← 返回首页
gh-102856: Clean some of the PEP 701 tokenizer implementation by pablogsal · Pull Request #103634 · python/cpython · GitHub
Skip to content

Navigation Menu

Toggle navigation
Sign in
Appearance settings
Search or jump to...

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Include my email address so I can be contacted
Cancel Submit feedback

Saved searches

Use saved searches to filter your results more quickly

Cancel Create saved search
Appearance settings
Resetting focus

gh-102856: Clean some of the PEP 701 tokenizer implementation #103634

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
pablogsal merged 2 commits into from
Apr 19, 2023
Conversation 1 Commits 2 Checks 0 Files changed
Merged

gh-102856: Clean some of the PEP 701 tokenizer implementation #103634

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
917ba81
gh-102856: Clean some of the PEP 701 tokenizer implementation
pablogsal Apr 19, 2023
a377562
fixup! gh-102856: Clean some of the PEP 701 tokenizer implementation
pablogsal Apr 19, 2023
File filter

Filter by extension

Filter by extension .c  (1) .h  (1) All 2 file types selected Viewed files
Conversations
Failed to load comments. Retry
Loading
Jump to
Jump to file
Failed to load files. Retry
Loading
Diff view
Unified
Split
Hide whitespace
Apply and reload
Show whitespace
Diff view
Unified
Split
Hide whitespace
Apply and reload
136 changes: 65 additions & 71 deletions Parser/tokenizer.c
Show comments
View file Edit file Delete file
Open in desktop
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters. Learn more about bidirectional Unicode characters
Show hidden characters
Original file line number Diff line number Diff line change
Expand Up @@ -11,11 +11,6 @@
#include "tokenizer.h"
#include "errcode.h"

#include "unicodeobject.h"
#include "bytesobject.h"
#include "fileobject.h"
#include "abstract.h"

/* Alternate tab spacing */
#define ALTTABSIZE 1

Expand Down Expand Up @@ -43,6 +38,8 @@
tok->lineno++; \
tok->col_offset = 0;

#define INSIDE_FSTRING(tok) (tok->tok_mode_stack_index > 0)
#define INSIDE_FSTRING_EXPR(tok) (tok->curly_bracket_expr_start_depth >= 0)
#ifdef Py_DEBUG
static inline tokenizer_mode* TOK_GET_MODE(struct tok_state* tok) {
assert(tok->tok_mode_stack_index >= 0);
Expand All @@ -54,15 +51,9 @@ static inline tokenizer_mode* TOK_NEXT_MODE(struct tok_state* tok) {
assert(tok->tok_mode_stack_index < MAXLEVEL);
return &(tok->tok_mode_stack[++tok->tok_mode_stack_index]);
}
static inline int *TOK_GET_BRACKET_MARK(tokenizer_mode* mode) {
assert(mode->bracket_mark_index >= 0);
assert(mode->bracket_mark_index < MAX_EXPR_NESTING);
return &(mode->bracket_mark[mode->bracket_mark_index]);
}
#else
#define TOK_GET_MODE(tok) (&(tok->tok_mode_stack[tok->tok_mode_stack_index]))
#define TOK_NEXT_MODE(tok) (&(tok->tok_mode_stack[++tok->tok_mode_stack_index]))
#define TOK_GET_BRACKET_MARK(mode) (&(mode->bracket_mark[mode->bracket_mark_index]))
#endif

