SRFI 250: Insertion-ordered hash tables

by John Cowan (shepherd, text), Will Clinger (text), Daphne Preston-Kendal (implementation)

Status

This SRFI is currently in draft status. Here is an explanation of each status that a SRFI can hold. To provide input on this SRFI, please send email to srfi-250@nospamsrfi.schemers.org. To subscribe to the list, follow these instructions. You can access previous messages via the mailing list archive.

Abstract

This SRFI defines an interface to hash tables, which are widely recognized as a fundamental data structure for a wide variety of applications. A hash table is a data structure that:

Is disjoint from all other types.

Provides a mapping from objects known as keys to corresponding objects known as values.

Keys may be any Scheme objects in some kinds of hash tables, but are restricted in other kinds.
Values may be any Scheme objects.

Provides an equality predicate which defines when a proposed key is the same as an existing key. No table may contain more than one value for a given key.

Provides a hash function which maps a candidate key into a non-negative exact integer.

Supports mutation as the primary means of setting the contents of a table.

Provides key lookup and destructive update in (expected) amortized constant time, provided that a satisfactory hash function is available.

Does not guarantee that whole-table operations work in the presence of concurrent mutation of the whole hash table. (Values may be safely mutated.)

Unlike the hash tables of SRFI 125, which is the direct ancestor of this specification, the hash tables described here are ordered by insertion: that is, associations inserted earlier in the history of the hash table appear earlier in the ordering. Advances in the implementations of hash tables, as provided by C++, Python, JavaScript, etc., make the provision of this new facility practical. As a result, the hash tables of this SRFI do not interoperate with the hash tables of SRFI 125, SRFI 126, or existing R6RS implementations.

Issues

It would be useful for hash-table-map to pass both the key and the value to proc, but that would make the API incompatible with SRFI 125. Rather than changing it in this way, a variant of hash-table-map could be added, under a different name, that passes both key and value.

There are no examples. SRFI 125, upon which this SRFI is based, didn't have examples.

Rationale

Hash tables themselves don't really need defending: almost all dynamically typed languages, from awk to JavaScript to Lua to Perl to Python to Common Lisp, and including many Scheme implementations, provide them in some form as a fundamental data structure. Therefore, what needs to be defended is not the data structure but the procedures. This SRFI supports a great many convenience procedures on top of the basic hash table interfaces provided by SRFI 69 and R6RS. Modulo the question of association ordering, nothing in it adds power to what those interfaces provide, but it does add convenience in the form of pre-debugged routines to do various common things, and even some things not so commonly done but useful.

This specification depends on SRFI 128 comparators, which package a type test, an equality predicate, and a hash function into a single bundle.

The relatively few hash table procedures in R6RS are all available in this SRFI under somewhat different names. This SRFI adopts SRFI 69's term hash-table rather than R6RS's hashtable, because of the universal use of "hash table" rather than "hashtable" in other computer languages and in technical prose generally. Besides, the English word hashtable obviously means something that can be ... hashted. It would be trivial to provide the R6RS names on top of this SRFI.

Common Lisp compatibility

As usual, the Common Lisp names are completely different from the Scheme names. Common Lisp provides the following capabilities that are not in this SRFI:

The constructor allows specifying the rehash size and rehash threshold of the new hash table. There are also accessors and mutators for these and for the current capacity (as opposed to size).

There are hash tables based on equalp (which does not exist in Scheme).

With-hash-table-iterator is a hash table external iterator implemented as a local macro.

Sxhash is an implementation-specific hash function for the equal predicate. It has the property that objects in different instantiations of the same Lisp implementation that are similar (a concept analogous to equal but defined across all instantiations of a Common Lisp program) always return the same value from sxhash; for example, the symbol xyz will have the same sxhash result in all instantiations.

Sources

The procedures in this SRFI are drawn primarily from SRFI 69 and R6RS. In addition, the following sources are acknowledged:

The hash-table-mutable? procedure and the second argument of hash-table-copy (which allows the creation of immutable hash tables) are from R6RS, renamed in the style of this SRFI.

The hash-table-intern! procedure is from Racket, renamed in the style of this SRFI.

The hash-table-find procedure is a modified version of table-search in Gambit.

The procedures hash-table-unfold and hash-table-count were suggested by SRFI 1.

The procedures hash-table=? and hash-table-map were suggested by Haskell's Data.Map.Strict module.

The procedure hash-table-map->list is from Guile.

The procedures hash-table-empty?, hash-table-empty-copy, hash-table-pop!, hash-table-map!, hash-table-intersection!, hash-table-difference!, and hash-table-xor! were added for convenience and completeness.

Pronunciation

Specification

The procedures in this SRFI are in the (srfi 250) library (or (srfi :250) on R6RS).

