view src/dict.c @ 32936:c517845bd10e v9.0.1776

patch 9.0.1776: No support for stable Python 3 ABI Commit: https://github.com/vim/vim/commit/c13b3d1350b60b94fe87f0761ea31c0e7fb6ebf3 Author: Yee Cheng Chin <ychin.git@gmail.com> Date: Sun Aug 20 21:18:38 2023 +0200 patch 9.0.1776: No support for stable Python 3 ABI Problem: No support for stable Python 3 ABI Solution: Support Python 3 stable ABI Commits: 1) Support Python 3 stable ABI to allow mixed version interoperatbility Vim currently supports embedding Python for use with plugins, and the "dynamic" linking option allows the user to specify a locally installed version of Python by setting `pythonthreedll`. However, one caveat is that the Python 3 libs are not binary compatible across minor versions, and mixing versions can potentially be dangerous (e.g. let's say Vim was linked against the Python 3.10 SDK, but the user sets `pythonthreedll` to a 3.11 lib). Usually, nothing bad happens, but in theory this could lead to crashes, memory corruption, and other unpredictable behaviors. It's also difficult for the user to tell something is wrong because Vim has no way of reporting what Python 3 version Vim was linked with. For Vim installed via a package manager, this usually isn't an issue because all the dependencies would already be figured out. For prebuilt Vim binaries like MacVim (my motivation for working on this), AppImage, and Win32 installer this could potentially be an issue as usually a single binary is distributed. This is more tricky when a new Python version is released, as there's a chicken-and-egg issue with deciding what Python version to build against and hard to keep in sync when a new Python version just drops and we have a mix of users of different Python versions, and a user just blindly upgrading to a new Python could lead to bad interactions with Vim. Python 3 does have a solution for this problem: stable ABI / limited API (see https://docs.python.org/3/c-api/stable.html). The C SDK limits the API to a set of functions that are promised to be stable across versions. This pull request adds an ifdef config that allows us to turn it on when building Vim. Vim binaries built with this option should be safe to freely link with any Python 3 libraies without having the constraint of having to use the same minor version. Note: Python 2 has no such concept and this doesn't change how Python 2 integration works (not that there is going to be a new version of Python 2 that would cause compatibility issues in the future anyway). --- Technical details: ====== The stable ABI can be accessed when we compile with the Python 3 limited API (by defining `Py_LIMITED_API`). The Python 3 code (in `if_python3.c` and `if_py_both.h`) would now handle this and switch to limited API mode. Without it set, Vim will still use the full API as before so this is an opt-in change. The main difference is that `PyType_Object` is now an opaque struct that we can't directly create "static types" out of, and we have to create type objects as "heap types" instead. This is because the struct is not stable and changes from version to version (e.g. 3.8 added a `tp_vectorcall` field to it). I had to change all the types to be allocated on the heap instead with just a pointer to them. Other functions are also simply missing in limited API, or they are introduced too late (e.g. `PyUnicode_AsUTF8AndSize` in 3.10) to it that we need some other ways to do the same thing, so I had to abstract a few things into macros, and sometimes re-implement functions like `PyObject_NEW`. One caveat is that in limited API, `OutputType` (used for replacing `sys.stdout`) no longer inherits from `PyStdPrinter_Type` which I don't think has any real issue other than minor differences in how they convert to a string and missing a couple functions like `mode()` and `fileno()`. Also fixed an existing bug where `tp_basicsize` was set incorrectly for `BufferObject`, `TabListObject, `WinListObject`. Technically, there could be a small performance drop, there is a little more indirection with accessing type objects, and some APIs like `PyUnicode_AsUTF8AndSize` are missing, but in practice I didn't see any difference, and any well-written Python plugin should try to avoid excessing callbacks to the `vim` module in Python anyway. I only tested limited API mode down to Python 3.7, which seemes to compile and work fine. I haven't tried earlier Python versions. 2) Fix PyIter_Check on older Python vers / type##Ptr unused warning For PyIter_Check, older versions exposed them as either macros (used in full API), or a function (for use in limited API). A previous change exposed PyIter_Check to the dynamic build because Python just moved it to function-only in 3.10 anyway. Because of that, just make sure we always grab the function in dynamic builds in earlier versions since that's what Python eventually did anyway. 3) Move Py_LIMITED_API define to configure script Can now use --with-python-stable-abi flag to customize what stable ABI version to target. Can also use an env var to do so as well. 4) Show +python/dyn-stable in :version, and allow has() feature query Not sure if the "/dyn-stable" suffix would break things, or whether we should do it another way. Or just don't show it in version and rely on has() feature checking. 5) Documentation first draft. Still need to implement v:python3_version 6) Fix PyIter_Check build breaks when compiling against Python 3.8 7) Add CI coverage stable ABI on Linux/Windows / make configurable on Windows This adds configurable options for Windows make files (both MinGW and MSVC). CI will also now exercise both traditional full API and stable ABI for Linux and Windows in the matrix for coverage. Also added a "dynamic" option to Linux matrix as a drive-by change to make other scripting languages like Ruby / Perl testable under both static and dynamic builds. 8) Fix inaccuracy in Windows docs Python's own docs are confusing but you don't actually want to use `python3.dll` for the dynamic linkage. 9) Add generated autoconf file 10) Add v:python3_version support This variable indicates the version of Python3 that Vim was built against (PY_VERSION_HEX), and will be useful to check whether the Python library you are loading in dynamically actually fits it. When built with stable ABI, it will be the limited ABI version instead (`Py_LIMITED_API`), which indicates the minimum version of Python 3 the user should have, rather than the exact match. When stable ABI is used, we won't be exposing PY_VERSION_HEX in this var because it just doesn't seem necessary to do so (the whole point of stable ABI is the promise that it will work across versions), and I don't want to confuse the user with too many variables. Also, cleaned up some documentation, and added help tags. 11) Fix Python 3.7 compat issues Fix a couple issues when using limited API < 3.8 - Crash on exit: In Python 3.7, if a heap-allocated type is destroyed before all instances are, it would cause a crash later. This happens when we destroyed `OptionsType` before calling `Py_Finalize` when using the limited API. To make it worse, later versions changed the semantics and now each instance has a strong reference to its own type and the recommendation has changed to have each instance de-ref its own type and have its type in GC traversal. To avoid dealing with these cross-version variations, we just don't free the heap type. They are static types in non-limited-API anyway and are designed to last through the entirety of the app, and we also don't restart the Python runtime and therefore do not need it to have absolutely 0 leaks. See: - https://docs.python.org/3/whatsnew/3.8.html#changes-in-the-c-api - https://docs.python.org/3/whatsnew/3.9.html#changes-in-the-c-api - PyIter_Check: This function is not provided in limited APIs older than 3.8. Previously I was trying to mock it out using manual PyType_GetSlot() but it was brittle and also does not actually work properly for static types (it will generate a Python error). Just return false. It does mean using limited API < 3.8 is not recommended as you lose the functionality to handle iterators, but from playing with plugins I couldn't find it to be an issue. - Fix loading of PyIter_Check so it will be done when limited API < 3.8. Otherwise loading a 3.7 Python lib will fail even if limited API was specified to use it. 12) Make sure to only load `PyUnicode_AsUTF8AndSize` in needed in limited API We don't use this function unless limited API >= 3.10, but we were loading it regardless. Usually it's ok in Unix-like systems where Python just has a single lib that we load from, but in Windows where there is a separate python3.dll this would not work as the symbol would not have been exposed in this more limited DLL file. This makes it much clearer under what condition is this function needed. closes: #12032 Signed-off-by: Christian Brabandt <cb@256bit.org> Co-authored-by: Yee Cheng Chin <ychin.git@gmail.com>
author Christian Brabandt <cb@256bit.org>
date Sun, 20 Aug 2023 21:30:04 +0200
parents 7a1a9ce831c3
children aceaf677dd92
line wrap: on
line source

/* vi:set ts=8 sts=4 sw=4 noet:
 *
 * VIM - Vi IMproved	by Bram Moolenaar
 *
 * Do ":help uganda"  in Vim to read copying and usage conditions.
 * Do ":help credits" in Vim to see a list of people who contributed.
 * See README.txt for an overview of the Vim source code.
 */

/*
 * dict.c: Dictionary support
 */

#include "vim.h"

#if defined(FEAT_EVAL) || defined(PROTO)

// List head for garbage collection. Although there can be a reference loop
// from partial to dict to partial, we don't need to keep track of the partial,
// since it will get freed when the dict is unused and gets freed.
static dict_T		*first_dict = NULL;

/*
 * Allocate an empty header for a dictionary.
 * Caller should take care of the reference count.
 */
    dict_T *
dict_alloc(void)
{
    dict_T *d;

    d = ALLOC_CLEAR_ONE(dict_T);
    if (d == NULL)
	return NULL;

    // Add the dict to the list of dicts for garbage collection.
    if (first_dict != NULL)
	first_dict->dv_used_prev = d;
    d->dv_used_next = first_dict;
    d->dv_used_prev = NULL;
    first_dict = d;

    hash_init(&d->dv_hashtab);
    d->dv_lock = 0;
    d->dv_scope = 0;
    d->dv_refcount = 0;
    d->dv_copyID = 0;
    return d;
}

/*
 * dict_alloc() with an ID for alloc_fail().
 */
    dict_T *
dict_alloc_id(alloc_id_T id UNUSED)
{
#ifdef FEAT_EVAL
    if (alloc_fail_id == id && alloc_does_fail(sizeof(list_T)))
	return NULL;
#endif
    return (dict_alloc());
}

    dict_T *
dict_alloc_lock(int lock)
{
    dict_T *d = dict_alloc();

    if (d != NULL)
	d->dv_lock = lock;
    return d;
}

/*
 * Allocate an empty dict for a return value.
 * Returns OK or FAIL.
 */
    int
rettv_dict_alloc(typval_T *rettv)
{
    dict_T	*d = dict_alloc_lock(0);

    if (d == NULL)
	return FAIL;

    rettv_dict_set(rettv, d);
    return OK;
}

/*
 * Set a dictionary as the return value
 */
    void
rettv_dict_set(typval_T *rettv, dict_T *d)
{
    rettv->v_type = VAR_DICT;
    rettv->vval.v_dict = d;
    if (d != NULL)
	++d->dv_refcount;
}

/*
 * Free a Dictionary, including all non-container items it contains.
 * Ignores the reference count.
 */
    void
dict_free_contents(dict_T *d)
{
    hashtab_free_contents(&d->dv_hashtab);
    free_type(d->dv_type);
    d->dv_type = NULL;
}

/*
 * Clear hashtab "ht" and dict items it contains.
 * If "ht" is not freed then you should call hash_init() next!
 */
    void
hashtab_free_contents(hashtab_T *ht)
{
    int		todo;
    hashitem_T	*hi;
    dictitem_T	*di;

    if (check_hashtab_frozen(ht, "clear dict"))
	return;

    // Lock the hashtab, we don't want it to resize while freeing items.
    hash_lock(ht);
    todo = (int)ht->ht_used;
    FOR_ALL_HASHTAB_ITEMS(ht, hi, todo)
    {
	if (!HASHITEM_EMPTY(hi))
	{
	    // Remove the item before deleting it, just in case there is
	    // something recursive causing trouble.
	    di = HI2DI(hi);
	    hash_remove(ht, hi, "clear dict");
	    dictitem_free(di);
	    --todo;
	}
    }

    // The hashtab is still locked, it has to be re-initialized anyway.
    hash_clear(ht);
}

    static void
dict_free_dict(dict_T *d)
{
    // Remove the dict from the list of dicts for garbage collection.
    if (d->dv_used_prev == NULL)
	first_dict = d->dv_used_next;
    else
	d->dv_used_prev->dv_used_next = d->dv_used_next;
    if (d->dv_used_next != NULL)
	d->dv_used_next->dv_used_prev = d->dv_used_prev;
    vim_free(d);
}

    static void
dict_free(dict_T *d)
{
    if (!in_free_unref_items)
    {
	dict_free_contents(d);
	dict_free_dict(d);
    }
}

/*
 * Unreference a Dictionary: decrement the reference count and free it when it
 * becomes zero.
 */
    void
dict_unref(dict_T *d)
{
    if (d != NULL && --d->dv_refcount <= 0)
	dict_free(d);
}

/*
 * Go through the list of dicts and free items without the copyID.
 * Returns TRUE if something was freed.
 */
    int
dict_free_nonref(int copyID)
{
    dict_T	*dd;
    int		did_free = FALSE;

    for (dd = first_dict; dd != NULL; dd = dd->dv_used_next)
	if ((dd->dv_copyID & COPYID_MASK) != (copyID & COPYID_MASK))
	{
	    // Free the Dictionary and ordinary items it contains, but don't
	    // recurse into Lists and Dictionaries, they will be in the list
	    // of dicts or list of lists.
	    dict_free_contents(dd);
	    did_free = TRUE;
	}
    return did_free;
}

    void
dict_free_items(int copyID)
{
    dict_T	*dd, *dd_next;

    for (dd = first_dict; dd != NULL; dd = dd_next)
    {
	dd_next = dd->dv_used_next;
	if ((dd->dv_copyID & COPYID_MASK) != (copyID & COPYID_MASK))
	    dict_free_dict(dd);
    }
}

/*
 * Allocate a Dictionary item.
 * The "key" is copied to the new item.
 * Note that the type and value of the item "di_tv" still needs to be
 * initialized!
 * Returns NULL when out of memory.
 */
    dictitem_T *
dictitem_alloc(char_u *key)
{
    dictitem_T *di;
    size_t len = STRLEN(key);

    di = alloc(offsetof(dictitem_T, di_key) + len + 1);
    if (di == NULL)
	return NULL;

    mch_memmove(di->di_key, key, len + 1);
    di->di_flags = DI_FLAGS_ALLOC;
    di->di_tv.v_lock = 0;
    di->di_tv.v_type = VAR_UNKNOWN;
    return di;
}

/*
 * Make a copy of a Dictionary item.
 */
    static dictitem_T *
dictitem_copy(dictitem_T *org)
{
    dictitem_T *di;
    size_t	len = STRLEN(org->di_key);

    di = alloc(offsetof(dictitem_T, di_key) + len + 1);
    if (di == NULL)
	return NULL;

    mch_memmove(di->di_key, org->di_key, len + 1);
    di->di_flags = DI_FLAGS_ALLOC;
    copy_tv(&org->di_tv, &di->di_tv);
    return di;
}

/*
 * Remove item "item" from Dictionary "dict" and free it.
 * "command" is used for the error message when the hashtab if frozen.
 */
    void
dictitem_remove(dict_T *dict, dictitem_T *item, char *command)
{
    hashitem_T	*hi;

    hi = hash_find(&dict->dv_hashtab, item->di_key);
    if (HASHITEM_EMPTY(hi))
	internal_error("dictitem_remove()");
    else
	hash_remove(&dict->dv_hashtab, hi, command);
    dictitem_free(item);
}

/*
 * Free a dict item.  Also clears the value.
 */
    void
dictitem_free(dictitem_T *item)
{
    clear_tv(&item->di_tv);
    if (item->di_flags & DI_FLAGS_ALLOC)
	vim_free(item);
}

/*
 * Make a copy of dict "d".  Shallow if "deep" is FALSE.
 * The refcount of the new dict is set to 1.
 * See item_copy() for "top" and "copyID".
 * Returns NULL when out of memory.
 */
    dict_T *
dict_copy(dict_T *orig, int deep, int top, int copyID)
{
    dict_T	*copy;
    dictitem_T	*di;
    int		todo;
    hashitem_T	*hi;

    if (orig == NULL)
	return NULL;

    copy = dict_alloc();
    if (copy == NULL)
	return NULL;

    if (copyID != 0)
    {
	orig->dv_copyID = copyID;
	orig->dv_copydict = copy;
    }
    if (orig->dv_type == NULL || top || deep)
	copy->dv_type = NULL;
    else
	copy->dv_type = alloc_type(orig->dv_type);

    todo = (int)orig->dv_hashtab.ht_used;
    for (hi = orig->dv_hashtab.ht_array; todo > 0 && !got_int; ++hi)
    {
	if (!HASHITEM_EMPTY(hi))
	{
	    --todo;

	    di = dictitem_alloc(hi->hi_key);
	    if (di == NULL)
		break;
	    if (deep)
	    {
		if (item_copy(&HI2DI(hi)->di_tv, &di->di_tv,
			    deep, FALSE, copyID) == FAIL)
		{
		    vim_free(di);
		    break;
		}
	    }
	    else
		copy_tv(&HI2DI(hi)->di_tv, &di->di_tv);
	    if (dict_add(copy, di) == FAIL)
	    {
		dictitem_free(di);
		break;
	    }
	}
    }

    ++copy->dv_refcount;
    if (todo > 0)
    {
	dict_unref(copy);
	copy = NULL;
    }

    return copy;
}

/*
 * Check for adding a function to g: or s: (in Vim9 script) or l:.
 * If the name is wrong give an error message and return TRUE.
 */
    int
dict_wrong_func_name(dict_T *d, typval_T *tv, char_u *name)
{
    return (d == get_globvar_dict()
		|| (in_vim9script() && SCRIPT_ID_VALID(current_sctx.sc_sid)
		   && d == &SCRIPT_ITEM(current_sctx.sc_sid)->sn_vars->sv_dict)
		|| &d->dv_hashtab == get_funccal_local_ht())
	    && (tv->v_type == VAR_FUNC || tv->v_type == VAR_PARTIAL)
	    && var_wrong_func_name(name, TRUE);
}

/*
 * Add item "item" to Dictionary "d".
 * Returns FAIL when out of memory and when key already exists.
 */
    int
dict_add(dict_T *d, dictitem_T *item)
{
    if (dict_wrong_func_name(d, &item->di_tv, item->di_key))
	return FAIL;
    return hash_add(&d->dv_hashtab, item->di_key, "add to dictionary");
}

/*
 * Add a number or special entry to dictionary "d".
 * Returns FAIL when out of memory and when key already exists.
 */
    static int
dict_add_number_special(dict_T *d, char *key, varnumber_T nr, vartype_T vartype)
{
    dictitem_T	*item;

    item = dictitem_alloc((char_u *)key);
    if (item == NULL)
	return FAIL;
    item->di_tv.v_type = vartype;
    item->di_tv.vval.v_number = nr;
    if (dict_add(d, item) == FAIL)
    {
	dictitem_free(item);
	return FAIL;
    }
    return OK;
}

/*
 * Add a number entry to dictionary "d".
 * Returns FAIL when out of memory and when key already exists.
 */
    int
dict_add_number(dict_T *d, char *key, varnumber_T nr)
{
    return dict_add_number_special(d, key, nr, VAR_NUMBER);
}

/*
 * Add a special entry to dictionary "d".
 * Returns FAIL when out of memory and when key already exists.
 */
    int
dict_add_bool(dict_T *d, char *key, varnumber_T nr)
{
    return dict_add_number_special(d, key, nr, VAR_BOOL);
}

/*
 * Add a string entry to dictionary "d".
 * Returns FAIL when out of memory and when key already exists.
 */
    int
dict_add_string(dict_T *d, char *key, char_u *str)
{
    return dict_add_string_len(d, key, str, -1);
}

