prism 1.2.0 → 1.4.0

data/src/regexp.c

@@ -158,6 +158,11 @@ pm_regexp_parse_range_quantifier(pm_regexp_parser_t *parser) {
     } state = PM_REGEXP_RANGE_QUANTIFIER_STATE_START;
 
     while (1) {
+        if (parser->cursor >= parser->end) {
+            parser->cursor = savepoint;
+            return true;
+        }
+
         switch (state) {
             case PM_REGEXP_RANGE_QUANTIFIER_STATE_START:
                 switch (*parser->cursor) {
@@ -594,8 +599,7 @@ pm_regexp_parse_group(pm_regexp_parser_t *parser, uint16_t depth) {
                 // If we hit a -, then we're done parsing options.
                 if (*parser->cursor != '-') break;
 
-                // Otherwise, fallthrough to the - case.
-                /* fallthrough */
+                PRISM_FALLTHROUGH
             case '-':
                 parser->cursor++;
                 while (!pm_regexp_char_is_eof(parser) && *parser->cursor != ':' && *parser->cursor != ')') {
@@ -707,7 +711,7 @@ pm_regexp_parse_item(pm_regexp_parser_t *parser, uint16_t depth) {
                 if (!pm_regexp_char_find(parser, '\n')) parser->cursor = parser->end;
                 return true;
             }
-        /* fallthrough */
+        PRISM_FALLTHROUGH
         default: {
             size_t width;
             if (!parser->encoding_changed) {

data/src/serialize.c

@@ -1706,6 +1706,12 @@ pm_serialize_node(pm_parser_t *parser, pm_node_t *node, pm_buffer_t *buffer) {
             } else {
                 pm_serialize_node(parser, (pm_node_t *)((pm_rescue_node_t *)node)->reference, buffer);
             }
+            if (((pm_rescue_node_t *)node)->then_keyword_loc.start == NULL) {
+                pm_buffer_append_byte(buffer, 0);
+            } else {
+                pm_buffer_append_byte(buffer, 1);
+                pm_serialize_location(parser, &((pm_rescue_node_t *)node)->then_keyword_loc, buffer);
+            }
             if (((pm_rescue_node_t *)node)->statements == NULL) {
                 pm_buffer_append_byte(buffer, 0);
             } else {
@@ -1916,6 +1922,12 @@ pm_serialize_node(pm_parser_t *parser, pm_node_t *node, pm_buffer_t *buffer) {
         case PM_UNTIL_NODE: {
             pm_buffer_append_varuint(buffer, (uint32_t) node->flags);
             pm_serialize_location(parser, &((pm_until_node_t *)node)->keyword_loc, buffer);
+            if (((pm_until_node_t *)node)->do_keyword_loc.start == NULL) {
+                pm_buffer_append_byte(buffer, 0);
+            } else {
+                pm_buffer_append_byte(buffer, 1);
+                pm_serialize_location(parser, &((pm_until_node_t *)node)->do_keyword_loc, buffer);
+            }
             if (((pm_until_node_t *)node)->closing_loc.start == NULL) {
                 pm_buffer_append_byte(buffer, 0);
             } else {
@@ -1954,6 +1966,12 @@ pm_serialize_node(pm_parser_t *parser, pm_node_t *node, pm_buffer_t *buffer) {
         case PM_WHILE_NODE: {
             pm_buffer_append_varuint(buffer, (uint32_t) node->flags);
             pm_serialize_location(parser, &((pm_while_node_t *)node)->keyword_loc, buffer);
+            if (((pm_while_node_t *)node)->do_keyword_loc.start == NULL) {
+                pm_buffer_append_byte(buffer, 0);
+            } else {
+                pm_buffer_append_byte(buffer, 1);
+                pm_serialize_location(parser, &((pm_while_node_t *)node)->do_keyword_loc, buffer);
+            }
             if (((pm_while_node_t *)node)->closing_loc.start == NULL) {
                 pm_buffer_append_byte(buffer, 0);
             } else {

data/src/static_literals.c

@@ -501,7 +501,7 @@ pm_static_literal_inspect_node(pm_buffer_t *buffer, const pm_static_literals_met
         case PM_FLOAT_NODE: {
             const double value = ((const pm_float_node_t *) node)->value;
 
