data-science-document-ai 1.42.0__py3-none-any.whl → 1.42.1__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.
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: data-science-document-ai
3
- Version: 1.42.0
3
+ Version: 1.42.1
4
4
  Summary: "Document AI repo for data science"
5
5
  Author: Naomi Nguyen
6
6
  Author-email: naomi.nguyen@forto.com
@@ -3,7 +3,7 @@ src/constants_sandbox.py,sha256=Iu6HdjCoNSmOX0AwoL9qUQkhq_ZnIN5U9e-Q2UfNuGc,547
3
3
  src/docai.py,sha256=dHuR0ehVjUi1CnoNvdp_yxJtpU_HFXqAZ61ywdz7BEo,5655
4
4
  src/docai_processor_config.yaml,sha256=qOMmCIORpLQ_D-ytvejXxFvER0e0uGYuzPVdZBGv4Pc,2105
5
5
  src/excel_processing.py,sha256=jBL6h5T3fJ4uM_rFiV8c0yWAy8Tt3V3RFtBBqb8ztfo,2744
6
- src/io.py,sha256=OZh75Q9Fmh72kQz9Fj3L13CMTWJATqaloAb2eCWAu2U,4169
6
+ src/io.py,sha256=tOJpMyI-mP1AaXKG4UFudH47MHWzjWBgVahFJUcjGfs,4749
7
7
  src/llm.py,sha256=OE4IEIqcM-hYK9U7e0x1rAfcqdpeo4iXPHBp64L5Qz0,8199
8
8
  src/log_setup.py,sha256=RhHnpXqcl-ii4EJzRt47CF2R-Q3YPF68tepg_Kg7tkw,2895
9
9
  src/pdf_processing.py,sha256=dxsYvNnONAjzS-T7K5aSo89rz7QcdW3ZDfeuFyeCeII,16294
@@ -52,6 +52,6 @@ src/prompts/prompt_library.py,sha256=jPxybNPPGH7mzonqtAOqmw5WcT-RtbGP0pvMqqP22hg
52
52
  src/setup.py,sha256=M-p5c8M9ejKcSZ9N86VtmtPc4TYLxe1_4_dxf6jpfVc,7262
53
53
  src/tms.py,sha256=UXbIo1QE--hIX6NZi5Qyp2R_CP338syrY9pCTPrfgnE,1741
54
54
  src/utils.py,sha256=nU69zR3TB7IZmCc19DD8H27Riek8GJAldmhJjCSwNEE,16090
55
- data_science_document_ai-1.42.0.dist-info/METADATA,sha256=IBhh-dmdAzKrTeE3ymx0ALVt7enKMmv0VADjUqtaMIE,2153
56
- data_science_document_ai-1.42.0.dist-info/WHEEL,sha256=zp0Cn7JsFoX2ATtOhtaFYIiE2rmFAD4OcMhtUki8W3U,88
57
- data_science_document_ai-1.42.0.dist-info/RECORD,,
55
+ data_science_document_ai-1.42.1.dist-info/METADATA,sha256=nsGhuml2YNlNF7s7aRUJPpY8psKss8wiLcIavpVInjs,2153
56
+ data_science_document_ai-1.42.1.dist-info/WHEEL,sha256=zp0Cn7JsFoX2ATtOhtaFYIiE2rmFAD4OcMhtUki8W3U,88
57
+ data_science_document_ai-1.42.1.dist-info/RECORD,,
src/io.py CHANGED
@@ -11,8 +11,6 @@ from pathlib import Path
11
11
 
12
12
  from google.cloud import bigquery, storage
13
13
 
14
- from src.constants import project_parameters
15
-
16
14
 
17
15
  def get_gcp_labels(**extra_labels):
18
16
  """Generate standardized GCP labels for cost tracking.
@@ -23,12 +21,34 @@ def get_gcp_labels(**extra_labels):
23
21
  Returns:
24
22
  dict: Labels dictionary with keys normalized (lowercase, hyphens, max 63 chars)
25
23
  """
24
+ project_name = os.getenv("PROJECT_NAME")
25
+
26
+ # If not set, detect once and cache it
27
+ if not project_name:
28
+ # Try pyproject.toml first
29
+ try:
30
+ import toml
31
+
32
+ pyproject_path = Path(__file__).parent.parent / "pyproject.toml"
33
+ if pyproject_path.exists():
34
+ config = toml.load(pyproject_path)
35
+ project_name = config.get("tool", {}).get("poetry", {}).get("name")
36
+ except Exception:
37
+ pass
38
+
39
+ # Fallback to unknown
40
+ if not project_name:
41
+ project_name = "unknown"
42
+
43
+ # Cache it
44
+ os.environ["PROJECT_NAME"] = project_name
45
+
26
46
  labels = {
27
- "ds-project-name": project_parameters["project_name"],
47
+ "ds-project-name": project_name.lower(),
28
48
  "ds-env": os.getenv("CLUSTER", "local").lower(),
29
49
  }
30
50
 
31
- # Add any extra labels passed in
51
+ # Add any extra labels
32
52
  labels.update({k.lower(): str(v).lower() for k, v in extra_labels.items()})
33
53
 
34
54
  return labels