diff --git a/resources/default_tweaks.py b/resources/default_tweaks.py index 32aeba9122..f1abfbe7ea 100644 --- a/resources/default_tweaks.py +++ b/resources/default_tweaks.py @@ -30,6 +30,13 @@ defaults. series_index_auto_increment = 'next' +# Should the completion separator be append +# to the end of the completed text to +# automatically begin a new completion operation. +# Can be either True or False +completer_append_separator = False + + # The algorithm used to copy author to author_sort # Possible values are: # invert: use "fn ln" -> "ln, fn" (the original algorithm) diff --git a/src/calibre/ebooks/txt/input.py b/src/calibre/ebooks/txt/input.py index 2399e599ae..e1392ef732 100644 --- a/src/calibre/ebooks/txt/input.py +++ b/src/calibre/ebooks/txt/input.py @@ -57,6 +57,7 @@ class TXTInput(InputFormatPlugin): log.debug('Reading text from file...') txt = stream.read() + # Get the encoding of the document. if options.input_encoding: ienc = options.input_encoding @@ -70,13 +71,16 @@ class TXTInput(InputFormatPlugin): log.debug('No input encoding specified and could not auto detect using %s' % ienc) txt = txt.decode(ienc, 'replace') + # Replace entities txt = _ent_pat.sub(xml_entity_to_unicode, txt) # Normalize line endings txt = normalize_line_endings(txt) + # Detect formatting if options.formatting_type == 'auto': options.formatting_type = detect_formatting_type(txt) + log.debug('Auto detected formatting as %s' % options.formatting_type) if options.formatting_type == 'heuristic': setattr(options, 'enable_heuristics', True) @@ -105,41 +109,43 @@ class TXTInput(InputFormatPlugin): docanalysis = DocAnalysis('txt', txt) length = docanalysis.line_length(.5) + # 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': + txt = separate_paragraphs_single_line(txt) + elif options.paragraph_type == 'print': + txt = separate_paragraphs_print_formatted(txt) + elif options.paragraph_type == 'unformatted': + from calibre.ebooks.conversion.utils import HeuristicProcessor + # unwrap lines based on punctuation + preprocessor = HeuristicProcessor(options, log=getattr(self, 'log', None)) + txt = preprocessor.punctuation_unwrap(length, txt, 'txt') + + # Process the text using the appropriate text processor. + html = '' if options.formatting_type == 'markdown': - log.debug('Running text though markdown conversion...') + log.debug('Running text through markdown conversion...') try: html = convert_markdown(txt, disable_toc=options.markdown_disable_toc) except RuntimeError: raise ValueError('This txt file has malformed markup, it cannot be' ' converted by calibre. See http://daringfireball.net/projects/markdown/syntax') elif options.formatting_type == 'textile': - log.debug('Running text though textile conversion...') + log.debug('Running text through textile conversion...') html = convert_textile(txt) else: - # Dehyphenate - dehyphenator = Dehyphenator(options.verbose, log=self.log) - txt = dehyphenator(txt,'txt', length) - - # 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': - txt = separate_paragraphs_single_line(txt) - elif options.paragraph_type == 'print': - txt = separate_paragraphs_print_formatted(txt) - - if options.paragraph_type == 'unformatted': - from calibre.ebooks.conversion.utils import HeuristicProcessor - # get length - - # unwrap lines based on punctuation - preprocessor = HeuristicProcessor(options, log=getattr(self, 'log', None)) - txt = preprocessor.punctuation_unwrap(length, txt, 'txt') + log.debug('Running text through basic conversion...') + if options.formatting_type == 'heuristic': + # Dehyphenate + dehyphenator = Dehyphenator(options.verbose, log=self.log) + txt = dehyphenator(txt,'txt', length) flow_size = getattr(options, 'flow_size', 0) html = convert_basic(txt, epub_split_size_kb=flow_size) + # Run the HTMLized text through the html processing plugin. from calibre.customize.ui import plugin_for_input_format html_input = plugin_for_input_format('html') for opt in html_input.options: @@ -158,6 +164,7 @@ class TXTInput(InputFormatPlugin): htmlfile.write(html.encode('utf-8')) odi = options.debug_pipeline options.debug_pipeline = None + # Generate oeb from htl conversion. oeb = html_input.convert(open(htmlfile.name, 'rb'), options, 'html', log, {}) options.debug_pipeline = odi diff --git a/src/calibre/gui2/complete.py b/src/calibre/gui2/complete.py index f589b30679..a013065690 100644 --- a/src/calibre/gui2/complete.py +++ b/src/calibre/gui2/complete.py @@ -10,6 +10,7 @@ from PyQt4.Qt import QLineEdit, QListView, QAbstractListModel, Qt, QTimer, \ QApplication, QPoint, QItemDelegate, QStyleOptionViewItem, \ QStyle, QEvent, pyqtSignal +from calibre.utils.config import tweaks from calibre.utils.icu import sort_key, lower from calibre.gui2 import NONE from calibre.gui2.widgets import EnComboBox @@ -231,12 +232,18 @@ class MultiCompleteLineEdit(QLineEdit): cursor_pos = self.cursorPosition() before_text = unicode(self.text())[:cursor_pos] after_text = unicode(self.text())[cursor_pos:] - after_parts = after_text.split(self.sep) - if len(after_parts) < 3 and not after_parts[-1].strip(): - after_text = u'' prefix_len = len(before_text.split(self.sep)[-1].lstrip()) - return prefix_len, \ - before_text[:cursor_pos - prefix_len] + text + after_text + if tweaks['completer_append_separator']: + prefix_len = len(before_text.split(self.sep)[-1].lstrip()) + completed_text = before_text[:cursor_pos - prefix_len] + text + self.sep + ' ' + after_text + prefix_len = prefix_len - len(self.sep) - 1 + if prefix_len < 0: + prefix_len = 0 + else: + prefix_len = len(before_text.split(self.sep)[-1].lstrip()) + completed_text = before_text[:cursor_pos - prefix_len] + text + after_text + return prefix_len, completed_text + def completion_selected(self, text): prefix_len, ctext = self.get_completed_text(text)