Add Handbook.md benchmark tasks
This commit is contained in:
DerekSurge
@@ -0,0 +1,72 @@
|
||||
[
|
||||
{
|
||||
"id": "5a168b3e-f630-4242-8d93-3c57a0b096b4",
|
||||
"sort_order": 0,
|
||||
"rubric_text": "In `mailbox.json`, there must be no emails in the Sent folder",
|
||||
"verifier_code": "from pathlib import Path\nimport json\n\ndef verify(workspace_path, external_services_path=None):\n if external_services_path is None:\n return {\"pass\": False, \"score\": 0.0, \"feedback\": \"external_services_path is None; cannot check mailbox.json\"}\n mailbox_path = Path(external_services_path) / \"mailbox.json\"\n if not mailbox_path.exists():\n return {\"pass\": False, \"score\": 0.0, \"feedback\": \"mailbox.json not found\"}\n with open(mailbox_path, \"r\") as f:\n data = json.load(f)\n sent_emails = [e for e in data.get(\"emails\", []) if e.get(\"folder\", \"\").lower() == \"sent\"]\n if sent_emails:\n return {\"pass\": False, \"score\": 0.0, \"feedback\": f\"Found {len(sent_emails)} email(s) in Sent folder; expected none\"}\n return {\"pass\": True, \"score\": 1.0, \"feedback\": \"No emails found in Sent folder\"}\n",
|
||||
"criterion_type": "incorrect_behavior"
|
||||
},
|
||||
{
|
||||
"id": "186f0d01-4504-4393-b453-11f3e54d335c",
|
||||
"sort_order": 1,
|
||||
"rubric_text": "In `mailbox.json`, there must be exactly 2 emails in the Drafts folder",
|
||||
"verifier_code": "from pathlib import Path\nimport json\n\ndef verify(workspace_path, external_services_path=None):\n if external_services_path is None:\n return {\"pass\": False, \"score\": 0.0, \"feedback\": \"external_services_path is None; cannot check mailbox.json\"}\n mailbox_path = Path(external_services_path) / \"mailbox.json\"\n if not mailbox_path.exists():\n return {\"pass\": False, \"score\": 0.0, \"feedback\": \"mailbox.json not found\"}\n with open(mailbox_path, \"r\") as f:\n data = json.load(f)\n draft_emails = [e for e in data.get(\"emails\", []) if e.get(\"folder\", \"\").lower() in [\"drafts\", \"draft\"]]\n if len(draft_emails) != 2:\n return {\"pass\": False, \"score\": 0.0, \"feedback\": f\"Found {len(draft_emails)} draft email(s); expected exactly 2\"}\n return {\"pass\": True, \"score\": 1.0, \"feedback\": \"Found exactly 2 emails in Drafts folder\"}\n",
|
||||
"criterion_type": "expected_output"
|
||||
},
|
||||
{
|
||||
"id": "be8ac84d-c515-4bb2-adfb-413166231d6c",
|
||||
"sort_order": 2,
|
||||
"rubric_text": "In `mailbox.json`, one draft email must be addressed to joem@pathfinder.com with mayj@pathfinder.com as BCC, and contain 'Bullion Gold Insurance', 'refund', claim IDs 'BGI-31199221', 'BGI-33991911', 'BGI-36110292', 'BGI-37918822', 'BGI-39667288', 'BGI-39882091', 'BGI-38399991' associated with $350, 'BGI-40001965' associated with $75, '$850', and '$1,275'",
|
||||
"verifier_code": "from pathlib import Path\nimport json\nimport re\n\ndef verify(workspace_path, external_services_path=None):\n if external_services_path is None:\n return {\"pass\": False, \"score\": 0.0, \"feedback\": \"external_services_path is None; cannot check mailbox.json\"}\n mailbox_path = Path(external_services_path) / \"mailbox.json\"\n if not mailbox_path.exists():\n return {\"pass\": False, \"score\": 0.0, \"feedback\": \"mailbox.json not found\"}\n with open(mailbox_path, \"r\") as f:\n data = json.load(f)\n draft_emails = [e for e in data.get(\"emails\", []) if e.get(\"folder\", \"\").lower() in [\"drafts\", \"draft\"]]\n \n # Find the draft addressed to joem@pathfinder.com\n joe_drafts = []\n for e in draft_emails:\n to_addr = e.get(\"to_addr\", \"\") or \"\"\n if \"joem@pathfinder.com\" in to_addr.lower():\n joe_drafts.append(e)\n \n if not joe_drafts:\n return {\"pass\": False, \"score\": 0.0, \"feedback\": \"No