From 4fb7a9830fac4be2352056264fe6bf1cd2057b12 Mon Sep 17 00:00:00 2001 From: Simon Brooke <simon@journeyman.cc> Date: Thu, 31 Oct 2013 08:54:35 +0000 Subject: [PATCH] Should have saved README before committing that last one! --- README.txt | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/README.txt b/README.txt index 5d05c3a..5cccef9 100644 --- a/README.txt +++ b/README.txt @@ -70,3 +70,7 @@ Having said all that, the parse tree is looking very good. I'm extracting the ru Right, StreamTokenizer was a poor choice. It seems to be legacy. But schlurping the whole text into a string and then using StringTokenizer or String.split() looks a bad choice too, since I don't know how long the string is. H'mmmm... This is a problem I kind of don't need, since it's not key to the project, and the .endsWith(PERIOD) hack works around it. Concentate on output. +New Git repository, and this time pushed out to Goldsmith, so that local power problems shouldn't affect it... + +Decluttered the TextGenerator class by moving the whole read stage into two new classes, Generator and Tokeniser. More declutter needed. +