Added the beginnings of hashmap but does not yet compile.

.vscode/settings.json

@@ -1,7 +0,0 @@
-{
-    "files.associations": {
-        "future": "cpp",
-        "system_error": "cpp",
-        "functional": "c"
-    }
-}

src/memory/conspage.h

@@ -13,7 +13,7 @@
 #ifndef __psse_conspage_h
 #define __psse_conspage_h
 
-#include "consspaceobject.h"
+#include "memory/consspaceobject.h"
 
 /**
  * the number of cons cells on a cons page. The maximum value this can

src/memory/consspaceobject.c

@@ -101,16 +101,16 @@ struct cons_pointer c_car( struct cons_pointer arg ) {
 struct cons_pointer c_cdr( struct cons_pointer arg ) {
     struct cons_pointer result = NIL;
 
-    struct cons_space_object cell = pointer2cell( arg );
+    struct cons_space_object *cell = &pointer2cell( arg );
 
-    switch (cell.tag.value) {
+    switch (cell->tag.value) {
         case CONSTV:
-        result = pointer2cell( arg ).payload.cons.cdr;
+        result = cell->payload.cons.cdr;
         break;
         case KEYTV:
         case STRINGTV:
         case SYMBOLTV:
-        result = pointer2cell( arg ).payload.string.cdr;
+        result = cell->payload.string.cdr;
         break;
     }
 
@@ -226,6 +226,36 @@ struct cons_pointer make_nlambda( struct cons_pointer args,
     return pointer;
 }
 
+/**
+ * Return a hash value for this string.
+ * 
+ * What's important here is that two strings with the same characters in the
+ * same order should have the same hash value, even if one was created using
+ * `"foobar"` and the other by `(append "foo" "bar")`. I *think* this function 
+ * has that property. I doubt that it's the most efficient hash function to 
+ * have that property.
+ */ 
+uint32_t calculate_hash( wint_t c, struct cons_pointer ptr) {
+struct cons_space_object *cell = &pointer2cell(ptr);
+    uint32_t result = 0;
+
+    switch (cell->tag.value)
+    {
+    case KEYTV:
+    case STRINGTV:
+    case SYMBOLTV:
+    if (nilp(ptr)) { 
+        result =(uint32_t) c; 
+    } else {
+        result = ((uint32_t)c * 
+            cell->payload.string.hash) & 
+            0xffffffff;
+    }
+    }
+
+    return result;
+}
+
 /**
  * Construct a string from this character (which later will be UTF) and
  * this tail. A string is implemented as a flat list of cells each of which
@@ -245,8 +275,10 @@ make_string_like_thing( wint_t c, struct cons_pointer tail, char *tag ) {
         cell->payload.string.cdr.page = tail.page;
         /* \todo There's a problem here. Sometimes the offsets on
          * strings are quite massively off. Fix is probably
-         * cell->payload.string.cdr = tsil */
+         * cell->payload.string.cdr = tail */
         cell->payload.string.cdr.offset = tail.offset;
+
+        cell->payload.string.hash = calculate_hash(c, tail);
     } else {
         // \todo should throw an exception!
         debug_printf( DEBUG_ALLOC,

src/memory/consspaceobject.h

@@ -20,7 +20,8 @@
 #include <wchar.h>
 #include <wctype.h>
 
-#include "fopen.h"
+#include "io/fopen.h"
+#include "memory/conspage.h"
 
 
 /**
@@ -557,8 +558,8 @@ struct stream_payload {
 struct string_payload {
     /** the actual character stored in this cell */
     wint_t character;
-    /** unused padding to word-align the cdr */
-    uint32_t padding;
+    /** a hash of the string value, computed at store time. */
+    uint32_t hash;
     /** the remainder of the string following this character. */
     struct cons_pointer cdr;
 };