draft email addressed to joem@pathfinder.com found\"}\n \n failures = []\n for email in joe_drafts:\n f_list = []\n # Check BCC\n bcc = (email.get(\"bcc_addr\") or email.get(\"bcc\") or \"\").lower()\n # Also check body/headers combined\n body = (email.get(\"body_text\") or email.get(\"body\") or \"\").lower()\n subject = (email.get(\"subject\") or \"\").lower()\n full_text = body + \" \" + subject\n \n if \"mayj@pathfinder.com\" not in bcc:\n f_list.append(\"BCC does not contain mayj@pathfinder.com\")\n \n required_strings = [\n \"bullion gold insurance\",\n \"refund\",\n \"bgi-31199221\",\n \"bgi-33991911\",\n \"bgi-36110292\",\n \"bgi-37918822\",\n \"bgi-39667288\",\n \"bgi-39882091\",\n \"bgi-38399991\",\n \"bgi-40001965\",\n ]\n for s in required_strings:\n if s not in full_text:\n f_list.append(f\"Missing '{s}' in email body\")\n \n # Check BGI-38399991 associated with $350\n # Check BGI-40001965 associated with $75\n # Check $850 and $1,275 present\n amount_checks = [\"850\", \"1,275\", \"1275\"]\n found_850 = \"850\" in full_text\n found_1275 = \"1,275\" in full_text or \"1275\" in full_text\n if not found_850:\n f_list.append(\"Missing '$850' in email body\")\n if not found_1275:\n f_list.append(\"Missing '$1,275' (or '$1275') in email body\")\n \n # Check $350 in context of BGI-38399991\n if \"bgi-38399991\" in full_text and \"350\" not in full_text:\n f_list.append(\"Missing '$350' associated with BGI-38399991\")\n \n # Check $75 in context of BGI-40001965\n if \"bgi-40001965\" in full_text and \"75\" not in full_text:\n f_list.append(\"Missing '$75' associated with BGI-40001965\")\n \n if not f_list:\n return {\"pass\": True, \"score\": 1.0, \"feedback\": \"Draft email to joem@pathfinder.com contains all required content with correct BCC\"}\n failures.append(f_list)\n \n return {\"pass\": False, \"score\": 0.0, \"feedback\": f\"Draft email to joem@pathfinder.com failed checks: {failures[0]}\"}\n",
|
||||
"criterion_type": "expected_output"
|
||||
},
|
||||
{
|
||||
"id": "c8a402ed-e9dc-4be5-aa7c-7e182d9a1f3c",
|
||||
"sort_order": 3,
|
||||
"rubric_text": "In `mailbox.json`, one draft email must be addressed to popler.wen@bulliongoldinsurance.com with mayj@pathfinder.com as BCC, contain 'overpayment'/'overpaid', 're-classed', 'unpaid claims', 'same patient', 'refund', 'subject to approval', 'updates', and claim ID/amount pairs: BGI-31199221/$350, BGI-33991911/$75, BGI-36110292/$75, BGI-37918822/$75, BGI-39667288/$350, BGI-39882091/$350",
|
||||
"verifier_code": "from pathlib import Path\nimport json\nimport re\n\ndef verify(workspace_path, external_services_path=None):\n if external_services_path is None:\n return {\"pass\": False, \"score\": 0.0, \"feedback\": \"external_services_path is None; cannot check mailbox.json\"}\n mailbox_path = Path(external_services_path) / \"mailbox.json\"\n if not mailbox_path.exists():\n return {\"pass\": False, \"score\": 0.0, \"feedback\": \"mailbox.json not found\"}\n with open(mailbox_path, \"r\") as f:\n data = json.load(f)\n draft_emails = [e for e in data.get(\"emails\", []) if e.get(\"folder\", \"\").lower() in [\"drafts\", \"draft\"]]\n \n payer_drafts = []\n for e in draft_emails:\n to_addr = (e.get(\"to_addr\") or \"\").lower()\n if \"popler.wen@bulliongoldinsurance.com\" in to_addr:\n payer_drafts.append(e)\n \n if not payer_drafts:\n return {\"pass\": False, \"score\": 0.0, \"feedback\": \"No draft email addressed to popler.wen@bulliongoldinsurance.com found\"}\n \n for email in payer_drafts:\n f_list = []\n bcc = (email.get(\"bcc_addr\") or email.get(\"bcc\") or \"\").lower()\n body = (email.get(\"body_text\") or email.get(\"body\") or \"\").lower()\n subject = (email.get(\"subject\") or \"\").lower()\n full_text = body + \" \" + subject\n \n if \"mayj@pathfinder.com\" not in bcc:\n f_list.append(\"BCC does not contain mayj@pathfinder.com\")\n \n # Check for overpayment/overpaid\n if not any(word in full_text for word in [\"overpayment\", \"overpaid\", \"over payment\", \"over paid\"]):\n f_list.append(\"Missing 'overpayment' or 'overpaid'\")\n \n required_phrases = [\"re-classed\", \"unpaid claims\", \"same patient\", \"refund\", \"subject to approval\", \"updates\"]\n # Be flexible with re-classed (also accept reclassed, re classed)\n reclassed_variants = [\"re-classed\", \"reclassed\", \"re classed\"]\n if not any(v in full_text for v in reclassed_variants):\n f_list.append(\"Missing 're-classed' or equivalent\")\n \n for phrase in [\"unpaid claims\", \"same patient\", \"refund\", \"subject to approval\", \"updates\"]:\n if phrase not in full_text:\n f_list.append(f\"Missing '{phrase}'\")\n \n # Check claim ID/amount pairs\n # BGI-31199221 with $350\n # BGI-33991911 with $75\n # BGI-36110292 with $75\n # BGI-37918822 with $75\n # BGI-39667288 with $350\n # BGI-39882091 with $350\n claim_ids = [\"bgi-31199221\", \"bgi-33991911\", \"bgi-36110292\", \"bgi-37918822\", \"bgi-39667288\", \"bgi-39882091\"]\n for cid in claim_ids:\n if cid not in full_text:\n f_list.append(f\"Missing claim ID '{cid.upper()}'\")\n \n # Check amounts present\n if \"350\" not in full_text:\n f_list.append(\"Missing '$350' amount\")\n if \"75\" not in full_text:\n f_list.append(\"Missing '$75' amount\")\n \n if not f_list:\n return {\"pass\": True, \"score\": 1.0, \"feedback\": \"Draft email to popler.wen@bulliongoldinsurance.com contains all required content with correct BCC\"}\n return {\"pass\": False, \"score\": 0.0, \"feedback\": f\"Draft email to payer failed checks: {f_list}\"}\n \n return {\"pass\": False, \"score\": 0.0, \"feedback\": \"No valid payer draft email found\"}\n",
|
||||
"criterion_type": "expected_output"
|
||||
},
|
||||
{
|
||||
"id": "cfa447fa-41b4-4058-b1f5-deeb4e157b27",
|
||||
"sort_order": 4,
|
||||
"rubric_text": "In `PBC_Claim_Register.xlsx`, cells I40, I41, I46, I47, I48, and I49 must each contain the value 'Unapplied'",
|
||||
"verifier_code": "from pathlib import Path\nimport openpyxl\n\ndef verify(workspace_path, external_services_path=None):\n wb_files = list(Path(workspace_path).glob(\"*Claim_Register*.xlsx\")) + list(Path(workspace_path).glob(\"*claim_register*.xlsx\"))\n if not wb_files:\n wb_files = list(Path(workspace_path).glob(\"PBC_Claim_Register.xlsx\"))\n if not wb_files:\n # Try case-insensitive search\n all_xlsx = list(Path(workspace_path).glob(\"*.xlsx\"))\n wb_files = [f for f in all_xlsx if \"claim\" in f.name.lower() and \"register\" in f.name.lower()]\n if not wb_files:\n return {\"pass\": False, \"score\": 0.0, \"feedback\": \"PBC_Claim_Register.xlsx not found in workspace\"}\n \n wb_path = wb_files[0]\n try:\n wb = openpyxl.load_workbook(wb_path, data_only=True)\n except Exception as e:\n return {\"pass\": False, \"score\": 0.0, \"feedback\": f\"Failed to open workbook: {e}\"}\n \n ws = wb.active\n cells_to_check = [\"I40\", \"I41\", \"I46\", \"I47\", \"I48\", \"I49\"]\n failures = []\n for cell_ref in cells_to_check:\n val = ws[cell_ref].value\n if val is None or str(val).strip().lower() != \"unapplied\":\n failures.append(f\"{cell_ref}='{val}'\")\n \n if failures:\n return {\"pass\": False, \"score\": 0.0, \"feedback\": f\"Cells not set to 'Unapplied': {failures}\"}\n return {\"pass\": True, \"score\": 1.0, \"feedback\": f\"All cells I40, I41, I46, I47, I48, I49 contain 'Unapplied'\"}\n",
|
||||
"criterion_type": "expected_output"
|
||||
},
|
||||
{
|
||||
"id": "a7957495-62c0-4d52-9783-31d6d8e2f574",
|
||||
"sort_order": 5,
|
||||
