bankstatementparser 0.0.18__tar.gz → 0.0.20__tar.gz
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.
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/PKG-INFO +78 -47
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/README.md +73 -46
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/__init__.py +67 -2
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/_llm_common.py +6 -1
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/additional_parsers.py +269 -275
- bankstatementparser-0.0.20/bankstatementparser/analytics.py +691 -0
- bankstatementparser-0.0.20/bankstatementparser/api.py +583 -0
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/bank_statement_parsers.py +40 -22
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/base_parser.py +77 -7
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/camt_parser.py +546 -249
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/cli.py +58 -49
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/enrichment/categorizer.py +47 -23
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/export/__init__.py +3 -0
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/export/ledger.py +33 -5
- bankstatementparser-0.0.20/bankstatementparser/export/parquet.py +161 -0
- bankstatementparser-0.0.20/bankstatementparser/field86_parser.py +186 -0
- bankstatementparser-0.0.20/bankstatementparser/forensics.py +337 -0
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/hybrid/llm_extractor.py +8 -6
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/hybrid/orchestrator.py +26 -6
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/hybrid/scanner.py +10 -6
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/hybrid/vision.py +174 -70
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/input_validator.py +13 -5
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/pain001_parser.py +281 -192
- bankstatementparser-0.0.20/bankstatementparser/parallel.py +162 -0
- bankstatementparser-0.0.20/bankstatementparser/plugins.py +165 -0
- bankstatementparser-0.0.20/bankstatementparser/privacy.py +73 -0
- bankstatementparser-0.0.20/bankstatementparser/py.typed +0 -0
- bankstatementparser-0.0.20/bankstatementparser/reconciliation.py +362 -0
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/record_types.py +12 -1
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/transaction_deduplicator.py +66 -41
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/transaction_models.py +35 -16
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/zip_security.py +110 -2
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/pyproject.toml +9 -5
- bankstatementparser-0.0.18/bankstatementparser/api.py +0 -326
- bankstatementparser-0.0.18/bankstatementparser/parallel.py +0 -121
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/LICENSE +0 -0
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/LICENSE-APACHE +0 -0
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/LICENSE-MIT +0 -0
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/_amounts.py +0 -0
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/enrichment/__init__.py +0 -0
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/enrichment/account_mapper.py +0 -0
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/exceptions.py +0 -0
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/hybrid/__init__.py +0 -0
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/hybrid/evaluation.py +0 -0
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/hybrid/ollama_direct.py +0 -0
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/hybrid/pdf_text.py +0 -0
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/hybrid/prompts.py +0 -0
- {bankstatementparser-0.0.18 → bankstatementparser-0.0.20}/bankstatementparser/hybrid/verification.py +0 -0
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
Metadata-Version: 2.4
|
|
2
2
|
Name: bankstatementparser
|
|
3
|
-
Version: 0.0.
|
|
3
|
+
Version: 0.0.20
|
|
4
4
|
Summary: BankStatementParser is your essential tool for easy bank statement management. Designed with finance and treasury experts in mind, it offers a simple way to handle CAMT (ISO 20022) formats and more. Get quick, accurate insights from your financial data and spend less time on processing. It's the smart, hassle-free way to stay on top of your transactions.
|
|
5
5
|
License: Apache-2.0 OR MIT
|
|
6
6
|
License-File: LICENSE
|
|
@@ -16,12 +16,14 @@ Classifier: Programming Language :: Python :: 3.11
|
|
|
16
16
|
Classifier: Programming Language :: Python :: 3.12
|
|
17
17
|
Classifier: Programming Language :: Python :: 3.13
|
|
18
18
|
Classifier: Programming Language :: Python :: 3.14
|
|
19
|
+
Classifier: Programming Language :: Python :: 3.15
|
|
19
20
|
Provides-Extra: api
|
|
20
21
|
Provides-Extra: enrichment
|
|
21
22
|
Provides-Extra: excel
|
|
22
23
|
Provides-Extra: hybrid
|
|
23
24
|
Provides-Extra: hybrid-plus
|
|
24
25
|
Provides-Extra: hybrid-vision
|
|
26
|
+
Provides-Extra: parquet
|
|
25
27
|
Provides-Extra: polars
|
|
26
28
|
Requires-Dist: defusedxml (>=0.7.1,<0.8.0)
|
|
27
29
|
Requires-Dist: fastapi (>=0.115.0) ; extra == "api"
|
|
@@ -31,15 +33,26 @@ Requires-Dist: openpyxl (>=3.1.5,<4.0.0) ; extra == "excel"
|
|
|
31
33
|
Requires-Dist: pandas (>=2.3.3,<3.0.0)
|
|
32
34
|
Requires-Dist: pdfplumber (>=0.11.0) ; extra == "hybrid-plus"
|
|
33
35
|
Requires-Dist: polars (>=1.32.0,<2.0.0) ; extra == "polars"
|
|
36
|
+
Requires-Dist: pyarrow (>=14.0.0) ; extra == "parquet"
|
|
34
37
|
Requires-Dist: pydantic (>=2.11.0,<3.0.0)
|
|
35
38
|
Requires-Dist: pypdf (>=4.0.0) ; extra == "hybrid" or extra == "hybrid-plus" or extra == "hybrid-vision"
|
|
36
39
|
Requires-Dist: pypdfium2 (>=4.30.0) ; extra == "hybrid-vision"
|
|
40
|
+
Requires-Dist: python-multipart (>=0.0.32) ; extra == "api"
|
|
37
41
|
Requires-Dist: uvicorn (>=0.34.0) ; extra == "api"
|
|
38
42
|
Project-URL: Homepage, https://bankstatementparser.com
|
|
39
43
|
Project-URL: Repository, https://github.com/sebastienrousseau/bankstatementparser
|
|
40
44
|
Description-Content-Type: text/markdown
|
|
41
45
|
|
|
42
|
-
<!-- SPDX-License-Identifier: Apache-2.0 OR MIT
|
|
46
|
+
<!-- SPDX-License-Identifier: Apache-2.0 OR MIT -->
|
|
47
|
+
|
|
48
|
+
<p align="center">
|
|
49
|
+
<img
|
|
50
|
+
src="https://cloudcdn.pro/bankstatementparser/v1/logos/bankstatementparser.svg"
|
|
51
|
+
alt="bankstatementparser logo"
|
|
52
|
+
width="120"
|
|
53
|
+
height="120"
|
|
54
|
+
/>
|
|
55
|
+
</p>
|
|
43
56
|
|
|
44
57
|
<h1 align="center">Bank Statement Parser</h1>
|
|
45
58
|
|
|
@@ -51,7 +64,7 @@ Description-Content-Type: text/markdown
|
|
|
51
64
|
|
|
52
65
|
<p align="center">
|
|
53
66
|
<a href="https://github.com/sebastienrousseau/bankstatementparser/actions"><img src="https://img.shields.io/github/actions/workflow/status/sebastienrousseau/bankstatementparser/quality-gates.yml?style=for-the-badge&logo=github" alt="Build" /></a>
|
|
54
|
-
<a href="https://pypi.org/project/bankstatementparser/"><img src="https://img.shields.io/pypi/pyversions/bankstatementparser.svg?style=for-the-badge&v=0.0.
