#include #include #include "tokenizer.h" #include "pegen.h" #include "string_parser.h" PyObject * _PyPegen_new_type_comment(Parser *p, char *s) { PyObject *res = PyUnicode_DecodeUTF8(s, strlen(s), NULL); if (res == NULL) { return NULL; } if (PyArena_AddPyObject(p->arena, res) < 0) { Py_DECREF(res); return NULL; } return res; } arg_ty _PyPegen_add_type_comment_to_arg(Parser *p, arg_ty a, Token *tc) { if (tc == NULL) { return a; } char *bytes = PyBytes_AsString(tc->bytes); if (bytes == NULL) { return NULL; } PyObject *tco = _PyPegen_new_type_comment(p, bytes); if (tco == NULL) { return NULL; } return arg(a->arg, a->annotation, tco, a->lineno, a->col_offset, a->end_lineno, a->end_col_offset, p->arena); } static int init_normalization(Parser *p) { if (p->normalize) { return 1; } PyObject *m = PyImport_ImportModuleNoBlock("unicodedata"); if (!m) { return 0; } p->normalize = PyObject_GetAttrString(m, "normalize"); Py_DECREF(m); if (!p->normalize) { return 0; } return 1; } /* Checks if the NOTEQUAL token is valid given the current parser flags 0 indicates success and nonzero indicates failure (an exception may be set) */ int _PyPegen_check_barry_as_flufl(Parser *p) { Token *t = p->tokens[p->fill - 1]; assert(t->bytes != NULL); assert(t->type == NOTEQUAL); char* tok_str = PyBytes_AS_STRING(t->bytes); if (p->flags & PyPARSE_BARRY_AS_BDFL && strcmp(tok_str, "<>") != 0) { RAISE_SYNTAX_ERROR("with Barry as BDFL, use '<>' instead of '!='"); return -1; } if (!(p->flags & PyPARSE_BARRY_AS_BDFL)) { return strcmp(tok_str, "!="); } return 0; } PyObject * _PyPegen_new_identifier(Parser *p, char *n) { PyObject *id = PyUnicode_DecodeUTF8(n, strlen(n), NULL); if (!id) { goto error; } /* PyUnicode_DecodeUTF8 should always return a ready string. */ assert(PyUnicode_IS_READY(id)); /* Check whether there are non-ASCII characters in the identifier; if so, normalize to NFKC. */ if (!PyUnicode_IS_ASCII(id)) { PyObject *id2; if (!init_normalization(p)) { Py_DECREF(id); goto error; } PyObject *form = PyUnicode_InternFromString("NFKC"); if (form == NULL) { Py_DECREF(id); goto error; } PyObject *args[2] = {form, id}; id2 = _PyObject_FastCall(p->normalize, args, 2); Py_DECREF(id); Py_DECREF(form); if (!id2) { goto error; } if (!PyUnicode_Check(id2)) { PyErr_Format(PyExc_TypeError, "unicodedata.normalize() must return a string, not " "%.200s", _PyType_Name(Py_TYPE(id2))); Py_DECREF(id2); goto error; } id = id2; } PyUnicode_InternInPlace(&id); if (PyArena_AddPyObject(p->arena, id) < 0) { Py_DECREF(id); goto error; } return id; error: p->error_indicator = 1; return NULL; } static PyObject * _create_dummy_identifier(Parser *p) { return _PyPegen_new_identifier(p, ""); } static inline Py_ssize_t byte_offset_to_character_offset(PyObject *line, Py_ssize_t col_offset) { const char *str = PyUnicode_AsUTF8(line); if (!str) { return 0; } assert(col_offset >= 0 && (unsigned long)col_offset <= strlen(str)); PyObject *text = PyUnicode_DecodeUTF8(str, col_offset, "replace"); if (!text) { return 0; } Py_ssize_t size = PyUnicode_GET_LENGTH(text); Py_DECREF(text); return size; } const char * _PyPegen_get_expr_name(expr_ty e) { assert(e != NULL); switch (e->kind) { case Attribute_kind: return "attribute"; case Subscript_kind: return "subscript"; case Starred_kind: return "starred"; case Name_kind: return "name"; case List_kind: return "list"; case Tuple_kind: return "tuple"; case Lambda_kind: return "lambda"; case Call_kind: return "function call"; case BoolOp_kind: case BinOp_kind: case UnaryOp_kind: return "operator"; case GeneratorExp_kind: return "generator expression"; case Yield_kind: case YieldFrom_kind: return "yield expression"; case Await_kind: return "await expression"; case ListComp_kind: return "list comprehension"; case SetComp_kind: return "set comprehension"; case DictComp_kind: return "dict comprehension"; case Dict_kind: return "dict display"; case Set_kind: return "set display"; case JoinedStr_kind: case FormattedValue_kind: return "f-string expression"; case Constant_kind: { PyObject *value = e->v.Constant.value; if (value == Py_None) { return "None"; } if (value == Py_False) { return "False"; } if (value == Py_True) { return "True"; } if (value == Py_Ellipsis) { return "Ellipsis"; } return "literal"; } case Compare_kind: return "comparison"; case IfExp_kind: return "conditional expression"; case NamedExpr_kind: return "named expression"; default: PyErr_Format(PyExc_SystemError, "unexpected expression in assignment %d (line %d)", e->kind, e->lineno); return NULL; } } static int raise_decode_error(Parser *p) { assert(PyErr_Occurred()); const char *errtype = NULL; if (PyErr_ExceptionMatches(PyExc_UnicodeError)) { errtype = "unicode error"; } else if (PyErr_ExceptionMatches(PyExc_ValueError)) { errtype = "value error"; } if (errtype) { PyObject *type; PyObject *value; PyObject *tback; PyObject *errstr; PyErr_Fetch(&type, &value, &tback); errstr = PyObject_Str(value); if (errstr) { RAISE_SYNTAX_ERROR("(%s) %U", errtype, errstr); Py_DECREF(errstr); } else { PyErr_Clear(); RAISE_SYNTAX_ERROR("(%s) unknown error", errtype); } Py_XDECREF(type); Py_XDECREF(value); Py_XDECREF(tback); } return -1; } static void raise_tokenizer_init_error(PyObject *filename) { if (!