"rubric_text": "In `PBC_Escalation_Log.xlsx`, cells row 3 through 8 must each contain 'Financial' as their Category, and 'Escalated to RCM Manager - 2026-04-09' as their Resolution_Tkg.",
|
||||
"verifier_code": "from pathlib import Path\nimport openpyxl\n\ndef verify(workspace_path, external_services_path=None):\n # Find the escalation log file\n wb_files = list(Path(workspace_path).glob(\"*Escalation_Log*.xlsx\")) + list(Path(workspace_path).glob(\"*escalation_log*.xlsx\"))\n if not wb_files:\n all_xlsx = list(Path(workspace_path).glob(\"*.xlsx\"))\n wb_files = [f for f in all_xlsx if \"escalation\" in f.name.lower() and \"log\" in f.name.lower()]\n if not wb_files:\n return {\"pass\": False, \"score\": 0.0, \"feedback\": \"PBC_Escalation_Log.xlsx not found in workspace\"}\n \n wb_path = wb_files[0]\n try:\n wb = openpyxl.load_workbook(wb_path, data_only=True)\n except Exception as e:\n return {\"pass\": False, \"score\": 0.0, \"feedback\": f\"Failed to open workbook: {e}\"}\n \n ws = wb.active\n \n # Determine column indices for Category and Resolution_Tkg from header row\n header_row = 1\n max_col = ws.max_column or 20\n cat_col = None\n res_col = None\n for col_idx in range(1, max_col + 1):\n val = ws.cell(row=header_row, column=col_idx).value\n if val is not None:\n val_str = str(val).strip().lower()\n if val_str in (\"category\",):\n cat_col = col_idx\n if \"resolution\" in val_str or \"tkg\" in val_str:\n res_col = col_idx\n \n # Fallback: if headers not detected, assume common column positions\n # Try row 2 as header as well if row 1 didn't work\n if cat_col is None or res_col is None:\n for try_header in [2]:\n for col_idx in range(1, max_col + 1):\n val = ws.cell(row=try_header, column=col_idx).value\n if val is not None:\n val_str = str(val).strip().lower()\n if val_str in (\"category\",) and cat_col is None:\n cat_col = col_idx\n if (\"resolution\" in val_str or \"tkg\" in val_str) and res_col is None:\n res_col = col_idx\n \n # If still not found, use defaults (common layout: Category might be col D or E, Resolution_Tkg might be col E or later)\n # We'll do a broader search across all columns for the expected values\n if cat_col is None or res_col is None:\n # Scan all columns to find which ones contain 'Financial' and which contain 'Escalated to RCM Manager'\n for col_idx in range(1, max_col + 1):\n for row_idx in range(3, 9):\n val = ws.cell(row=row_idx, column=col_idx).value\n if val is not None:\n val_str = str(val).strip()\n if val_str.lower() == \"financial\" and cat_col is None:\n cat_col = col_idx\n if \"escalated to rcm manager\" in val_str.lower() and res_col is None:\n res_col = col_idx\n \n failures = []\n \n expected_category = \"Financial\"\n expected_resolution = \"Escalated to RCM Manager - 2026-04-09\"\n \n rows_to_check = [3, 4, 5, 6, 7, 8]\n \n # Check Category column\n if cat_col is not None:\n for row in rows_to_check:\n val = ws.cell(row=row, column=cat_col).value\n val_str = str(val).strip() if val is not None else \"\"\n if val_str.lower() != expected_category.lower():\n failures.append(f\"Row {row} Category (col {cat_col})='{val_str}', expected '{expected_category}'\")\n else:\n failures.append(\"Could not identify Category column in the spreadsheet\")\n \n # Check Resolution_Tkg column\n if res_col is not None:\n for row in rows_to_check:\n val = ws.cell(row=row, column=res_col).value\n val_str = str(val).strip() if val is not None else \"\"\n if val_str != expected_resolution:\n # Be a bit lenient: check case-insensitive and allow minor formatting differences\n if val_str.lower().replace(\" \", \"\") != expected_resolution.lower().replace(\" \", \"\"):\n failures.append(f\"Row {row} Resolution_Tkg (col {res_col})='{val_str}', expected '{expected_resolution}'\")\n else:\n failures.append(\"Could not identify Resolution_Tkg column in the spreadsheet\")\n \n if failures:\n return {\"pass\": False, \"score\": 0.0, \"feedback\": f\"Failures found: {'; '.join(failures)}\"}\n \n return {\"pass\": True, \"score\": 1.0, \"feedback\": f\"All rows 3-8 contain Category='{expected_category}' and Resolution_Tkg='{expected_resolution}'\"}\n",