|
|
67
|
+
<a href="https://pypi.org/project/bankstatementparser/"><img src="https://img.shields.io/pypi/pyversions/bankstatementparser.svg?style=for-the-badge&v=0.0.20" alt="PyPI" /></a>
|
|
55
68
|
<a href="https://pypi.org/project/bankstatementparser/"><img src="https://img.shields.io/pypi/dm/bankstatementparser.svg?style=for-the-badge" alt="PyPI Downloads" /></a>
|
|
56
69
|
<a href="https://codecov.io/github/sebastienrousseau/bankstatementparser?branch=main"><img src="https://img.shields.io/codecov/c/github/sebastienrousseau/bankstatementparser?style=for-the-badge" alt="Codecov" /></a>
|
|
57
70
|
<a href="LICENSE"><img src="https://img.shields.io/github/license/sebastienrousseau/bankstatementparser?style=for-the-badge" alt="License" /></a>
|
|
@@ -228,26 +241,28 @@ from bankstatementparser.hybrid import smart_ingest
|
|
|
228
241
|
|
|
229
242
|
# Path A — deterministic parser (free, fastest, $0)
|
|
230
243
|
result = smart_ingest("statement.xml")
|
|
231
|
-
print(result.source_method)
|
|
244
|
+
print(result.source_method) # "deterministic"
|
|
232
245
|
|
|
233
246
|
# Path B — text-LLM for digital PDFs (set BSP_HYBRID_MODEL=ollama/llama3)
|
|
234
247
|
result = smart_ingest("statement.pdf")
|
|
235
|
-
print(result.source_method)
|
|
236
|
-
print(
|
|
248
|
+
print(result.source_method) # "llm"
|
|
249
|
+
print(
|
|
250
|
+
result.verification.status
|
|
251
|
+
) # VERIFIED | DISCREPANCY | UNVERIFIABLE | FAILED
|
|
237
252
|
|
|
238
253
|
# Path C — multimodal vision for scanned PDFs (set BSP_HYBRID_VISION_MODEL)
|
|
239
254
|
# auto-routed when pypdf cannot extract enough text
|
|
240
255
|
result = smart_ingest("scan.pdf")
|
|
241
|
-
print(result.source_method)
|
|
256
|
+
print(result.source_method) # "vision"
|
|
242
257
|
```
|
|
243
258
|
|
|
244
259
|
Every row carries:
|
|
245
260
|
|
|
246
261
|
- `source_method` — `"deterministic"`, `"llm"`, or `"vision"` for
|
|
247
262
|
full audit provenance
|
|
248
|
-
- `transaction_hash` —
|
|
249
|
-
|
|
250
|
-
re-ingestion
|
|
263
|
+
- `transaction_hash` — versioned SHA-256 fingerprint scoped to account,
|
|
264
|
+
currency, date, description, amount, and available bank identity.
|
|
265
|
+
Rebuild persisted pre-v2 keys before re-ingestion
|
|
251
266
|
- `confidence` — float between 0 and 1 for LLM rows, `None` for
|
|
252
267
|
deterministic
|
|
253
268
|
- `raw_source_text` — best-effort source-text slice for the v0.0.6
|
|
@@ -276,7 +291,9 @@ For ZIP archives, use `iter_secure_xml_entries()`.
|
|
|
276
291
|
from bankstatementparser import CamtParser, iter_secure_xml_entries
|
|
277
292
|
|
|
278
293
|
for entry in iter_secure_xml_entries("statements.zip"):
|
|
279
|
-
parser = CamtParser.from_bytes(
|
|
294
|
+
parser = CamtParser.from_bytes(
|
|
295
|
+
entry.xml_bytes, source_name=entry.source_name
|
|
296
|
+
)
|
|
280
297
|
transactions = parser.parse()
|
|
281
298
|
print(entry.source_name, len(transactions), "transactions")
|
|
282
299
|
```
|
|
@@ -298,23 +315,23 @@ Gemini, …).
|
|
|
298
315
|
|
|
299
316
|
```mermaid
|
|
300
317
|
flowchart TD
|
|
301
|
-
A[smart_ingest
|
|
302
|
-
B
|
|
303
|
-
C --> Z[IngestResult<br/>source_method='deterministic']
|
|
318
|
+
A["smart_ingest(path)"] --> B{"detect_statement_format"}
|
|
319
|
+
B -->|CAMT/PAIN/OFX/MT940/CSV| C["Path A: deterministic parser<br/>$0, fastest"]
|
|
320
|
+
C --> Z["IngestResult<br/>source_method='deterministic'"]
|
|
304
321
|
|
|
305
|
-
B
|
|
306
|
-
D --> E{text
|
|
322
|
+
B -->|pdf or unknown| D["pypdf extract_text"]
|
|
323
|
+
D --> E{"text length >= 50?"}
|
|
307
324
|
|
|
308
|
-
E
|
|
309
|
-
F --> Y[IngestResult<br/>source_method='llm']
|
|
325
|
+
E -->|yes| F["Path B: text-LLM<br/>default ollama/llama3"]
|
|
326
|
+
F --> Y["IngestResult<br/>source_method='llm'"]
|
|
310
327
|
|
|
311
|
-
E
|
|
312
|
-
G --> X[IngestResult<br/>source_method='vision']
|
|
328
|
+
E -->|no| G["Path C: vision-LLM<br/>opt-in via BSP_HYBRID_VISION_MODEL"]
|
|
329
|
+
G --> X["IngestResult<br/>source_method='vision'"]
|
|
313
330
|
|
|
314
|
-
Z --> V[verify_balance<br/>Golden Rule]
|
|
331
|
+
Z --> V["verify_balance<br/>Golden Rule"]
|
|
315
332
|
Y --> V
|
|
316
333
|
X --> V
|
|
317
|
-
V --> R[VERIFIED / DISCREPANCY / UNVERIFIABLE / FAILED]
|
|
334
|
+
V --> R["VERIFIED / DISCREPANCY / UNVERIFIABLE / FAILED"]
|
|
318
335
|
```
|
|
319
336
|
|
|
320
337
|
Every extracted row carries an immutable `transaction_hash`, an
|
|
@@ -354,7 +371,7 @@ for the full surface.