-            if (isinf(value)) {
+            if (PRISM_ISINF(value)) {
                 if (*node->location.start == '-') {
                     pm_buffer_append_byte(buffer, '-');
                 }

data/src/util/pm_buffer.c

@@ -172,6 +172,46 @@ pm_buffer_append_double(pm_buffer_t *buffer, double value) {
     pm_buffer_append(buffer, source, sizeof(double));
 }
 
+/**
+ * Append a unicode codepoint to the buffer.
+ */
+bool
+pm_buffer_append_unicode_codepoint(pm_buffer_t *buffer, uint32_t value) {
+    if (value <= 0x7F) {
+        pm_buffer_append_byte(buffer, (uint8_t) value); // 0xxxxxxx
+        return true;
+    } else if (value <= 0x7FF) {
+        uint8_t bytes[] = {
+            (uint8_t) (0xC0 | ((value >> 6) & 0x3F)), // 110xxxxx
+            (uint8_t) (0x80 | (value & 0x3F))         // 10xxxxxx
+        };
+
+        pm_buffer_append_bytes(buffer, bytes, 2);
+        return true;
+    } else if (value <= 0xFFFF) {
+        uint8_t bytes[] = {
+            (uint8_t) (0xE0 | ((value >> 12) & 0x3F)), // 1110xxxx
+            (uint8_t) (0x80 | ((value >> 6) & 0x3F)),  // 10xxxxxx
+            (uint8_t) (0x80 | (value & 0x3F))          // 10xxxxxx
+        };
+
+        pm_buffer_append_bytes(buffer, bytes, 3);
+        return true;
+    } else if (value <= 0x10FFFF) {
+        uint8_t bytes[] = {
+            (uint8_t) (0xF0 | ((value >> 18) & 0x3F)), // 11110xxx
+            (uint8_t) (0x80 | ((value >> 12) & 0x3F)), // 10xxxxxx
+            (uint8_t) (0x80 | ((value >> 6) & 0x3F)),  // 10xxxxxx
+            (uint8_t) (0x80 | (value & 0x3F))          // 10xxxxxx
+        };
+
+        pm_buffer_append_bytes(buffer, bytes, 4);
+        return true;
+    } else {
+        return false;
+    }
+}
+
 /**
  * Append a slice of source code to the buffer.
  */

data/src/util/pm_char.c

@@ -185,7 +185,7 @@ pm_strspn_number_kind_underscores(const uint8_t *string, ptrdiff_t length, const
         size++;
     }
 
-    if (string[size - 1] == '_') *invalid = string + size - 1;
+    if (size > 0 && string[size - 1] == '_') *invalid = string + size - 1;
     return size;
 }
 

data/src/util/pm_constant_pool.c

@@ -15,8 +15,12 @@ pm_constant_id_list_init(pm_constant_id_list_t *list) {
  */
 void
 pm_constant_id_list_init_capacity(pm_constant_id_list_t *list, size_t capacity) {
-    list->ids = xcalloc(capacity, sizeof(pm_constant_id_t));
-    if (list->ids == NULL) abort();
+    if (capacity) {
+        list->ids = xcalloc(capacity, sizeof(pm_constant_id_t));
+        if (list->ids == NULL) abort();
+    } else {
+        list->ids = NULL;
+    }
 
     list->size = 0;
     list->capacity = capacity;

data/src/util/pm_string.c

@@ -189,6 +189,7 @@ pm_string_mapped_init(pm_string_t *string, const char *filepath) {
 
     source = mmap(NULL, size, PROT_READ, MAP_PRIVATE, fd, 0);
     if (source == MAP_FAILED) {
+        close(fd);
         return PM_STRING_INIT_ERROR_GENERIC;
     }
 

data/src/util/pm_strncasecmp.c

@@ -1,5 +1,17 @@
 #include "prism/util/pm_strncasecmp.h"
 
