data-science-document-ai 1.60.2__py3-none-any.whl → 1.60.3__py3-none-any.whl

{data_science_document_ai-1.60.2.dist-info → data_science_document_ai-1.60.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: data-science-document-ai
-Version: 1.60.2
+Version: 1.60.3
 Summary: "Document AI repo for data science"
 Author: Naomi Nguyen
 Author-email: naomi.nguyen@forto.com

{data_science_document_ai-1.60.2.dist-info → data_science_document_ai-1.60.3.dist-info}/RECORD

@@ -7,7 +7,7 @@ src/io.py,sha256=rYjXVLlriEacw1uNuPIYhg12bXNu48Qs9GYMY2YcVTE,5563
 src/llm.py,sha256=a7UYA4ITUNjzct_2fHgM-bma_XWc28VC0FV71g9tnUI,7137
 src/log_setup.py,sha256=RhHnpXqcl-ii4EJzRt47CF2R-Q3YPF68tepg_Kg7tkw,2895
 src/pdf_processing.py,sha256=Fx-Glb9niEUU3WUCrBZ02ZYV-E2vWoUM0ifN7-0A1Q4,19961
-src/postprocessing/common.py,sha256=dagAg0hZGuZc03bXdfOolxekewMEVUfz917IGCiAtWI,26118
+src/postprocessing/common.py,sha256=gcHZTuj_TRG4E0eTTpr7lUjTYKG9N4IiClvqiEoXGZ8,26591
 src/postprocessing/postprocess_booking_confirmation.py,sha256=nK32eDiBNbauyQz0oCa9eraysku8aqzrcoRFoWVumDU,4827
 src/postprocessing/postprocess_commercial_invoice.py,sha256=3I8ijluTZcOs_sMnFZxfkAPle0UFQ239EMuvZfDZVPg,1028
 src/postprocessing/postprocess_partner_invoice.py,sha256=WuaTQK5D09dV_QNrh29ZoKX9IvQn2Ub-WnAMyRjCsvI,14240
@@ -27,24 +27,24 @@ src/prompts/library/bookingConfirmation/other/placeholders.json,sha256=zkpcKzOTC
 src/prompts/library/bookingConfirmation/other/prompt.txt,sha256=EMmKLDX7ir96KcU7Ca12B7k5-crdioU-Sw3VebX33HY,7867
 src/prompts/library/bookingConfirmation/yangming/placeholders.json,sha256=2fu7q7J4QxrYfmZJlpAfDs9wIHKTheyysIw7i7UmOR0,6254
 src/prompts/library/bookingConfirmation/yangming/prompt.txt,sha256=c_qHnP7f2JTCumDRN0aQ9QpXmglLPRuaMEiDXq3wMJ0,4605
-src/prompts/library/bundeskasse/other/placeholders.json,sha256=7xKzi_ypkIICO9nrEl45W9G7-h33uWVRVWnpg2b5lUg,4288
-src/prompts/library/bundeskasse/other/prompt.txt,sha256=miNYoqRZEd6Z1LNisTahX1-tenzr5kEpRA6gvPH7NCw,3316
+src/prompts/library/bundeskasse/other/placeholders.json,sha256=6VvXUluc_shZNg8jb40JiaDjovNhuoEHpNSfsTzoRs4,4456
+src/prompts/library/bundeskasse/other/prompt.txt,sha256=prmk2BIGlSZ_1vhi5WcBUm91D5zGJJT3fbbLEQ3YeTk,3487
 src/prompts/library/commercialInvoice/other/placeholders.json,sha256=zUK2mg9MnHiEQRYF6VgTiUiq68WGy5f7_4qL63CWyR0,4700
 src/prompts/library/commercialInvoice/other/prompt.txt,sha256=CJapcVrmcvynJUanETDklkzU-0N9hHdhq5wL4MK7OIY,2683
 src/prompts/library/customsAssessment/other/placeholders.json,sha256=scIV--C9HNWAQbU9zEz3GT_FoAvJqbfuY85YUtt7t-Q,3850
 src/prompts/library/customsAssessment/other/prompt.txt,sha256=z3FuoHZ588Pz1WBJDW7ISAC3J6n7hPJCcS92CdHDTFw,2494
-src/prompts/library/customsInvoice/other/placeholders.json,sha256=hZb6Gl_rUMzmuiXuB39TBHwL08h2EDXa2AASNPgZEeM,12169
-src/prompts/library/customsInvoice/other/prompt.txt,sha256=N1kzaxOlgzc779V4ZjPO9s_djMyiVY5I7uBx7-ZMxUA,10225