/*
 * Add a string entry to dictionary "d".
 * "str" will be copied to allocated memory.
 * When "len" is -1 use the whole string, otherwise only this many bytes.
 * Returns FAIL when out of memory and when key already exists.
 */
    int
dict_add_string_len(dict_T *d, char *key, char_u *str, int len)
{
    dictitem_T	*item;
    char_u	*val = NULL;

    item = dictitem_alloc((char_u *)key);
    if (item == NULL)
	return FAIL;
    item->di_tv.v_type = VAR_STRING;
    if (str != NULL)
    {
	if (len == -1)
	    val = vim_strsave(str);
	else
	    val = vim_strnsave(str, len);
    }
    item->di_tv.vval.v_string = val;
    if (dict_add(d, item) == FAIL)
    {
	dictitem_free(item);
	return FAIL;
    }
    return OK;
}

/*
 * Add a list entry to dictionary "d".
 * Returns FAIL when out of memory and when key already exists.
 */
    int
dict_add_list(dict_T *d, char *key, list_T *list)
{
    dictitem_T	*item;

    item = dictitem_alloc((char_u *)key);
    if (item == NULL)
	return FAIL;
    item->di_tv.v_type = VAR_LIST;
    item->di_tv.vval.v_list = list;
    ++list->lv_refcount;
    if (dict_add(d, item) == FAIL)
    {
	dictitem_free(item);
	return FAIL;
    }
    return OK;
}

/*
 * Add a typval_T entry to dictionary "d".
 * Returns FAIL when out of memory and when key already exists.
 */
    int
dict_add_tv(dict_T *d, char *key, typval_T *tv)
{
    dictitem_T	*item;

    item = dictitem_alloc((char_u *)key);
    if (item == NULL)
	return FAIL;
    copy_tv(tv, &item->di_tv);
    if (dict_add(d, item) == FAIL)
    {
	dictitem_free(item);
	return FAIL;
    }
    return OK;
}

/*
 * Add a callback to dictionary "d".
 * Returns FAIL when out of memory and when key already exists.
 */
    int
dict_add_callback(dict_T *d, char *key, callback_T *cb)
{
    dictitem_T	*item;

    item = dictitem_alloc((char_u *)key);
    if (item == NULL)
	return FAIL;
    put_callback(cb, &item->di_tv);
    if (dict_add(d, item) == FAIL)
    {
	dictitem_free(item);
	return FAIL;
    }
    return OK;
}

/*
 * Initializes "iter" for iterating over dictionary items with
 * dict_iterate_next().
 * If "var" is not a Dict or an empty Dict then there will be nothing to
 * iterate over, no error is given.
 * NOTE: The dictionary must not change until iterating is finished!
 */
    void
dict_iterate_start(typval_T *var, dict_iterator_T *iter)
{
    if (var->v_type != VAR_DICT || var->vval.v_dict == NULL)
	iter->dit_todo = 0;
    else
    {
	dict_T	*d = var->vval.v_dict;

	iter->dit_todo = d->dv_hashtab.ht_used;
	iter->dit_hi = d->dv_hashtab.ht_array;
    }
}

/*
 * Iterate over the items referred to by "iter".  It should be initialized with
 * dict_iterate_start().
 * Returns a pointer to the key.
 * "*tv_result" is set to point to the value for that key.
 * If there are no more items, NULL is returned.
 */
    char_u *
dict_iterate_next(dict_iterator_T *iter, typval_T **tv_result)
{
    dictitem_T	*di;
    char_u      *result;

    if (iter->dit_todo == 0)
	return NULL;

    while (HASHITEM_EMPTY(iter->dit_hi))
	++iter->dit_hi;

    di = HI2DI(iter->dit_hi);
    result = di->di_key;
    *tv_result = &di->di_tv;

    --iter->dit_todo;
    ++iter->dit_hi;
    return result;
}

/*
 * Add a dict entry to dictionary "d".
 * Returns FAIL when out of memory and when key already exists.
 */
    int
dict_add_dict(dict_T *d, char *key, dict_T *dict)
{
    dictitem_T	*item;

    item = dictitem_alloc((char_u *)key);
    if (item == NULL)
	return FAIL;
    item->di_tv.v_type = VAR_DICT;
    item->di_tv.vval.v_dict = dict;
    ++dict->dv_refcount;
    if (dict_add(d, item) == FAIL)
    {
	dictitem_free(item);
	return FAIL;
    }
    return OK;
}

/*
 * Get the number of items in a Dictionary.
 */
    long
dict_len(dict_T *d)
{
    if (d == NULL)
	return 0L;
    return (long)d->dv_hashtab.ht_used;
}

/*
 * Find item "key[len]" in Dictionary "d".
 * If "len" is negative use strlen(key).
 * Returns NULL when not found.
 */
    dictitem_T *
dict_find(dict_T *d, char_u *key, int len)
{
#define AKEYLEN 200
    char_u	buf[AKEYLEN];
    char_u	*akey;
    char_u	*tofree = NULL;
    hashitem_T	*hi;

    if (d == NULL)
	return NULL;
    if (len < 0)
	akey = key;
    else if (len >= AKEYLEN)
    {
	tofree = akey = vim_strnsave(key, len);
	if (akey == NULL)
	    return NULL;
    }
    else
    {
	// Avoid a malloc/free by using buf[].
	vim_strncpy(buf, key, len);
	akey = buf;
    }

    hi = hash_find(&d->dv_hashtab, akey);
    vim_free(tofree);
    if (HASHITEM_EMPTY(hi))
	return NULL;
    return HI2DI(hi);
}

/*
 * Returns TRUE if "key" is present in Dictionary "d".
 */
    int
dict_has_key(dict_T *d, char *key)
{
    return dict_find(d, (char_u *)key, -1) != NULL;
}

/*
 * Get a typval_T item from a dictionary and copy it into "rettv".
 * Returns FAIL if the entry doesn't exist or out of memory.
 */
    int
dict_get_tv(dict_T *d, char *key, typval_T *rettv)
{
    dictitem_T	*di;

    di = dict_find(d, (char_u *)key, -1);
    if (di == NULL)
	return FAIL;
    copy_tv(&di->di_tv, rettv);
    return OK;
}

/*
 * Get a string item from a dictionary.
 * When "save" is TRUE allocate memory for it.
 * When FALSE a shared buffer is used, can only be used once!
 * Returns NULL if the entry doesn't exist or out of memory.
 */
    char_u *
dict_get_string(dict_T *d, char *key, int save)
{
    dictitem_T	*di;
    char_u	*s;

    di = dict_find(d, (char_u *)key, -1);
    if (di == NULL)
	return NULL;
    s = tv_get_string(&di->di_tv);
    if (save && s != NULL)
	s = vim_strsave(s);
    return s;
}

