Problem:
Currently due to multiple variations of NSDL, CDSL statements, the parsing accuracy takes a hit. Also, schemes are not parsed correctly in either of them.
Proposed Solution:
As in both we have well defined tables, we can mitigate this by using PyMuPDF Tables which, though comparatively slow, provides much accurate results.
We can set this approach as part of separate function/flow so that users can switch between both as per requirement.
Problem:
Currently due to multiple variations of NSDL, CDSL statements, the parsing accuracy takes a hit. Also, schemes are not parsed correctly in either of them.
Proposed Solution:
As in both we have well defined tables, we can mitigate this by using
PyMuPDF Tableswhich, though comparatively slow, provides much accurate results.We can set this approach as part of separate function/flow so that users can switch between both as per requirement.