+src/prompts/library/customsInvoice/other/placeholders.json,sha256=Y3nztiKvaqll7Oq2Vo1MjhjqNBG_HyiB_4lYtwCOB0g,12263
+src/prompts/library/customsInvoice/other/prompt.txt,sha256=i1ox7ZuWuK69DOllzYIB8sKduFYmxyI-OAtSP_tf6IQ,10297
 src/prompts/library/deliveryOrder/other/placeholders.json,sha256=j-9F4V3yDg4610PPsOwU3oOj_S9vAvAB9Ix155WGIwc,3827
 src/prompts/library/deliveryOrder/other/prompt.txt,sha256=RD076vq0x0IjoEVQfh-G0u4nxITCpgKZGrwMlR9YAvk,2695
 src/prompts/library/draftMbl/other/placeholders.json,sha256=Gn8kQ8cMmrzRGLSFH7_8wO1_j2jxhqHd4zeivZP2SjU,4304
-src/prompts/library/draftMbl/other/prompt.txt,sha256=4RjlGT2OFmcBCUJhuCnO9GtmCn3vVesXHi_ml2g3dK8,2386
+src/prompts/library/draftMbl/other/prompt.txt,sha256=Z4uH7lL01ok0GajHWQ0feCKJqRIRj_lyFMmc3-wqfpU,2458
 src/prompts/library/finalMbL/other/placeholders.json,sha256=Gn8kQ8cMmrzRGLSFH7_8wO1_j2jxhqHd4zeivZP2SjU,4304
-src/prompts/library/finalMbL/other/prompt.txt,sha256=cyeKjK94sepqXiLEeZKB4VpmT0-nqXALP4dih-B67M8,2386
+src/prompts/library/finalMbL/other/prompt.txt,sha256=68IHh7DDAoxQAID2dsRmKPUVQmX9vxxmMcUqnDN8-j0,2458
 src/prompts/library/packingList/other/placeholders.json,sha256=cGUUvEFoi4Lm0BAiyD29KbNFbUgzO1s7eit_qK3F0ig,4478
 src/prompts/library/packingList/other/prompt.txt,sha256=6Q9d0KBG6YWmNtzFivvmtQmitaUE2jytfwwc5YwsUgQ,2872
-src/prompts/library/partnerInvoice/other/placeholders.json,sha256=igfmag8KGQ7106aSG7UHZEksO4LB-qWeq8SS8VDzckM,10894
-src/prompts/library/partnerInvoice/other/prompt.txt,sha256=QegtgBgB6RmeAhBDguuczJ7UIqOfLnCAboYNaB7fLFI,8411
+src/prompts/library/partnerInvoice/other/placeholders.json,sha256=S5zWxVM6Ex23drBg6wGKLk_Vd2KJon6mdK1ag3E4RBs,10988
+src/prompts/library/partnerInvoice/other/prompt.txt,sha256=Qw27Tj9Kd3exJi1Wd57Ne_sIaFa8j5qSPwI2y1xj55M,8483
 src/prompts/library/postprocessing/port_code/placeholders.json,sha256=2TiXf3zSzrglOMPtDOlCntIa5RSvyZQAKG2-IgrCY5A,22
 src/prompts/library/postprocessing/port_code/prompt_port_code.txt,sha256=--1wunSqEr2ox958lEhjO-0JFBfOLzA3qfKYIzG_Iok,884
 src/prompts/library/preprocessing/carrier/placeholders.json,sha256=tQeVDtvembhVqvel9vGoy4qcKp1hOvg-bLCgZRdQj0g,192
@@ -55,6 +55,6 @@ src/prompts/prompt_library.py,sha256=VJWHeXN-s501C2GiidIIvQQuZdU6T1R27hE2dKBiI40
 src/setup.py,sha256=8-vZWjC8Iwa3xxdk3iR4412VCjtNtgzVqkXcFon7UBE,7309
 src/tms.py,sha256=UXbIo1QE--hIX6NZi5Qyp2R_CP338syrY9pCTPrfgnE,1741
 src/utils.py,sha256=Ow5_Jals88o8mbZ1BoHfZpHZoCfig_UQb5aalH-mpWE,17278
-data_science_document_ai-1.60.2.dist-info/METADATA,sha256=OchQC8yfBcQyJrk-zv2uTqC3DWlXYnQMQ7SansDP5eU,2152
-data_science_document_ai-1.60.2.dist-info/WHEEL,sha256=zp0Cn7JsFoX2ATtOhtaFYIiE2rmFAD4OcMhtUki8W3U,88
-data_science_document_ai-1.60.2.dist-info/RECORD,,
+data_science_document_ai-1.60.3.dist-info/METADATA,sha256=FyamkknqoADm8IVibk-09-6grFI8qfofMblJTHabohE,2152
+data_science_document_ai-1.60.3.dist-info/WHEEL,sha256=zp0Cn7JsFoX2ATtOhtaFYIiE2rmFAD4OcMhtUki8W3U,88
+data_science_document_ai-1.60.3.dist-info/RECORD,,