/*
 * Get a number item from a dictionary.
 * Returns 0 if the entry doesn't exist.
 */
    varnumber_T
dict_get_number(dict_T *d, char *key)
{
    return dict_get_number_def(d, key, 0);
}

/*
 * Get a number item from a dictionary.
 * Returns "def" if the entry doesn't exist.
 */
    varnumber_T
dict_get_number_def(dict_T *d, char *key, int def)
{
    dictitem_T	*di;

    di = dict_find(d, (char_u *)key, -1);
    if (di == NULL)
	return def;
    return tv_get_number(&di->di_tv);
}

/*
 * Get a number item from a dictionary.
 * Returns 0 if the entry doesn't exist.
 * Give an error if the entry is not a number.
 */
    varnumber_T
dict_get_number_check(dict_T *d, char_u *key)
{
    dictitem_T	*di;

    di = dict_find(d, key, -1);
    if (di == NULL)
	return 0;
    if (di->di_tv.v_type != VAR_NUMBER)
    {
	semsg(_(e_invalid_argument_str), tv_get_string(&di->di_tv));
	return 0;
    }
    return tv_get_number(&di->di_tv);
}

/*
 * Get a bool item (number or true/false) from a dictionary.
 * Returns "def" if the entry doesn't exist.
 */
    varnumber_T
dict_get_bool(dict_T *d, char *key, int def)
{
    dictitem_T	*di;

    di = dict_find(d, (char_u *)key, -1);
    if (di == NULL)
	return def;
    return tv_get_bool(&di->di_tv);
}

/*
 * Return an allocated string with the string representation of a Dictionary.
 * May return NULL.
 */
    char_u *
dict2string(typval_T *tv, int copyID, int restore_copyID)
{
    garray_T	ga;
    int		first = TRUE;
    char_u	*tofree;
    char_u	numbuf[NUMBUFLEN];
    hashitem_T	*hi;
    char_u	*s;
    dict_T	*d;
    int		todo;

    if ((d = tv->vval.v_dict) == NULL)
	return NULL;
    ga_init2(&ga, sizeof(char), 80);
    ga_append(&ga, '{');

    todo = (int)d->dv_hashtab.ht_used;
    FOR_ALL_HASHTAB_ITEMS(&d->dv_hashtab, hi, todo)
    {
	if (!HASHITEM_EMPTY(hi))
	{
	    --todo;

	    if (first)
		first = FALSE;
	    else
		ga_concat(&ga, (char_u *)", ");

	    tofree = string_quote(hi->hi_key, FALSE);
	    if (tofree != NULL)
	    {
		ga_concat(&ga, tofree);
		vim_free(tofree);
	    }
	    ga_concat(&ga, (char_u *)": ");
	    s = echo_string_core(&HI2DI(hi)->di_tv, &tofree, numbuf, copyID,
						 FALSE, restore_copyID, TRUE);
	    if (s != NULL)
		ga_concat(&ga, s);
	    vim_free(tofree);
	    if (s == NULL || did_echo_string_emsg)
		break;
	    line_breakcheck();

	}
    }
    if (todo > 0)
    {
	vim_free(ga.ga_data);
	return NULL;
    }

    ga_append(&ga, '}');
    ga_append(&ga, NUL);
    return (char_u *)ga.ga_data;
}

/*
 * Advance over a literal key, including "-".  If the first character is not a
 * literal key character then "key" is returned.
 */
    static char_u *
skip_literal_key(char_u *key)
{
    char_u *p;

    for (p = key; ASCII_ISALNUM(*p) || *p == '_' || *p == '-'; ++p)
	;
    return p;
}

/*
 * Get the key for #{key: val} into "tv" and advance "arg".
 * Return FAIL when there is no valid key.
 */
    static int
get_literal_key_tv(char_u **arg, typval_T *tv)
{
    char_u *p = skip_literal_key(*arg);

    if (p == *arg)
	return FAIL;
    tv->v_type = VAR_STRING;
    tv->vval.v_string = vim_strnsave(*arg, p - *arg);

    *arg = p;
    return OK;
}

/*
 * Get a literal key for a Vim9 dict:
 * {"name": value},
 * {'name': value},
 * {name: value} use "name" as a literal key
 * Return the key in allocated memory or NULL in the case of an error.
 * "arg" is advanced to just after the key.
 */
    char_u *
get_literal_key(char_u **arg)
{
    char_u	*key;
    char_u	*end;
    typval_T	rettv;

    if (**arg == '\'')
    {
	if (eval_lit_string(arg, &rettv, TRUE, FALSE) == FAIL)
	    return NULL;
	key = rettv.vval.v_string;
    }
    else if (**arg == '"')
    {
	if (eval_string(arg, &rettv, TRUE, FALSE) == FAIL)
	    return NULL;
	key = rettv.vval.v_string;
    }
    else
    {
	end = skip_literal_key(*arg);
	if (end == *arg)
	{
	    semsg(_(e_invalid_key_str), *arg);
	    return NULL;
	}
	key = vim_strnsave(*arg, end - *arg);
	*arg = end;
    }
    return key;
}

/*
 * Allocate a variable for a Dictionary and fill it from "*arg".
 * "*arg" points to the "{".
 * "literal" is TRUE for #{key: val}
 * Return OK or FAIL.  Returns NOTDONE for {expr}.
 */
    int
eval_dict(char_u **arg, typval_T *rettv, evalarg_T *evalarg, int literal)
{
    int		evaluate = evalarg == NULL ? FALSE
				       : (evalarg->eval_flags & EVAL_EVALUATE);
    dict_T	*d = NULL;
    typval_T	tvkey;
    typval_T	tv;
    char_u	*key = NULL;
    dictitem_T	*item;
    char_u	*curly_expr = skipwhite(*arg + 1);
    char_u	buf[NUMBUFLEN];
    int		vim9script = in_vim9script();
    int		had_comma;

    // First check if it's not a curly-braces expression: {expr}.
    // Must do this without evaluating, otherwise a function may be called
    // twice.  Unfortunately this means we need to call eval1() twice for the
    // first item.
    // "{}" is an empty Dictionary.
    // "#{abc}" is never a curly-braces expression.
    if (!vim9script
	    && *curly_expr != '}'
	    && !literal
	    && eval1(&curly_expr, &tv, NULL) == OK
	    && *skipwhite(curly_expr) == '}')
	return NOTDONE;

    if (evaluate)
    {
	d = dict_alloc();
	if (d == NULL)
	    return FAIL;
    }
    tvkey.v_type = VAR_UNKNOWN;
    tv.v_type = VAR_UNKNOWN;

