create user tests

any_parser/__init__.py

@@ -1,4 +1,6 @@
-from any_parser.base import AnyParser
+"""AnyParser module for parsing data."""
+
+from any_parser.any_parser import AnyParser
 
 __all__ = ["AnyParser"]
 

any_parser/any_parser_rt.py → any_parser/any_parser.py

@@ -12,6 +12,7 @@
 TIMEOUT = 60
 SUPPORTED_FILE_EXTENSIONS = ["pdf", "doc", "docx", "ppt", "pptx"]
 
+
 class AnyParser:
     """AnyParser RT: Real-time parser for any data format."""
 
@@ -70,7 +71,10 @@ def extract(
         # Send the POST request
         start_time = time.time()
         response = requests.post(
-            self._sync_url, headers=self._headers, data=json.dumps(payload), timeout=TIMEOUT
+            self._sync_url,
+            headers=self._headers,
+            data=json.dumps(payload),
+            timeout=TIMEOUT,
         )
         end_time = time.time()
 
@@ -186,4 +190,4 @@ def async_fetch(
             return "\n".join(markdown_list)
         elif response.status_code == 202:
             return None
-        return f"Error: {response.status_code}"
+        return f"Error: {response.status_code}"

examples/test_async_example.py

@@ -0,0 +1,23 @@
+"""Testing Aynchronous Extraction"""
+
+import sys
+
+from any_parser import AnyParser
+
+sys.path.append(".")
+sys.path.append("..")
+sys.path.append("../..")
+
+# define API key and file path
+API_KEY = "U1xoe4F7qQ9MGJkiMFtR240r3r2iF6NQaAMUEYYc"
+FILE_PATH = "./sample_data/stoxx_index_guide_0003.pdf"
+
+# create instance of AnyParser
+ap = AnyParser(API_KEY)
+
+# asynchronous extract job (extraction and result fetching are separate)
+file_id = ap.async_extract(FILE_PATH)
+
+# fetch results (5s polling)
+md = ap.async_fetch(file_id=file_id)
+print(md)

examples/test_sync_example.py

@@ -0,0 +1,21 @@
+"""Testing Syncrhonous Extraction"""
+
+import sys
+
+from any_parser import AnyParser
+
+sys.path.append(".")
+sys.path.append("..")
+sys.path.append("../..")
+
+# define API key and file path
+API_KEY = "U1xoe4F7qQ9MGJkiMFtR240r3r2iF6NQaAMUEYYc"
+FILE_PATH = "./sample_data/stoxx_index_guide_0003.pdf"
+
+# create instance of AnyParser
+ap = AnyParser(API_KEY)
+
+# synchronous extract job (extraction and result fetching together)
+md_output, total_time = ap.extract(FILE_PATH)
+print(md_output)
+print(total_time)