May 2, 2022
12ced2c
Accept only one string, make it a raw bytes
gvanrossum May 2, 2022
73f3c34
Parse the string as f-string
gvanrossum May 2, 2022
67175b0
Start calling tag(*args)
gvanrossum May 2, 2022
1c36df7
Pass lambdas (not quite thunks, but a step in the right direction)
gvanrossum May 2, 2022
d5a0e64
Lambdafy the expression only, exclusing conversion, spec
gvanrossum May 3, 2022
94c7335
Generate a (func, rawstr, conv, spec) tuple for thunks
gvanrossum May 3, 2022
399eaba
Restore original signature of _PyPegen_FstringParser_Init
gvanrossum May 3, 2022
c78e4a1
Fix unparsing of tagstring
gvanrossum May 4, 2022
1ec2fbe
Insist tag and str are flush together
gvanrossum May 4, 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
Prev Previous commit
Next Next commit
Parse the string as f-string
  • Loading branch information
@gvanrossum
gvanrossum committed May 3, 2022
commit 73f3c34429e6d5838d8c61081768942407bb32fe
4 changes: 2 additions & 2 deletions Grammar/python.gram
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -796,7 +796,7 @@ slice[expr_ty]:
| a=named_expression { a }

atom[expr_ty]:
| a=NAME b=STRING { _PyPegen_tag_string(p, a, b) }
| a=NAME b=STRING { _PyPegen_tag_string(p, a, (Token *)b) }
| NAME
| 'True' { _PyAST_Constant(Py_True, NULL, EXTRA) }
| 'False' { _PyAST_Constant(Py_False, NULL, EXTRA) }
Expand Down Expand Up @@ -871,7 +871,7 @@ lambda_param[arg_ty]: a=NAME { _PyAST_arg(a->v.Name.id, NULL, NULL, EXTRA) }
# LITERALS
# ========

strings[expr_ty] (memo): a=STRING+ { _PyPegen_concatenate_strings(p, a) }
strings[expr_ty] (memo): a=STRING+ { _PyPegen_concatenate_strings(p, a, 0) }

list[expr_ty]:
| '[' a=[star_named_expressions] ']' { _PyAST_List(a, Load, EXTRA) }
Expand Down
18 changes: 11 additions & 7 deletions Parser/action_helpers.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -870,14 +870,18 @@ _PyPegen_seq_delete_starred_exprs(Parser *p, asdl_seq *kwargs)
}

expr_ty
_PyPegen_tag_string(Parser *p, expr_ty tag, Token *t)
_PyPegen_tag_string(Parser *p, expr_ty tag, Token *tok)
{
// No prefixes (f, r, b, u)
// Parse like fstring
// Create a node similar to f-string AST
expr_ty str = _PyAST_Constant(t->bytes, NULL, t->lineno,
t->col_offset, t->end_lineno,
t->end_col_offset, p->arena);
asdl_generic_seq *tokens = _Py_asdl_generic_seq_new(1, p->arena);
if (tokens == NULL)
return NULL;
asdl_seq_SET(tokens, 0, tok);
expr_ty str = _PyPegen_concatenate_strings(p, (asdl_seq *)tokens, 1);
if (str == NULL)
return NULL;
return _PyAST_TagString(tag, str,
tag->lineno, tag->col_offset, str->end_lineno, str->end_col_offset,
p->arena);
Expand All @@ -886,7 +890,7 @@ _PyPegen_tag_string(Parser *p, expr_ty tag, Token *t)


expr_ty
_PyPegen_concatenate_strings(Parser *p, asdl_seq *strings)
_PyPegen_concatenate_strings(Parser *p, asdl_seq *strings, int tagged)
{
Py_ssize_t len = asdl_seq_LEN(strings);
assert(len > 0);
Expand All @@ -898,7 +902,7 @@ _PyPegen_concatenate_strings(Parser *p, asdl_seq *strings)
PyObject *bytes_str = NULL;

FstringParser state;
_PyPegen_FstringParser_Init(&state);
_PyPegen_FstringParser_Init(&state, tagged);

for (Py_ssize_t i = 0; i < len; i++) {
Token *t = asdl_seq_GET_UNTYPED(strings, i);
Expand All @@ -909,7 +913,7 @@ _PyPegen_concatenate_strings(Parser *p, asdl_seq *strings)
const char *fstr;
Py_ssize_t fstrlen = -1;

if (_PyPegen_parsestr(p, &this_bytesmode, &this_rawmode, &s, &fstr, &fstrlen, t) != 0) {
if (_PyPegen_parsestr(p, &this_bytesmode, &this_rawmode, &s, &fstr, &fstrlen, t, tagged) != 0) {
goto error;
}

Expand Down
4 changes: 2 additions & 2 deletions Parser/parser.c
View file
Open in desktop

