diff options
| author | Thibaut Horel <thibaut.horel@gmail.com> | 2014-09-07 15:55:27 -0400 |
|---|---|---|
| committer | Thibaut Horel <thibaut.horel@gmail.com> | 2014-09-07 15:55:27 -0400 |
| commit | d28394833d54a68f5ca13d2edaa261128f6c5170 (patch) | |
| tree | ab00b100b99066b80979613b06fca2c9b7087701 /web/utils.py | |
| parent | dfcd65c8f10aa94f19fe40940565681ab9a73e44 (diff) | |
| download | ocr-layer-curation-d28394833d54a68f5ca13d2edaa261128f6c5170.tar.gz | |
Compute alignment on the final (html formatted) text
Diffstat (limited to 'web/utils.py')
| -rw-r--r-- | web/utils.py | 15 |
1 files changed, 7 insertions, 8 deletions
diff --git a/web/utils.py b/web/utils.py index 8522841..1947f8b 100644 --- a/web/utils.py +++ b/web/utils.py @@ -1,21 +1,20 @@ import djvu_utils as du import sys import string_utils as su -from wikisource import get_page +from wikisource import get_page2 + def gen_html(book, page_number): doc = du.get_document("../" + book) - page = doc.pages[int(page_number)-1] + page = doc.pages[int(page_number) - 1] d = du.parse_page(page) - corrected_text = get_page(book, int(page_number)) - corrected_words = su.simplify(corrected_text).split() + elem, corrected_text = get_page2(open("test.txt").read()) if d: words, coords = zip(*d) - C = su.align(corrected_words, list(words), list(coords)) - r = su.alignment_to_sexp(corrected_text.split(), words, coords, C[1]) - corrected_words, coords = zip(*r) + C = su.align(corrected_text.split(), list(words), list(coords)) + coords = [coords[e[0]] for e in C[1]] coords_html = du.convert_to_htmlcoord(coords, page.size[1]) - return (list(enumerate(coords_html)), list(enumerate(corrected_words))) + return (list(enumerate(coords_html)), str(elem)) if __name__ == "__main__": gen_html(*sys.argv[1:3]) |
