datamule 2.2.9__tar.gz → 2.3.0__tar.gz

{datamule-2.2.9 → datamule-2.3.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: datamule
-Version: 2.2.9
+Version: 2.3.0
 Summary: Work with SEC submissions at scale.
 Home-page: https://github.com/john-friedman/datamule-python
 Author: John Friedman

{datamule-2.2.9 → datamule-2.3.0}/datamule/document/document.py

@@ -7,8 +7,6 @@ from doc2dict import html2dict, visualize_dict, get_title, unnest_dict, pdf2dict
 from ..mapping_dicts.txt_mapping_dicts import dict_10k, dict_10q, dict_8k, dict_13d, dict_13g
 from ..mapping_dicts.xml_mapping_dicts import dict_345
 from ..mapping_dicts.html_mapping_dicts import *
-from selectolax.parser import HTMLParser
-
 from pathlib import Path
 import webbrowser
 from secsgml.utils import bytes_to_str
@@ -294,7 +292,6 @@ class Document:
             return bool(re.search(pattern, self.content))
         return False
     
-    # Note: this method will be heavily modified in the future
     def parse(self):
         # check if we have already parsed the content
         if self._data:
@@ -384,6 +381,8 @@ class Document:
                 dct = html2dict(content=self.content, mapping_dict=mapping_dict)
             elif self.extension in ['.txt']:
                 dct = txt2dict(content=self.content, mapping_dict=mapping_dict)
+            elif self.extension == '.pdf':
+                dct = pdf2dict(content=self.content, mapping_dict=mapping_dict)
             else:
                 dct = {}
             
@@ -391,10 +390,8 @@ class Document:
         elif self.extension == '.xml':
             if self.type in ['3', '4', '5', '3/A', '4/A', '5/A']:
                 mapping_dict = dict_345
-            
             self._data = xml2dict(content=self.content, mapping_dict=mapping_dict)
-        elif self.extension == '.pdf':
-            self._data = pdf2dict(content=self.content, mapping_dict=mapping_dict)
+
         else:
             pass
 
@@ -409,6 +406,12 @@ class Document:
             
             if not isinstance(self._data, DataWithTags):
                 self._data = DataWithTags(self._data, self)
+        elif self.extension == '.xml':
+            if self._data is None:
+                self.parse()
+
+            if self._data is None:
+                self._data = {}
             
         return self._data
     
@@ -444,19 +447,46 @@ class Document:
             json.dump(self.data, f, indent=2)
 
     def parse_tables(self,must_exist_in_mapping=True):
-        if self.extension != '.xml':
-            self._tables = []
+        """Must exist in mapping means columns must occur in mapping schema."""
+        if self.extension == '.xml':
+            tables = Tables(document_type = self.type, accession=self.accession)
+            tables.parse_tables(data=self.data,must_exist_in_mapping=must_exist_in_mapping)
+            self._tables = tables
+
+        elif self._data_bool:
+            tables = Tables(document_type = self.type, accession=self.accession)
+            data_tuples = self.data_tuples
+            
+            for i, (id, type, content, level) in enumerate(data_tuples):
+                if type == "table" and i > 0:
+                    description = None
+                    
+                    # Look at previous element
+                    prev_id, prev_type, prev_content, prev_level = data_tuples[i-1]
+                    
+                    # Case 1: Same level + text content
+                    if prev_level == level and prev_type in ["text", "textsmall"]:
+                        description = prev_content
+                    
+                    # Case 2: Higher level (lower number) + title
+                    elif prev_level < level and prev_type == "title":
+                        description = prev_content
+                    
+                    # Case 3: No matching description - add table without description
+                    # (description remains None)
+                    
+                    tables.add_table(data=content, description=description, name="extracted_table")
+
+            self._tables = tables
+
         else:
-            # Use the property to trigger parsing if needed
-            data = self.data
-            tables = Tables(document_type = self.type, accession=self.accession, data=data,must_exist_in_mapping=must_exist_in_mapping)
-            self._tables = tables.tables
+            self._tables = []
 