All references to "executing in expected amortized constant time" presuppose that a satisfactory hash function is available. Arbitrary or impure hash functions can make a hash of any implementation.

Hash tables are allowed to cache the results of calling the equality predicate and hash function, so programs cannot rely on the hash function being called exactly once for every primitive hash table operation: it may be called zero, one, or more times.

It is an error if the procedure argument of hash-table-find, hash-table-count, hash-table-map, hash-table-for-each, hash-table-map!, hash-table-map->list, hash-table-fold or hash-table-prune! mutates the hash table being walked.

It is an error to pass two hash tables that have different (in the sense of eq?) comparators to any of the procedures of this SRFI.

Implementations are permitted to ignore user-specified hash functions in certain circumstances. Specifically, if the equality predicate, whether passed as part of a comparator or explicitly, is more fine-grained (in the sense of R7RS-small section 6.1) than equal?, the implementation is free — indeed, is encouraged — to ignore the user-specified hash function and use something implementation-dependent. This allows the use of addresses as hashes, in which case the keys must be rehashed if they are moved by the garbage collector. Such a hash function is unsafe to use outside the context of implementation-provided hash tables. It can of course be exposed by an implementation as an extension, with suitable warnings against inappropriate uses.

It is an error to mutate a key during or after its insertion into a hash table in such a way that the hash function of the table will return a different result when applied to that key.

Index

Constructors: make-hash-table, hash-table, hash-table-unfold, alist->hash-table

Predicates: hash-table?, hash-table-contains?, hash-table-empty?, hash-table=?, hash-table-mutable?

Accessors: hash-table-ref, hash-table-ref/default, hash-table-comparator

Mutators: hash-table-set!, hash-table-delete!, hash-table-intern!, hash-table-update!, hash-table-update!/default, hash-table-pop!, hash-table-clear!

The whole hash table: hash-table-size, hash-table-keys, hash-table-values, hash-table-entries, hash-table-key-vector, hash-table-value-vector, hash-table-entry-vectors, hash-table-size, hash-table-keys, hash-table-values, hash-table-entries, hash-table-find, hash-table-count

Mapping and folding: hash-table-map, hash-table-for-each, hash-table-map!, hash-table-map->list, hash-table-fold, hash-table-prune!

Copying and conversion: hash-table-copy, hash-table-empty-copy, hash-table->alist

Hash tables as sets: hash-table-union!, hash-table-intersection!, hash-table-difference!, hash-table-xor!

Constructors

Note that the argument k is a positive integer representing the initial capacity of the hashtable being created (that is, the number of associations it can hold without having to grow). If not present, the initial capacity is implementation-dependent.

Returns a newly allocated hash table whose equality predicate and hash function are extracted from comparator.

As mentioned above, implementations are free to use an appropriate implementation-dependent hash function instead of the specified hash function, provided that the specified equality predicate is a refinement of the equal? predicate. This applies whether the hash function and equality predicate are passed as separate arguments or packaged up into a comparator.

The constraints on equality predicates and hash functions are given in SRFI 128.

(R6RS make-eq-hashtable, make-eqv-hashtable, and make-hashtable; Common Lisp make-hash-table)

Returns a newly allocated hash table, created as if by make-hash-table using comparator. For each pair of arguments, an association is added to the new hash table with key as its key and value as its value. This procedure returns an immutable hash table. If the same key (in the sense of the equality predicate) is specified more than once, it is an error.

Create a new hash table as if by make-hash-table using comparator and the args. If the result of applying the predicate stop? to seed is true, return the hash table. Otherwise, apply the procedure mapper to seed. Mapper returns two values, which are inserted into the hash table as the key and the value respectively. Then get a new seed by applying the procedure successor to seed, and repeat this algorithm.

Returns a newly allocated hash-table as if by make-hash-table using comparator and the args. It is then initialized from the associations of alist. Associations earlier in the list take precedence over those that come later.

Predicates

Returns #t if obj is a hash table, and #f otherwise. (R6RS hashtable?; Common Lisp hash-table-p)

Returns #t if there is any association to key in hash-table, and #f otherwise. Must execute in expected amortized constant time. (R6RS hashtable-contains?)

Returns #t if hash-table₁ and hash-table₂ have the same keys (in the sense of their common equality predicate) and each key has the same value (in the sense of value-comparator), and #f otherwise.

Accessors

Extracts the value associated to key in hash-table, invokes the procedure success on it, and returns its result; if success is not provided, then the value itself is returned. If key is not contained in hash-table and failure is supplied, then failure is invoked on no arguments and its result is returned. Otherwise, it is an error. Must execute in expected amortized constant time, not counting the time to call the procedures.

Returns a comparator equivalent to the comparator with which hash-table was created. R6RS hashtable-equivalence-function and hashtable-hash-function.