(PyErr_ExceptionMatches(PyExc_LookupError) || PyErr_ExceptionMatches(PyExc_ValueError) || PyErr_ExceptionMatches(PyExc_UnicodeDecodeError))) { return; } PyObject *errstr = NULL; PyObject *tuple = NULL; PyObject *type; PyObject *value; PyObject *tback; PyErr_Fetch(&type, &value, &tback); errstr = PyObject_Str(value); if (!errstr) { goto error; } PyObject *tmp = Py_BuildValue("(OiiO)", filename, 0, -1, Py_None); if (!tmp) { goto error; } tuple = PyTuple_Pack(2, errstr, tmp); Py_DECREF(tmp); if (!value) { goto error; } PyErr_SetObject(PyExc_SyntaxError, tuple); error: Py_XDECREF(type); Py_XDECREF(value); Py_XDECREF(tback); Py_XDECREF(errstr); Py_XDECREF(tuple); } static int tokenizer_error(Parser *p) { if (PyErr_Occurred()) { return -1; } const char *msg = NULL; PyObject* errtype = PyExc_SyntaxError; switch (p->tok->done) { case E_TOKEN: msg = "invalid token"; break; case E_EOFS: RAISE_SYNTAX_ERROR("EOF while scanning triple-quoted string literal"); return -1; case E_EOLS: RAISE_SYNTAX_ERROR("EOL while scanning string literal"); return -1; case E_EOF: RAISE_SYNTAX_ERROR("unexpected EOF while parsing"); return -1; case E_DEDENT: RAISE_INDENTATION_ERROR("unindent does not match any outer indentation level"); return -1; case E_INTR: if (!PyErr_Occurred()) { PyErr_SetNone(PyExc_KeyboardInterrupt); } return -1; case E_NOMEM: PyErr_NoMemory(); return -1; case E_TABSPACE: errtype = PyExc_TabError; msg = "inconsistent use of tabs and spaces in indentation"; break; case E_TOODEEP: errtype = PyExc_IndentationError; msg = "too many levels of indentation"; break; case E_LINECONT: msg = "unexpected character after line continuation character"; break; default: msg = "unknown parsing error"; } PyErr_Format(errtype, msg); // There is no reliable column information for this error PyErr_SyntaxLocationObject(p->tok->filename, p->tok->lineno, 0); return -1; } void * _PyPegen_raise_error(Parser *p, PyObject *errtype, const char *errmsg, ...) { Token *t = p->known_err_token != NULL ? p->known_err_token : p->tokens[p->fill - 1]; Py_ssize_t col_offset; if (t->col_offset == -1) { col_offset = Py_SAFE_DOWNCAST(p->tok->cur - p->tok->buf, intptr_t, int); } else { col_offset = t->col_offset + 1; } va_list va; va_start(va, errmsg); _PyPegen_raise_error_known_location(p, errtype, t->lineno, col_offset, errmsg, va); va_end(va); return NULL; } void * _PyPegen_raise_error_known_location(Parser *p, PyObject *errtype, Py_ssize_t lineno, Py_ssize_t col_offset, const char *errmsg, va_list va) { PyObject *value = NULL; PyObject *errstr = NULL; PyObject *error_line = NULL; PyObject *tmp = NULL; p->error_indicator = 1; if (p->start_rule == Py_fstring_input) { const char *fstring_msg = "f-string: "; Py_ssize_t len = strlen(fstring_msg) + strlen(errmsg); char *new_errmsg = PyMem_Malloc(len + 1); // Lengths of both strings plus NULL character if (!new_errmsg) { return (void *) PyErr_NoMemory(); } // Copy both strings into new buffer memcpy(new_errmsg, fstring_msg, strlen(fstring_msg)); memcpy(new_errmsg + strlen(fstring_msg), errmsg, strlen(errmsg)); new_errmsg[len] = 0; errmsg = new_errmsg; } errstr = PyUnicode_FromFormatV(errmsg, va); if (!errstr) { goto error; } if (p->start_rule == Py_file_input) { error_line = PyErr_ProgramTextObject(p->tok->filename, (int) lineno); } if (!error_line) { Py_ssize_t size = p->tok->inp - p->tok->buf; error_line = PyUnicode_DecodeUTF8(p->tok->buf, size, "replace"); if (!error_line) { goto error; } } if (p->start_rule == Py_fstring_input) { col_offset -= p->starting_col_offset; } Py_ssize_t col_number = col_offset; if (p->tok->encoding != NULL) { col_number = byte_offset_to_character_offset(error_line, col_offset); } tmp = Py_BuildValue("(OiiN)", p->tok->filename, lineno, col_number, error_line); if (!tmp) { goto error; } value = PyTuple_Pack(2, errstr, tmp); Py_DECREF(tmp); if (!value) { goto error; } PyErr_SetObject(errtype, value); Py_DECREF(errstr); Py_DECREF(value); if (p->start_rule == Py_fstring_input) { PyMem_Free((void *)errmsg); } return NULL; error: Py_XDECREF(errstr); Py_XDECREF(error_line); if (p->start_rule == Py_fstring_input) { PyMem_Free((void *)errmsg); } return NULL; } #if 0 static const char * token_name(int type) { if (0 <= type && type <= N_TOKENS) { return _PyParser_TokenNames[type]; } return ""; } #endif // Here, mark is the start of the node, while p->mark is the end. // If node==NULL, they should be the same. int _PyPegen_insert_memo(Parser *p, int mark, int type, void *node) { // Insert in front Memo *m = PyArena_Malloc(p->arena, sizeof(Memo)); if (m == NULL) { return -1; } m->type = type; m->node = node; m->mark = p->mark; m->next = p->tokens[mark]->memo; p->tokens[mark]->memo = m; return 0; } // Like _PyPegen_insert_memo(), but updates an existing node if found. int _PyPegen_update_memo(Parser *p, int mark, int type, void *node) { for (Memo *m = p->tokens[mark]->memo; m != NULL; m = m->next) { if (m->type == type) { // Update existing node. m->node = node; m->mark = p->mark; return 0; } } // Insert new node. return _PyPegen_insert_memo(p, mark, type, node); } // Return dummy NAME. void * _PyPegen_dummy_name(Parser *p, ...) { static void *cache = NULL; if (cache != NULL) { return cache; } PyObject *id = _create_dummy_identifier(p); if (!id) { return NULL; } cache = Name(id, Load, 1, 0, 1, 0, p->arena); return cache; } static int _get_keyword_or_name_type(Parser *p, const char *name, int name_len) { assert(name_len > 0); if (name_len >= p->n_keyword_lists || p->keywords[name_len] == NULL || p->keywords[name_len]->type == -1) { return NAME; } for (KeywordToken *k = p->keywords[name_len]; k != NULL && k->type != -1; k++) { if (strncmp(k->str, name, name_len) == 0) { return k->type; } } return NAME; } static int growable_comment_array_init(growable_comment_array *arr, size_t initial_size) { assert(initial_size > 0); arr->items = PyMem_Malloc(initial_size * sizeof(*arr->items)); arr->size = initial_size; arr->num_items = 0; return arr->items != NULL; } static int growable_comment_array_add(growable_comment_array *arr, int lineno, char *comment) { if (arr->num_items >= arr->size) { size_t new_size = arr->size * 2; void *new_items_array = PyMem_Realloc(arr->items, new_size * sizeof(*arr->items)); if (!new_items_array) { return 0; } arr->items = new_items_array; arr->size = new_size; } arr->items[arr->num_items].lineno = lineno; arr->items[arr->num_items].comment = comment; // Take ownership arr->num_items++; return 1; } static void growable_comment_array_deallocate(growable_comment_array *arr) { for (unsigned i = 0; i < arr->num_items; i++) { PyMem_Free(arr->items[i].comment); } PyMem_Free(arr->items); } int _PyPegen_fill_token(Parser *p) { const char *start; const char *end; int type = PyTokenizer_Get(p->tok, &start, &end); // Record and skip '# type: ignore' comments while (type == TYPE_IGNORE) { Py_ssize_t len = end - start; char *tag = PyMem_Malloc(len + 1); if (tag == NULL) { PyErr_NoMemory(); return -1; } strncpy(tag, start, len); tag[len] = '\0'; // Ownership of tag passes to the growable array if (!growable_comment_array_add(&p->type_ignore_comments, p->tok->lineno, tag)) { PyErr_NoMemory(); return -1; } type = PyTokenizer_Get(p->tok, &start, &end); } if (type == ENDMARKER && p->start_rule == Py_single_input && p->parsing_started) { type = NEWLINE; /* Add an extra newline */ p->parsing_started = 0; if (p->tok->indent && !