datamule 2.2.3__tar.gz → 2.2.5__tar.gz

{datamule-2.2.3 → datamule-2.2.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: datamule
-Version: 2.2.3
+Version: 2.2.5
 Summary: Work with SEC submissions at scale.
 Home-page: https://github.com/john-friedman/datamule-python
 Author: John Friedman

{datamule-2.2.3 → datamule-2.2.5}/datamule/datamule/downloader.py

@@ -287,7 +287,7 @@ class Downloader:
                     keepalive_timeout=60
                 )
 
-                async with aiohttp.ClientSession(connector=connector, timeout=aiohttp.ClientTimeout(total=30)) as session:
+                async with aiohttp.ClientSession(connector=connector, timeout=aiohttp.ClientTimeout(total=600)) as session:
                     tasks = [
                         self.download_and_process(
                             session, url, semaphore, decompression_pool, 

{datamule-2.2.3 → datamule-2.2.5}/datamule/document/document.py

@@ -221,7 +221,6 @@ class Tags(TextAnalysisBase):
         if not hasattr(self, '_persons'):
             self._persons = []
             sources = self._get_text_sources()
-            
             for source in sources:
                 if '8k_2024_persons' in self.processors:
                     results = get_full_names_dictionary_lookup(source['text'], self.processors['8k_2024_persons'])
@@ -289,6 +288,12 @@ class Document:
         self._data = None
         self._tables = None
         self._text = None
+
+        # booleans
+        self._text_bool = self.extension in ('.htm', '.html','.txt')
+        self._data_bool = self.extension in ('.htm', '.html','.txt')
+        self._visualize_bool = self._data_bool
+        self._tables_bool = self.extension in ('.xml')
         
 
 
@@ -544,6 +549,17 @@ class Document:
                     writer.writeheader()
                     writer.writerows(table.data)
 
+    def reset_nlp(self):
+        """Reset all NLP analysis by creating fresh wrapper objects"""
+        # Reset data wrapper
+        if hasattr(self, '_data') and self._data is not None:
+            raw_data = dict(self._data)  # Extract the underlying dict
+            self._data = DataWithTags(raw_data, self)
+        
+        # Reset text wrapper
+        if hasattr(self, '_text') and self._text is not None:
+            raw_text = str(self._text)  # Extract the underlying string
+            self._text = TextWithTags(raw_text, self)
         
     def _document_to_section_text(self, document_data, parent_key=''):
         items = []

{datamule-2.2.3 → datamule-2.2.5}/datamule/submission.py

@@ -121,6 +121,16 @@ class Submission:
             self.accession = self.metadata.content['accession-number']
             self.filing_date= f"{self.metadata.content['filing-date'][:4]}-{self.metadata.content['filing-date'][4:6]}-{self.metadata.content['filing-date'][6:8]}"
 
+
+        # booleans
+        self._has_xbrl = any(
+                doc['type'] in ('EX-100.INS', 'EX-101.INS') or 
+                doc.get('filename', '').endswith('_htm.xml')
+                for doc in self.metadata.content['documents']
+            )
+        
+        self._has_fundamentals = self._has_xbrl
+        
     def _load_document_by_index(self, idx):
         """Load a document by its index in the metadata documents list."""
         doc = self.metadata.content['documents'][idx]

{datamule-2.2.3 → datamule-2.2.5}/datamule/tags/config.py

@@ -3,6 +3,12 @@ from .dictionaries import download_dictionary, load_dictionary
 _active_dictionaries = []
 _loaded_dictionaries = {}
 
+def clear_dictionaries():
+    """Remove all active dictionaries"""
+    global _active_dictionaries, _loaded_dictionaries
+    _active_dictionaries = []
+    _loaded_dictionaries = {}
+
 def set_dictionaries(dictionaries, overwrite=False):
     """Set active dictionaries and load them into memory"""
     global _active_dictionaries, _loaded_dictionaries

{datamule-2.2.3 → datamule-2.2.5}/datamule/tags/utils.py

@@ -67,25 +67,31 @@ def get_ticker_regex_dict():
     return dict(ticker_regex_list)
 
 # will change in future to accomodate other datasets
-def validate_full_name(full_name,keywords):
+def validate_full_name(full_name, keywords):
     if len(full_name) == 1:
         return False
-    # check all is upper
-    if all(word.isupper() for word in full_name):
+    
+    # Clean punctuation before validation
+    cleaned_name = [word.rstrip(".,;:!?()[]") for word in full_name]
+    
+    # Skip validation if cleaning removed everything
+    if not all(cleaned_name):
         return False
-    # check if any number in word
-    if any(any(char.isdigit() for char in word) for word in full_name):
+    
+    # Apply existing checks to cleaned words
+    if all(word.isupper() for word in cleaned_name):
         return False
-    if any(any(char in ".,;:!?()[]" for char in word) for word in full_name):
+    
+    # check if any number in word
+    if any(any(char.isdigit() for char in word) for word in cleaned_name):
         return False
     
     # add optional set lookups
     if keywords is not None:
         # return false if first word is not in keywords set
-        if full_name[0] not in keywords:
+        if cleaned_name[0] not in keywords:
             return False
     
-    
     return True
 
 def get_full_names(text,keywords=None):

{datamule-2.2.3 → datamule-2.2.5}/datamule.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: datamule
-Version: 2.2.3
+Version: 2.2.5
 Summary: Work with SEC submissions at scale.
 Home-page: https://github.com/john-friedman/datamule-python
 Author: John Friedman

{datamule-2.2.3 → datamule-2.2.5}/setup.py

@@ -32,7 +32,7 @@ if not os.path.exists(file_path):
 setup(
     name="datamule",
     author="John Friedman",
-    version="2.2.3",
+    version="2.2.5",
     description="Work with SEC submissions at scale.",
     packages=find_packages(include=['datamule', 'datamule.*']),
     url="https://github.com/john-friedman/datamule-python",