/* Forward */
Expand Down Expand Up @@ -398,20 +389,7 @@ update_fstring_expr(struct tok_state *tok, char cur)
tokenizer_mode *tok_mode = TOK_GET_MODE(tok);

switch (cur) {
case '{':
if (tok_mode->last_expr_buffer != NULL) {
PyMem_Free(tok_mode->last_expr_buffer);
}
tok_mode->last_expr_buffer = PyMem_Malloc(size);
if (tok_mode->last_expr_buffer == NULL) {
tok->done = E_NOMEM;
return 0;
}
tok_mode->last_expr_size = size;
tok_mode->last_expr_end = -1;
strncpy(tok_mode->last_expr_buffer, tok->cur, size);
break;
case 0:
case 0:
if (!tok_mode->last_expr_buffer || tok_mode->last_expr_end >= 0) {
return 1;
}
Expand All @@ -421,23 +399,38 @@ update_fstring_expr(struct tok_state *tok, char cur)
);
if (new_buffer == NULL) {
PyMem_Free(tok_mode->last_expr_buffer);
tok->done = E_NOMEM;
return 0;
goto error;
}
tok_mode->last_expr_buffer = new_buffer;
strncpy(tok_mode->last_expr_buffer + tok_mode->last_expr_size, tok->cur, size);
tok_mode->last_expr_size += size;
break;
case '{':
if (tok_mode->last_expr_buffer != NULL) {
PyMem_Free(tok_mode->last_expr_buffer);
}
tok_mode->last_expr_buffer = PyMem_Malloc(size);
if (tok_mode->last_expr_buffer == NULL) {
goto error;
}
tok_mode->last_expr_size = size;
tok_mode->last_expr_end = -1;
strncpy(tok_mode->last_expr_buffer, tok->cur, size);
break;
case '}':
case '!':
case ':':
if (tok_mode->last_expr_end == -1) {
tok_mode->last_expr_end = strlen(tok->start);
}
break;
default:
Py_UNREACHABLE();
}

return 1;
error:
tok->done = E_NOMEM;
return 0;
}

static void
Expand Down Expand Up @@ -1766,7 +1759,7 @@ tok_get_normal_mode(struct tok_state *tok, tokenizer_mode* current_tok, struct t
/* Skip comment, unless it's a type comment */
if (c == '#') {

if (tok->tok_mode_stack_index > 0) {
if (INSIDE_FSTRING(tok)) {
return MAKE_TOKEN(syntaxerror(tok, "f-string expression part cannot include '#'"));
}

Expand Down Expand Up @@ -2208,32 +2201,31 @@ tok_get_normal_mode(struct tok_state *tok, tokenizer_mode* current_tok, struct t

p_start = tok->start;
p_end = tok->cur;
tokenizer_mode *current_tok = TOK_NEXT_MODE(tok);
current_tok->kind = TOK_FSTRING_MODE;
current_tok->f_string_quote = quote;
current_tok->f_string_quote_size = quote_size;
current_tok->f_string_start = tok->start;
current_tok->f_string_multi_line_start = tok->line_start;
current_tok->last_expr_buffer = NULL;
current_tok->last_expr_size = 0;
current_tok->last_expr_end = -1;
tokenizer_mode *the_current_tok = TOK_NEXT_MODE(tok);
the_current_tok->kind = TOK_FSTRING_MODE;
the_current_tok->f_string_quote = quote;
the_current_tok->f_string_quote_size = quote_size;
the_current_tok->f_string_start = tok->start;
the_current_tok->f_string_multi_line_start = tok->line_start;
the_current_tok->last_expr_buffer = NULL;
the_current_tok->last_expr_size = 0;
the_current_tok->last_expr_end = -1;

switch (*tok->start) {
case 'F':
case 'f':
current_tok->f_string_raw = tolower(*(tok->start + 1)) == 'r';
the_current_tok->f_string_raw = tolower(*(tok->start + 1)) == 'r';
break;
case 'R':
case 'r':
current_tok->f_string_raw = 1;
the_current_tok->f_string_raw = 1;
break;
default:
Py_UNREACHABLE();
}

current_tok->bracket_stack = 0;
current_tok->bracket_mark[0] = 0;
current_tok->bracket_mark_index = -1;
the_current_tok->curly_bracket_depth = 0;
the_current_tok->curly_bracket_expr_start_depth = -1;
return MAKE_TOKEN(FSTRING_START);
}