(p->flags & PyPARSE_DONT_IMPLY_DEDENT)) { p->tok->pendin = -p->tok->indent; p->tok->indent = 0; } } else { p->parsing_started = 1; } if (p->fill == p->size) { int newsize = p->size * 2; Token **new_tokens = PyMem_Realloc(p->tokens, newsize * sizeof(Token *)); if (new_tokens == NULL) { PyErr_NoMemory(); return -1; } p->tokens = new_tokens; for (int i = p->size; i < newsize; i++) { p->tokens[i] = PyMem_Malloc(sizeof(Token)); if (p->tokens[i] == NULL) { p->size = i; // Needed, in order to cleanup correctly after parser fails PyErr_NoMemory(); return -1; } memset(p->tokens[i], '\0', sizeof(Token)); } p->size = newsize; } Token *t = p->tokens[p->fill]; t->type = (type == NAME) ? _get_keyword_or_name_type(p, start, (int)(end - start)) : type; t->bytes = PyBytes_FromStringAndSize(start, end - start); if (t->bytes == NULL) { return -1; } PyArena_AddPyObject(p->arena, t->bytes); int lineno = type == STRING ? p->tok->first_lineno : p->tok->lineno; const char *line_start = type == STRING ? p->tok->multi_line_start : p->tok->line_start; int end_lineno = p->tok->lineno; int col_offset = -1; int end_col_offset = -1; if (start != NULL && start >= line_start) { col_offset = (int)(start - line_start); } if (end != NULL && end >= p->tok->line_start) { end_col_offset = (int)(end - p->tok->line_start); } t->lineno = p->starting_lineno + lineno; t->col_offset = p->tok->lineno == 1 ? p->starting_col_offset + col_offset : col_offset; t->end_lineno = p->starting_lineno + end_lineno; t->end_col_offset = p->tok->lineno == 1 ? p->starting_col_offset + end_col_offset : end_col_offset; p->fill += 1; if (type == ERRORTOKEN) { if (p->tok->done == E_DECODE) { return raise_decode_error(p); } return tokenizer_error(p); } return 0; } // Instrumentation to count the effectiveness of memoization. // The array counts the number of tokens skipped by memoization, // indexed by type. #define NSTATISTICS 2000 static long memo_statistics[NSTATISTICS]; void _PyPegen_clear_memo_statistics() { for (int i = 0; i < NSTATISTICS; i++) { memo_statistics[i] = 0; } } PyObject * _PyPegen_get_memo_statistics() { PyObject *ret = PyList_New(NSTATISTICS); if (ret == NULL) { return NULL; } for (int i = 0; i < NSTATISTICS; i++) { PyObject *value = PyLong_FromLong(memo_statistics[i]); if (value == NULL) { Py_DECREF(ret); return NULL; } // PyList_SetItem borrows a reference to value. if (PyList_SetItem(ret, i, value) < 0) { Py_DECREF(ret); return NULL; } } return ret; } int // bool _PyPegen_is_memoized(Parser *p, int type, void *pres) { if (p->mark == p->fill) { if (_PyPegen_fill_token(p) < 0) { p->error_indicator = 1; return -1; } } Token *t = p->tokens[p->mark]; for (Memo *m = t->memo; m != NULL; m = m->next) { if (m->type == type) { if (0 <= type && type < NSTATISTICS) { long count = m->mark - p->mark; // A memoized negative result counts for one. if (count <= 0) { count = 1; } memo_statistics[type] += count; } p->mark = m->mark; *(void **)(pres) = m->node; return 1; } } return 0; } int _PyPegen_lookahead_with_name(int positive, expr_ty (func)(Parser *), Parser *p) { int mark = p->mark; void *res = func(p); p->mark = mark; return (res != NULL) == positive; } int _PyPegen_lookahead_with_string(int positive, expr_ty (func)(Parser *, const char*), Parser *p, const char* arg) { int mark = p->mark; void *res = func(p, arg); p->mark = mark; return (res != NULL) == positive; } int _PyPegen_lookahead_with_int(int positive, Token *(func)(Parser *, int), Parser *p, int arg) { int mark = p->mark; void *res = func(p, arg); p->mark = mark; return (res != NULL) == positive; } int _PyPegen_lookahead(int positive, void *(func)(Parser *), Parser *p) { int mark = p->mark; void *res = (void*)func(p); p->mark = mark; return (res != NULL) == positive; } Token * _PyPegen_expect_token(Parser *p, int type) { if (p->mark == p->fill) { if (_PyPegen_fill_token(p) < 0) { p->error_indicator = 1; return NULL; } } Token *t = p->tokens[p->mark]; if (t->type != type) { return NULL; } p->mark += 1; return t; } expr_ty _PyPegen_expect_soft_keyword(Parser *p, const char *keyword) { if (p->mark == p->fill) { if (_PyPegen_fill_token(p) < 0) { p->error_indicator = 1; return NULL; } } Token *t = p->tokens[p->mark]; if (t->type != NAME) { return NULL; } char *s = PyBytes_AsString(t->bytes); if (!s) { p->error_indicator = 1; return NULL; } if (strcmp(s, keyword) != 0) { return NULL; } return _PyPegen_name_token(p); } Token * _PyPegen_get_last_nonnwhitespace_token(Parser *p) { assert(p->mark >= 0); Token *token = NULL; for (int m = p->mark - 1; m >= 0; m--) { token = p->tokens[m]; if (token->type != ENDMARKER && (token->type < NEWLINE || token->type > DEDENT)) { break; } } return token; } expr_ty _PyPegen_name_token(Parser *p) { Token *t = _PyPegen_expect_token(p, NAME); if (t == NULL) { return NULL; } char* s = PyBytes_AsString(t->bytes); if (!s) { p->error_indicator = 1; return NULL; } PyObject *id = _PyPegen_new_identifier(p, s); if (id == NULL) { p->error_indicator = 1; return NULL; } return Name(id, Load, t->lineno, t->col_offset, t->end_lineno, t->end_col_offset, p->arena); } void * _PyPegen_string_token(Parser *p) { return _PyPegen_expect_token(p, STRING); } static PyObject * parsenumber_raw(const char *s) { const char *end; long x; double dx; Py_complex compl; int imflag; assert(s != NULL); errno = 0; end = s + strlen(s) - 1; imflag = *end == 'j' || *end == 'J'; if (s[0] == '0') { x = (long)PyOS_strtoul(s, (char **)&end, 0); if (x < 0 && errno == 0) { return PyLong_FromString(s, (char **)0, 0); } } else { x = PyOS_strtol(s, (char **)&end, 0); } if (*end == '\0') { if (errno != 0) { return PyLong_FromString(s, (char **)0, 0); } return PyLong_FromLong(x); } /* XXX Huge floats may silently fail */ if (imflag) { compl.real = 0.; compl.imag = PyOS_string_to_double(s, (char **)&end, NULL); if (compl.imag == -1.0 && PyErr_Occurred()) { return NULL; } return PyComplex_FromCComplex(compl); } dx = PyOS_string_to_double(s, NULL, NULL); if (dx == -1.0 && PyErr_Occurred()) { return NULL; } return PyFloat_FromDouble(dx); } static PyObject * parsenumber(const char *s) { char *dup; char *end; PyObject *res = NULL; assert(s != NULL); if (strchr(s, '_') == NULL) { return parsenumber_raw(s); } /* Create a duplicate without underscores. */ dup = PyMem_Malloc(strlen(s) + 1); if (dup == NULL) { return PyErr_NoMemory(); } end = dup; for (; *s; s++) { if (*s != '_') { *end++ = *s; } } *end = '\0'; res = parsenumber_raw(dup); PyMem_Free(dup); return res; } expr_ty _PyPegen_number_token(Parser *p) { Token *t = _PyPegen_expect_token(p, NUMBER); if (t == NULL) { return NULL; } char *num_raw = PyBytes_AsString(t->bytes); if (num_raw == NULL) { p->error_indicator = 1; return NULL; } if (p->feature_version < 6 && strchr(num_raw, '_') != NULL) { p->error_indicator = 1; return RAISE_SYNTAX_ERROR("Underscores in numeric literals are only supported " "in Python 3.6 and greater"); } PyObject *c = parsenumber(num_raw); if (c == NULL) { p->error_indicator = 1; return NULL; } if (PyArena_AddPyObject(p->arena, c) < 0) { Py_DECREF(c); p->error_indicator = 1; return NULL; } return Constant(c, NULL, t->lineno, t->col_offset, t->end_lineno, t->end_col_offset, p->arena); } static int // bool newline_in_string(Parser *p, const char *cur) { for (const char *c = cur; c >= p->tok->buf; c--) { if (*c == '\'' || *c == '"') { return 1; } } return 0; } /* Check that the source for a single input statement really is a single statement by looking at what is left in the buffer after parsing. Trailing whitespace and comments are OK. */ static int // bool bad_single_statement(Parser *p) { const char *cur = strchr(p->tok->buf, '\n'); /* Newlines are allowed if preceded by a line continuation character or if they appear inside a string. */ if (!cur || *(cur - 1) == '\\' || newline_in_string(p, cur)) { return 0; } char c = *cur; for (;;) { while (c == ' ' || c == '\t' || c == '\n' || c == '\014') { c = *++cur; } if (!c) { return 0; } if (c != '#') { return 1; } /* Suck up comment. */ while (c && c != '\n') { c = *++cur; } } } void _PyPegen_Parser_Free(Parser *p) { Py_XDECREF(p->normalize); for (int i = 0; i < p->size; i++) { PyMem_Free(p->tokens[i]); } PyMem_Free(p->tokens); growable_comment_array_deallocate(&p->type_ignore_comments); PyMem_Free(p); } static int compute_parser_flags(PyCompilerFlags *flags) { int parser_flags = 0; if (!flags) { return 0; } if (flags->cf_flags & PyCF_DONT_IMPLY_DEDENT) { parser_flags |= PyPARSE_DONT_IMPLY_DEDENT; } if (flags->cf_flags & PyCF_IGNORE_COOKIE) { parser_flags |= PyPARSE_IGNORE_COOKIE; } if (flags->cf_flags & CO_FUTURE_BARRY_AS_BDFL) { parser_flags |= PyPARSE_BARRY_AS_BDFL; } if (flags->cf_flags & PyCF_TYPE_COMMENTS) { parser_flags |= PyPARSE_TYPE_COMMENTS; } if ((flags->cf_flags & PyCF_ONLY_AST) && flags->cf_feature_version < 7) { parser_flags |= PyPARSE_ASYNC_HACKS; } return parser_flags; } Parser * _PyPegen_Parser_New(struct tok_state *tok, int start_rule, int flags, int feature_version, int *errcode, PyArena *arena) { Parser *p = PyMem_Malloc(sizeof(Parser)); if (p == NULL) { return (Parser *) PyErr_NoMemory(); } assert(tok != NULL); tok->type_comments = (flags & PyPARSE_TYPE_COMMENTS) > 0; tok->async_hacks = (flags & PyPARSE_ASYNC_HACKS) > 0; p->tok = tok; p->keywords = NULL; p->n_keyword_lists = -1; p->tokens = PyMem_Malloc(sizeof(Token *)); if (!p->tokens) { PyMem_Free(p); return (Parser *) PyErr_NoMemory(); } p->tokens[0] = PyMem_Calloc(1, sizeof(Token)); if (!p->tokens) { PyMem_Free(p->tokens); PyMem_Free(p); return (Parser *) PyErr_NoMemory(); } if (!growable_comment_array_init(&p->type_ignore_comments, 10)) { PyMem_Free(p->tokens[0]); PyMem_Free(p->tokens); PyMem_Free(p); return (Parser *) PyErr_NoMemory(); } p->mark = 0; p->fill = 0; p->size = 1; p->errcode = errcode; p->arena = arena; p->start_rule = start_rule; p->parsing_started = 0; p->normalize = NULL; p->error_indicator = 0; p->starting_lineno = 0; p->starting_col_offset = 0; p->flags = flags; p->feature_version = feature_version; p->known_err_token = NULL; p->level = 0; return p; } void * _PyPegen_run_parser(Parser *p) { void *res = _PyPegen_parse(p); if (res == NULL) { if (PyErr_Occurred()) { return NULL; } if (p->fill == 0) { RAISE_SYNTAX_ERROR("error at start before reading any input"); } else if (p->tok->done == E_EOF) { RAISE_SYNTAX_ERROR("unexpected EOF while parsing"); } else { if (p->tokens[p->fill-1]->type == INDENT) { RAISE_INDENTATION_ERROR("unexpected indent"); } else if (p->tokens[p->fill-1]->type == DEDENT) { RAISE_INDENTATION_ERROR("unexpected unindent"); } else { RAISE_SYNTAX_ERROR("invalid syntax"); } } return NULL; } if (p->start_rule == Py_single_input && bad_single_statement(p)) { p->tok->done = E_BADSINGLE; // This is not necessary for now, but might be in the future return RAISE_SYNTAX_ERROR("multiple statements found while compiling a single statement"); } return res; } mod_ty _PyPegen_run_parser_from_file_pointer(FILE *fp, int start_rule, PyObject *filename_ob, const char *enc, const char *ps1, const char *ps2, PyCompilerFlags *flags, int *errcode, PyArena *arena) { struct tok_state *tok = PyTokenizer_FromFile(fp, enc, ps1, ps2); if (tok == NULL) { if (PyErr_Occurred()) { raise_tokenizer_init_error(filename_ob); return NULL; } return NULL; } // This transfers the ownership to the tokenizer tok->filename = filename_ob; Py_INCREF(filename_ob); // From here on we need to clean up even if there's an error mod_ty result = NULL; int parser_flags = compute_parser_flags(flags); Parser *p = _PyPegen_Parser_New(tok, start_rule, parser_flags, PY_MINOR_VERSION, errcode, arena); if (p == NULL) { goto error; } result = _PyPegen_run_parser(p); _PyPegen_Parser_Free(p); error: PyTokenizer_Free(tok); return result; } mod_ty _PyPegen_run_parser_from_file(const char *filename, int start_rule, PyObject *filename_ob, PyCompilerFlags *flags, PyArena *arena) { FILE *fp = fopen(filename, "rb"); if (fp == NULL) { PyErr_SetFromErrnoWithFilename(PyExc_OSError, filename); return NULL; } mod_ty result = _PyPegen_run_parser_from_file_pointer(fp, start_rule, filename_ob, NULL, NULL, NULL, flags, NULL, arena); fclose(fp); return result; } mod_ty _PyPegen_run_parser_from_string(const char *str, int start_rule, PyObject *filename_ob, PyCompilerFlags *flags, PyArena *arena) { int exec_input = start_rule == Py_file_input; struct tok_state *tok; if (flags == NULL || flags->cf_flags & PyCF_IGNORE_COOKIE) { tok = PyTokenizer_FromUTF8(str, exec_input); } else { tok = PyTokenizer_FromString(str, exec_input); } if (tok == NULL) { if (PyErr_Occurred()) { raise_tokenizer_init_error(filename_ob); } return NULL; } // This transfers the ownership to the tokenizer tok->filename = filename_ob; Py_INCREF(filename_ob); // We need to clear up from here on mod_ty result = NULL; int parser_flags = compute_parser_flags(flags); int feature_version = flags && (flags->cf_flags & PyCF_ONLY_AST) ? flags->cf_feature_version : PY_MINOR_VERSION; Parser *p = _PyPegen_Parser_New(tok, start_rule, parser_flags, feature_version, NULL, arena); if (p == NULL) { goto error; } result = _PyPegen_run_parser(p); _PyPegen_Parser_Free(p); error: PyTokenizer_Free(tok); return result; } void * _PyPegen_interactive_exit(Parser *p) { if (p->errcode) { *(p->errcode) = E_EOF; } return NULL; } /* Creates a single-element asdl_seq* that contains a */ asdl_seq * _PyPegen_singleton_seq(Parser *p, void *a) { assert(a != NULL); asdl_seq *seq = _Py_asdl_seq_new(1, p->arena); if (!seq) { return NULL; } asdl_seq_SET(seq, 0, a); return seq; } /* Creates a copy of seq and prepends a to it */ asdl_seq * _PyPegen_seq_insert_in_front(Parser *p, void *a, asdl_seq *seq) { assert(a != NULL); if (!seq) { return _PyPegen_singleton_seq(p, a); } asdl_seq *new_seq = _Py_asdl_seq_new(asdl_seq_LEN(seq) + 1, p->arena); if (!new_seq) { return NULL; } asdl_seq_SET(new_seq, 0, a); for (Py_ssize_t i = 1, l = asdl_seq_LEN(new_seq); i < l; i++) { asdl_seq_SET(new_seq, i, asdl_seq_GET(seq, i - 1)); } return new_seq; } /* Creates a copy of seq and appends a to it */ asdl_seq * _PyPegen_seq_append_to_end(Parser *p, asdl_seq *seq, void *a) { assert(a != NULL); if (!seq) { return _PyPegen_singleton_seq(p, a); } asdl_seq *new_seq = _Py_asdl_seq_new(asdl_seq_LEN(seq) + 1, p->arena); if (!new_seq) { return NULL; } for (Py_ssize_t i = 0, l = asdl_seq_LEN(new_seq); i + 1 < l; i++) { asdl_seq_SET(new_seq, i, asdl_seq_GET(seq, i)); } asdl_seq_SET(new_seq, asdl_seq_LEN(new_seq) - 1, a); return new_seq; } static Py_ssize_t _get_flattened_seq_size(asdl_seq *seqs) { Py_ssize_t size = 0; for (Py_ssize_t i = 0, l = asdl_seq_LEN(seqs); i < l; i++) { asdl_seq *inner_seq = asdl_seq_GET(seqs, i); size += asdl_seq_LEN(inner_seq); } return size; } /* Flattens an asdl_seq* of asdl_seq*s */ asdl_seq * _PyPegen_seq_flatten(Parser *p, asdl_seq *seqs) { Py_ssize_t flattened_seq_size = _get_flattened_seq_size(seqs); assert(flattened_seq_size > 0); asdl_seq *flattened_seq = _Py_asdl_seq_new(flattened_seq_size, p->arena); if (!flattened_seq) { return NULL; } int flattened_seq_idx = 0; for (Py_ssize_t i = 0, l = asdl_seq_LEN(seqs); i < l; i++) { asdl_seq *inner_seq = asdl_seq_GET(seqs, i); for (Py_ssize_t j = 0, li = asdl_seq_LEN(inner_seq); j < li; j++) { asdl_seq_SET(flattened_seq, flattened_seq_idx++, asdl_seq_GET(inner_seq, j)); } } assert(flattened_seq_idx == flattened_seq_size); return flattened_seq; } /* Creates a new name of the form . */ expr_ty _PyPegen_join_names_with_dot(Parser *p, expr_ty first_name, expr_ty second_name) { assert(first_name != NULL && second_name != NULL); PyObject *first_identifier = first_name->v.Name.id; PyObject *second_identifier = second_name->v.Name.id; if (PyUnicode_READY(first_identifier) == -1) { return NULL; } if (PyUnicode_READY(second_identifier) == -1) { return NULL; } const char *first_str = PyUnicode_AsUTF8(first_identifier); if (!first_str) { return NULL; } const char *second_str = PyUnicode_AsUTF8(second_identifier); if (!second_str) { return NULL; } Py_ssize_t len = strlen(first_str) + strlen(second_str) + 1; // +1 for the dot PyObject *str = PyBytes_FromStringAndSize(NULL, len); if (!str) { return NULL; } char *s = PyBytes_AS_STRING(str); if (!s) { return NULL; } strcpy(s, first_str); s += strlen(first_str); *s++ = '.'; strcpy(s, second_str); s += strlen(second_str); *s = '\0'; PyObject *uni = PyUnicode_DecodeUTF8(PyBytes_AS_STRING(str), PyBytes_GET_SIZE(str), NULL); Py_DECREF(str); if (!uni) { return NULL; } PyUnicode_InternInPlace(&uni); if (PyArena_AddPyObject(p->arena, uni) < 0) { Py_DECREF(uni); return NULL; } return _Py_Name(uni, Load, EXTRA_EXPR(first_name, second_name)); } /* Counts the total number of dots in seq's tokens */ int _PyPegen_seq_count_dots(asdl_seq *seq) { int number_of_dots = 0; for (Py_ssize_t i = 0, l = asdl_seq_LEN(seq); i < l; i++) { Token *current_expr = asdl_seq_GET(seq, i); switch (current_expr->type) { case ELLIPSIS: number_of_dots += 3; break; case DOT: number_of_dots += 1; break; default: Py_UNREACHABLE(); } } return number_of_dots; } /* Creates an alias with '*' as the identifier name */ alias_ty _PyPegen_alias_for_star(Parser *p) { PyObject *str = PyUnicode_InternFromString("*"); if (!str) { return NULL; } if (PyArena_AddPyObject(p->arena, str) < 0) { Py_DECREF(str); return NULL; } return alias(str, NULL, p->arena); } /* Creates a new asdl_seq* with the identifiers of all the names in seq */ asdl_seq * _PyPegen_map_names_to_ids(Parser *p, asdl_seq *seq) { Py_ssize_t len = asdl_seq_LEN(seq); assert(len > 0); asdl_seq *new_seq = _Py_asdl_seq_new(len, p->arena); if (!new_seq) { return NULL; } for (Py_ssize_t i = 0; i < len; i++) { expr_ty e = asdl_seq_GET(seq, i); asdl_seq_SET(new_seq, i, e->v.Name.id); } return new_seq; } /* Constructs a CmpopExprPair */ CmpopExprPair * _PyPegen_cmpop_expr_pair(Parser *p, cmpop_ty cmpop, expr_ty expr) { assert(expr != NULL); CmpopExprPair *a = PyArena_Malloc(p->arena, sizeof(CmpopExprPair)); if (!a) { return NULL; } a->cmpop = cmpop; a->expr = expr; return a; } asdl_int_seq * _PyPegen_get_cmpops(Parser *p, asdl_seq *seq) { Py_ssize_t len = asdl_seq_LEN(seq); assert(len > 0); asdl_int_seq *new_seq = _Py_asdl_int_seq_new(len, p->arena); if (!new_seq) { return NULL; } for (Py_ssize_t i = 0; i < len; i++) { CmpopExprPair *pair = asdl_seq_GET(seq, i); asdl_seq_SET(new_seq, i, pair->cmpop); } return new_seq; } asdl_seq * _PyPegen_get_exprs(Parser *p, asdl_seq *seq) { Py_ssize_t len = asdl_seq_LEN(seq); assert(len > 0); asdl_seq *new_seq = _Py_asdl_seq_new(len, p->arena); if (!new_seq) { return NULL; } for (Py_ssize_t i = 0; i < len; i++) { CmpopExprPair *pair = asdl_seq_GET(seq, i); asdl_seq_SET(new_seq, i, pair->expr); } return new_seq; } /* Creates an asdl_seq* where all the elements have been changed to have ctx as context */ static asdl_seq * _set_seq_context(Parser *p, asdl_seq *seq, expr_context_ty ctx) { Py_ssize_t len = asdl_seq_LEN(seq); if (len == 0) { return NULL; } asdl_seq *new_seq = _Py_asdl_seq_new(len, p->arena); if (!new_seq) { return NULL; } for (Py_ssize_t i = 0; i < len; i++) { expr_ty e = asdl_seq_GET(seq, i); asdl_seq_SET(new_seq, i, _PyPegen_set_expr_context(p, e, ctx)); } return new_seq; } static expr_ty _set_name_context(Parser *p, expr_ty e, expr_context_ty ctx) { return _Py_Name(e->v.Name.id, ctx, EXTRA_EXPR(e, e)); } static expr_ty _set_tuple_context(Parser *p, expr_ty e, expr_context_ty ctx) { return _Py_Tuple(_set_seq_context(p, e->v.Tuple.elts, ctx), ctx, EXTRA_EXPR(e, e)); } static expr_ty _set_list_context(Parser *p, expr_ty e, expr_context_ty ctx) { return _Py_List(_set_seq_context(p, e->v.List.elts, ctx), ctx, EXTRA_EXPR(e, e)); } static expr_ty _set_subscript_context(Parser *p, expr_ty e, expr_context_ty ctx) { return _Py_Subscript(e->v.Subscript.value, e->v.Subscript.slice, ctx, EXTRA_EXPR(e, e)); } static expr_ty _set_attribute_context(Parser *p, expr_ty e, expr_context_ty ctx) { return _Py_Attribute(e->v.Attribute.value, e->v.Attribute.attr, ctx, EXTRA_EXPR(e, e)); } static expr_ty _set_starred_context(Parser *p, expr_ty e, expr_context_ty ctx) { return _Py_Starred(_PyPegen_set_expr_context(p, e->v.Starred.value, ctx), ctx, EXTRA_EXPR(e, e)); } /* Creates an `expr_ty` equivalent to `expr` but with `ctx` as context */ expr_ty _PyPegen_set_expr_context(Parser *p, expr_ty expr, expr_context_ty ctx) { assert(expr != NULL); expr_ty new = NULL; switch (expr->kind) { case Name_kind: new = _set_name_context(p, expr, ctx); break; case Tuple_kind: new = _set_tuple_context(p, expr, ctx); break; case List_kind: new = _set_list_context(p, expr, ctx); break; case Subscript_kind: new = _set_subscript_context(p, expr, ctx); break; case Attribute_kind: new = _set_attribute_context(p, expr, ctx); break; case Starred_kind: new = _set_starred_context(p, expr, ctx); break; default: new = expr; } return new; } /* Constructs a KeyValuePair that is used when parsing a dict's key value pairs */ KeyValuePair * _PyPegen_key_value_pair(Parser *p, expr_ty key, expr_ty value) { KeyValuePair *a = PyArena_Malloc(p->arena, sizeof(KeyValuePair)); if (!a) { return NULL; } a->key = key; a->value = value; return a; } /* Extracts all keys from an asdl_seq* of KeyValuePair*'s */ asdl_seq * _PyPegen_get_keys(Parser *p, asdl_seq *seq) { Py_ssize_t len = asdl_seq_LEN(seq); asdl_seq *new_seq = _Py_asdl_seq_new(len, p->arena); if (!new_seq) { return NULL; } for (Py_ssize_t i = 0; i < len; i++) { KeyValuePair *pair = asdl_seq_GET(seq, i); asdl_seq_SET(new_seq, i, pair->key); } return new_seq; } /* Extracts all values from an asdl_seq* of KeyValuePair*'s */ asdl_seq * _PyPegen_get_values(Parser *p, asdl_seq *seq) { Py_ssize_t len = asdl_seq_LEN(seq); asdl_seq *new_seq = _Py_asdl_seq_new(len, p->arena); if (!new_seq) { return NULL; } for (Py_ssize_t i = 0; i < len; i++) { KeyValuePair *pair = asdl_seq_GET(seq, i); asdl_seq_SET(new_seq, i, pair->value); } return new_seq; } /* Constructs a NameDefaultPair */ NameDefaultPair * _PyPegen_name_default_pair(Parser *p, arg_ty arg, expr_ty value, Token *tc) { NameDefaultPair *a = PyArena_Malloc(p->arena, sizeof(NameDefaultPair)); if (!a) { return NULL; } a->arg = _PyPegen_add_type_comment_to_arg(p, arg, tc); a->value = value; return a; } /* Constructs a SlashWithDefault */ SlashWithDefault * _PyPegen_slash_with_default(Parser *p, asdl_seq *plain_names, asdl_seq *names_with_defaults) { SlashWithDefault *a = PyArena_Malloc(p->arena, sizeof(SlashWithDefault)); if (!a) { return NULL; } a->plain_names = plain_names; a->names_with_defaults = names_with_defaults; return a; } /* Constructs a StarEtc */ StarEtc * _PyPegen_star_etc(Parser *p, arg_ty vararg, asdl_seq *kwonlyargs, arg_ty kwarg) { StarEtc *a = PyArena_Malloc(p->arena, sizeof(StarEtc)); if (!a) { return NULL; } a->vararg = vararg; a->kwonlyargs = kwonlyargs; a->kwarg = kwarg; return a; } asdl_seq * _PyPegen_join_sequences(Parser *p, asdl_seq *a, asdl_seq *b) { Py_ssize_t first_len = asdl_seq_LEN(a); Py_ssize_t second_len = asdl_seq_LEN(b); asdl_seq *new_seq = _Py_asdl_seq_new(first_len + second_len, p->arena); if (!new_seq) { return NULL; } int k = 0; for (Py_ssize_t i = 0; i < first_len; i++) { asdl_seq_SET(new_seq, k++, asdl_seq_GET(a, i)); } for (Py_ssize_t i = 0; i < second_len; i++) { asdl_seq_SET(new_seq, k++, asdl_seq_GET(b, i)); } return new_seq; } static asdl_seq * _get_names(Parser *p, asdl_seq *names_with_defaults) { Py_ssize_t len = asdl_seq_LEN(names_with_defaults); asdl_seq *seq = _Py_asdl_seq_new(len, p->arena); if (!seq) { return NULL; } for (Py_ssize_t i = 0; i < len; i++) { NameDefaultPair *pair = asdl_seq_GET(names_with_defaults, i); asdl_seq_SET(seq, i, pair->arg); } return seq; } static asdl_seq * _get_defaults(Parser *p, asdl_seq *names_with_defaults) { Py_ssize_t len = asdl_seq_LEN(names_with_defaults); asdl_seq *seq = _Py_asdl_seq_new(len, p->arena); if (!seq) { return NULL; } for (Py_ssize_t i = 0; i < len; i++) { NameDefaultPair *pair = asdl_seq_GET(names_with_defaults, i); asdl_seq_SET(seq, i, pair->value); } return seq; } /* Constructs an arguments_ty object out of all the parsed constructs in the