feat: Add staging brick for Datasaur token-based tasks (#50)

* feat: Add staging brick for Datasaur token-based tasks

* Added doc string and formatting with flake8,mypy and black

* docs: Added documentation for stage_for_datasaur

* fix: version sync correction

* fix: Corrections to docs fror stage_for_datasaur

* fix: changes in naming of example variables

* Update docs/source/bricks.rst

Co-authored-by: Matt Robinson <[email protected]>

Author: benjats07

CHANGELOG.md

@@ -1,3 +1,7 @@
+## 0.2.2-dev0
+
+* Add staging brick for Datasaur
+
 ## 0.2.1
 
 * Added brick to convert an ISD dictionary to a list of elements

docs/source/bricks.rst

@@ -750,3 +750,18 @@ files to an S3 bucket.
 
   upload_staged_files()
 
+``stage_for_datasaur``
+--------------------------
+Formats a list of ``Text`` elements as input to token based tasks in Datasaur.
+
+Example: 
+
+.. code:: python
+
+  from unstructured.staging.datasaur import stage_for_datasaur
+  elements  = [Text("Text1"),Text("Text2")]
+  datasaur_data = stage_for_datasaur(elements)
+
+The output is a list of dictionaries, each one with two keys:
+"text" with the content of the element and 
+"entities" with an empty list.

test_unstructured/staging/test_datasaur.py

@@ -0,0 +1,14 @@
+import unstructured.staging.datasaur as datasaur
+
+from unstructured.documents.elements import Text
+
+
+def test_stage_for_datasaur():
+    elements = [Text("Text 1"), Text("Text 2"), Text("Text 3")]
+    result = datasaur.stage_for_datasaur(elements)
+    assert result[0]["text"] == "Text 1"
+    assert result[0]["entities"] == []
+    assert result[1]["text"] == "Text 2"
+    assert result[1]["entities"] == []
+    assert result[2]["text"] == "Text 3"
+    assert result[2]["entities"] == []

unstructured/__version__.py

@@ -1 +1 @@
-__version__ = "0.2.1"  # pragma: no cover
+__version__ = "0.2.2-dev0"  # pragma: no cover

unstructured/staging/datasaur.py

@@ -0,0 +1,12 @@
+from typing import Dict, List, Any
+from unstructured.documents.elements import Text
+
+
+def stage_for_datasaur(elements: List[Text]) -> List[Dict[str, Any]]:
+    """Convert a list of elements into a list of dictionaries for use in Datasaur"""
+    result: List[Dict[str, Any]] = list()
+    for item in elements:
+        data = dict(text=item.text, entities=[])
+        result.append(data)
+
+    return result