|
|
|
354
371
|
|---|---|
|
|
355
372
|
| **Golden Rule verification** *(v0.0.5)* | Every result carries `opening + credits − debits == closing` status: `VERIFIED`, `DISCREPANCY`, `UNVERIFIABLE`, or `FAILED`. |
|
|
356
373
|
| **Multi-currency verification** *(v0.0.8)* | `verify_balance_multi_currency()` groups transactions by currency and runs the Golden Rule independently per group — no more false `DISCREPANCY` on multi-currency statements. |
|
|
357
|
-
| **Idempotent dedup** *(v0.0.5)* | Every `Transaction` carries a stable `transaction_hash` (
|
|
374
|
+
| **Idempotent dedup** *(v0.0.5)* | Every `Transaction` carries a stable `transaction_hash` (versioned SHA-256 of account, currency, date, description, amount, and bank identity). `Deduplicator.dedupe_by_hash()` makes incremental ingestion safe to re-run. |
|
|
358
375
|
| **Interactive review** *(v0.0.6)* | `--type review` CLI walks through discrepancies with accept/edit/skip/delete/quit. `IngestResult.to_json()` / `.from_json()` for stable round-trip with embedded audit trail. |
|
|
359
376
|
|
|
360
377
|
### Enrichment & export
|
|
@@ -373,17 +390,18 @@ for the full surface.
|
|
|
373
390
|
|---|---|
|
|
374
391
|
| **PII redaction** | Names, IBANs, and addresses masked by default — opt in with `--show-pii` |
|
|
375
392
|
| **Secure ZIP** | `iter_secure_xml_entries()` rejects zip bombs, encrypted entries, and suspicious compression ratios |
|
|
376
|
-
| **Tested** |
|
|
393
|
+
| **Tested** | 1007 tests, coverage gated at 100% in CI, property-based fuzzing with Hypothesis |
|
|
377
394
|
|
|
378
395
|
---
|
|
379
396
|
|
|
380
397
|
## PII Redaction
|
|
381
398
|
|
|
382
399
|
PII (names, IBANs, addresses) is **redacted by default** in
|
|
383
|
-
console output
|
|
400
|
+
CLI console output. Python parser methods return full records unless
|
|
401
|
+
`redact_pii=True` is supplied.
|
|
384
402
|
|
|
385
403
|
```python
|
|
386
|
-
#
|
|
404
|
+
# Request redaction explicitly in Python
|
|
387
405
|
for tx in parser.parse_streaming(redact_pii=True):
|
|
388
406
|
print(tx) # Names and addresses show as ***REDACTED***
|
|
389
407
|
|
|
@@ -392,8 +410,9 @@ for tx in parser.parse_streaming(redact_pii=False):
|
|
|
392
410
|
print(tx)
|
|
393
411
|
```
|
|
394
412
|
|
|
395
|
-
|
|
396
|
-
|
|
413
|
+
Regular and hybrid file exports contain full records. Legacy CLI streaming
|
|
414
|
+
exports follow `--show-pii`; Python exports reflect the records supplied.
|
|
415
|
+
Choose redaction explicitly before sharing an exported file.
|
|
397
416
|
|
|
398
417
|
---
|
|
399
418
|
|
|
@@ -440,11 +459,13 @@ Process multiple files simultaneously across CPU cores:
|
|
|
440
459
|
```python
|
|
441
460
|
from bankstatementparser import parse_files_parallel
|
|
442
461
|
|
|
443
|
-
results = parse_files_parallel(
|
|
444
|
-
|
|
445
|
-
|
|
446
|
-
|
|
447
|
-
|
|
462
|
+
results = parse_files_parallel(
|
|
463
|
+
[
|
|
464
|
+
"statements/jan.xml",
|
|
465
|
+
"statements/feb.xml",
|
|
466
|
+
"statements/mar.xml",
|
|
467
|
+
]
|
|
468
|
+
)
|
|
448
469
|
|
|
449
470
|
for r in results:
|
|
450
471
|
print(r.path, r.status, len(r.transactions), "rows")
|
|
@@ -693,13 +714,22 @@ versioned companion packages build on it — each is its own repository
|
|
|
693
714
|
and Python package, so you install only what you need and the core
|
|
694
715
|
stays dependency-light.
