usc-sail / mica-text-script-parser Goto Github PK

salvation_road.pdf

I get the following error when I run the parse_script_noindent on the above pdf with the abridged flag on -

Traceback (most recent call last):
  File "/proj/sbaruah/movie/mica-text-script-parser/parse_scripts_noindent.py", line 351, in <module>
    parse(file_orig, save_dir, abr_flag, tag_flag)
  File "/proj/sbaruah/movie/mica-text-script-parser/parse_scripts_noindent.py", line 342, in parse
    dial_str = ' '.join(abridged_script[i + 1].split('D:')[1].split())
IndexError: list index out of range

Can you please look into it? It seems the program expects D tag to always follow a C tag, which does not happen at some point for the linked pdf.

None of the code is commented. I am trying to run this parser on a standalone pdf - and there are no instructions on where to begin with. Can you please add comments and give a few use cases, such as fetching from a webpage, or running on a pdf, etc.

Scripts can be .txt files.
Clients might parse the pdf on their side and give text files.
Please allow for the input of .txt files in parse_scripts_noindent.py

In some scripts they have scene number before or after the scene boundary. And causing the output to be parsed as an utterance.

like this --
" 141 141 => All the denizens of Warm Springs are crowded around the radio."