Expand Down Expand Up @@ -2282,15 +2274,15 @@ tok_get_normal_mode(struct tok_state *tok, tokenizer_mode* current_tok, struct t
int start = tok->lineno;
tok->lineno = tok->first_lineno;

if (tok->tok_mode_stack_index > 0) {
if (INSIDE_FSTRING(tok)) {
/* When we are in an f-string, before raising the
* unterminated string literal error, check whether
* does the initial quote matches with f-strings quotes
* and if it is, then this must be a missing '}' token
* so raise the proper error */
tokenizer_mode *current_tok = TOK_GET_MODE(tok);
if (current_tok->f_string_quote == quote &&
current_tok->f_string_quote_size == quote_size) {
tokenizer_mode *the_current_tok = TOK_GET_MODE(tok);
if (the_current_tok->f_string_quote == quote &&
the_current_tok->f_string_quote_size == quote_size) {
return MAKE_TOKEN(syntaxerror(tok, "f-string: expecting '}'", start));
}
}
Expand Down Expand Up @@ -2339,18 +2331,17 @@ tok_get_normal_mode(struct tok_state *tok, tokenizer_mode* current_tok, struct t

/* Punctuation character */
int is_punctuation = (c == ':' || c == '}' || c == '!' || c == '{');
if (is_punctuation && tok->tok_mode_stack_index > 0 && current_tok->bracket_mark_index >= 0) {
int mark = *TOK_GET_BRACKET_MARK(current_tok);
/* This code block gets executed before the bracket_stack is incremented
if (is_punctuation && INSIDE_FSTRING(tok) && INSIDE_FSTRING_EXPR(current_tok)) {
/* This code block gets executed before the curly_bracket_depth is incremented
* by the `{` case, so for ensuring that we are on the 0th level, we need
* to adjust it manually */
int cursor = current_tok->bracket_stack - (c != '{');
int cursor = current_tok->curly_bracket_depth - (c != '{');

if (cursor == 0 && !update_fstring_expr(tok, c)) {
return MAKE_TOKEN(ENDMARKER);
}

if (c == ':' && cursor == mark) {
if (c == ':' && cursor == current_tok->curly_bracket_expr_start_depth) {
current_tok->kind = TOK_FSTRING_MODE;
p_start = tok->start;
p_end = tok->cur;
Expand Down Expand Up @@ -2390,16 +2381,15 @@ tok_get_normal_mode(struct tok_state *tok, tokenizer_mode* current_tok, struct t
tok->parenlinenostack[tok->level] = tok->lineno;
tok->parencolstack[tok->level] = (int)(tok->start - tok->line_start);
tok->level++;

if (tok->tok_mode_stack_index > 0) {
current_tok->bracket_stack++;
if (INSIDE_FSTRING(tok)) {
current_tok->curly_bracket_depth++;
}
break;
case ')':
case ']':
case '}':
if (!tok->level) {
if (tok->tok_mode_stack_index > 0 && !current_tok->bracket_stack && c == '}') {
if (INSIDE_FSTRING(tok) && !current_tok->curly_bracket_depth && c == '}') {
return MAKE_TOKEN(syntaxerror(tok, "f-string: single '}' is not allowed"));
}
return MAKE_TOKEN(syntaxerror(tok, "unmatched '%c'", c));
Expand All @@ -2415,10 +2405,10 @@ tok_get_normal_mode(struct tok_state *tok, tokenizer_mode* current_tok, struct t
nested expression, then instead of matching a different
syntactical construct with it; we'll throw an unmatched
parentheses error. */
if (tok->tok_mode_stack_index > 0 && opening == '{') {
assert(current_tok->bracket_stack >= 0);
int previous_bracket = current_tok->bracket_stack - 1;
if (previous_bracket == *TOK_GET_BRACKET_MARK(current_tok)) {
if (INSIDE_FSTRING(tok) && opening == '{') {
assert(current_tok->curly_bracket_depth >= 0);
int previous_bracket = current_tok->curly_bracket_depth - 1;
if (previous_bracket == current_tok->curly_bracket_expr_start_depth) {
return MAKE_TOKEN(syntaxerror(tok, "f-string: unmatched '%c'", c));
}
}
Expand All @@ -2436,14 +2426,16 @@ tok_get_normal_mode(struct tok_state *tok, tokenizer_mode* current_tok, struct t
}
}