src/memory/hashmap.c

@@ -0,0 +1,61 @@
+/*
+ * hashmap.c
+ *
+ * Basic implementation of a hashmap.
+ *
+ * (c) 2021 Simon Brooke <simon@journeyman.cc>
+ * Licensed under GPL version 2.0, or, at your option, any later version.
+ */
+
+#include "arith/integer.h"
+#include "memory/consspaceobject.h"
+#include "memory/hashmap.h"
+
+/**
+ * Get the hash value for the cell indicated by this `ptr`; currently only
+ * implemented for string like things.
+ */
+uint32_t get_hash(struct cons_pointer ptr)
+{
+    struct cons_space_object *cell = &pointer2cell(ptr);
+    uint32_t result = 0;
+
+    switch (cell->tag.value)
+    {
+    case KEYTV:
+    case STRINGTV:
+    case SYMBOLTV:
+        result = cell->payload.string.hash;
+    default:
+        // TODO: Not Yet Implemented
+        result = 0;
+    }
+
+    return result;
+}
+
+/**
+ * A lisp function signature conforming wrapper around get_hash, q.v..
+ */
+struct cons_pointer lisp_get_hash(struct stack_frame *frame,
+                                  struct cons_pointer frame_pointer,
+                                  struct cons_pointer env)
+{
+    return make_integer(get_hash(frame->arg[0]), NIL);
+}
+
+/**
+ * Make a hashmap with this number of buckets.
+ */
+struct cons_pointer make_hashmap( uint32_t n_buckets) {
+    struct cons_pointer result = make_vso(HASHTAG,
+    (sizeof(struct cons_pointer) * (n_buckets + 1)) +
+    (sizeof(uint32_t) * 2));
+
+    //  TODO: fill in the payload!
+
+    struct hashmap_payload *payload  = 
+        (struct hashmap_payload *) &pointer_to_vso(result)->payload;
+
+    return result;
+}

src/memory/hashmap.h

@@ -0,0 +1,36 @@
+/*
+ * hashmap.h
+ *
+ * Basic implementation of a hashmap.
+ *
+ * (c) 2021 Simon Brooke <simon@journeyman.cc>
+ * Licensed under GPL version 2.0, or, at your option, any later version.
+ */
+
+#ifndef __psse_hashmap_h
+#define __psse_hashmap_h
+
+#include "memory/consspaceobject.h"
+#include "memory/vectorspace.h"
+
+/**
+ * The payload of a hashmap. The number of buckets is assigned at run-time,
+ * and is stored in n_buckets. Each bucket is something ASSOC can consume: 
+ * i.e. either an assoc list or a further hashmap.
+ */
+struct hashmap_payload {
+    struct cons_pointer hash_fn;
+    uint32_t n_buckets;
+    uint32_t unused; /* for word alignment and possible later expansion */
+    struct cons_pointer buckets[];
+};
+
+uint32_t get_hash(struct cons_pointer ptr);
+
+struct cons_pointer lisp_get_hash(struct stack_frame *frame,
+                                  struct cons_pointer frame_pointer,
+                                  struct cons_pointer env);
+
+struct cons_pointer make_hashmap( uint32_t n_buckets, struct cons_pointer hash_fn);
+
+#endif

src/ops/equal.c

@@ -10,11 +10,11 @@
 #include <math.h>
 #include <stdbool.h>
 
-#include "conspage.h"
-#include "consspaceobject.h"
-#include "integer.h"
-#include "peano.h"
-#include "ratio.h"
+#include "memory/conspage.h"
+#include "memory/consspaceobject.h"
+#include "arith/integer.h"
+#include "arith/peano.h"
+#include "arith/ratio.h"
 
 /**
  * Shallow, and thus cheap, equality: true if these two objects are
@@ -69,6 +69,9 @@ bool equal(struct cons_pointer a, struct cons_pointer b)
         case CONSTV:
         case LAMBDATV:
         case NLAMBDATV:
+            /* TODO: it is not OK to do this on the stack since list-like 
+            * structures can be of indefinite extent. It *must* be done by 
+            * iteration (and even that is problematic) */
             result =
                 equal(cell_a->payload.cons.car, cell_b->payload.cons.car) && equal(cell_a->payload.cons.cdr,
                                                                                    cell_b->payload.cons.cdr);
@@ -76,11 +79,13 @@ bool equal(struct cons_pointer a, struct cons_pointer b)
         case KEYTV:
         case STRINGTV:
         case SYMBOLTV:
-            /*
-                 * slightly complex because a string may or may not have a '\0'
+            /* slightly complex because a string may or may not have a '\0'
             * cell at the end, but I'll ignore that for now. I think in
             * practice only the empty string will.
             */
+            /* TODO: it is not OK to do this on the stack since list-like 
+             * structures can be of indefinite extent. It *must* be done by 
+             * iteration (and even that is problematic) */
             result =
                 cell_a->payload.string.character ==
                     cell_b->payload.string.character &&