src/postprocessing/common.py

@@ -723,6 +723,17 @@ async def format_all_entities(result, document_type_code, params, mime_type):
     if document_type_code in ["partnerInvoice", "bundeskasse"]:
         await process_partner_invoice(params, aggregated_data, document_type_code)
 
+    # TODO: This is a temporary change until the terminal codes are updated
+    if document_type_code == "bookingConfirmation":
+        if "gateInTerminalCode" in aggregated_data:
+            aggregated_data["gateInTerminal"] = aggregated_data.pop(
+                "gateInTerminalCode"
+            )
+        if "pickUpTerminalCode" in aggregated_data:
+            aggregated_data["pickUpTerminal"] = aggregated_data.pop(
+                "pickUpTerminalCode"
+            )
+
     logger.info("Data Extraction completed successfully")
     return aggregated_data
 

src/prompts/library/bundeskasse/other/placeholders.json

@@ -93,14 +93,14 @@
       "invoiceNumber": {
         "type": "STRING",
         "nullable": true,
-        "description": "Invoice Number is a unique identifier for the invoice, it starts with \"ATC\", \"AT-C\", or \"AT/C\" only (e.g., ATC40, AT-C-40-, AT/C/40/....). Do NOT extract \"NIZZA-Registrierkennzeichen number."
+        "description": "Invoice Number is a unique identifier for the invoice, it starts with ATC, AT-C, or AT/C only (e.g., ATC40, AT-C-40-, AT/C/40/....) It can be found just below the title of the invoice or in the top section of the invoice. Do NOT extract NIZZA-Registrierkennzeichen number (e.g. ATC0040M00...)."
       },
       "containerNumber": {
         "type": "ARRAY",
         "items": {
           "type": "STRING",
           "nullable": true,
-          "description": "The unique identifier for each container. It always starts with 4 capital letters and followed by 7 digits. Example: TEMU7972458."
+          "description": "The unique identifier for each container. It always starts with 4 capital letters and followed by 7 digits. Example: TEMU7972458. Do not get confused between 0 vs O in the 7 digits of container number."
       }
   },
       "creditNoteInvoiceNumber": {

src/prompts/library/bundeskasse/other/prompt.txt

@@ -16,7 +16,7 @@ Your role is to accurately extract specific entities from these Customs invoices
 - The amount and the currency is always in EUR both for grandTotal and line items.
 
