Add tf.strings.unicode_script, which detects the script of a unicode codepoint

based on standard ranges.

PiperOrigin-RevId: 214796357

2 files changed, 34 insertions, 0 deletions

diff --git a/tensorflow/core/api_def/base_api/api_def_UnicodeScript.pbtxt b/tensorflow/core/api_def/base_api/api_def_UnicodeScript.pbtxt
new file mode 100644
index 0000000000..7898fe8d6b
--- /dev/null
+++ b/tensorflow/core/api_def/base_api/api_def_UnicodeScript.pbtxt
@@ -0,0 +1,28 @@
+op {
+  graph_op_name: "UnicodeScript"
+  endpoint {
+    name: "UnicodeScript"
+  }
+  in_arg {
+    name: "input"
+    description: <<END
+A Tensor of int32 Unicode code points.
+END
+  }
+  out_arg {
+    name: "output"
+    description: <<END
+A Tensor of int32 script codes corresponding to each input code point.
+END
+  }
+  summary: <<END
+Determine the script codes of a given tensor of Unicode integer code points.
+END
+  description: <<END
+This operation converts Unicode code points to script codes corresponding to
+each code point. Script codes correspond to International Components for
+Unicode (ICU) UScriptCode values. See http://icu-project.org/apiref/icu4c/uscript_8h.html.
+Returns -1 (USCRIPT_INVALID_CODE) for invalid codepoints. Output shape will
+match input shape.
+END
+}
diff --git a/tensorflow/core/api_def/python_api/api_def_UnicodeScript.pbtxt b/tensorflow/core/api_def/python_api/api_def_UnicodeScript.pbtxt
new file mode 100644
index 0000000000..a884a46143
--- /dev/null
+++ b/tensorflow/core/api_def/python_api/api_def_UnicodeScript.pbtxt
@@ -0,0 +1,6 @@
+op {
+  graph_op_name: "UnicodeScript"
+  endpoint {
+    name: "strings.unicode_script"
+  }
+}