if (tok->tok_mode_stack_index > 0) {
current_tok->bracket_stack--;
if (c == '}' && current_tok->bracket_stack == *TOK_GET_BRACKET_MARK(current_tok)) {
current_tok->bracket_mark_index--;
if (INSIDE_FSTRING(tok)) {
current_tok->curly_bracket_depth--;
if (c == '}' && current_tok->curly_bracket_depth == current_tok->curly_bracket_expr_start_depth) {
current_tok->curly_bracket_expr_start_depth--;
current_tok->kind = TOK_FSTRING_MODE;
}
}
break;
default:
break;
}

if (!Py_UNICODE_ISPRINTABLE(c)) {
Expand Down Expand Up @@ -2479,11 +2471,10 @@ tok_get_fstring_mode(struct tok_state *tok, tokenizer_mode* current_tok, struct

if ((start_char == '{' && peek1 != '{') || (start_char == '}' && peek1 != '}')) {
if (start_char == '{') {
current_tok->bracket_mark_index++;
if (current_tok->bracket_mark_index >= MAX_EXPR_NESTING) {
current_tok->curly_bracket_expr_start_depth++;
if (current_tok->curly_bracket_expr_start_depth >= MAX_EXPR_NESTING) {
return MAKE_TOKEN(syntaxerror(tok, "f-string: expressions nested too deeply"));
}
*TOK_GET_BRACKET_MARK(current_tok) = current_tok->bracket_stack;
}
TOK_GET_MODE(tok)->kind = TOK_REGULAR_MODE;
return tok_get_normal_mode(tok, current_tok, token);
Expand Down Expand Up @@ -2544,17 +2535,20 @@ tok_get_fstring_mode(struct tok_state *tok, tokenizer_mode* current_tok, struct
end_quote_size = 0;
}

int in_format_spec = current_tok->last_expr_end != -1 && current_tok->bracket_mark_index >= 0;
int in_format_spec = (
current_tok->last_expr_end != -1
&&
INSIDE_FSTRING_EXPR(current_tok)
);
if (c == '{') {
int peek = tok_nextc(tok);
if (peek != '{' || in_format_spec) {
tok_backup(tok, peek);
tok_backup(tok, c);
current_tok->bracket_mark_index++;
if (current_tok->bracket_mark_index >= MAX_EXPR_NESTING) {
current_tok->curly_bracket_expr_start_depth++;
if (current_tok->curly_bracket_expr_start_depth >= MAX_EXPR_NESTING) {
return MAKE_TOKEN(syntaxerror(tok, "f-string: expressions nested too deeply"));
}
*TOK_GET_BRACKET_MARK(current_tok) = current_tok->bracket_stack;
TOK_GET_MODE(tok)->kind = TOK_REGULAR_MODE;
p_start = tok->start;
p_end = tok->cur;
Expand Down
5 changes: 2 additions & 3 deletions Parser/tokenizer.h
Show comments
View file Edit file Delete file
Open in desktop
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters. Learn more about bidirectional Unicode characters
Show hidden characters
Original file line number Diff line number Diff line change
Expand Up @@ -43,9 +43,8 @@ enum tokenizer_mode_kind_t {
typedef struct _tokenizer_mode {
enum tokenizer_mode_kind_t kind;

int bracket_stack;
int bracket_mark[MAX_EXPR_NESTING];
int bracket_mark_index;
int curly_bracket_depth;
int curly_bracket_expr_start_depth;

char f_string_quote;
int f_string_quote_size;
Expand Down
Toggle all file notes Toggle all file annotations

Footer

© 2026 GitHub, Inc.