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

2 changes: 1 addition & 1 deletion Parser/pegen.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -316,7 +316,7 @@ asdl_keyword_seq *_PyPegen_seq_delete_starred_exprs(Parser *, asdl_seq *);
expr_ty _PyPegen_collect_call_seqs(Parser *, asdl_expr_seq *, asdl_seq *,
int lineno, int col_offset, int end_lineno,
int end_col_offset, PyArena *arena);
expr_ty _PyPegen_concatenate_strings(Parser *p, asdl_seq *);
expr_ty _PyPegen_concatenate_strings(Parser *p, asdl_seq *, int);
expr_ty _PyPegen_tag_string(Parser *p, expr_ty, Token *);
expr_ty _PyPegen_ensure_imaginary(Parser *p, expr_ty);
expr_ty _PyPegen_ensure_real(Parser *p, expr_ty);
Expand Down
16 changes: 10 additions & 6 deletions Parser/string_parser.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -166,7 +166,7 @@ decode_bytes_with_escapes(Parser *p, const char *s, Py_ssize_t len, Token *t)
string object. Return 0 if no errors occurred. */
int
_PyPegen_parsestr(Parser *p, int *bytesmode, int *rawmode, PyObject **result,
const char **fstr, Py_ssize_t *fstrlen, Token *t)
const char **fstr, Py_ssize_t *fstrlen, Token *t, int tagged)
{
const char *s = PyBytes_AsString(t->bytes);
if (s == NULL) {
Expand All @@ -175,12 +175,16 @@ _PyPegen_parsestr(Parser *p, int *bytesmode, int *rawmode, PyObject **result,

size_t len;
int quote = Py_CHARMASK(*s);
int fmode = 0;
int fmode = tagged;
*bytesmode = 0;
*rawmode = 0;
*rawmode = tagged;
*result = NULL;
*fstr = NULL;
if (Py_ISALPHA(quote)) {
if (tagged) {
RAISE_SYNTAX_ERROR("Cannot combine tag and letter prefix");
return -1;
}
while (!*bytesmode || !*rawmode) {
if (quote == 'b' || quote == 'B') {
quote =(unsigned char)*++s;
Expand Down Expand Up @@ -1030,10 +1034,10 @@ FstringParser_check_invariants(FstringParser *state)
#endif

void
_PyPegen_FstringParser_Init(FstringParser *state)
_PyPegen_FstringParser_Init(FstringParser *state, int fmode)
{
state->last_str = NULL;
state->fmode = 0;
state->fmode = fmode;
ExprList_Init(&state->expr_list);
FstringParser_check_invariants(state);
}
Expand Down Expand Up @@ -1245,7 +1249,7 @@ fstring_parse(Parser *p, const char **str, const char *end, int raw,
{
FstringParser state;

_PyPegen_FstringParser_Init(&state);
_PyPegen_FstringParser_Init(&state, 0);
if (_PyPegen_FstringParser_ConcatFstring(p, &state, str, end, raw, recurse_lvl,
first_token, t, last_token) < 0) {
_PyPegen_FstringParser_Dealloc(&state);
Expand Down
4 changes: 2 additions & 2 deletions Parser/string_parser.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -33,9 +33,9 @@ typedef struct {
int fmode;
} FstringParser;

void _PyPegen_FstringParser_Init(FstringParser *);
void _PyPegen_FstringParser_Init(FstringParser *, int);
int _PyPegen_parsestr(Parser *, int *, int *, PyObject **,
const char **, Py_ssize_t *, Token *);
const char **, Py_ssize_t *, Token *, int);
int _PyPegen_FstringParser_ConcatFstring(Parser *, FstringParser *, const char **,
const char *, int, int, Token *, Token *,
Token *);
Expand Down
32 changes: 14 additions & 18 deletions Python/compile.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4915,24 +4915,20 @@ compiler_tag_string(struct compiler *c, expr_ty e)
expr_ty tag = e->v.TagString.tag;
expr_ty str = e->v.TagString.str;
if (tag->kind == Name_kind) {
if (str->kind == Constant_kind) {
PyObject *value = str->v.Constant.value;
PyObject *kind = str->v.Constant.kind;
if (kind == NULL && PyBytes_CheckExact(value)) {
// Generate code for tag(value)
asdl_expr_seq *args =
_Py_asdl_expr_seq_new(1, c->c_arena);
if (args == NULL)
return 0;
asdl_seq_SET(args, 0, str);
asdl_keyword_seq *keywords =
_Py_asdl_keyword_seq_new(0, c->c_arena);
if (keywords == NULL)
return 0;
ADDOP(c, PUSH_NULL);
VISIT(c, expr, tag);
return compiler_call_helper(c, 0, args, keywords);
}
if (str->kind == JoinedStr_kind) {
// Generate code for tag(str)
asdl_expr_seq *args =
_Py_asdl_expr_seq_new(1, c->c_arena);
if (args == NULL)
return 0;
asdl_seq_SET(args, 0, str);
asdl_keyword_seq *keywords =
_Py_asdl_keyword_seq_new(0, c->c_arena);
if (keywords == NULL)
return 0;
ADDOP(c, PUSH_NULL);
VISIT(c, expr, tag);
return compiler_call_helper(c, 0, args, keywords);
}
}
}
Expand Down