    *arg = skipwhite_and_linebreak(*arg + 1, evalarg);
    while (**arg != '}' && **arg != NUL)
    {
	int	has_bracket = vim9script && **arg == '[';

	if (literal)
	{
	    if (get_literal_key_tv(arg, &tvkey) == FAIL)
		goto failret;
	}
	else if (vim9script && !has_bracket)
	{
	    tvkey.vval.v_string = get_literal_key(arg);
	    if (tvkey.vval.v_string == NULL)
		goto failret;
	    tvkey.v_type = VAR_STRING;
	}
	else
	{
	    if (has_bracket)
		*arg = skipwhite(*arg + 1);
	    if (eval1(arg, &tvkey, evalarg) == FAIL)	// recursive!
		goto failret;
	    if (has_bracket)
	    {
		*arg = skipwhite(*arg);
		if (**arg != ']')
		{
		    emsg(_(e_missing_matching_bracket_after_dict_key));
		    clear_tv(&tvkey);
		    return FAIL;
		}
		++*arg;
	    }
	}

	// the colon should come right after the key, but this wasn't checked
	// previously, so only require it in Vim9 script.
	if (!vim9script)
	    *arg = skipwhite(*arg);
	if (**arg != ':')
	{
	    if (*skipwhite(*arg) == ':')
		semsg(_(e_no_white_space_allowed_before_str_str), ":", *arg);
	    else
		semsg(_(e_missing_colon_in_dictionary_str), *arg);
	    clear_tv(&tvkey);
	    goto failret;
	}
	if (evaluate)
	{
	    if (tvkey.v_type == VAR_FLOAT)
	    {
		tvkey.vval.v_string = typval_tostring(&tvkey, TRUE);
		tvkey.v_type = VAR_STRING;
	    }
	    key = tv_get_string_buf_chk(&tvkey, buf);
	    if (key == NULL)
	    {
		// "key" is NULL when tv_get_string_buf_chk() gave an errmsg
		clear_tv(&tvkey);
		goto failret;
	    }
	}
	if (vim9script && (*arg)[1] != NUL && !VIM_ISWHITE((*arg)[1]))
	{
	    semsg(_(e_white_space_required_after_str_str), ":", *arg);
	    clear_tv(&tvkey);
	    goto failret;
	}

	*arg = skipwhite_and_linebreak(*arg + 1, evalarg);
	if (eval1(arg, &tv, evalarg) == FAIL)	// recursive!
	{
	    if (evaluate)
		clear_tv(&tvkey);
	    goto failret;
	}
	if (evaluate)
	{
	    item = dict_find(d, key, -1);
	    if (item != NULL)
	    {
		semsg(_(e_duplicate_key_in_dictionary_str), key);
		clear_tv(&tvkey);
		clear_tv(&tv);
		goto failret;
	    }
	    item = dictitem_alloc(key);
	    if (item != NULL)
	    {
		item->di_tv = tv;
		item->di_tv.v_lock = 0;
		if (dict_add(d, item) == FAIL)
		    dictitem_free(item);
	    }
	}
	clear_tv(&tvkey);

	// the comma should come right after the value, but this wasn't checked
	// previously, so only require it in Vim9 script.
	if (!vim9script)
	    *arg = skipwhite(*arg);
	had_comma = **arg == ',';
	if (had_comma)
	{
	    if (vim9script && (*arg)[1] != NUL && !VIM_ISWHITE((*arg)[1]))
	    {
		semsg(_(e_white_space_required_after_str_str), ",", *arg);
		goto failret;
	    }
	    *arg = skipwhite(*arg + 1);
	}

	// the "}" can be on the next line
	*arg = skipwhite_and_linebreak(*arg, evalarg);
	if (**arg == '}')
	    break;
	if (!had_comma)
	{
	    if (**arg == ',')
		semsg(_(e_no_white_space_allowed_before_str_str), ",", *arg);
	    else
		semsg(_(e_missing_comma_in_dictionary_str), *arg);
	    goto failret;
	}
    }

    if (**arg != '}')
    {
	if (evalarg != NULL)
	    semsg(_(e_missing_dict_end_str), *arg);
failret:
	if (d != NULL)
	    dict_free(d);
	return FAIL;
    }

    *arg = *arg + 1;
    if (evaluate)
	rettv_dict_set(rettv, d);

    return OK;
}

/*
 * Go over all entries in "d2" and add them to "d1".
 * When "action" is "error" then a duplicate key is an error.
 * When "action" is "force" then a duplicate key is overwritten.
 * When "action" is "move" then move items instead of copying.
 * Otherwise duplicate keys are ignored ("action" is "keep").
 * "func_name" is used for reporting where an error occurred.
 */
    void
dict_extend(dict_T *d1, dict_T *d2, char_u *action, char *func_name)
{
    dictitem_T	*di1;
    int		todo;
    char_u	*arg_errmsg = (char_u *)N_("extend() argument");
    type_T	*type;

    if (check_hashtab_frozen(&d1->dv_hashtab, "extend"))
	return;

    if (*action == 'm')
    {
	if (check_hashtab_frozen(&d2->dv_hashtab, "extend"))
	    return;
	hash_lock(&d2->dv_hashtab);  // don't rehash on hash_remove()
    }

    if (d1->dv_type != NULL && d1->dv_type->tt_member != NULL)
	type = d1->dv_type->tt_member;
    else
	type = NULL;

    todo = (int)d2->dv_hashtab.ht_used;
    hashitem_T *hi2;
    FOR_ALL_HASHTAB_ITEMS(&d2->dv_hashtab, hi2, todo)
    {
	if (!HASHITEM_EMPTY(hi2))
	{
	    --todo;
	    di1 = dict_find(d1, hi2->hi_key, -1);
	    // Check the key to be valid when adding to any scope.
	    if (d1->dv_scope != 0 && !valid_varname(hi2->hi_key, -1, TRUE))
		break;

	    if (type != NULL
		     && check_typval_arg_type(type, &HI2DI(hi2)->di_tv,
							 func_name, 0) == FAIL)
		break;

	    if (di1 == NULL)
	    {
		if (*action == 'm')
		{
		    // Cheap way to move a dict item from "d2" to "d1".
		    // If dict_add() fails then "d2" won't be empty.
		    di1 = HI2DI(hi2);
		    if (dict_add(d1, di1) == OK)
			hash_remove(&d2->dv_hashtab, hi2, "extend");
		}
		else
		{
		    di1 = dictitem_copy(HI2DI(hi2));
		    if (di1 != NULL && dict_add(d1, di1) == FAIL)
			dictitem_free(di1);
		}
	    }
	    else if (*action == 'e')
	    {
		semsg(_(e_key_already_exists_str), hi2->hi_key);
		break;
	    }
	    else if (*action == 'f' && HI2DI(hi2) != di1)
	    {
		if (value_check_lock(di1->di_tv.v_lock, arg_errmsg, TRUE)
			|| var_check_ro(di1->di_flags, arg_errmsg, TRUE))
		    break;
		// Disallow replacing a builtin function.
		if (dict_wrong_func_name(d1, &HI2DI(hi2)->di_tv, hi2->hi_key))
		    break;
		clear_tv(&di1->di_tv);
		copy_tv(&HI2DI(hi2)->di_tv, &di1->di_tv);
	    }
	}
    }

    if (*action == 'm')
	hash_unlock(&d2->dv_hashtab);
}

/*
 * Return the dictitem that an entry in a hashtable points to.
 */
    dictitem_T *
dict_lookup(hashitem_T *hi)
{
    return HI2DI(hi);
}

