diff --git a/src/calibre/ebooks/pdb/input.py b/src/calibre/ebooks/pdb/input.py
index 3ad1a6121c..8f4751b42b 100644
--- a/src/calibre/ebooks/pdb/input.py
+++ b/src/calibre/ebooks/pdb/input.py
@@ -22,6 +22,12 @@ class PDBInput(InputFormatPlugin):
help=_('Normally calibre treats blank lines as paragraph markers. '
'With this option it will assume that every line represents '
'a paragraph instead.')),
+ OptionRecommendation(name='print_formatted_paras', recommended_value=False,
+ help=_('Normally calibre treats blank lines as paragraph markers. '
+ 'With this option it will assume that every line starting with '
+ 'an indent (either a tab or 2+ spaces) represents a paragraph.'
+ 'Paragraphs end when the next line that starts with an indent '
+ 'is reached.')),
])
def convert(self, stream, options, file_ext, log,
diff --git a/src/calibre/ebooks/pdb/palmdoc/reader.py b/src/calibre/ebooks/pdb/palmdoc/reader.py
index 8992382597..0a57e3f51a 100644
--- a/src/calibre/ebooks/pdb/palmdoc/reader.py
+++ b/src/calibre/ebooks/pdb/palmdoc/reader.py
@@ -13,8 +13,8 @@ import struct
from calibre.ebooks.compression.palmdoc import decompress_doc
from calibre.ebooks.pdb.formatreader import FormatReader
-from calibre.ebooks.txt.processor import convert_basic, separate_paragraphs, \
- opf_writer
+from calibre.ebooks.txt.processor import convert_basic, opf_writer, \
+ separate_paragraphs_single_line, separate_paragraphs_print_formatted
class HeaderRecord(object):
'''
@@ -36,6 +36,7 @@ class Reader(FormatReader):
self.log = log
self.encoding = options.input_encoding
self.single_line_paras = options.single_line_paras
+ self.print_formatted_paras = options.print_formatted_paras
self.sections = []
for i in range(header.num_sections):
@@ -63,7 +64,9 @@ class Reader(FormatReader):
self.log.info('Converting text to OEB...')
if self.single_line_paras:
- txt = separate_paragraphs(txt)
+ txt = separate_paragraphs_single_line(txt)
+ if self.print_formatted_paras:
+ txt = separate_paragraphs_print_formatted(txt)
html = convert_basic(txt)
with open(os.path.join(output_dir, 'index.html'), 'wb') as index:
index.write(html.encode('utf-8'))
diff --git a/src/calibre/ebooks/pdb/ztxt/reader.py b/src/calibre/ebooks/pdb/ztxt/reader.py
index 664f498bee..5cac283264 100644
--- a/src/calibre/ebooks/pdb/ztxt/reader.py
+++ b/src/calibre/ebooks/pdb/ztxt/reader.py
@@ -12,8 +12,8 @@ import os, struct, zlib
from calibre.ebooks.pdb.formatreader import FormatReader
from calibre.ebooks.pdb.ztxt import zTXTError
-from calibre.ebooks.txt.processor import convert_basic, separate_paragraphs, \
- opf_writer
+from calibre.ebooks.txt.processor import convert_basic, opf_writer, \
+ separate_paragraphs_single_line, separate_paragraphs_print_formatted
SUPPORTED_VERSION = (1, 40)
@@ -31,22 +31,23 @@ class HeaderRecord(object):
self.size, = struct.unpack('>L', raw[4:8])
self.record_size, = struct.unpack('>H', raw[8:10])
self.flags, = struct.unpack('>B', raw[18:19])
-
-
+
+
class Reader(FormatReader):
-
+
def __init__(self, header, stream, log, options):
self.stream = stream
self.log = log
self.encoding = options.input_encoding
self.single_line_paras = options.single_line_paras
-
+ self.print_formatted_paras = options.print_formatted_paras
+
self.sections = []
for i in range(header.num_sections):
self.sections.append(header.section_data(i))
self.header_record = HeaderRecord(self.section_data(0))
-
+
vmajor = (self.header_record.version & 0x0000FF00) >> 8
vminor = self.header_record.version & 0x000000FF
if vmajor < 1 or (vmajor == 1 and vminor < 40):
@@ -71,7 +72,7 @@ class Reader(FormatReader):
def extract_content(self, output_dir):
txt = ''
-
+
self.log.info('Decompressing text...')
for i in range(1, self.header_record.num_records + 1):
self.log.debug('\tDecompressing text section %i' % i)
@@ -79,16 +80,18 @@ class Reader(FormatReader):
self.log.info('Converting text to OEB...')
