1434 - support all unicode spaces

7 files changed, 128 insertions, 21 deletions

diff --git a/002test.cc b/002test.cc
index 76bf02cf..ea75db82 100644
--- a/002test.cc
+++ b/002test.cc
@@ -88,6 +88,7 @@ long long int to_integer(string n) {
   char* end = NULL;
   // safe because string.c_str() is guaranteed to be null-terminated
   long long int result = strtoll(n.c_str(), &end, /*any base*/0);
+  if (*end != '\0') raise << "tried to convert " << n << " to number\n";
   assert(*end == '\0');
   return result;
 }
diff --git a/031address.cc b/031address.cc
index cee1d0e8..052eec37 100644
--- a/031address.cc
+++ b/031address.cc
@@ -102,14 +102,14 @@ base = canonize(base);
 
 :(code)
 bool has_property(reagent x, string name) {
-  for (long long int i = 0; i < SIZE(x.properties); ++i) {
+  for (long long int i = /*skip name:type*/1; i < SIZE(x.properties); ++i) {
     if (x.properties.at(i).first == name) return true;
   }
   return false;
 }
 
 vector<string> property(const reagent& r, const string& name) {
-  for (long long int p = 0; p != SIZE(r.properties); ++p) {
+  for (long long int p = /*skip name:type*/1; p != SIZE(r.properties); ++p) {
     if (r.properties.at(p).first == name)
       return r.properties.at(p).second;
   }
diff --git a/041name.cc b/041name.cc
index 0b9e2270..300a9c60 100644
--- a/041name.cc
+++ b/041name.cc
@@ -70,8 +70,7 @@ bool disqualified(/*mutable*/ reagent& x) {
   if (is_raw(x)) return true;
   if (isa_literal(x)) return true;
   if (is_integer(x.name)) return true;
-  if (x.name == "default-space")
-    x.initialized = true;
+  // End Disqualified Reagents
   if (x.initialized) return true;
   return false;
 }
@@ -125,12 +124,11 @@ bool is_raw(const reagent& r) {
 
 bool is_special_name(const string& s) {
   if (s == "_") return true;
-  // lexical scopes
-  if (s == "default-space") return true;
   if (s == "0") return true;
   // tests will use these in later layers even though tests will mostly use numeric addresses
   if (s == "screen") return true;
   if (s == "keyboard") return true;
+  // End is_special_name Cases
   return false;
 }
 
@@ -142,14 +140,6 @@ recipe main [
 +name: assign x 1
 -name: assign _ 1
 
-//: one reserved word that we'll need later
-:(scenario convert_names_passes_default_space)
-recipe main [
-  default-space:number, x:number <- copy 0:literal, 1:literal
-]
-+name: assign x 1
--name: assign default-space 1
-
 //: an escape hatch to suppress name conversion that we'll use later
 :(scenario convert_names_passes_raw)
 recipe main [
diff --git a/043space.cc b/043space.cc
index 37683ab8..e1100f89 100644
--- a/043space.cc
+++ b/043space.cc
@@ -25,6 +25,21 @@ recipe main [
 ]
 +mem: storing 34 in location 8
 
+//:: first disable name conversion for 'default-space'
+:(scenario convert_names_passes_default_space)
+recipe main [
+  default-space:number, x:number <- copy 0:literal, 1:literal
+]
++name: assign x 1
+-name: assign default-space 1
+
+:(before "End Disqualified Reagents")
+if (x.name == "default-space")
+  x.initialized = true;
+:(before "End is_special_name Cases")
+if (s == "default-space") return true;
+
+//:: now implement space support
 :(before "End call Fields")
 long long int default_space;
 :(replace "call(recipe_number r) :running_recipe(r)")
diff --git a/044space_surround.cc b/044space_surround.cc
index 54c92973..61521013 100644
--- a/044space_surround.cc
+++ b/044space_surround.cc
@@ -27,11 +27,13 @@ recipe main [
 