|
|
695
716
|
|
|
696
|
-
| Package | PyPI | Role |
|
|
697
|
-
|
|
698
|
-
| [`bankstatementparser
|
|
699
|
-
| [`bankstatementparser-
|
|
700
|
-
| [`bankstatementparser-
|
|
701
|
-
| [`bankstatementparser-
|
|
702
|
-
| [`bankstatementparser-
|
|
717
|
+
| Package | GitHub Repository | PyPI | Role | Description |
|
|
718
|
+
|---|---|---|---|---|
|
|
719
|
+
| [`bankstatementparser`](https://github.com/sebastienrousseau/bankstatementparser) | [`sebastienrousseau/bankstatementparser`](https://github.com/sebastienrousseau/bankstatementparser) | [](https://pypi.org/project/bankstatementparser/) | Core Engine | Unified parser for CAMT (052/053), PAIN.001, CSV, OFX, QFX, MT940, and PDF statements |
|
|
720
|
+
| [`bankstatementparser-mcp`](https://github.com/sebastienrousseau/bankstatementparser-mcp) | [`sebastienrousseau/bankstatementparser-mcp`](https://github.com/sebastienrousseau/bankstatementparser-mcp) | [](https://pypi.org/project/bankstatementparser-mcp/) | AI Protocol | Model Context Protocol (MCP) server exposing statement tools to LLMs & AI agents |
|
|
721
|
+
| [`bankstatementparser-lsp`](https://github.com/sebastienrousseau/bankstatementparser-lsp) | [`sebastienrousseau/bankstatementparser-lsp`](https://github.com/sebastienrousseau/bankstatementparser-lsp) | [](https://pypi.org/project/bankstatementparser-lsp/) | Developer Tooling | Language Server Protocol (LSP) with live SWIFT MT940 statement validation & diagnostics |
|
|
722
|
+
| [`bankstatementparser-transport-ebics`](https://github.com/sebastienrousseau/bankstatementparser-transport-ebics) | [`sebastienrousseau/bankstatementparser-transport-ebics`](https://github.com/sebastienrousseau/bankstatementparser-transport-ebics) | [](https://pypi.org/project/bankstatementparser-transport-ebics/) | Transport | Automated bank statement retrieval over EBICS 3.0 (`H005`) and 2.5 (`H004`) protocols |
|
|
723
|
+
| [`bankstatementparser-writer-xlsx`](https://github.com/sebastienrousseau/bankstatementparser-writer-xlsx) | [`sebastienrousseau/bankstatementparser-writer-xlsx`](https://github.com/sebastienrousseau/bankstatementparser-writer-xlsx) | [](https://pypi.org/project/bankstatementparser-writer-xlsx/) | Output Writer | Formats and exports parsed banking transactions into styled Microsoft Excel (`.xlsx`) workbooks |
|
|
724
|
+
| [`bankstatementparser-writer-qif`](https://github.com/sebastienrousseau/bankstatementparser-writer-qif) | [`sebastienrousseau/bankstatementparser-writer-qif`](https://github.com/sebastienrousseau/bankstatementparser-writer-qif) | [](https://pypi.org/project/bankstatementparser-writer-qif/) | Output Writer | Serializes transactions into standard Quicken Interchange Format (`.qif`) exchange files |
|
|
725
|
+
| [`bankstatementparser-writer-ofx`](https://github.com/sebastienrousseau/bankstatementparser-writer-ofx) | [`sebastienrousseau/bankstatementparser-writer-ofx`](https://github.com/sebastienrousseau/bankstatementparser-writer-ofx) | [](https://pypi.org/project/bankstatementparser-writer-ofx/) | Output Writer | Serializes transactions into standard Open Financial Exchange (`.ofx`) XML/SGML files |
|
|
726
|
+
| [`bankstatementparser-writer-swift`](https://github.com/sebastienrousseau/bankstatementparser-writer-swift) | [`sebastienrousseau/bankstatementparser-writer-swift`](https://github.com/sebastienrousseau/bankstatementparser-writer-swift) | [](https://pypi.org/project/bankstatementparser-writer-swift/) | Output Writer | Exports transactions to SWIFT MT940 customer statements and MT942 interim reports |
|
|
727
|
+
| [`bankstatementparser-loader-bai2`](https://github.com/sebastienrousseau/bankstatementparser-loader-bai2) | [`sebastienrousseau/bankstatementparser-loader-bai2`](https://github.com/sebastienrousseau/bankstatementparser-loader-bai2) | [](https://pypi.org/project/bankstatementparser-loader-bai2/) | Input Loader | Parses BAI2 cash-management and account balance statements |
|
|
728
|
+
| [`bankstatementparser-loader-mt942`](https://github.com/sebastienrousseau/bankstatementparser-loader-mt942) | [`sebastienrousseau/bankstatementparser-loader-mt942`](https://github.com/sebastienrousseau/bankstatementparser-loader-mt942) | [](https://pypi.org/project/bankstatementparser-loader-mt942/) | Input Loader | Parses SWIFT MT942 interim transaction reports with credit/debit summary reconciliation |
|
|
729
|
+
| [`bankstatementparser-loader-cfonb`](https://github.com/sebastienrousseau/bankstatementparser-loader-cfonb) | [`sebastienrousseau/bankstatementparser-loader-cfonb`](https://github.com/sebastienrousseau/bankstatementparser-loader-cfonb) | [](https://pypi.org/project/bankstatementparser-loader-cfonb/) | Input Loader | Parses French CFONB 120 / AFB120 120-byte fixed-width banking statement files |
|
|
730
|
+
| [`bankstatementparser-loader-camt054`](https://github.com/sebastienrousseau/bankstatementparser-loader-camt054) | [`sebastienrousseau/bankstatementparser-loader-camt054`](https://github.com/sebastienrousseau/bankstatementparser-loader-camt054) | [](https://pypi.org/project/bankstatementparser-loader-camt054/) | Input Loader | Ingests ISO 20022 CAMT.054 real-time debit/credit notification stream XML |
|
|
731
|
+
| [`bankstatementparser-loader-sepa`](https://github.com/sebastienrousseau/bankstatementparser-loader-sepa) | [`sebastienrousseau/bankstatementparser-loader-sepa`](https://github.com/sebastienrousseau/bankstatementparser-loader-sepa) | [](https://pypi.org/project/bankstatementparser-loader-sepa/) | Input Loader | Ingests ISO 20022 SEPA PAIN.002 payment status reports and PAIN.008 direct debit mandates |
|
|
732
|
+
| [`bankstatementparser-loader-bacs`](https://github.com/sebastienrousseau/bankstatementparser-loader-bacs) | [`sebastienrousseau/bankstatementparser-loader-bacs`](https://github.com/sebastienrousseau/bankstatementparser-loader-bacs) | [](https://pypi.org/project/bankstatementparser-loader-bacs/) | Input Loader | Parses UK BACS Standard 18 / Faster Payments 106-byte fixed-width transmission files |
|
|
703
733
|
|
|
704
734
|
Loaders turn an additional source format into the same unified
|
|
705
735
|
`Transaction` model; writers take parsed data back out to another
|
|
@@ -707,8 +737,8 @@ target. Each companion pins `bankstatementparser` as a dependency and
|
|
|
707
737
|
ships its own 100%-coverage test suite.
