Last active
May 28, 2018 22:34
-
-
Save alexshpilkin/71502afd060d2566b3eb29db4605c0e4 to your computer and use it in GitHub Desktop.
Convert Georgian election PDFs into TSV
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| #!/bin/sh -eu | |
| set -o pipefail # FIXME bashism | |
| pdftotext -q -layout "$1" - \ | |
| | iconv -t iso8859-1//translit | iconv -f georgian-ps \ | |
| | "$0.post" |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| #!/usr/bin/env python3 | |
| from collections import namedtuple | |
| from sys import exit, stderr, stdin | |
| Column = namedtuple('Column', 'start end name') | |
| lines = iter(stdin) | |
| line = next(lines) | |
| print('# ' + line.strip()) # title | |
| data = [] | |
| prevcols = None | |
| while line != '\f': | |
| # FIXME This assumes lines inside a single heading go from longest to | |
| # shortest. Good enough for now. | |
| heads = [] | |
| for line in lines: | |
| assert not line.startswith('\f') | |
| if len(line.strip()) == 0: continue | |
| if all(s.isdigit() for s in line.split()): break | |
| line = line.rstrip() | |
| heads.append(Column(len(line)-len(line.lstrip()), len(line), | |
| line.lstrip())) | |
| heads.sort(key=lambda x: x.start) | |
| cols = [Column(None, -1, None)] | |
| for s, e, n in heads: | |
| ps, pe, pn = cols[-1] | |
| if pe >= e: | |
| cols[-1] = Column(ps, pe, pn+' '+n) | |
| else: | |
| cols.append(Column(s, e, n)) | |
| cols = cols[1:] | |
| assert (prevcols is None or | |
| all(pc.name == c.name for pc, c in zip(prevcols, cols))) | |
| prevcols = cols | |
| data.append(line.split()) | |
| for line in lines: | |
| row = line.split() | |
| if len(row) == 0: break # end of page | |
| start = len(line) - len(line.lstrip()); shift = 0 | |
| while start > cols[shift].end: shift += 1 | |
| assert shift == 0 or shift == 2 | |
| if shift + len(row) == len(cols): | |
| data.append([''] * shift + row) | |
| else: | |
| data.append(row) | |
| print("warning: row {} has {} entries instead of {}" | |
| .format(len(data), len(row), len(cols)), | |
| file=stderr) | |
| for line in lines: | |
| if len(line.strip()) != 0: break | |
| assert line.strip().isdigit() # page number | |
| line = next(lines, None) | |
| assert line.startswith('\f') # next page title | |
| print('\t'.join(c.name for c in cols)) | |
| for row in data: | |
| assert(all('\t' not in field for field in row)) | |
| print('\t'.join(row)) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment