Qumo/Interview-Analysis
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

cpc1 afwijking

Browse Source
This commit is contained in:
Luigi Maiorano
2025-12-14 20:02:40 +01:00
parent c2a5c12794
commit a5ffd8315e
2 changed files with 7 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
01_Taguette-Pre-Process.py
View File

@@ -55,7 +55,7 @@ def _(file_dropdown, jpmc_transcript_to_md, mo):
preview = mo.md("") preview = mo.md("")
if file_dropdown.value: if file_dropdown.value:
md_content = jpmc_transcript_to_md(file_dropdown.value) md_content = jpmc_transcript_to_md(file_dropdown.value)
preview = mo.md(md_content) preview = mo.md(md_content[:1000])
preview preview
return return

6
utils/transcript_utils.py
View File

@@ -106,6 +106,12 @@ def cpc_smb_to_markdown(cpc_path: Path) -> str:
for line in content.splitlines(): for line in content.splitlines():
line = line.strip().replace('\n', ' ') line = line.strip().replace('\n', ' ')
# Handle edge case: "CPC1, (She/ Her,) LOCATION: Hello." -> "CPC1: Hello."
match = re.match(r'^"?([A-Za-z0-9]+),\s*\(.*?\)\s*LOCATION:\s*(.*?)"?$', line)
if match:
line = f"{match.group(1)}: {match.group(2)}"
# Remove surrounding quotes # Remove surrounding quotes
if line.startswith('"') and line.endswith('"'): if line.startswith('"') and line.endswith('"'):
line = line[1:-1].strip() line = line[1:-1].strip()