parameters rule */ arguments_ty _PyPegen_make_arguments(Parser *p, asdl_seq *slash_without_default, SlashWithDefault *slash_with_default, asdl_seq *plain_names, asdl_seq *names_with_default, StarEtc *star_etc) { asdl_seq *posonlyargs; if (slash_without_default != NULL) { posonlyargs = slash_without_default; } else if (slash_with_default != NULL) { asdl_seq *slash_with_default_names = _get_names(p, slash_with_default->names_with_defaults); if (!slash_with_default_names) { return NULL; } posonlyargs = _PyPegen_join_sequences(p, slash_with_default->plain_names, slash_with_default_names); if (!posonlyargs) { return NULL; } } else { posonlyargs = _Py_asdl_seq_new(0, p->arena); if (!posonlyargs) { return NULL; } } asdl_seq *posargs; if (plain_names != NULL && names_with_default != NULL) { asdl_seq *names_with_default_names = _get_names(p, names_with_default); if (!names_with_default_names) { return NULL; } posargs = _PyPegen_join_sequences(p, plain_names, names_with_default_names); if (!posargs) { return NULL; } } else if (plain_names == NULL && names_with_default != NULL) { posargs = _get_names(p, names_with_default); if (!posargs) { return NULL; } } else if (plain_names != NULL && names_with_default == NULL) { posargs = plain_names; } else { posargs = _Py_asdl_seq_new(0, p->arena); if (!posargs) { return NULL; } } asdl_seq *posdefaults; if (slash_with_default != NULL && names_with_default != NULL) { asdl_seq *slash_with_default_values = _get_defaults(p, slash_with_default->names_with_defaults); if (!slash_with_default_values) { return NULL; } asdl_seq *names_with_default_values = _get_defaults(p, names_with_default); if (!names_with_default_values) { return NULL; } posdefaults = _PyPegen_join_sequences(p, slash_with_default_values, names_with_default_values); if (!posdefaults) { return NULL; } } else if (slash_with_default == NULL && names_with_default != NULL) { posdefaults = _get_defaults(p, names_with_default); if (!posdefaults) { return NULL; } } else if (slash_with_default != NULL && names_with_default == NULL) { posdefaults = _get_defaults(p, slash_with_default->names_with_defaults); if (!posdefaults) { return NULL; } } else { posdefaults = _Py_asdl_seq_new(0, p->arena); if (!posdefaults) { return NULL; } } arg_ty vararg = NULL; if (star_etc != NULL && star_etc->vararg != NULL) { vararg = star_etc->vararg; } asdl_seq *kwonlyargs; if (star_etc != NULL && star_etc->kwonlyargs != NULL) { kwonlyargs = _get_names(p, star_etc->kwonlyargs); if (!kwonlyargs) { return NULL; } } else { kwonlyargs = _Py_asdl_seq_new(0, p->arena); if (!kwonlyargs) { return NULL; } } asdl_seq *kwdefaults; if (star_etc != NULL && star_etc->kwonlyargs != NULL) { kwdefaults = _get_defaults(p, star_etc->kwonlyargs); if (!kwdefaults) { return NULL; } } else { kwdefaults = _Py_asdl_seq_new(0, p->arena); if (!kwdefaults) { return NULL; } } arg_ty kwarg = NULL; if (star_etc != NULL && star_etc->kwarg != NULL) { kwarg = star_etc->kwarg; } return _Py_arguments(posonlyargs, posargs, vararg, kwonlyargs, kwdefaults, kwarg, posdefaults, p->arena); } /* Constructs an empty arguments_ty object, that gets used when a function accepts no * arguments. */ arguments_ty _PyPegen_empty_arguments(Parser *p) { asdl_seq *posonlyargs = _Py_asdl_seq_new(0, p->arena); if (!posonlyargs) { return NULL; } asdl_seq *posargs = _Py_asdl_seq_new(0, p->arena); if (!posargs) { return NULL; } asdl_seq *posdefaults = _Py_asdl_seq_new(0, p->arena); if (!posdefaults) { return NULL; } asdl_seq *kwonlyargs = _Py_asdl_seq_new(0, p->arena); if (!kwonlyargs) { return NULL; } asdl_seq *kwdefaults = _Py_asdl_seq_new(0, p->arena); if (!kwdefaults) { return NULL; } return _Py_arguments(posonlyargs, posargs, NULL, kwonlyargs, kwdefaults, NULL, kwdefaults, p->arena); } /* Encapsulates the value of an operator_ty into an AugOperator struct */ AugOperator * _PyPegen_augoperator(Parser *p, operator_ty kind) { AugOperator *a = PyArena_Malloc(p->arena, sizeof(AugOperator)); if (!a) { return NULL; } a->kind = kind; return a; } /* Construct a FunctionDef equivalent to function_def, but with decorators */ stmt_ty _PyPegen_function_def_decorators(Parser *p, asdl_seq *decorators, stmt_ty function_def) { assert(function_def != NULL); if (function_def->kind == AsyncFunctionDef_kind) { return _Py_AsyncFunctionDef( function_def->v.FunctionDef.name, function_def->v.FunctionDef.args, function_def->v.FunctionDef.body, decorators, function_def->v.FunctionDef.returns, function_def->v.FunctionDef.type_comment, function_def->lineno, function_def->col_offset, function_def->end_lineno, function_def->end_col_offset, p->arena); } return _Py_FunctionDef(function_def->v.FunctionDef.name, function_def->v.FunctionDef.args, function_def->v.FunctionDef.body, decorators, function_def->v.FunctionDef.returns, function_def->v.FunctionDef.type_comment, function_def->lineno, function_def->col_offset, function_def->end_lineno, function_def->end_col_offset, p->arena); } /* Construct a ClassDef equivalent to class_def, but with decorators */ stmt_ty _PyPegen_class_def_decorators(Parser *p, asdl_seq *decorators, stmt_ty class_def) { assert(class_def != NULL); return _Py_ClassDef(class_def->v.ClassDef.name, class_def->v.ClassDef.bases, class_def->v.ClassDef.keywords, class_def->v.ClassDef.body, decorators, class_def->lineno, class_def->col_offset, class_def->end_lineno, class_def->end_col_offset, p->arena); } /* Construct a KeywordOrStarred */ KeywordOrStarred * _PyPegen_keyword_or_starred(Parser *p, void *element, int is_keyword) { KeywordOrStarred *a = PyArena_Malloc(p->arena, sizeof(KeywordOrStarred)); if (!a) { return NULL; } a->element = element; a->is_keyword = is_keyword; return a; } /* Get the number of starred expressions in an asdl_seq* of KeywordOrStarred*s */ static int _seq_number_of_starred_exprs(asdl_seq *seq) { int n = 0; for (Py_ssize_t i = 0, l = asdl_seq_LEN(seq); i < l; i++) { KeywordOrStarred *k = asdl_seq_GET(seq, i); if (!k->is_keyword) { n++; } } return n; } /* Extract the starred expressions of an asdl_seq* of KeywordOrStarred*s */ asdl_seq * _PyPegen_seq_extract_starred_exprs(Parser *p, asdl_seq *kwargs) { int new_len = _seq_number_of_starred_exprs(kwargs); if (new_len == 0) { return NULL; } asdl_seq *new_seq = _Py_asdl_seq_new(new_len, p->arena); if (!new_seq) { return NULL; } int idx = 0; for (Py_ssize_t i = 0, len = asdl_seq_LEN(kwargs); i < len; i++) { KeywordOrStarred *k = asdl_seq_GET(kwargs, i); if (!k->is_keyword) { asdl_seq_SET(new_seq, idx++, k->element); } } return new_seq; } /* Return a new asdl_seq* with only the keywords in kwargs */ asdl_seq * _PyPegen_seq_delete_starred_exprs(Parser *p, asdl_seq *kwargs) { Py_ssize_t len = asdl_seq_LEN(kwargs); Py_ssize_t new_len = len - _seq_number_of_starred_exprs(kwargs); if (new_len == 0) { return NULL; } asdl_seq *new_seq = _Py_asdl_seq_new(new_len, p->arena); if (!new_seq) { return NULL; } int idx = 0; for (Py_ssize_t i = 0; i < len; i++) { KeywordOrStarred *k = asdl_seq_GET(kwargs, i); if (k->is_keyword) { asdl_seq_SET(new_seq, idx++, k->element); } } return new_seq; } expr_ty _PyPegen_concatenate_strings(Parser *p, asdl_seq *strings) { Py_ssize_t len = asdl_seq_LEN(strings); assert(len > 0); Token *first = asdl_seq_GET(strings, 0); Token *last = asdl_seq_GET(strings, len - 1); int bytesmode = 0; PyObject *bytes_str = NULL; FstringParser state; _PyPegen_FstringParser_Init(&state); for (Py_ssize_t i = 0; i < len; i++) { Token *t = asdl_seq_GET(strings, i); int this_bytesmode; int this_rawmode; PyObject *s; const char *fstr; Py_ssize_t fstrlen = -1; if (_PyPegen_parsestr(p, &this_bytesmode, &this_rawmode, &s, &fstr, &fstrlen, t) != 0) { goto error; } /* Check that we are not mixing bytes with unicode. */ if (i != 0 && bytesmode != this_bytesmode) { RAISE_SYNTAX_ERROR("cannot mix bytes and nonbytes literals"); Py_XDECREF(s); goto error; } bytesmode = this_bytesmode; if (fstr != NULL) { assert(s == NULL && !bytesmode); int result = _PyPegen_FstringParser_ConcatFstring(p, &state, &fstr, fstr + fstrlen, this_rawmode, 0, first, t, last); if (result < 0) { goto error; } } else { /* String or byte string. */ assert(s != NULL && fstr == NULL); assert(bytesmode ? PyBytes_CheckExact(s) : PyUnicode_CheckExact(s)); if (bytesmode) { if (i == 0) { bytes_str = s; } else { PyBytes_ConcatAndDel(&bytes_str, s); if (!bytes_str) { goto error; } } } else { /* This is a regular string. Concatenate it. */ if (_PyPegen_FstringParser_ConcatAndDel(&state, s) < 0) { goto error; } } } } if (bytesmode) { if (PyArena_AddPyObject(p->arena, bytes_str) < 0) { goto error; } return Constant(bytes_str, NULL, first->lineno, first->col_offset, last->end_lineno, last->end_col_offset, p->arena); } return _PyPegen_FstringParser_Finish(p, &state, first, last); error: Py_XDECREF(bytes_str); _PyPegen_FstringParser_Dealloc(&state); if (PyErr_Occurred()) { raise_decode_error(p); } return NULL; } mod_ty _PyPegen_make_module(Parser *p, asdl_seq *a) { asdl_seq *type_ignores = NULL; Py_ssize_t num = p->type_ignore_comments.num_items; if (num > 0) { // Turn the raw (comment, lineno) pairs into TypeIgnore objects in the arena type_ignores = _Py_asdl_seq_new(num, p->arena); if (type_ignores == NULL) { return NULL; } for (int i = 0; i < num; i++) { PyObject *tag = _PyPegen_new_type_comment(p, p->type_ignore_comments.items[i].comment); if (tag == NULL) { return NULL; } type_ignore_ty ti = TypeIgnore(p->type_ignore_comments.items[i].lineno, tag, p->arena); if (ti == NULL) { return NULL; } asdl_seq_SET(type_ignores, i, ti); } } return Module(a, type_ignores, p->arena); } // Error reporting helpers expr_ty _PyPegen_get_invalid_target(expr_ty e, TARGETS_TYPE targets_type) { if (e == NULL) { return NULL; } #define VISIT_CONTAINER(CONTAINER, TYPE) do { \ Py_ssize_t len = asdl_seq_LEN(CONTAINER->v.TYPE.elts);\ for (Py_ssize_t i = 0; i < len; i++) {\ expr_ty other = asdl_seq_GET(CONTAINER->v.TYPE.elts, i);\ expr_ty child = _PyPegen_get_invalid_target(other, targets_type);\ if (child != NULL) {\ return child;\ }\ }\ } while (0) // We only need to visit List and Tuple nodes recursively as those // are the only ones that can contain valid names in targets when // they are parsed as expressions. Any other kind of expression // that is a container (like Sets or Dicts) is directly invalid and // we don't need to visit it recursively. switch (e->kind) { case List_kind: VISIT_CONTAINER(e, List); return NULL; case Tuple_kind: VISIT_CONTAINER(e, Tuple); return NULL; case Starred_kind: if (targets_type == DEL_TARGETS) { return e; } return _PyPegen_get_invalid_target(e->v.Starred.value, targets_type); case Compare_kind: // This is needed, because the `a in b` in `for a in b` gets parsed // as a comparison, and so we need to search the left side of the comparison // for invalid targets. if (targets_type == FOR_TARGETS) { cmpop_ty cmpop = (cmpop_ty) asdl_seq_GET(e->v.Compare.ops, 0); if (cmpop == In) { return _PyPegen_get_invalid_target(e->v.Compare.left, targets_type); } return NULL; } return e; case Name_kind: case Subscript_kind: case Attribute_kind: return NULL; default: return e; } } void *_PyPegen_arguments_parsing_error(Parser *p, expr_ty e) { int kwarg_unpacking = 0; for (Py_ssize_t i = 0, l = asdl_seq_LEN(e->v.Call.keywords); i < l; i++) { keyword_ty keyword = asdl_seq_GET(e->v.Call.keywords, i); if (!keyword->arg) { kwarg_unpacking = 1; } } const char *msg = NULL; if (kwarg_unpacking) { msg = "positional argument follows keyword argument unpacking"; } else { msg = "positional argument follows keyword argument"; } return RAISE_SYNTAX_ERROR(msg); } void * _PyPegen_nonparen_genexp_in_call(Parser *p, expr_ty args) { /* The rule that calls this function is 'args for_if_clauses'. For the input f(L, x for x in y), L and x are in args and the for is parsed as a for_if_clause. We have to check if len <= 1, so that input like dict((a, b) for a, b in x) gets successfully parsed and then we pass the last argument (x in the above example) as the location of the error */ Py_ssize_t len = asdl_seq_LEN(args->v.Call.args); if (len <= 1) { return NULL; } return RAISE_SYNTAX_ERROR_KNOWN_LOCATION( (expr_ty) asdl_seq_GET(args->v.Call.args, len - 1), "Generator expression must be parenthesized" ); }