+/**
+ * A locale-insensitive version of `tolower(3)`
+ */
+static inline int
+pm_tolower(int c)
+{
+    if ('A' <= c && c <= 'Z') {
+        return c | 0x20;
+    }
+    return c;
+}
+
 /**
  * Compare two strings, ignoring case, up to the given length. Returns 0 if the
  * strings are equal, a negative number if string1 is less than string2, or a
@@ -16,7 +28,7 @@ pm_strncasecmp(const uint8_t *string1, const uint8_t *string2, size_t length) {
 
     while (offset < length && string1[offset] != '\0') {
         if (string2[offset] == '\0') return string1[offset];
-        if ((difference = tolower(string1[offset]) - tolower(string2[offset])) != 0) return difference;
+        if ((difference = pm_tolower(string1[offset]) - pm_tolower(string2[offset])) != 0) return difference;
         offset++;
     }
 

metadata

@@ -1,16 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: prism
 version: !ruby/object:Gem::Version
-  version: 1.2.0
+  version: 1.4.0
 platform: ruby
 authors:
 - Shopify
-autorequire:
 bindir: bin
 cert_chain: []
-date: 2024-10-10 00:00:00.000000000 Z
+date: 2025-03-18 00:00:00.000000000 Z
 dependencies: []
-description:
 email:
 - ruby@shopify.com
 executables: []
@@ -39,6 +37,7 @@ files:
 - docs/parser_translation.md
 - docs/parsing_rules.md
 - docs/releasing.md
+- docs/relocation.md
 - docs/ripper_translation.md
 - docs/ruby_api.md
 - docs/ruby_parser_translation.md
@@ -90,16 +89,21 @@ files:
 - lib/prism/parse_result/errors.rb
 - lib/prism/parse_result/newlines.rb
 - lib/prism/pattern.rb
+- lib/prism/polyfill/append_as_bytes.rb
 - lib/prism/polyfill/byteindex.rb
 - lib/prism/polyfill/unpack1.rb
 - lib/prism/reflection.rb
+- lib/prism/relocation.rb
 - lib/prism/serialize.rb
+- lib/prism/string_query.rb
 - lib/prism/translation.rb
 - lib/prism/translation/parser.rb
+- lib/prism/translation/parser/builder.rb
 - lib/prism/translation/parser/compiler.rb
 - lib/prism/translation/parser/lexer.rb
 - lib/prism/translation/parser33.rb
 - lib/prism/translation/parser34.rb
+- lib/prism/translation/parser35.rb
 - lib/prism/translation/ripper.rb
 - lib/prism/translation/ripper/sexp.rb
 - lib/prism/translation/ripper/shim.rb
@@ -114,9 +118,11 @@ files:
 - rbi/prism/node_ext.rbi
 - rbi/prism/parse_result.rbi
 - rbi/prism/reflection.rbi
+- rbi/prism/string_query.rbi
 - rbi/prism/translation/parser.rbi
 - rbi/prism/translation/parser33.rbi
 - rbi/prism/translation/parser34.rbi
+- rbi/prism/translation/parser35.rbi
 - rbi/prism/translation/ripper.rbi
 - rbi/prism/visitor.rbi
 - sig/prism.rbs
@@ -133,7 +139,9 @@ files:
 - sig/prism/parse_result.rbs
 - sig/prism/pattern.rbs
 - sig/prism/reflection.rbs
+- sig/prism/relocation.rbs
 - sig/prism/serialize.rbs
+- sig/prism/string_query.rbs
 - sig/prism/visitor.rbs
 - src/diagnostic.c
 - src/encoding.c
@@ -163,7 +171,6 @@ metadata:
   allowed_push_host: https://rubygems.org
   source_code_uri: https://github.com/ruby/prism
   changelog_uri: https://github.com/ruby/prism/blob/main/CHANGELOG.md
-post_install_message:
 rdoc_options: []
 require_paths:
 - lib
@@ -178,8 +185,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.6.0.dev
-signing_key:
+rubygems_version: 3.6.2
 specification_version: 4
 summary: Prism Ruby parser
 test_files: []