 - containerNumber:
-    - Container Number consists of 4 capital letters followed by 7 digits (e.g., TEMU7972458, CAIU7222892).
+    - Container Number consists of 4 capital letters followed by 7 digits (e.g., TEMU7972458, CAIU7222892). Do not get confused between 0 vs O in the 7 digits of container number.
     - Few invoices contains multiple container numbers, in that case, all container numbers should be captured.
 
 - shipmentID:
@@ -26,7 +26,8 @@ Your role is to accurately extract specific entities from these Customs invoices
 
 - invoiceNumber:
     - Invoice Number is a unique identifier for the invoice, it starts with "ATC", "AT-C", or "AT/C" only (e.g., ATC40..., AT-C-40-..., AT/C/40/....).
-    - Do NOT extract if the text is about vehicle registrations, license plates, or location identifiers (e.g., "NIZZA-Registrierkennzeichen: ATC0040....")
+    - It can be found just below the title of the invoice or in the top section of the invoice.
+    - Do NOT extract if the text is about vehicle registrations, license plates, or location identifiers (e.g., "NIZZA-Registrierkennzeichen: ATC0040M00....")
 
 - creditNoteInvoiceNumber:
     - Credit Note Invoice Number is a unique identifier for the credit note, it starts with "ATS" only (e.g., ATS.....).

src/prompts/library/customsInvoice/other/placeholders.json

@@ -93,7 +93,8 @@
                         "description": "The percentage rate of VAT applied to the totalAmount of the line item. This is used to calculate the vatAmount."
                     },
                     "containerNumber": {"type": "STRING", "nullable": true,
-                        "description": "The container number associated with the line item. containerNumber MUST start with 4 letters followed by 7 digits (e.g., CMAU1234567)"},
+                        "description": "The container number associated with the line item. containerNumber MUST start with 4 letters followed by 7 digits (e.g., CMAU1234567). Do not get confused between 0 vs O in the 7 digits of container number."
+                    },
                     "containerSize": {"type": "STRING", "nullable": true,
                         "description": "The size of the container associated with the containerNumber, such as 20ft, 40ft, 40HC, 20DC etc."}
                 }

src/prompts/library/customsInvoice/other/prompt.txt

@@ -59,7 +59,7 @@ Your role is to accurately extract specific entities from these invoices to supp
     - totalAmount: The total amount for the item. It can be in different currencies, so ensure to capture the currency as well for the totalAmountCurrency.
     - totalAmountEuro: Few line items contains a total amount in Euro. You can find it by looking for the term "Total EUR" or "Amount in Euro" in the line item but it's always in the EURO / € currency. Sometimes, it can be same as totalAmount if the line item is already in Euro.
     - quantity: The quantity of the item or service provided in the line item. Pay attention to 2 x 40HC or 2x40HC. It means, quantity is 2 and 40HC is containerSize but not 240.
-    - containerNumber: Container Number always starts with 4 letters and is followed by 7 digits (e.g., ABCD1234567).
+    - containerNumber: Container Number always starts with 4 letters and is followed by 7 digits (e.g., ABCD1234567). Do not get confused between 0 vs O in the 7 digits of container number.
 
 - hblNumber and mblNumber:
     - The Master Bill of Lading number. Commonly known as "Bill of Lading Number", "BILL OF LADING NO.", "BL Number", "BL No.", "B/L No.", "BL-Nr.", "B/L", or "HBL No.".

src/prompts/library/draftMbl/other/prompt.txt

@@ -28,7 +28,7 @@ Your role is to accurately extract specific entities from these draftMBLs to sup
     - Vessel Name is the name of the ship carrying the cargo. It can be referred to as "Vessel", "Ship Name", "Schiff", "Schiffsname", "Nave", or "Vessel/Flight No.".
 
