From ed3b2866cfb846c7ccbb39e64711fb37d56e927c Mon Sep 17 00:00:00 2001 From: John Schember Date: Sun, 6 Feb 2011 08:35:07 -0500 Subject: [PATCH] Sync ldolse preprocessing changes. --- src/calibre/ebooks/txt/input.py | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/src/calibre/ebooks/txt/input.py b/src/calibre/ebooks/txt/input.py index 75bafc7cef..a07b423ebb 100644 --- a/src/calibre/ebooks/txt/input.py +++ b/src/calibre/ebooks/txt/input.py @@ -57,6 +57,7 @@ class TXTInput(InputFormatPlugin): def convert(self, stream, options, file_ext, log, accelerators): self.log = log + length = None log.debug('Reading text from file...') txt = stream.read() @@ -109,7 +110,7 @@ class TXTInput(InputFormatPlugin): # Reformat paragraphs to block formatting based on the detected type. # We don't check for block because the processor assumes block. # single and print at transformed to block for processing. - if options.paragraph_type == 'single' or options.paragraph_type == 'unformatted': + if options.paragraph_type == 'single': txt = separate_paragraphs_single_line(txt) elif options.paragraph_type == 'print': txt = separate_paragraphs_print_formatted(txt) @@ -120,10 +121,12 @@ class TXTInput(InputFormatPlugin): length = docanalysis.line_length(.5) preprocessor = HeuristicProcessor(options, log=getattr(self, 'log', None)) txt = preprocessor.punctuation_unwrap(length, txt, 'txt') + txt = separate_paragraphs_single_line(txt) if getattr(options, 'enable_heuristics', False) and getattr(options, 'dehyphenate', False): docanalysis = DocAnalysis('txt', txt) - length = docanalysis.line_length(.5) + if not length: + length = docanalysis.line_length(.5) dehyphenator = Dehyphenator(options.verbose, log=self.log) txt = dehyphenator(txt,'txt', length)