|
|
708
738
|
|
|
709
739
|
```bash
|
|
710
|
-
# Mix and match: e.g.
|
|
711
|
-
pip install bankstatementparser-
|
|
740
|
+
# Mix and match: e.g. fetch via EBICS 3.0, parse CAMT.053, and export to SWIFT MT940 or Excel
|
|
741
|
+
pip install bankstatementparser-transport-ebics bankstatementparser-writer-swift bankstatementparser-writer-xlsx
|
|
712
742
|
```
|
|
713
743
|
|
|
714
744
|
Loaders hand back the same `Transaction` model the core parsers
|
|
@@ -720,15 +750,15 @@ cleanly — see each companion's README for runnable examples.
|
|
|
720
750
|
## Project Layout
|
|
721
751
|
|
|
722
752
|
```text
|
|
723
|
-
bankstatementparser/ Source code (
|
|
753
|
+
bankstatementparser/ Source code (39 modules)
|
|
724
754
|
bankstatementparser/hybrid/ PDF pipeline: orchestrator, llm_extractor, vision, scanner, ollama_direct, verification
|
|
725
755
|
bankstatementparser/enrichment/ Categorizer, AccountMapper, EnrichedTransaction
|
|
726
|
-
bankstatementparser/export/ hledger + beancount journal export
|
|
756
|
+
bankstatementparser/export/ hledger + beancount journal export, Apache Parquet columnar export
|
|
727
757
|
bankstatementparser/api.py REST API microservice (FastAPI)
|
|
728
758
|
docs/compliance/ ISO 13485 validation, risk register, traceability matrix
|
|
729
759
|
examples/ 14 deterministic + 9 hybrid runnable example scripts
|
|
730
760
|
scripts/ SBOM generation, checksums, signature verification
|
|
731
|
-
tests/
|
|
761
|
+
tests/ 1007 tests (unit, integration, property-based, security, hybrid mocks)
|
|
732
762
|
```
|
|
733
763
|
|
|
734
764
|
---
|
|
@@ -800,8 +830,9 @@ No. Zero network calls. XML parsers enforce `no_network=True`. No
|
|
|
800
830
|
cloud, no telemetry.
|
|
801
831
|
|
|
802
832
|
**Is PII redacted automatically?**
|
|
803
|
-
|
|
804
|
-
|
|
833
|
+
CLI console output masks personal fields by default. Python callers
|
|
834
|
+
must request `redact_pii=True`. Regular file exports retain full data;
|
|
835
|
+
legacy streaming exports follow `--show-pii`.
|
|
805
836
|
|
|
806
837
|
**Is the extraction deterministic?**
|
|
807
838
|
Yes. Same input produces byte-identical output. Critical for
|
|
@@ -1,4 +1,13 @@
|
|
|
1
|
-
<!-- SPDX-License-Identifier: Apache-2.0 OR MIT
|
|
1
|
+
<!-- SPDX-License-Identifier: Apache-2.0 OR MIT -->
|
|
2
|
+
|
|
3
|
+
<p align="center">
|
|
4
|
+
<img
|
|
5
|
+
src="https://cloudcdn.pro/bankstatementparser/v1/logos/bankstatementparser.svg"
|
|
6
|
+
alt="bankstatementparser logo"
|
|
7
|
+
width="120"
|
|
8
|
+
height="120"
|
|
9
|
+
/>
|
|
10
|
+
</p>
|
|
2
11
|
|
|
3
12
|
<h1 align="center">Bank Statement Parser</h1>
|
|
4
13
|
|
|
@@ -10,7 +19,7 @@
|
|
|
10
19
|
|
|
11
20
|
<p align="center">
|
|
12
21
|
<a href="https://github.com/sebastienrousseau/bankstatementparser/actions"><img src="https://img.shields.io/github/actions/workflow/status/sebastienrousseau/bankstatementparser/quality-gates.yml?style=for-the-badge&logo=github" alt="Build" /></a>
|
|
13
|
-
<a href="https://pypi.org/project/bankstatementparser/"><img src="https://img.shields.io/pypi/pyversions/bankstatementparser.svg?style=for-the-badge&v=0.0.
|
|
22
|
+
<a href="https://pypi.org/project/bankstatementparser/"><img src="https://img.shields.io/pypi/pyversions/bankstatementparser.svg?style=for-the-badge&v=0.0.20" alt="PyPI" /></a>
|
|
14
23
|
<a href="https://pypi.org/project/bankstatementparser/"><img src="https://img.shields.io/pypi/dm/bankstatementparser.svg?style=for-the-badge" alt="PyPI Downloads" /></a>
|
|
15
24
|
<a href="https://codecov.io/github/sebastienrousseau/bankstatementparser?branch=main"><img src="https://img.shields.io/codecov/c/github/sebastienrousseau/bankstatementparser?style=for-the-badge" alt="Codecov" /></a>
|
|
16
25
|
<a href="LICENSE"><img src="https://img.shields.io/github/license/sebastienrousseau/bankstatementparser?style=for-the-badge" alt="License" /></a>
|
|
@@ -187,26 +196,28 @@ from bankstatementparser.hybrid import smart_ingest
|
|
|
187
196
|
|
|
188
197
|
# Path A — deterministic parser (free, fastest, $0)
|
|
189
198
|
result = smart_ingest("statement.xml")
|
|
190
|
-
print(result.source_method)
|
|
199
|
+
print(result.source_method) # "deterministic"
|
|
191
200
|
|
|
192
201
|
# Path B — text-LLM for digital PDFs (set BSP_HYBRID_MODEL=ollama/llama3)
|
|
193
202
|
result = smart_ingest("statement.pdf")
|
|
194
|
-
print(result.source_method)
|
|
195
|
-
print(
|
|
203
|
+
print(result.source_method) # "llm"
|
|
204
|
+
print(
|
|
205
|
+
result.verification.status
|
|
206
|
+
) # VERIFIED | DISCREPANCY | UNVERIFIABLE | FAILED
|
|
196
207
|
|
|
197
208
|
# Path C — multimodal vision for scanned PDFs (set BSP_HYBRID_VISION_MODEL)
|
|
198
209
|
# auto-routed when pypdf cannot extract enough text
|
|
199
210
|
result = smart_ingest("scan.pdf")
|
|
200
|
-
print(result.source_method)
|
|
211
|
+
print(result.source_method) # "vision"
|
|
201
212
|
```
|
|
202
213
|
|
|
203
214
|
Every row carries:
|
|
204
215
|
|
|
205
216
|
- `source_method` — `"deterministic"`, `"llm"`, or `"vision"` for
|
|
206
217
|
full audit provenance
|
|
207
|
-
- `transaction_hash` —
|
|
208
|
-
|
|
209
|
-
re-ingestion
|
|
218
|
+
- `transaction_hash` — versioned SHA-256 fingerprint scoped to account,
|
|
219
|
+
currency, date, description, amount, and available bank identity.
