lutaml-path 0.2.2 → 0.2.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 4a72cb3f56ae5e29135bdd7842172d4c3bc187e66a63828da8f698d11b1c6e10
-  data.tar.gz: c5e410dc1cfbade92c7508245333a97cf8622c8803220b8aa0053ff422d14293
+  metadata.gz: 8e786faa427ee58eeb9f9dcf52685d2f777ccbb74933329773256a3c37ddaa26
+  data.tar.gz: a72d4e6d1d6746bd40d43366b1439c51f99e25c623a5558e8de40e19cf9ec37d
 SHA512:
-  metadata.gz: 605a1be485a4c239615849214699268f27d444577ef725c1c784f69ad8957eb46310a9fdd556673fcabc20ac56752128172bbb22d0ac47d2013f4c0e777d3604
-  data.tar.gz: b42c5f16e33900a164f7ce3cd8f9297789410cb8ce427264d1fbca8b196a206aa8749819fc3dbf8fd9e07864635a891f43effec2e7c32219f066d5add114abd9
+  metadata.gz: 2b04e70ae8817a53590387a721bd0b98768b137c6e8ebcbfa5e5c00c8068f7fb73400ee0159877d6468c6ebf8fb7c727a4a210fbc935ce99cf8034dbeea9ab60
+  data.tar.gz: 0e2b4b3379e8cdf975698bd154b28c01dd3320e3bb2e83204d261549d212667ab519290b61587c3525de1a623725944dc74704d0d35ae6c912fe8d47de153cd7

data/lib/lutaml/path/transformer.rb

@@ -51,12 +51,25 @@ module Lutaml
       # and a leaf cannot know whether the whole path has dot-steps or filters,
       # so the root rule -- the only place that knows -- decides the type and
       # unwraps Step -> PathSegment when it picks ElementPath.
+      #
+      # parsanol hands captured leaves back as binary-tagged strings, which
+      # raises ArgumentError (invalid byte sequence) at the first downstream
+      # regex on non-ASCII input. The boundary re-tag fixes mistagged-valid
+      # captures; byte loss inside a multibyte character is a parsanol defect
+      # (parsanol/parsanol-ruby#410) and scrubs to replacement content until
+      # it ships. Rule blocks run under the transform engine's self, so the
+      # helper is a class method.
+      def self.utf8(content)
+        content.to_s.dup.force_encoding("UTF-8").scrub
+      end
+
       rule(content: simple(:content)) do |dict|
-        Step.new(PathSegment.new(dict[:content].to_s))
+        Step.new(PathSegment.new(Transformer.utf8(dict[:content])))
       end
 
       rule(content: simple(:content), condition: subtree(:cond)) do |dict|
-        Step.new(PathSegment.new(dict[:content].to_s), condition: dict[:cond])
+        Step.new(PathSegment.new(Transformer.utf8(dict[:content])),
+                 condition: dict[:cond])
       end
 
       rule(segment: subtree(:segment)) { segment }

data/lib/lutaml/path/version.rb

@@ -2,6 +2,6 @@
 
 module Lutaml
   module Path
-    VERSION = "0.2.2"
+    VERSION = "0.2.3"
   end
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: lutaml-path
 version: !ruby/object:Gem::Version
-  version: 0.2.2
+  version: 0.2.3
 platform: ruby
 authors:
 - Ribose Inc.