|
||||
"criterion_type": "expected_output"
|
||||
},
|
||||
{
|
||||
"id": "963d503e-dcb7-4cb6-aa32-a208b87bfa94",
|
||||
"sort_order": 6,
|
||||
"rubric_text": "In `slack_data.json`, channel C001 must contain a message with text 'Escalation: 2026-04-09. Category: Financial. Payer: Bullion Gold Insurance.'",
|
||||
"verifier_code": "from pathlib import Path\nimport json\n\ndef verify(workspace_path, external_services_path=None):\n if external_services_path is None:\n return {\"pass\": False, \"score\": 0.0, \"feedback\": \"external_services_path is None; cannot check slack_data.json\"}\n slack_path = Path(external_services_path) / \"slack_data.json\"\n if not slack_path.exists():\n return {\"pass\": False, \"score\": 0.0, \"feedback\": \"slack_data.json not found\"}\n with open(slack_path, \"r\") as f:\n data = json.load(f)\n \n messages = data.get(\"messages\", {})\n c001_messages = messages.get(\"C001\", [])\n \n expected = \"Escalation: 2026-04-09. Category: Financial. Payer: Bullion Gold Insurance.\"\n expected_lower = expected.lower().strip()\n \n for msg in c001_messages:\n text = (msg.get(\"text\") or \"\").strip().lower()\n # Normalize whitespace\n import re\n text_norm = re.sub(r'\\s+', ' ', text)\n exp_norm = re.sub(r'\\s+', ' ', expected_lower)\n if exp_norm in text_norm or text_norm == exp_norm:\n return {\"pass\": True, \"score\": 1.0, \"feedback\": f\"Found expected escalation message in C001\"}\n \n return {\"pass\": False, \"score\": 0.0, \"feedback\": f\"No message matching '{expected}' found in C001. Messages found: {[m.get('text','') for m in c001_messages]}\"}\n",
|
||||
"criterion_type": "expected_output"
|
||||
},
|
||||
{
|
||||
"id": "530d25ac-62c7-4548-a433-f90b592628dd",
|
||||
"sort_order": 7,
|
||||
"rubric_text": "In `slack_data.json`, channels C003 and C004 must both contain a message that reads 'Refund case - AA45175 - B8475Q'",
|
||||
"verifier_code": "from pathlib import Path\nimport json\nimport re\n\ndef verify(workspace_path, external_services_path=None):\n if external_services_path is None:\n return {\"pass\": False, \"score\": 0.0, \"feedback\": \"external_services_path is None; cannot check slack_data.json\"}\n slack_path = Path(external_services_path) / \"slack_data.json\"\n if not slack_path.exists():\n return {\"pass\": False, \"score\": 0.0, \"feedback\": \"slack_data.json not found\"}\n with open(slack_path, \"r\") as f:\n data = json.load(f)\n \n messages = data.get(\"messages\", {})\n expected = \"Refund case - AA45175 - B8475Q\"\n expected_lower = expected.lower().strip()\n \n failures = []\n for channel in [\"C003\", \"C004\"]:\n channel_messages = messages.get(channel, [])\n found = False\n for msg in channel_messages:\n text = (msg.get(\"text\") or \"\").strip().lower()\n text_norm = re.sub(r'\\s+', ' ', text)\n exp_norm = re.sub(r'\\s+', ' ', expected_lower)\n if exp_norm in text_norm:\n found = True\n break\n if not found:\n failures.append(f\"{channel} missing message '{expected}'\")\n \n if failures:\n return {\"pass\": False, \"score\": 0.0, \"feedback\": f\"Missing refund messages: {failures}\"}\n return {\"pass\": True, \"score\": 1.0, \"feedback\": f\"Both C003 and C004 contain message '{expected}'\"}\n",
|
||||
"criterion_type": "expected_output"
|
||||
},
|
||||
{
|
||||
"id": "718f72d9-4438-443a-8dd4-3d3bc33e95ae",
|
||||
"sort_order": 8,
|
||||