/*
 * Return TRUE when two dictionaries have exactly the same key/values.
 */
    int
dict_equal(
    dict_T	*d1,
    dict_T	*d2,
    int		ic,	    // ignore case for strings
    int		recursive)  // TRUE when used recursively
{
    hashitem_T	*hi;
    dictitem_T	*item2;
    int		todo;

    if (d1 == d2)
	return TRUE;
    if (dict_len(d1) != dict_len(d2))
	return FALSE;
    if (dict_len(d1) == 0)
	// empty and NULL dicts are considered equal
	return TRUE;
    if (d1 == NULL || d2 == NULL)
	return FALSE;

    todo = (int)d1->dv_hashtab.ht_used;
    FOR_ALL_HASHTAB_ITEMS(&d1->dv_hashtab, hi, todo)
    {
	if (!HASHITEM_EMPTY(hi))
	{
	    item2 = dict_find(d2, hi->hi_key, -1);
	    if (item2 == NULL)
		return FALSE;
	    if (!tv_equal(&HI2DI(hi)->di_tv, &item2->di_tv, ic, recursive))
		return FALSE;
	    --todo;
	}
    }
    return TRUE;
}

/*
 * Count the number of times item "needle" occurs in Dict "d". Case is ignored
 * if "ic" is TRUE.
 */
    long
dict_count(dict_T *d, typval_T *needle, int ic)
{
    int		todo;
    hashitem_T	*hi;
    long	n = 0;

    if (d == NULL)
	return 0;

    todo = (int)d->dv_hashtab.ht_used;
    FOR_ALL_HASHTAB_ITEMS(&d->dv_hashtab, hi, todo)
    {
	if (!HASHITEM_EMPTY(hi))
	{
	    --todo;
	    if (tv_equal(&HI2DI(hi)->di_tv, needle, ic, FALSE))
		++n;
	}
    }

    return n;
}

/*
 * extend() a Dict. Append Dict argvars[1] to Dict argvars[0] and return the
 * resulting Dict in "rettv".  "is_new" is TRUE for extendnew().
 */
    void
dict_extend_func(
	typval_T	*argvars,
	type_T		*type,
	char		*func_name,
	char_u		*arg_errmsg,
	int		is_new,
	typval_T	*rettv)
{
    dict_T	*d1, *d2;
    char_u	*action;
    int	i;

    d1 = argvars[0].vval.v_dict;
    if (d1 == NULL)
    {
	emsg(_(e_cannot_extend_null_dict));
	return;
    }
    d2 = argvars[1].vval.v_dict;
    if (d2 == NULL)
	return;

    if (!is_new && value_check_lock(d1->dv_lock, arg_errmsg, TRUE))
	return;

    if (is_new)
    {
	d1 = dict_copy(d1, FALSE, TRUE, get_copyID());
	if (d1 == NULL)
	    return;
    }

    // Check the third argument.
    if (argvars[2].v_type != VAR_UNKNOWN)
    {
	static char *(av[]) = {"keep", "force", "error"};

	action = tv_get_string_chk(&argvars[2]);
	if (action == NULL)
	    return;
	for (i = 0; i < 3; ++i)
	    if (STRCMP(action, av[i]) == 0)
		break;
	if (i == 3)
	{
	    semsg(_(e_invalid_argument_str), action);
	    return;
	}
    }
    else
	action = (char_u *)"force";

    if (type != NULL && check_typval_arg_type(type, &argvars[1],
							 func_name, 2) == FAIL)
	return;
    dict_extend(d1, d2, action, func_name);

    if (is_new)
    {
	rettv->v_type = VAR_DICT;
	rettv->vval.v_dict = d1;
	rettv->v_lock = FALSE;
    }
    else
	copy_tv(&argvars[0], rettv);
}

/*
 * Implementation of map() and filter() for a Dict.  Apply "expr" to every
 * item in Dict "d" and return the result in "rettv".
 */
    void
dict_filter_map(
	dict_T		*d,
	filtermap_T	filtermap,
	type_T		*argtype,
	char		*func_name,
	char_u		*arg_errmsg,
	typval_T	*expr,
	typval_T	*rettv)
{
    dict_T	*d_ret = NULL;
    hashtab_T	*ht;
    hashitem_T	*hi;
    dictitem_T	*di;
    int		todo;
    int		rem;
    typval_T	newtv;
    funccall_T	*fc;

    if (filtermap == FILTERMAP_MAPNEW)
    {
	rettv->v_type = VAR_DICT;
	rettv->vval.v_dict = NULL;
    }
    if (d == NULL
	  || (filtermap == FILTERMAP_FILTER
			&& value_check_lock(d->dv_lock, arg_errmsg, TRUE)))
	return;

    if (filtermap == FILTERMAP_MAPNEW)
    {
	if (rettv_dict_alloc(rettv) == FAIL)
	    return;
	d_ret = rettv->vval.v_dict;
    }

    // Create one funccall_T for all eval_expr_typval() calls.
    fc = eval_expr_get_funccal(expr, &newtv);

    int prev_lock = d->dv_lock;
    if (d->dv_lock == 0)
	d->dv_lock = VAR_LOCKED;
    ht = &d->dv_hashtab;
    hash_lock(ht);
    todo = (int)ht->ht_used;
    FOR_ALL_HASHTAB_ITEMS(ht, hi, todo)
    {
	if (!HASHITEM_EMPTY(hi))
	{
	    int		r;

	    --todo;
	    di = HI2DI(hi);
	    if (filtermap == FILTERMAP_MAP
		    && (value_check_lock(di->di_tv.v_lock,
			    arg_errmsg, TRUE)
			|| var_check_ro(di->di_flags,
			    arg_errmsg, TRUE)))
		break;
	    set_vim_var_string(VV_KEY, di->di_key, -1);
	    newtv.v_type = VAR_UNKNOWN;
	    r = filter_map_one(&di->di_tv, expr, filtermap, fc, &newtv, &rem);
	    clear_tv(get_vim_var_tv(VV_KEY));
	    if (r == FAIL || did_emsg)
	    {
		clear_tv(&newtv);
		break;
	    }
	    if (filtermap == FILTERMAP_MAP)
	    {
		if (argtype != NULL && check_typval_arg_type(
			     argtype->tt_member, &newtv, func_name, 0) == FAIL)
		{
		    clear_tv(&newtv);
		    break;
		}
		// map(): replace the dict item value
		clear_tv(&di->di_tv);
		newtv.v_lock = 0;
		di->di_tv = newtv;
	    }
	    else if (filtermap == FILTERMAP_MAPNEW)
	    {
		// mapnew(): add the item value to the new dict
		r = dict_add_tv(d_ret, (char *)di->di_key, &newtv);
		clear_tv(&newtv);
		if (r == FAIL)
		    break;
	    }
	    else if (filtermap == FILTERMAP_FILTER && rem)
	    {
		// filter(false): remove the item from the dict
		if (var_check_fixed(di->di_flags, arg_errmsg, TRUE)
			|| var_check_ro(di->di_flags, arg_errmsg, TRUE))
		    break;
		dictitem_remove(d, di, "filter");
	    }
	}
    }
    hash_unlock(ht);
    d->dv_lock = prev_lock;
    if (fc != NULL)
	remove_funccal();
}

/*
 * "remove({dict})" function
 */
    void
dict_remove(typval_T *argvars, typval_T *rettv, char_u *arg_errmsg)
{
    dict_T	*d;
    char_u	*key;
    dictitem_T	*di;

    if (argvars[2].v_type != VAR_UNKNOWN)
    {
	semsg(_(e_too_many_arguments_for_function_str), "remove()");
	return;
    }

    d = argvars[0].vval.v_dict;
    if (d == NULL || value_check_lock(d->dv_lock, arg_errmsg, TRUE))
	return;

    key = tv_get_string_chk(&argvars[1]);
    if (key == NULL)
	return;

    di = dict_find(d, key, -1);
    if (di == NULL)
    {
	semsg(_(e_key_not_present_in_dictionary_str), key);
	return;
    }

    if (var_check_fixed(di->di_flags, arg_errmsg, TRUE)
	    || var_check_ro(di->di_flags, arg_errmsg, TRUE))
	return;

    *rettv = di->di_tv;
    init_tv(&di->di_tv);
    dictitem_remove(d, di, "remove()");
}

typedef enum {
    DICT2LIST_KEYS,
    DICT2LIST_VALUES,
    DICT2LIST_ITEMS,
} dict2list_T;

/*
 * Turn a dict into a list.
 */
    static void
dict2list(typval_T *argvars, typval_T *rettv, dict2list_T what)
{
    list_T	*l2;
    dictitem_T	*di;
    hashitem_T	*hi;
    listitem_T	*li;
    dict_T	*d;
    int		todo;

    if (rettv_list_alloc(rettv) == FAIL)
	return;

    if ((what == DICT2LIST_ITEMS
		? check_for_string_or_list_or_dict_arg(argvars, 0)
		: check_for_dict_arg(argvars, 0)) == FAIL)
	return;

    d = argvars[0].vval.v_dict;
    if (d == NULL)
	// NULL dict behaves like an empty dict
	return;

    todo = (int)d->dv_hashtab.ht_used;
    FOR_ALL_HASHTAB_ITEMS(&d->dv_hashtab, hi, todo)
    {
	if (!HASHITEM_EMPTY(hi))
	{
	    --todo;
	    di = HI2DI(hi);

	    li = listitem_alloc();
	    if (li == NULL)
		break;
	    list_append(rettv->vval.v_list, li);

	    if (what == DICT2LIST_KEYS)
	    {
		// keys()
		li->li_tv.v_type = VAR_STRING;
		li->li_tv.v_lock = 0;
		li->li_tv.vval.v_string = vim_strsave(di->di_key);
	    }
	    else if (what == DICT2LIST_VALUES)
	    {
		// values()
		copy_tv(&di->di_tv, &li->li_tv);
	    }
	    else
	    {
		// items()
		l2 = list_alloc();
		li->li_tv.v_type = VAR_LIST;
		li->li_tv.v_lock = 0;
		li->li_tv.vval.v_list = l2;
		if (l2 == NULL)
		    break;
		++l2->lv_refcount;

		if (list_append_string(l2, di->di_key, -1) == FAIL
			|| list_append_tv(l2, &di->di_tv) == FAIL)
		    break;
	    }
	}
    }
}

/*
 * "items(dict)" function
 */
    void
f_items(typval_T *argvars, typval_T *rettv)
{
    if (argvars[0].v_type == VAR_STRING)
	string2items(argvars, rettv);
    else if (argvars[0].v_type == VAR_LIST)
	list2items(argvars, rettv);
    else
	dict2list(argvars, rettv, DICT2LIST_ITEMS);
}

/*
 * "keys()" function
 */
    void
f_keys(typval_T *argvars, typval_T *rettv)
{
    dict2list(argvars, rettv, DICT2LIST_KEYS);
}

/*
 * "values(dict)" function
 */
    void
f_values(typval_T *argvars, typval_T *rettv)
{
    dict2list(argvars, rettv, DICT2LIST_VALUES);
}

/*
 * Make each item in the dict readonly (not the value of the item).
 */
    void
dict_set_items_ro(dict_T *di)
{
    int		todo = (int)di->dv_hashtab.ht_used;
    hashitem_T	*hi;

    // Set readonly
    FOR_ALL_HASHTAB_ITEMS(&di->dv_hashtab, hi, todo)
    {
	if (HASHITEM_EMPTY(hi))
	    continue;
	--todo;
	HI2DI(hi)->di_flags |= DI_FLAGS_RO | DI_FLAGS_FIX;
    }
}

/*
 * "has_key()" function
 */
    void
f_has_key(typval_T *argvars, typval_T *rettv)
{
    if (in_vim9script()
	    && (check_for_dict_arg(argvars, 0) == FAIL
		|| check_for_string_or_number_arg(argvars, 1) == FAIL))
	return;

    if (check_for_dict_arg(argvars, 0) == FAIL)
	return;

    if (argvars[0].vval.v_dict == NULL)
	return;

    rettv->vval.v_number = dict_has_key(argvars[0].vval.v_dict,
				(char *)tv_get_string(&argvars[1]));
}

#endif // defined(FEAT_EVAL)