|
|
220
|
+
Rebuild persisted pre-v2 keys before re-ingestion
|
|
210
221
|
- `confidence` — float between 0 and 1 for LLM rows, `None` for
|
|
211
222
|
deterministic
|
|
212
223
|
- `raw_source_text` — best-effort source-text slice for the v0.0.6
|
|
@@ -235,7 +246,9 @@ For ZIP archives, use `iter_secure_xml_entries()`.
|
|
|
235
246
|
from bankstatementparser import CamtParser, iter_secure_xml_entries
|
|
236
247
|
|
|
237
248
|
for entry in iter_secure_xml_entries("statements.zip"):
|
|
238
|
-
parser = CamtParser.from_bytes(
|
|
249
|
+
parser = CamtParser.from_bytes(
|
|
250
|
+
entry.xml_bytes, source_name=entry.source_name
|
|
251
|
+
)
|
|
239
252
|
transactions = parser.parse()
|
|
240
253
|
print(entry.source_name, len(transactions), "transactions")
|
|
241
254
|
```
|
|
@@ -257,23 +270,23 @@ Gemini, …).
|
|
|
257
270
|
|
|
258
271
|
```mermaid
|
|
259
272
|
flowchart TD
|
|
260
|
-
A[smart_ingest
|
|
261
|
-
B
|
|
262
|
-
C --> Z[IngestResult<br/>source_method='deterministic']
|
|
273
|
+
A["smart_ingest(path)"] --> B{"detect_statement_format"}
|
|
274
|
+
B -->|CAMT/PAIN/OFX/MT940/CSV| C["Path A: deterministic parser<br/>$0, fastest"]
|
|
275
|
+
C --> Z["IngestResult<br/>source_method='deterministic'"]
|
|
263
276
|
|
|
264
|
-
B
|
|
265
|
-
D --> E{text
|
|
277
|
+
B -->|pdf or unknown| D["pypdf extract_text"]
|
|
278
|
+
D --> E{"text length >= 50?"}
|
|
266
279
|
|
|
267
|
-
E
|
|
268
|
-
F --> Y[IngestResult<br/>source_method='llm']
|
|
280
|
+
E -->|yes| F["Path B: text-LLM<br/>default ollama/llama3"]
|
|
281
|
+
F --> Y["IngestResult<br/>source_method='llm'"]
|
|
269
282
|
|
|
270
|
-
E
|
|
271
|
-
G --> X[IngestResult<br/>source_method='vision']
|
|
283
|
+
E -->|no| G["Path C: vision-LLM<br/>opt-in via BSP_HYBRID_VISION_MODEL"]
|
|
284
|
+
G --> X["IngestResult<br/>source_method='vision'"]
|
|
272
285
|
|
|
273
|
-
Z --> V[verify_balance<br/>Golden Rule]
|
|
286
|
+
Z --> V["verify_balance<br/>Golden Rule"]
|
|
274
287
|
Y --> V
|
|
275
288
|
X --> V
|
|
276
|
-
V --> R[VERIFIED / DISCREPANCY / UNVERIFIABLE / FAILED]
|
|
289
|
+
V --> R["VERIFIED / DISCREPANCY / UNVERIFIABLE / FAILED"]
|
|
277
290
|
```
|
|
278
291
|
|
|
279
292
|
Every extracted row carries an immutable `transaction_hash`, an
|
|
@@ -313,7 +326,7 @@ for the full surface.
|
|
|
313
326
|
|---|---|
|
|
314
327
|
| **Golden Rule verification** *(v0.0.5)* | Every result carries `opening + credits − debits == closing` status: `VERIFIED`, `DISCREPANCY`, `UNVERIFIABLE`, or `FAILED`. |
|
|
315
328
|
| **Multi-currency verification** *(v0.0.8)* | `verify_balance_multi_currency()` groups transactions by currency and runs the Golden Rule independently per group — no more false `DISCREPANCY` on multi-currency statements. |
|
|
316
|
-
| **Idempotent dedup** *(v0.0.5)* | Every `Transaction` carries a stable `transaction_hash` (
|
|
329
|
+
| **Idempotent dedup** *(v0.0.5)* | Every `Transaction` carries a stable `transaction_hash` (versioned SHA-256 of account, currency, date, description, amount, and bank identity). `Deduplicator.dedupe_by_hash()` makes incremental ingestion safe to re-run. |
|
|
317
330
|
| **Interactive review** *(v0.0.6)* | `--type review` CLI walks through discrepancies with accept/edit/skip/delete/quit. `IngestResult.to_json()` / `.from_json()` for stable round-trip with embedded audit trail. |
|
|
318
331
|
|
|
319
332
|
### Enrichment & export
|
|
@@ -332,17 +345,18 @@ for the full surface.
|
|
|
332
345
|
|---|---|
|
|
333
346
|
| **PII redaction** | Names, IBANs, and addresses masked by default — opt in with `--show-pii` |
|
|
334
347
|
| **Secure ZIP** | `iter_secure_xml_entries()` rejects zip bombs, encrypted entries, and suspicious compression ratios |
|
|
335
|
-
| **Tested** |
|
|
348
|
+
| **Tested** | 1007 tests, coverage gated at 100% in CI, property-based fuzzing with Hypothesis |
|
|
336
349
|
|
|
337
350
|
---
|
|
338
351
|
|
|
339
352
|
## PII Redaction
|
|
340
353
|
|
|
341
354
|
PII (names, IBANs, addresses) is **redacted by default** in
|
|
342
|
-
console output
|
|
355
|
+
CLI console output. Python parser methods return full records unless
|
|
356
|
+
`redact_pii=True` is supplied.
