From b51137379fab2d5579e0caeada5389c682ec3cc5 Mon Sep 17 00:00:00 2001
From: Thibaut Horel <thibaut.horel@gmail.com>
Date: Mon, 5 Aug 2013 22:29:24 +0200
Subject: Use a Needleman-Wunsch type algorithm for text alignment

---
 compare.py | 15 +++++++++------
 1 file changed, 9 insertions(+), 6 deletions(-)

(limited to 'compare.py')

diff --git a/compare.py b/compare.py
index aeb9092..c37cde9 100644
--- a/compare.py
+++ b/compare.py
@@ -1,12 +1,15 @@
+# -*- coding: utf-8 -*-
+
 from wikisource import get_page
 from parsedjvutext import parse_page_sexp
-from string_utils import LCS, printDiff
+import string_utils as su
 
-wikibook = "Villiers de L'Isle-Adam - Tribulat Bonhomet, 1908.djvu"
+wikibook = "Bloy - Le Sang du pauvre, Stock, 1932.djvu".replace(" ", "_")
+# wikibook = "Villiers de L'Isle-Adam - Tribulat Bonhomet, 1908.djvu".replace(" ", "_")
 
-n = 42
+n = 79
 ocrpage = parse_page_sexp(wikibook, n)
 l1 = ocrpage['words']
-l2 = get_page(wikibook, n).split()
-C = LCS(l1, l2)
-printDiff(C, l1, l2, len(l1), len(l2))
+l2 = get_page(wikibook, n).replace(u"’", u"'").split()
+C = su.align(l2, l1)
+su.print_alignment(l2, l1, C[1])
-- 
cgit v1.2.3-70-g09d2