     @property
     def tables(self):
         if self._tables is None:
             self.parse_tables()
-        return self._tables
+        return self._tables.tables
 
 
     def write_csv(self, output_folder):
@@ -547,6 +577,7 @@ class Document:
             webbrowser.open('file://' + temp_path)
         else:
             print(f"Cannot open files with extension {self.extension}")
+
     def get_section(self, title=None, title_regex=None,title_class=None, format='dict'):
         if self._data_bool:
             if not self.data:
@@ -557,3 +588,9 @@ class Document:
                 return [item[1] for item in result]
             else:
                 return [flatten_dict(item[1],format) for item in result]
+
+    # TODO
+    def get_tables(self,description_regex=None,name=None):
+        # make sure tables is initialized
+        self.tables
+        return self._tables.get_tables(description_regex=description_regex, name=name)

{datamule-2.2.9 → datamule-2.3.0}/datamule/document/tables/tables.py

@@ -6,8 +6,10 @@ from .tables_npx import config_npx
 from .tables_sbsef import config_sbsef
 from .tables_sdr import config_sdr
 from .tables_proxyvotingrecord import config_proxyvotingrecord
+from doc2dict.utils.format_dict import _format_table
  
 from .utils import safe_get, flatten_dict
+import re
 # will add filing date param later? or extension
 all_tables_dict = {
     '3' : config_ownership,
@@ -93,25 +95,30 @@ def apply_mapping(flattened_data, mapping_dict, accession, must_exist_in_mapping
 
 # should have table type, accession, data
 class Table:
-    def __init__(self,data,name,accession):
+    def __init__(self,data,name,accession,description = None):
         self.data = data
         self.name = name
         self.accession = accession
+        self.description = description
+
+    # TODO MADE IN A HURRY #
+    def __str__(self):
+        formatted_table = _format_table(self.data)
+        if isinstance(formatted_table, list):
+            table_str = '\n'.join(formatted_table)
+        else:
+            table_str = str(formatted_table)
+        return f"Table '{self.name}' ({self.accession}) - {len(self.data) if isinstance(self.data, list) else 'N/A'} rows\ndescription: {self.description if self.description else ''}\n{table_str}"
 
 
 class Tables():
-    def __init__(self,document_type,accession,data,must_exist_in_mapping=True):
+    def __init__(self,document_type,accession):
         self.document_type = document_type
         self.accession = accession
-        self.data = data
-
-        # to fill in
         self.tables = []
 
-        self.parse_tables(must_exist_in_mapping=must_exist_in_mapping)
-
-    def parse_tables(self,must_exist_in_mapping=True):
-        # first select dict
+    def parse_tables(self,data,must_exist_in_mapping=True):
+        self.data = data
 
         try:
             tables_dict = all_tables_dict[self.document_type]
@@ -120,11 +127,32 @@ class Tables():
         
         # now get the dicts from the data
         data_dicts = seperate_data(tables_dict,self.data)
-        
+
         # now flatten
         data_dicts = [(x,flatten_dict(y)) for x,y in data_dicts]
         
         for table_name, flattened_data in data_dicts:
             mapping_dict = tables_dict[table_name]['mapping']
             mapped_data = apply_mapping(flattened_data, mapping_dict, self.accession,must_exist_in_mapping)
-            self.tables.append(Table(mapped_data, table_name, self.accession))
+            self.tables.append(Table(mapped_data, table_name, self.accession))
+        
+    def add_table(self,data,name,description=None):
+        self.tables.append(Table(data=data,name=name,accession=self.accession,description=description))
+
+    def get_tables(self, description_regex=None, name=None):
+        matching_tables = []
+        
+        for table in self.tables:
+            # Check name match (exact match)
+            if name is not None:
+                if table.name == name:
+                    matching_tables.append(table)
+                    continue
+            
+            # Check description regex match
+            if description_regex is not None and table.description is not None:
+                if re.search(description_regex, table.description):
+                    matching_tables.append(table)
+                    continue
+        
+        return matching_tables

{datamule-2.2.9 → datamule-2.3.0}/datamule.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: datamule
-Version: 2.2.9
+Version: 2.3.0
 Summary: Work with SEC submissions at scale.
 Home-page: https://github.com/john-friedman/datamule-python
 Author: John Friedman

{datamule-2.2.9 → datamule-2.3.0}/setup.py

@@ -32,7 +32,7 @@ if not os.path.exists(file_path):
 setup(
     name="datamule",
     author="John Friedman",
-    version="2.2.9",
+    version="2.3.0",
     description="Work with SEC submissions at scale.",
     packages=find_packages(include=['datamule', 'datamule.*']),
     url="https://github.com/john-friedman/datamule-python",