 :(replace{} "long long int space_base(const reagent& x)")
 long long int space_base(const reagent& x) {
+//?   cerr << "space_base: " << x.to_string() << '\n'; //? 1
   return space_base(x, space_index(x), Current_routine->calls.front().default_space);
 }
 
 long long int space_base(const reagent& x, long long int space_index, long long int base) {
 //?   trace("foo") << "base of space " << space_index << '\n'; //? 1
+//?   cerr << "space_base sub: " << x.to_string() << '\n'; //? 1
   if (space_index == 0) {
 //?     trace("foo") << "base of space " << space_index << " is " << base << '\n'; //? 1
     return base;
@@ -42,7 +44,8 @@ long long int space_base(const reagent& x, long long int space_index, long long
 }
 
 long long int space_index(const reagent& x) {
-  for (long long int i = 0; i < SIZE(x.properties); ++i) {
+//?   cerr << "space_index: " << x.to_string() << '\n'; //? 1
+  for (long long int i = /*skip name:type*/1; i < SIZE(x.properties); ++i) {
     if (x.properties.at(i).first == "space") {
       assert(SIZE(x.properties.at(i).second) == 1);
       return to_integer(x.properties.at(i).second.at(0));
@@ -50,3 +53,8 @@ long long int space_index(const reagent& x) {
   }
   return 0;
 }
+
+:(scenario permit_space_as_variable_name)
+recipe main [
+  space:number <- copy 0:literal
+]
diff --git a/050scenario.cc b/050scenario.cc
index 1d7366cd..8df08e77 100644
--- a/050scenario.cc
+++ b/050scenario.cc
@@ -252,13 +252,31 @@ void check_type(const string& lhs, istream& in) {
 
 void check_string(long long int address, const string& literal) {
   trace(Primitive_recipe_depth, "run") << "checking string length at " << address;
-  if (Memory[address] != SIZE(literal))
-    raise << "expected location " << address << " to contain length " << SIZE(literal) << " of string [" << literal << "] but saw " << Memory[address] << '\n';
+  if (Memory[address] != SIZE(literal)) {
+    if (Current_scenario && !Hide_warnings)
+      raise << "\nF - " << Current_scenario->name << ": expected location " << address << " to contain length " << SIZE(literal) << " of string [" << literal << "] but saw " << Memory[address] << '\n';
+    else
+      raise << "expected location " << address << " to contain length " << SIZE(literal) << " of string [" << literal << "] but saw " << Memory[address] << '\n';
+    if (!Hide_warnings) {
+      Passed = false;
+      ++Num_failures;
+    }
+    return;
+  }
   ++address;  // now skip length
   for (long long int i = 0; i < SIZE(literal); ++i) {
     trace(Primitive_recipe_depth, "run") << "checking location " << address+i;
-    if (Memory[address+i] != literal.at(i))
-      raise << "expected location " << (address+i) << " to contain " << literal.at(i) << " but saw " << Memory[address+i] << '\n';
+    if (Memory[address+i] != literal.at(i)) {
+      if (Current_scenario && !Hide_warnings)
+        raise << "\nF - " << Current_scenario->name << ": expected location " << (address+i) << " to contain " << literal.at(i) << " but saw " << Memory[address+i] << '\n';
+      else
+        raise << "expected location " << (address+i) << " to contain " << literal.at(i) << " but saw " << Memory[address+i] << '\n';
+      if (!Hide_warnings) {
+        Passed = false;
+        ++Num_failures;
+      }
+      return;
+    }
   }
 }
 
diff --git a/060string.mu b/060string.mu
index f0b3ad18..338748fc 100644
--- a/060string.mu
+++ b/060string.mu
@@ -494,6 +494,69 @@ scenario interpolate-at-end [
   ]
 ]
 
+recipe space? [
+  default-space:array:address:location <- new location:type, 30:literal
+  c:character <- next-ingredient
+  result:boolean <- equal c:character, 32:literal/space
+  jump-if result:boolean, +reply:label
+  result:boolean <- equal c:character, 10:literal/newline
+  jump-if result:boolean, +reply:label
+  result:boolean <- equal c:character, 9:literal/tab
+  jump-if result:boolean, +reply:label
+  result:boolean <- equal c:character, 13:literal/carriage-return
+  # remaining uncommon cases in sorted order
+  # http://unicode.org code-points in unicode-set Z and Pattern_White_Space
+  jump-if result:boolean, +reply:label
+  result:boolean <- equal c:character, 11:literal/ctrl-k
+  jump-if result:boolean, +reply:label
+  result:boolean <- equal c:character, 12:literal/ctrl-l
+  jump-if result:boolean, +reply:label
+  result:boolean <- equal c:character, 133:literal/ctrl-0085
+  jump-if result:boolean, +reply:label
+  result:boolean <- equal c:character, 160:literal/no-break-space
+  jump-if result:boolean, +reply:label
+  result:boolean <- equal c:character, 5760:literal/ogham-space-mark
+  jump-if result:boolean, +reply:label
+  result:boolean <- equal c:character, 8192:literal/en-quad
+  jump-if result:boolean, +reply:label
+  result:boolean <- equal c:character, 8193:literal/em-quad
+  jump-if result:boolean, +reply:label
+  result:boolean <- equal c:character, 8194:literal/en-space
+  jump-if result:boolean, +reply:label
+  result:boolean <- equal c:character, 8195:literal/em-space
+  jump-if result:boolean, +reply:label
+  result:boolean <- equal c:character, 8196:literal/three-per-em-space
+  jump-if result:boolean, +reply:label
+  result:boolean <- equal c:character, 8197:literal/four-per-em-space
+  jump-if result:boolean, +reply:label
+  result:boolean <- equal c:character, 8198:literal/six-per-em-space
+  jump-if result:boolean, +reply:label
+  result:boolean <- equal c:character, 8199:literal/figure-space
+  jump-if result:boolean, +reply:label
+  result:boolean <- equal c:character, 8200:literal/punctuation-space
+  jump-if result:boolean, +reply:label
+  result:boolean <- equal c:character, 8201:literal/thin-space
+  jump-if result:boolean, +reply:label
+  result:boolean <- equal c:character, 8202:literal/hair-space
+  jump-if result:boolean, +reply:label
+  result:boolean <- equal c:character, 8206:literal/left-to-right
+  jump-if result:boolean, +reply:label
+  result:boolean <- equal c:character, 8207:literal/right-to-left
+  jump-if result:boolean, +reply:label
+  result:boolean <- equal c:character, 8232:literal/line-separator
+  jump-if result:boolean, +reply:label
+  result:boolean <- equal c:character, 8233:literal/paragraph-separator
+  jump-if result:boolean, +reply:label
+  result:boolean <- equal c:character, 8239:literal/narrow-no-break-space
+  jump-if result:boolean, +reply:label
+  result:boolean <- equal c:character, 8287:literal/medium-mathematical-space
+  jump-if result:boolean, +reply:label
+  result:boolean <- equal c:character, 12288:literal/ideographic-space
+  jump-if result:boolean, +reply:label
+  +reply
+  reply result:boolean
+]
+
 recipe trim [
   default-space:array:address:location <- new location:type, 30:literal
   s:address:array:character <- next-ingredient
@@ -508,7 +571,7 @@ recipe trim [
       reply result:address:array:character
     }
     curr:character <- index s:address:array:character/deref, start:number
-    whitespace?:boolean <- equal curr:character, 32:literal/space
+    whitespace?:boolean <- space? curr:character
     break-unless whitespace?:boolean
     start:number <- add start:number, 1:literal
     loop
@@ -519,7 +582,7 @@ recipe trim [
     not-at-start?:boolean <- greater-than end:number, start:number
     assert not-at-start?:boolean [end ran up against start]
     curr:character <- index s:address:array:character/deref, end:number
-    whitespace?:boolean <- equal curr:character, 32:literal/space
+    whitespace?:boolean <- space? curr:character
     break-unless whitespace?:boolean
     end:number <- subtract end:number, 1:literal
     loop
@@ -589,3 +652,15 @@ scenario trim-left-right [
     3:string <- [abc]
   ]
 ]
+
+scenario trim-newline-tab [
+  run [
+    1:address:array:character <- new [	abc
+]
+    2:address:array:character <- trim 1:address:array:character
+    3:array:character <- copy 2:address:array:character/deref
+  ]
+  memory-should-contain [
+    3:string <- [abc]
+  ]
+]