|
|
343
357
|
|
|
344
358
|
```python
|
|
345
|
-
#
|
|
359
|
+
# Request redaction explicitly in Python
|
|
346
360
|
for tx in parser.parse_streaming(redact_pii=True):
|
|
347
361
|
print(tx) # Names and addresses show as ***REDACTED***
|
|
348
362
|
|
|
@@ -351,8 +365,9 @@ for tx in parser.parse_streaming(redact_pii=False):
|
|
|
351
365
|
print(tx)
|
|
352
366
|
```
|
|
353
367
|
|
|
354
|
-
|
|
355
|
-
|
|
368
|
+
Regular and hybrid file exports contain full records. Legacy CLI streaming
|
|
369
|
+
exports follow `--show-pii`; Python exports reflect the records supplied.
|
|
370
|
+
Choose redaction explicitly before sharing an exported file.
|
|
356
371
|
|
|
357
372
|
---
|
|
358
373
|
|
|
@@ -399,11 +414,13 @@ Process multiple files simultaneously across CPU cores:
|
|
|
399
414
|
```python
|
|
400
415
|
from bankstatementparser import parse_files_parallel
|
|
401
416
|
|
|
402
|
-
results = parse_files_parallel(
|
|
403
|
-
|
|
404
|
-
|
|
405
|
-
|
|
406
|
-
|
|
417
|
+
results = parse_files_parallel(
|
|
418
|
+
[
|
|
419
|
+
"statements/jan.xml",
|
|
420
|
+
"statements/feb.xml",
|
|
421
|
+
"statements/mar.xml",
|
|
422
|
+
]
|
|
423
|
+
)
|
|
407
424
|
|
|
408
425
|
for r in results:
|
|
409
426
|
print(r.path, r.status, len(r.transactions), "rows")
|
|
@@ -652,13 +669,22 @@ versioned companion packages build on it — each is its own repository
|
|
|
652
669
|
and Python package, so you install only what you need and the core
|
|
653
670
|
stays dependency-light.
|
|
654
671
|
|
|
655
|
-
| Package | PyPI | Role |
|
|
656
|
-
|
|
657
|
-
| [`bankstatementparser
|
|
658
|
-
| [`bankstatementparser-
|
|
659
|
-
| [`bankstatementparser-
|
|
660
|
-
| [`bankstatementparser-
|
|
661
|
-
| [`bankstatementparser-
|
|
672
|
+
| Package | GitHub Repository | PyPI | Role | Description |
|
|
673
|
+
|---|---|---|---|---|
|
|
674
|
+
| [`bankstatementparser`](https://github.com/sebastienrousseau/bankstatementparser) | [`sebastienrousseau/bankstatementparser`](https://github.com/sebastienrousseau/bankstatementparser) | [](https://pypi.org/project/bankstatementparser/) | Core Engine | Unified parser for CAMT (052/053), PAIN.001, CSV, OFX, QFX, MT940, and PDF statements |
|
|
675
|
+
| [`bankstatementparser-mcp`](https://github.com/sebastienrousseau/bankstatementparser-mcp) | [`sebastienrousseau/bankstatementparser-mcp`](https://github.com/sebastienrousseau/bankstatementparser-mcp) | [](https://pypi.org/project/bankstatementparser-mcp/) | AI Protocol | Model Context Protocol (MCP) server exposing statement tools to LLMs & AI agents |
|
|
676
|
+
| [`bankstatementparser-lsp`](https://github.com/sebastienrousseau/bankstatementparser-lsp) | [`sebastienrousseau/bankstatementparser-lsp`](https://github.com/sebastienrousseau/bankstatementparser-lsp) | [](https://pypi.org/project/bankstatementparser-lsp/) | Developer Tooling | Language Server Protocol (LSP) with live SWIFT MT940 statement validation & diagnostics |
|
|
677
|
+
| [`bankstatementparser-transport-ebics`](https://github.com/sebastienrousseau/bankstatementparser-transport-ebics) | [`sebastienrousseau/bankstatementparser-transport-ebics`](https://github.com/sebastienrousseau/bankstatementparser-transport-ebics) | [](https://pypi.org/project/bankstatementparser-transport-ebics/) | Transport | Automated bank statement retrieval over EBICS 3.0 (`H005`) and 2.5 (`H004`) protocols |
|
|
678
|
+
| [`bankstatementparser-writer-xlsx`](https://github.com/sebastienrousseau/bankstatementparser-writer-xlsx) | [`sebastienrousseau/bankstatementparser-writer-xlsx`](https://github.com/sebastienrousseau/bankstatementparser-writer-xlsx) | [](https://pypi.org/project/bankstatementparser-writer-xlsx/) | Output Writer | Formats and exports parsed banking transactions into styled Microsoft Excel (`.xlsx`) workbooks |
|
|
679
|
+
| [`bankstatementparser-writer-qif`](https://github.com/sebastienrousseau/bankstatementparser-writer-qif) | [`sebastienrousseau/bankstatementparser-writer-qif`](https://github.com/sebastienrousseau/bankstatementparser-writer-qif) | [](https://pypi.org/project/bankstatementparser-writer-qif/) | Output Writer | Serializes transactions into standard Quicken Interchange Format (`.qif`) exchange files |
|
|
680
|
+
| [`bankstatementparser-writer-ofx`](https://github.com/sebastienrousseau/bankstatementparser-writer-ofx) | [`sebastienrousseau/bankstatementparser-writer-ofx`](https://github.com/sebastienrousseau/bankstatementparser-writer-ofx) | [](https://pypi.org/project/bankstatementparser-writer-ofx/) | Output Writer | Serializes transactions into standard Open Financial Exchange (`.ofx`) XML/SGML files |
|
|
681
|
+
| [`bankstatementparser-writer-swift`](https://github.com/sebastienrousseau/bankstatementparser-writer-swift) | [`sebastienrousseau/bankstatementparser-writer-swift`](https://github.com/sebastienrousseau/bankstatementparser-writer-swift) | [](https://pypi.org/project/bankstatementparser-writer-swift/) | Output Writer | Exports transactions to SWIFT MT940 customer statements and MT942 interim reports |
