remove exact_search

bellingcat · Mar 3, 2024 · 3ad45ab · 3ad45ab
1 parent 5f86aa1
commit 3ad45ab
Show file tree

Hide file tree

Showing 3 changed files with 4 additions and 15 deletions.
diff --git a/README.md b/README.md
@@ -108,11 +108,14 @@ python main.py text_search --help
 # Basic usage (defaults to searching the last 5 years of records)
 python main.py text_search John Doe
 
+# Basic usage with a combination of exact and partial search parameters
+python main.py text_search \"John Doe\" Pharmaceuticals Chemicals
+
 # Usage with date range and export to custom CSV file
 python main.py text_search Tsunami Hazards --start_date "2021-01-01" --end_date "2021-12-31" --output "results.csv"
 
 # More advanced usage specifying more arguments, with export to JSON
-python main.py text_search Volcano Monitoring --start_date "2021-01-01" --end_date "2021-12-31" --exact_search --output "results.json"\
+python main.py text_search Volcano Monitoring --start_date "2021-01-01" --end_date "2021-12-31" --output "results.json"\
           --filing_type "all_annual_quarterly_and_current_reports" --entity_id "0001030717" \
           --min_wait 5.0 --max_wait 7.0 --retries 3 --browser "firefox" --headless
 

diff --git a/src/cli.py b/src/cli.py
@@ -61,7 +61,6 @@ def text_search(
         output: str = f"edgar_search_results_{datetime.now().strftime('%d%m%Y_%H%M%S')}.csv",
         entity_id: Optional[str] = None,
         filing_type: Optional[str] = None,
-        exact_search: bool = False,
         start_date: str = (date.today() - timedelta(days=365 * 5)).strftime("%Y-%m-%d"),
         end_date: str = date.today().strftime("%Y-%m-%d"),
         min_wait: float = 5.0,
@@ -78,7 +77,6 @@ def text_search(
         :param output: Name of the output file to save the results to
         :param entity_id: CIK or name or ticker of the company to search for
         :param filing_type: Type of filing to search for
-        :param exact_search: Whether to exactly match the sequence of keywords or not
         :param start_date: Start date of the search
         :param end_date: End date of the search
         :param min_wait: Minimum wait time for the request to complete before checking the page or retrying a request
@@ -89,7 +87,6 @@ def text_search(
         """
         try:
             keywords = list(keywords)
-            exact_search = bool(exact_search)
             start_date = datetime.strptime(start_date, "%Y-%m-%d")
             end_date = datetime.strptime(end_date, "%Y-%m-%d")
             min_wait = float(min_wait)
@@ -115,7 +112,6 @@ def text_search(
                 keywords=keywords,
                 entity_id=entity_id,
                 filing_type=filing_type,
-                exact_search=exact_search,
                 start_date=start_date,
                 end_date=end_date,
                 min_wait_seconds=min_wait,

diff --git a/src/text_search.py b/src/text_search.py
@@ -189,7 +189,6 @@ def _generate_request_args(
         keywords: List[str],
         entity_id: Optional[str],
         filing_type: Optional[str],
-        exact_search: bool,
         start_date: date,
         end_date: date,
         page_number: int,
@@ -200,7 +199,6 @@ def _generate_request_args(
         :param keywords: Search keywords to input in the "Document word or phrase" field
         :param entity_id: Entity/Person name, ticker, or CIK number to input in the "Company name, ticker, or CIK" field
         :param filing_type: Filing category to select from the dropdown menu, defaults to None
-        :param exact_search: Whether to perform an exact search on the search_keywords argument or not, defaults to False in order to return the maximum amount of search results by default
         :param start_date: Start date for the custom date range, defaults to 5 years ago to replicate the default behavior of the SEC website
         :param end_date: End date for the custom date range, defaults to current date in order to replicate the default behavior of the SEC website
         :param page_number: Page number to request, defaults to 1
@@ -214,7 +212,6 @@ def _generate_request_args(
 
         # Join search keywords into a single string
         keywords = " ".join(keywords)
-        keywords = f'"{keywords}"' if exact_search else keywords
 
         # Generate request arguments
         request_args = {
@@ -310,7 +307,6 @@ def _generate_search_requests(
         keywords: List[str],
         entity_id: Optional[str],
         filing_type: Optional[str],
-        exact_search: bool,
         start_date: date,
         end_date: date,
         min_wait_seconds: float,
@@ -324,7 +320,6 @@ def _generate_search_requests(
         :param keywords: Search keywords to input in the "Document word or phrase" field
         :param entity_id: Entity/Person name, ticker, or CIK number to input in the "Company name, ticker, or CIK" field
         :param filing_type: Filing category to select from the dropdown menu
-        :param exact_search: Whether to perform an exact search on the search_keywords argument or not
         :param start_date: Start date for the custom date range
         :param end_date: End date for the custom date range
         :param min_wait_seconds: Minimum number of seconds to wait for the request to complete
@@ -337,7 +332,6 @@ def _generate_search_requests(
             keywords=keywords,
             entity_id=entity_id,
             filing_type=filing_type,
-            exact_search=exact_search,
             start_date=start_date,
             end_date=end_date,
             page_number=1,
@@ -385,7 +379,6 @@ def _generate_search_requests(
                         keywords=keywords,
                         entity_id=entity_id,
                         filing_type=filing_type,
-                        exact_search=exact_search,
                         start_date=start,
                         end_date=end,
                         min_wait_seconds=min_wait_seconds,
@@ -400,7 +393,6 @@ def text_search(
         keywords: List[str],
         entity_id: Optional[str],
         filing_type: Optional[str],
-        exact_search: bool,
         start_date: date,
         end_date: date,
         min_wait_seconds: float,
@@ -414,7 +406,6 @@ def text_search(
         :param keywords: Search keywords to input in the "Document word or phrase" field
         :param entity_id: Entity/Person name, ticker, or CIK number to input in the "Company name, ticker, or CIK" field
         :param filing_type: Filing category to select from the dropdown menu
-        :param exact_search: Whether to perform an exact search on the search_keywords argument or not
         :param start_date: Start date for the custom date range
         :param end_date: End date for the custom date range
         :param min_wait_seconds: Minimum number of seconds to wait for the request to complete
@@ -427,7 +418,6 @@ def text_search(
             keywords=keywords,
             entity_id=entity_id,
             filing_type=filing_type,
-            exact_search=exact_search,
             start_date=start_date,
             end_date=end_date,
             min_wait_seconds=min_wait_seconds,