Mutators

The following procedures alter the associations in a hash table either unconditionally or conditionally on the presence or absence of a specified key. It is an error to add an association to a hash table whose key does not satisfy the type test predicate of the comparator used to create the hash table.

Repeatedly mutates hash-table, creating new associations in it by processing the arguments from left to right. The args alternate between keys and values. Whenever there is a previous association for a key, it is deleted. It is an error if the type check procedure of the comparator of hash-table, when invoked on a key, does not return #t. Likewise, it is an error if a key is not a valid argument to the equality predicate of hash-table. Returns an unspecified value. Must execute in expected amortized constant time per key. R6RS hashtable-set! and Common Lisp (setf gethash) do not handle multiple associations.

Deletes any association to each key in hash-table and returns the number of keys that had associations. Must execute in expected amortized constant time per key. R6RS hashtable-delete! and Common Lisp remhash do not handle multiple associations.

Effectively invokes hash-table-ref with the given arguments and returns what it returns. If key was not found in hash-table, its value is set to the result of calling failure. Must execute in expected amortized constant time.

Chooses the first association from hash-table and removes it, returning the key and value as two values.

Delete all the associations from hash-table. (R6RS hashtable-clear!; Common Lisp clrhash)

The whole hash table

Returns the number of associations in hash-table as an exact integer. Must execute in constant time. (R6RS hashtable-size; Common Lisp hash-table-count.)

Returns a newly allocated list/vector of all the keys in hash-table. R6RS hashtable-keys returns a vector.

(hash-table-values hash-table)
(hash-table-value-vector hash-table) R6RS hashtable-values returns a vector.

(hash-table-entries hash-table)
(hash-table-entry-vectors hash-table)

Returns two values, a newly allocated list/vector of all the keys in hash-table and a newly allocated list/vector of all the values in hash-table in the corresponding order. R6RS hash-table-entries returns vectors.

For each association of hash-table, invoke proc on its key and value. If proc returns true, then hash-table-find returns what proc returns. If all the calls to proc return #f, return the result of invoking the thunk failure.

For each association of hash-table, invoke pred on its key and value. Return the number of calls to pred which returned true.

Mapping and folding

Returns a newly allocated hash table as if by (make-hash-table comparator). Calls proc for every association in hash-table with the value of the association. The key of the association and the result of invoking proc are entered into the new hash table. Note that this is not the result of lifting mapping over the domain of hash tables, but it is considered more useful.

If comparator recognizes multiple keys in the hash-table as equivalent, any one of such associations is taken.

Calls proc for every association in hash-table with two arguments: the key of the association and the value of the association. The value returned by proc is discarded. Returns an unspecified value.

Calls proc for every association in hash-table with two arguments: the key of the association and the value of the association. The value returned by proc is used to update the value of the association. Returns an unspecified value.

Calls proc for every association in hash-table with two arguments: the key of the association and the value of the association. The values returned by the invocations of proc are accumulated into a list, which is returned.

Calls proc for every association in hash-table with three arguments: the key of the association, the value of the association, and an accumulated value val. Val is seed for the first invocation of procedure, and for subsequent invocations of proc, the returned value of the previous invocation. The value returned by hash-table-fold is the return value of the last invocation of proc.

Calls proc for every association in hash-table with two arguments, the key and the value of the association, and removes all associations from hash-table for which proc returns true. Returns an unspecified value.

Copying and conversion

Returns a newly allocated hash table with the same properties and associations as hash-table. If the second argument is present and is true, the new hash table is mutable. Otherwise it is immutable provided that the implementation supports immutable hash tables. (R6RS hashtable-copy)

Returns a newly allocated mutable hash table with the same properties as hash-table, but with no associations.

Hash tables as sets

Adds the associations of hash-table₂ to hash-table₁ and returns hash-table₁. If a key appears in both hash tables, its value is set to the value appearing in hash-table₁. Returns hash-table₁.

Deletes the associations from hash-table₁ whose keys don't also appear in hash-table₂ and returns hash-table₁.

Deletes the associations of hash-table₁ whose keys are also present in hash-table₂ and returns hash-table₁.

Deletes the associations of hash-table₁ whose keys are also present in hash-table₂, and then adds the associations of hash-table₂ whose keys are not present in hash-table₁ to hash-table₁. Returns hash-table₁.

Implementation

The current sample implementation is unfinished, and is here. When it is finished, it will be moved into this SRFI's repo. It relies upon SRFI 128.

Acknowledgements

Some of the language of this SRFI is copied from SRFI 69 with thanks to its author, Panu Kalliokoski. However, he is not responsible for what I have done with it.

I also acknowledge the members of the SRFI 250, 125, 126, and 128 mailing lists, especially Takashi Kato, Alex Shinn, Shiro Kawai, and Per Bothner.

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice (including the next paragraph) shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

250: Insertion-ordered hash tables