Resolve "Read not only .txt files"

Closes #2 (closed)

src/easymatcher_worker.py

@@ -21,23 +21,45 @@ class EasymatcherWorker(nlp_ws.NLPWorker):
     https://gitlab.clarin-pl.eu/knowledge-extraction/tools/easymatcher
     """
 
+    @staticmethod
+    def is_jsonl(
+            document_path: str | Path
+    ) -> bool:
+        """Validates whether text file has json/jsonl structure."""
+        try:
+            with open(document_path, 'r', encoding="utf-8") as file:
+                for line in file:
+                    json_obj = json.loads(line)
+                    if "text" not in json_obj:
+                        return False
+            return True
+        except (json.JSONDecodeError, FileNotFoundError):
+            return False
+
     @staticmethod
     def prepare_and_append_document(
-        file_path: str | Path, document_path: str | Path
+            file_path: str | Path, document_path: str | Path
     ) -> None:
         """Formats and appends plain texts into jsonl file."""
         document = {}
-        with open(document_path, "r", encoding="utf-8") as _df:
-            document["text"] = _df.read()
+        if EasymatcherWorker.is_jsonl(document_path):
+            with open(file_path, "a", encoding="utf-8") as _f:
+                with open(document_path, "r", encoding="utf-8") as _df:
+                    for line in _df:
+                        line_data = json.loads(line)
+                        _f.write(json.dumps(line_data) + "\n")
+        else:
+            with open(document_path, "r", encoding="utf-8") as _df:
+                document["text"] = _df.read()
 
-        with open(file_path, "a", encoding="utf-8") as _f:
-            _f.write(json.dumps(document) + "\n")
+            with open(file_path, "a", encoding="utf-8") as _f:
+                _f.write(json.dumps(document) + "\n")
 
     def process(
-        self,
-        input_path: str,
-        task_options: dict[str, str | int | float],
-        output_path: str,
+            self,
+            input_path: str,
+            task_options: dict[str, str | int | float],
+            output_path: str,
     ) -> None:
         """Called for each request made to the worker.
 
@@ -83,5 +105,8 @@ class EasymatcherWorker(nlp_ws.NLPWorker):
         os.unlink(tmpf.name)
 
         with open(output_path, "w", encoding="utf-8") as _f:
-            for out_document in out_documents:
-                _f.write(json.dumps(out_document) + "\n")
+            for out_document, document in zip(out_documents, documents):
+                # We want to keep content of the original labeled documents
+                document['label'] = out_document['label']
+                document['text'] = out_document['text']
+                _f.write(json.dumps(document) + "\n")