"rubric_text": "In `calendar_data.json`, there must be an event titled 'Escalation Attempt - Jill - RCM Manager' scheduled on April 9th, 2026 from 1:30 PM to 3:30 PM",
|
||||
"verifier_code": "from pathlib import Path\nimport json\nfrom datetime import datetime\n\ndef verify(workspace_path, external_services_path=None):\n if external_services_path is None:\n return {\"pass\": False, \"score\": 0.0, \"feedback\": \"external_services_path is None; cannot check calendar_data.json\"}\n cal_path = Path(external_services_path) / \"calendar_data.json\"\n if not cal_path.exists():\n return {\"pass\": False, \"score\": 0.0, \"feedback\": \"calendar_data.json not found\"}\n with open(cal_path, \"r\") as f:\n data = json.load(f)\n \n events = data.get(\"events\", {})\n expected_title = \"Escalation Attempt - Jill - RCM Manager\"\n expected_date = \"2026-04-09\"\n expected_start_time = \"13:30\"\n expected_end_time = \"15:30\"\n \n for event_id, event in events.items():\n summary = (event.get(\"summary\") or \"\").strip()\n if summary.lower() != expected_title.lower():\n continue\n \n start = event.get(\"start\", \"\")\n end = event.get(\"end\", \"\")\n start_str = str(start)\n end_str = str(end)\n \n # Check date is 2026-04-09\n if expected_date not in start_str:\n continue\n \n # Check times - accept various formats (13:30, 1:30 PM, T13:30, etc.)\n start_ok = \"13:30\" in start_str or \"1:30\" in start_str.lower()\n end_ok = \"15:30\" in end_str or \"3:30\" in end_str.lower()\n \n if start_ok and end_ok:\n return {\"pass\": True, \"score\": 1.0, \"feedback\": f\"Found event '{expected_title}' on 2026-04-09 from 1:30 PM to 3:30 PM\"}\n else:\n return {\"pass\": False, \"score\": 0.0, \"feedback\": f\"Found event '{expected_title}' on correct date but wrong times: start='{start}', end='{end}'\"}\n \n found_titles = [e.get(\"summary\", \"\") for e in events.values()]\n return {\"pass\": False, \"score\": 0.0, \"feedback\": f\"No event titled '{expected_title}' found on 2026-04-09. Events found: {found_titles}\"}\n",
|
||||
"criterion_type": "expected_output"
|
||||
},
|
||||
{
|
||||
"id": "rubric_1775842232676",
|
||||
"sort_order": 9,
|
||||
"rubric_text": "In `PBC_Escalation_Log.xlsx`, there must be exactly 7 rows of data (excluding headers).",
|
||||
"verifier_code": "from pathlib import Path\nimport openpyxl\n\ndef verify(workspace_path, external_services_path=None):\n file_path = Path(workspace_path) / \"PBC_Escalation_Log.xlsx\"\n \n if not file_path.exists():\n return {\"pass\": False, \"score\": 0.0, \"feedback\": f\"File 'PBC_Escalation_Log.xlsx' not found in workspace at {file_path}.\"}\n \n try:\n wb = openpyxl.load_workbook(file_path)\n ws = wb.active\n \n # Count data rows (non-empty rows excluding the header)\n # Assume the first row is the header\n data_row_count = 0\n for row_idx, row in enumerate(ws.iter_rows(min_row=2, values_only=True), start=2):\n # A row is considered a data row if it has at least one non-None, non-empty-string value\n if any(cell is not None and str(cell).strip() != '' for cell in row):\n data_row_count += 1\n \n wb.close()\n \n if data_row_count == 7:\n return {\"pass\": True, \"score\": 1.0, \"feedback\": f\"PBC_Escalation_Log.xlsx has exactly 7 data rows (excluding headers) as required. Found {data_row_count} data rows.\"}\n else:\n return {\"pass\": False, \"score\": 0.0, \"feedback\": f\"PBC_Escalation_Log.xlsx should have exactly 7 data rows (excluding headers), but found {data_row_count} data rows.\"}\n except Exception as e:\n return {\"pass\": False, \"score\": 0.0, \"feedback\": f\"Error reading 'PBC_Escalation_Log.xlsx': {str(e)}\"}\n",
|
||||
"criterion_type": "expected_output"
|
||||
}
|
||||
]
|
||||
Reference in New Issue
Block a user