if self.single_line_paras:
- txt = separate_paragraphs(txt)
+ txt = separate_paragraphs_single_line(txt)
+ if self.print_formatted_paras:
+ txt = separate_paragraphs_print_formatted(txt)
html = convert_basic(txt)
with open(os.path.join(output_dir, 'index.html'), 'wb') as index:
index.write(html.encode('utf-8'))
-
+
from calibre.ebooks.metadata.meta import get_metadata
mi = get_metadata(self.stream, 'pdb')
manifest = [('index.html', None)]
spine = ['index.html']
opf_writer(output_dir, 'metadata.opf', manifest, spine, mi)
-
+
return os.path.join(output_dir, 'metadata.opf')
diff --git a/src/calibre/ebooks/txt/input.py b/src/calibre/ebooks/txt/input.py
index f52bfa6fb5..eb86113f7a 100644
--- a/src/calibre/ebooks/txt/input.py
+++ b/src/calibre/ebooks/txt/input.py
@@ -8,7 +8,7 @@ import os
from calibre.customize.conversion import InputFormatPlugin, OptionRecommendation
from calibre.ebooks.txt.processor import convert_basic, convert_markdown, \
- separate_paragraphs
+ separate_paragraphs_single_line, separate_paragraphs_print_formatted
class TXTInput(InputFormatPlugin):
@@ -22,6 +22,12 @@ class TXTInput(InputFormatPlugin):
help=_('Normally calibre treats blank lines as paragraph markers. '
'With this option it will assume that every line represents '
'a paragraph instead.')),
+ OptionRecommendation(name='print_formatted_paras', recommended_value=False,
+ help=_('Normally calibre treats blank lines as paragraph markers. '
+ 'With this option it will assume that every line starting with '
+ 'an indent (either a tab or 2+ spaces) represents a paragraph.'
+ 'Paragraphs end when the next line that starts with an indent '
+ 'is reached.')),
OptionRecommendation(name='markdown', recommended_value=False,
help=_('Run the text input through the markdown pre-processor. To '
'learn more about markdown see')+' http://daringfireball.net/projects/markdown/'),
@@ -35,8 +41,11 @@ class TXTInput(InputFormatPlugin):
log.debug('Reading text from file...')
txt = stream.read().decode(ienc, 'replace')
+ # Adjust paragraph formatting as requested
if options.single_line_paras:
- txt = separate_paragraphs(txt)
+ txt = separate_paragraphs_single_line(txt)
+ if options.print_formatted_paras:
+ txt = separate_paragraphs_print_formatted(txt)
if options.markdown:
log.debug('Running text though markdown conversion...')
diff --git a/src/calibre/ebooks/txt/processor.py b/src/calibre/ebooks/txt/processor.py
index f6503c0bc5..9d0e1283c1 100644
--- a/src/calibre/ebooks/txt/processor.py
+++ b/src/calibre/ebooks/txt/processor.py
@@ -45,12 +45,16 @@ def convert_markdown(txt, title=''):
safe_mode=False,)
return HTML_TEMPLATE % (title, md.convert(txt))
-def separate_paragraphs(txt):
+def separate_paragraphs_single_line(txt):
txt = txt.replace('\r\n', '\n')
txt = txt.replace('\r', '\n')
txt = re.sub(u'(?<=.)\n(?=.)', u'\n\n', txt)
return txt
+def separate_paragraphs_print_formatted(txt):
+ txt = re.sub('(?miu)^(\t+|[ ]{2,})(?=.)', '\n\t', txt)
+ return txt
+
def opf_writer(path, opf_name, manifest, spine, mi):
opf = OPFCreator(path, mi)
opf.create_manifest(manifest)
diff --git a/src/calibre/gui2/convert/pdb_input.py b/src/calibre/gui2/convert/pdb_input.py
index 4b0ba73fda..058f589856 100644
--- a/src/calibre/gui2/convert/pdb_input.py
+++ b/src/calibre/gui2/convert/pdb_input.py
@@ -14,6 +14,6 @@ class PluginWidget(Widget, Ui_Form):
def __init__(self, parent, get_option, get_help, db=None, book_id=None):
Widget.__init__(self, parent, 'pdb_input',
- ['single_line_paras'])
+ ['single_line_paras', 'print_formatted_paras'])
self.db, self.book_id = db, book_id
self.initialize_options(get_option, get_help, db, book_id)
diff --git a/src/calibre/gui2/convert/pdb_input.ui b/src/calibre/gui2/convert/pdb_input.ui
index 191e749833..2b632b1a33 100644
--- a/src/calibre/gui2/convert/pdb_input.ui
+++ b/src/calibre/gui2/convert/pdb_input.ui
@@ -14,7 +14,7 @@
Form
- -
+
-
Qt::Vertical
@@ -34,6 +34,13 @@
+ -
+
+
+ Assume print formatting
+
+
+
diff --git a/src/calibre/gui2/convert/txt_input.py b/src/calibre/gui2/convert/txt_input.py
index 3d17eefe0d..505a916f81 100644
--- a/src/calibre/gui2/convert/txt_input.py
+++ b/src/calibre/gui2/convert/txt_input.py
@@ -14,6 +14,6 @@ class PluginWidget(Widget, Ui_Form):
def __init__(self, parent, get_option, get_help, db=None, book_id=None):
Widget.__init__(self, parent, 'txt_input',
- ['single_line_paras', 'markdown'])
+ ['single_line_paras', 'print_formatted_paras', 'markdown'])
self.db, self.book_id = db, book_id
self.initialize_options(get_option, get_help, db, book_id)
diff --git a/src/calibre/gui2/convert/txt_input.ui b/src/calibre/gui2/convert/txt_input.ui
index 8c22ff721e..94cc1ad25f 100644
--- a/src/calibre/gui2/convert/txt_input.ui
+++ b/src/calibre/gui2/convert/txt_input.ui
@@ -14,7 +14,7 @@
Form
- -
+
-
Qt::Vertical
@@ -34,14 +34,14 @@
- -
+
-
Process using markdown
- -
+
-
<p>Markdown is a simple markup language for text files, that allows for advanced formatting. To learn more visit <a href="http://daringfireball.net/projects/markdown">markdown</a>.
@@ -51,6 +51,13 @@
+ -
+
+
+ Assume print formatting
+
+
+