|
|
682
|
+
| [`bankstatementparser-loader-bai2`](https://github.com/sebastienrousseau/bankstatementparser-loader-bai2) | [`sebastienrousseau/bankstatementparser-loader-bai2`](https://github.com/sebastienrousseau/bankstatementparser-loader-bai2) | [](https://pypi.org/project/bankstatementparser-loader-bai2/) | Input Loader | Parses BAI2 cash-management and account balance statements |
|
|
683
|
+
| [`bankstatementparser-loader-mt942`](https://github.com/sebastienrousseau/bankstatementparser-loader-mt942) | [`sebastienrousseau/bankstatementparser-loader-mt942`](https://github.com/sebastienrousseau/bankstatementparser-loader-mt942) | [](https://pypi.org/project/bankstatementparser-loader-mt942/) | Input Loader | Parses SWIFT MT942 interim transaction reports with credit/debit summary reconciliation |
|
|
684
|
+
| [`bankstatementparser-loader-cfonb`](https://github.com/sebastienrousseau/bankstatementparser-loader-cfonb) | [`sebastienrousseau/bankstatementparser-loader-cfonb`](https://github.com/sebastienrousseau/bankstatementparser-loader-cfonb) | [](https://pypi.org/project/bankstatementparser-loader-cfonb/) | Input Loader | Parses French CFONB 120 / AFB120 120-byte fixed-width banking statement files |
|
|
685
|
+
| [`bankstatementparser-loader-camt054`](https://github.com/sebastienrousseau/bankstatementparser-loader-camt054) | [`sebastienrousseau/bankstatementparser-loader-camt054`](https://github.com/sebastienrousseau/bankstatementparser-loader-camt054) | [](https://pypi.org/project/bankstatementparser-loader-camt054/) | Input Loader | Ingests ISO 20022 CAMT.054 real-time debit/credit notification stream XML |
|
|
686
|
+
| [`bankstatementparser-loader-sepa`](https://github.com/sebastienrousseau/bankstatementparser-loader-sepa) | [`sebastienrousseau/bankstatementparser-loader-sepa`](https://github.com/sebastienrousseau/bankstatementparser-loader-sepa) | [](https://pypi.org/project/bankstatementparser-loader-sepa/) | Input Loader | Ingests ISO 20022 SEPA PAIN.002 payment status reports and PAIN.008 direct debit mandates |
|
|
687
|
+
| [`bankstatementparser-loader-bacs`](https://github.com/sebastienrousseau/bankstatementparser-loader-bacs) | [`sebastienrousseau/bankstatementparser-loader-bacs`](https://github.com/sebastienrousseau/bankstatementparser-loader-bacs) | [](https://pypi.org/project/bankstatementparser-loader-bacs/) | Input Loader | Parses UK BACS Standard 18 / Faster Payments 106-byte fixed-width transmission files |
|
|
662
688
|
|
|
663
689
|
Loaders turn an additional source format into the same unified
|
|
664
690
|
`Transaction` model; writers take parsed data back out to another
|
|
@@ -666,8 +692,8 @@ target. Each companion pins `bankstatementparser` as a dependency and
|
|
|
666
692
|
ships its own 100%-coverage test suite.
|
|
667
693
|
|
|
668
694
|
```bash
|
|
669
|
-
# Mix and match: e.g.
|
|
670
|
-
pip install bankstatementparser-
|
|
695
|
+
# Mix and match: e.g. fetch via EBICS 3.0, parse CAMT.053, and export to SWIFT MT940 or Excel
|
|
696
|
+
pip install bankstatementparser-transport-ebics bankstatementparser-writer-swift bankstatementparser-writer-xlsx
|
|
671
697
|
```
|
|
672
698
|
|
|
673
699
|
Loaders hand back the same `Transaction` model the core parsers
|
|
@@ -679,15 +705,15 @@ cleanly — see each companion's README for runnable examples.
|
|
|
679
705
|
## Project Layout
|
|
680
706
|
|
|
681
707
|
```text
|
|
682
|
-
bankstatementparser/ Source code (
|
|
708
|
+
bankstatementparser/ Source code (39 modules)
|
|
683
709
|
bankstatementparser/hybrid/ PDF pipeline: orchestrator, llm_extractor, vision, scanner, ollama_direct, verification
|
|
684
710
|
bankstatementparser/enrichment/ Categorizer, AccountMapper, EnrichedTransaction
|
|
685
|
-
bankstatementparser/export/ hledger + beancount journal export
|
|
711
|
+
bankstatementparser/export/ hledger + beancount journal export, Apache Parquet columnar export
|
|
686
712
|
bankstatementparser/api.py REST API microservice (FastAPI)
|
|
687
713
|
docs/compliance/ ISO 13485 validation, risk register, traceability matrix
|
|
688
714
|
examples/ 14 deterministic + 9 hybrid runnable example scripts
|
|
689
715
|
scripts/ SBOM generation, checksums, signature verification
|
|
690
|
-
tests/
|
|
716
|
+
tests/ 1007 tests (unit, integration, property-based, security, hybrid mocks)
|
|
691
717
|
```
|
|
692
718
|
|
|
693
719
|
---
|
|
@@ -759,8 +785,9 @@ No. Zero network calls. XML parsers enforce `no_network=True`. No
|
|
|
759
785
|
cloud, no telemetry.
|
|
760
786
|
|
|
761
787
|
**Is PII redacted automatically?**
|
|
762
|
-
|
|
763
|
-
|
|
788
|
+
CLI console output masks personal fields by default. Python callers
|
|
789
|
+
must request `redact_pii=True`. Regular file exports retain full data;
|
|
790
|
+
legacy streaming exports follow `--show-pii`.
|
|
764
791
|
|
|
765
792
|
**Is the extraction deterministic?**
|
|
766
793
|
Yes. Same input produces byte-identical output. Critical for
|