edgar-parser 0.1.0__py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (13) hide show
  1. edgar_parser/__init__.py +4 -0
  2. edgar_parser/config.py +35 -0
  3. edgar_parser/earnings_8k.py +670 -0
  4. edgar_parser/enrich.py +132 -0
  5. edgar_parser/matching.py +1155 -0
  6. edgar_parser/pipeline.py +2034 -0
  7. edgar_parser/section_parser.py +680 -0
  8. edgar_parser/tools.py +793 -0
  9. edgar_parser/utils.py +626 -0
  10. edgar_parser-0.1.0.dist-info/METADATA +113 -0
  11. edgar_parser-0.1.0.dist-info/RECORD +13 -0
  12. edgar_parser-0.1.0.dist-info/WHEEL +4 -0
  13. edgar_parser-0.1.0.dist-info/licenses/LICENSE +21 -0
edgar_parser/__init__.py ADDED
@@ -0,0 +1,4 @@
1
+ """edgar-parser: structured financial data from SEC EDGAR filings."""
2
+
3
+ from .pipeline import FilingNotFoundError, enrich_filing, parse_filing
4
+ from .matching import match_filing
edgar_parser/config.py ADDED
@@ -0,0 +1,35 @@
1
+ #!/usr/bin/env python
2
+ # coding: utf-8
3
+
4
+ from dotenv import load_dotenv
5
+ load_dotenv()
6
+
7
+ # === CONFIG & SETUP ==========================================
8
+
9
+ # === HEADERS ======
10
+ HEADERS = {
11
+ "User-Agent": "edgar-parser (Python library; github.com/henrysouchien/edgar-parser)",
12
+ "Accept-Encoding": "gzip, deflate",
13
+ }
14
+
15
+ TICKER_CIK_URL = "https://www.sec.gov/files/company_tickers.json"
16
+
17
+ # === NUMBER OF FILINGS TO PULL ======
18
+ N_10Q = 12
19
+ N_10K = 4
20
+
21
+ # === EXTRA FILING PULLS ===
22
+ N_10Q_EXTRA = 0
23
+ N_10K_EXTRA = 0
24
+
25
+ # === SAFE LIMIT TIMES ===
26
+ REQUEST_DELAY = 1 # in seconds
27
+
28
+ # === EXPORTS ===
29
+ OUTPUT_METRICS_DIR = "metrics"
30
+ EXPORT_UPDATER_DIR = "exports"
31
+
32
+ # 8-K earnings release extraction
33
+ ANTHROPIC_MODEL_8K = "claude-sonnet-4-20250514"
34
+ MAX_8K_HTML_BYTES = 500_000
35
+