 - containers: Details of each container on the draftMBL. Make sure to extract each container information separately.
-    - containerNumber: Container Number consists of 4 capital letters followed by 7 digits (e.g., TEMU7972458, CAIU 7222892).
+    - containerNumber: Container Number consists of 4 capital letters followed by 7 digits (e.g., TEMU7972458, CAIU 7222892). Do not get confused between 0 vs O in the 7 digits of container number.
     - sealNumber: Seal numbers are unique identifiers for shipping seals. They are usually mentioned as seal numbers in the document but they are definitely not container numbers.
 
 <INSTRUCTIONS>

src/prompts/library/finalMbL/other/prompt.txt

@@ -28,7 +28,7 @@ Your role is to accurately extract specific entities from these finalMBLs to sup
     - Vessel Name is the name of the ship carrying the cargo. It can be referred to as "Vessel", "Ship Name", "Schiff", "Schiffsname", "Nave", or "Vessel/Flight No.".
 
 - containers: Details of each container on the finalMBL. Make sure to extract each container information separately.
-    - containerNumber: Container Number consists of 4 capital letters followed by 7 digits (e.g., TEMU7972458, CAIU 7222892).
+    - containerNumber: Container Number consists of 4 capital letters followed by 7 digits (e.g., TEMU7972458, CAIU 7222892). Do not get confused between 0 vs O in the 7 digits of container number.
     - sealNumber: Seal numbers are unique identifiers for shipping seals. They are usually mentioned as seal numbers in the document but they are definitely not container numbers.
 
 <INSTRUCTIONS>

src/prompts/library/partnerInvoice/other/placeholders.json

@@ -91,7 +91,8 @@
                         "description": "The percentage rate of VAT applied to the totalAmount of the line item. This is used to calculate the vatAmount."
                     },
                     "containerNumber": {"type": "STRING", "nullable": true,
-                        "description": "The container number associated with the line item. containerNumber MUST start with 4 letters followed by 7 digits (e.g., CMAU1234567)"},
+                        "description": "The container number associated with the line item. containerNumber MUST start with 4 letters followed by 7 digits (e.g., CMAU1234567). Do not get confused between 0 vs O in the 7 digits of container number."
+                    },
                     "containerSize": {"type": "STRING", "nullable": true,
                         "description": "The size of the container associated with the containerNumber, such as 20ft, 40ft, 40HC, 20DC etc."}
                 }

src/prompts/library/partnerInvoice/other/prompt.txt

@@ -57,7 +57,7 @@ Your role is to accurately extract specific entities from these invoices to supp
     - totalAmount: The total amount for the item. It can be in different currencies, so ensure to capture the currency as well for the totalAmountCurrency.
     - totalAmountEuro: Few line items contains a total amount in Euro. You can find it by looking for the term "Total EUR" or "Amount in Euro" in the line item but it's always in the EURO / € currency. Sometimes, it can be same as totalAmount if the line item is already in Euro.
     - quantity: The quantity of the item or service provided in the line item. Pay attention to 2 x 40HC or 2x40HC. It means, quantity is 2 and 40HC is containerSize but not 240.
-    - containerNumber: Container Number always starts with 4 letters and is followed by 7 digits (e.g., ABCD1234567, XALU 8593678).
+    - containerNumber: Container Number always starts with 4 letters and is followed by 7 digits (e.g., ABCD1234567, XALU 8593678). Do not get confused between 0 vs O in the 7 digits of container number.
 
 - hblNumber and mblNumber:
     - The Master Bill of Lading number. Commonly known as "Bill of Lading Number", "BILL OF LADING NO.", "BL Number", "BL No.", "B/L No.", "BL-Nr.", "B/L", or "HBL No.".