Pull from driver-dev

src/calibre/ebooks/conversion/preprocess.py

@@ -32,6 +32,39 @@ def chap_head(match):
                return '<h1>'+chap+'<br/>'+title+'</h1><br/>'
 
 
+def line_length(raw, percent):
+    '''
+    raw is the raw text to find the line length to use for wrapping.
+    percentage is a decimal number, 0 - 1 which is used to determine
+    how far in the list of line lengths to use.
+    '''
+    raw = raw.replace('&nbsp;', ' ')
+    linere = re.compile('(?<=<br>).*?(?=<br>)', re.DOTALL)
+    lines = linere.findall(raw)
+
+    lengths = []
+    for line in lines:
+        if len(line) > 0:
+            lengths.append(len(line))
+    total = sum(lengths)
+    avg = total / len(lengths)
+    max_line = avg * 2
+    
+    lengths = sorted(lengths)
+    for i in range(len(lengths) - 1, -1, -1):
+        if lengths[i] > max_line:
+            del lengths[i]
+    
+    if percent > 1:
+        percent = 1
+    if percent < 0:
+        percent = 0
+
+    index = int(len(lengths) * percent) - 1
+    
+    return lengths[index]
+
+
 class CSSPreProcessor(object):
 
     PAGE_PAT   = re.compile(r'@page[^{]*?{[^}]*?}')
@@ -129,7 +162,12 @@ class HTMLPreProcessor(object):
         elif self.is_book_designer(html):
             rules = self.BOOK_DESIGNER
         elif self.is_pdftohtml(html):
-            rules = self.PDFTOHTML
+            # Add rules that require matching line length here
+            #line_length_rules = [
+            #    (re.compile('%i' % line_length(html, .85)), lambda match:)
+            #]
+            
+            rules = self.PDFTOHTML # + line_length_rules
         else:
             rules = []
         for rule in self.PREPROCESS + rules:

src/calibre/ebooks/pdb/ereader/pmlconverter.py

@@ -12,10 +12,12 @@ import re
 
 from calibre.ebooks.htmlsymbols import HTML_SYMBOLS
 
+from BeautifulSoup import BeautifulSoup
+
 PML_HTML_RULES = [
     (re.compile(r'\\p'), lambda match: '<br /><br style="page-break-after: always;" />'),
     (re.compile(r'\\x(?P<text>.+?)\\x', re.DOTALL), lambda match: '<h1 style="page-break-before: always;">%s</h1>' % match.group('text')),
-    (re.compile(r'\\X(?P<val>[0-4])(?P<text>.+?)\\X[0-4]', re.DOTALL), lambda match: '<h%i style="page-break-before: always;">%s</h%i>' % (int(match.group('val')) + 1, match.group('text'), int(match.group('val')) + 1)),
+    (re.compile(r'\\X(?P<val>[0-4])(?P<text>.+?)\\X[0-4]', re.DOTALL), lambda match: '<h%s style="page-break-before: always;">%s</h%s>' % (int(match.group('val')) + 1, match.group('text'), int(match.group('val')) + 1)),
     (re.compile(r'\\C\d=".+"'), lambda match: ''), # This should be made to create a TOC entry
     (re.compile(r'\\c(?P<text>.+?)\\c', re.DOTALL), lambda match: '<div style="text-align: center; display: block; margin: auto;">%s</div>' % match.group('text')),
     (re.compile(r'\\r(?P<text>.+?)\\r', re.DOTALL), lambda match: '<div style="text-align: right; display: block;">%s</div>' % match.group('text')),
@@ -24,8 +26,8 @@ PML_HTML_RULES = [
     (re.compile(r'\\o(?P<text>.+?)\\o', re.DOTALL), lambda match: '<del>%s</del>' % match.group('text')),
     (re.compile(r'\\v(?P<text>.+?)\\v', re.DOTALL), lambda match: '<!-- %s -->' % match.group('text')),
     (re.compile(r'\\t(?P<text>.+?)\\t', re.DOTALL), lambda match: '<div style="margin-left: 5%%;">%s</div>' % match.group('text')),
-    (re.compile(r'\\T="(?P<val>\d+)%%*"(?P<text>.+?)$', re.MULTILINE), lambda match: '<div style="margin-left: %i%%;">%s</div>' % (match.group('val'), match.group('text'))),
+    (re.compile(r'\\T="(?P<val>\d+)%*"(?P<text>.+?)$', re.MULTILINE), lambda match: r'<div style="margin-left: %s%%;">%s</div>' % (match.group('val'), match.group('text'))),
-    (re.compile(r'\\w="(?P<val>\d+)%%"'), lambda match: '<hr width="%s%%" />' % match.group('val')),
+    (re.compile(r'\\w="(?P<val>\d+)%"'), lambda match: '<hr width="%s%%" />' % match.group('val')),
     (re.compile(r'\\n'), lambda match: ''),
     (re.compile(r'\\s'), lambda match: ''),
     (re.compile(r'\\b(?P<text>.+?)\\b', re.DOTALL), lambda match: '<b>%s</b>' % match.group('text')), # \b is deprecated; \B should be used instead.
@@ -34,10 +36,10 @@ PML_HTML_RULES = [
     (re.compile(r'\\Sp(?P<text>.+?)\\Sp', re.DOTALL), lambda match: '<sup>%s</sup>' % match.group('text')),
     (re.compile(r'\\Sb(?P<text>.+?)\\Sb', re.DOTALL), lambda match: '<sub>%s</sub>' % match.group('text')),
     (re.compile(r'\\k(?P<text>.+?)\\k', re.DOTALL), lambda match: '<small>%s</small>' % match.group('text')),
-    (re.compile(r'\\a(?P<num>\d\d\d)'), lambda match: '&#%i;' % match.group('num')),
+    (re.compile(r'\\a(?P<num>\d\d\d)'), lambda match: '&#%s;' % match.group('num')),
-    (re.compile(r'\\U(?P<num>\d\d\d\d)'), lambda match: '&#%i;' % int(match.group('num'))),
+    (re.compile(r'\\U(?P<num>\d+)'), lambda match: '%s' % unichr(int(match.group('num'), 16))),
     (re.compile(r'\\m="(?P<name>.+?)"'), lambda match: '<img src="images/%s" />' % match.group('name')),
-    (re.compile(r'\\q="(?P<target>#.+?)"(?P<text>)\\q', re.DOTALL), lambda match: '<a href="%s">%s</a>' % (match.group('target'), match.group('text'))),
+    (re.compile(r'\\q="(?P<target>#.+?)"(?P<text>.+?)\\q', re.DOTALL), lambda match: '<a href="%s">%s</a>' % (match.group('target'), match.group('text'))),
     (re.compile(r'\\Q="(?P<target>.+?)"'), lambda match: '<div id="%s"></div>' % match.group('target')),
     (re.compile(r'\\-'), lambda match: ''),
     (re.compile(r'\\Fn="(?P<target>.+?)"(?P<text>.+?)\\Fn'), lambda match: '<a href="#footnote-%s">%s</a>' % (match.group('target'), match.group('text'))),
@@ -58,14 +60,6 @@ PML_HTML_RULES = [
     (re.compile(r'\\\\'), lambda match: '\\'),
 ]
 
-FOOTNOTE_HTML_RULES = [
-    (re.compile('<footnote id="(?P<id>.+?)">(?P<text>.+?)</footnote>', re.DOTALL), lambda match: '<div id="footnote-%s">%s</div>')
-]
-
-SIDEBAR_HTML_RULES = [
-    (re.compile('<sidebar id="(?P<id>.+?)">(?P<text>.+?)</sidebar>', re.DOTALL), lambda match: '<div id="sidebar-%s">%s</div>')
-]
-
 HTML_PML_RULES = [
     (re.compile(r'\\'), lambda match: '\\\\'),
     (re.compile('(?<=[^\n])[ ]*<p.*?>'), lambda match: '\n<p>'),
@@ -75,23 +69,23 @@ HTML_PML_RULES = [
     (re.compile('<div.*?id="(?P<target>.+?).*?"></div>'), lambda match: '\\\\Q="%s"' % match.group('target')),
     (re.compile('<a.*?href="(?P<target>#.+?).*?">(?P<text>)</a>', re.DOTALL), lambda match: '\\q="%s"%s\\q' % (match.group('target'), match.group('text'))),
     (re.compile('<img.*?src="images/(?P<name>.+?)".*?>'), lambda match: '\\m="%s"' % match.group('name')),
-    (re.compile('&#(?P<num>\d\d\d\d);'), lambda match: '\\U%i' % int(match.group('num'))),
+    #(re.compile('&#(?P<num>\d\d\d\d);'), lambda match: '\\U%s' % int(match.group('num'))),
-    (re.compile('&#(?P<num>\d\d\d);'), lambda match: '\\a%i' % match.group('num')),
+    (re.compile('&#(?P<num>\d\d\d);'), lambda match: '\\a%s' % match.group('num')),
     (re.compile('<small.*?>(?P<text>.+?)</small>', re.DOTALL), lambda match: '\\k%s\\k' % match.group('text')),
     (re.compile('<sub.*?>(?P<text>.+?)</sub>', re.DOTALL), lambda match: '\\Sb%s\\Sb' % match.group('text')),
     (re.compile('<sup.*?>(?P<text>.+?)</sup>', re.DOTALL), lambda match: '\\Sp%s\\Sp' % match.group('text')),
     (re.compile('<b.*?>(?P<text>.+?)</b>', re.DOTALL), lambda match: '\\B%s\\B' % match.group('text')),
     (re.compile('<big.*?>(?P<text>.+?)</big>', re.DOTALL), lambda match: '\\l%s\\l' % match.group('text')),
-    (re.compile('<hr.*?width="(?P<val>\d+)%%".*?>'), lambda match: '\\w="%s%%"' % match.group('val')),
+    (re.compile('<hr.*?width="(?P<val>\d+)%".*?>'), lambda match: '\\w="%s%%"' % match.group('val')),
-    (re.compile('<div.*?style.*?margin-left: (?P<val>\d+)%%*;.*?>(?P<text>.+?)</div>', re.MULTILINE), lambda match: '\\T="%i%%"%s$' % (match.group('val'), match.group('text'))),
+    (re.compile('<div.*?style.*?margin-left: (?P<val>\d+)%*;.*?>(?P<text>.+?)</div>', re.MULTILINE), lambda match: '\\T="%s%%"%s$' % (match.group('val'), match.group('text'))),
-    (re.compile('<div.*?style.*?margin-left: \d{1,3}%%;.*?>(?P<text>.+?)</div>', re.DOTALL), lambda match: '\\t%s\\t' % match.group('text')),
+    (re.compile('<div.*?style.*?margin-left: \d{1,3}%;.*?>(?P<text>.+?)</div>', re.DOTALL), lambda match: '\\t%s\\t' % match.group('text')),
     (re.compile('<!-- (?P<text>.+?) -->', re.DOTALL), lambda match: '\\v%s\\v' % match.group('text')),
     (re.compile('<del.*?>(?P<text>.+?)</del>', re.DOTALL), lambda match: '\\o%s\\o' % match.group('text')),
     (re.compile('<div.*?style.*?text-decoration: underline;.*?>(?P<text>.+?)</div>', re.DOTALL), lambda match: '\\u%s\\u' % match.group('text')),
     (re.compile('<i.*?>(?P<text>.+?)</i>', re.DOTALL), lambda match: '\\\\i%s\\i' % match.group('text')),
     (re.compile('<div.*?style.*?text-align: right;.*?>(?P<text>.+?)</div>', re.DOTALL), lambda match: '\\r%s\\r' % match.group('text')),
     (re.compile('<div.*?style.*?text-align: center;.*?".*?>(?P<text>.+?)</div>', re.DOTALL), lambda match: '\\c%s\\c' % match.group('text')),
-    (re.compile('<h(?P<val>[0-4]).*?>(?P<text>.+?)</h[0-4]>', re.DOTALL), lambda match: '\\X%i%s\\X%i' % (int(match.group('val')) + 1, match.group('text'), int(match.group('val')) + 1)),
+    (re.compile('<h(?P<val>[0-4]).*?>(?P<text>.+?)</h[0-4]>', re.DOTALL), lambda match: '\\X%s%s\\X%s' % (int(match.group('val')) + 1, match.group('text'), int(match.group('val')) + 1)),
     (re.compile('<h1.*?>(?P<text>.+?)</h1>', re.DOTALL), lambda match: '\\x%s\\x' % match.group('text')),
     (re.compile('<br.*?>'), lambda match: '\\p'),
     (re.compile('<.*?>'), lambda match: ''),
@@ -109,28 +103,20 @@ def pml_to_html(pml):
         
     return html
 
-def footnote_to_html(footnotes):
+def footnote_sidebar_to_html(id, pml):
-    html = footnotes
+    html = '<div id="sidebar-%s"><dt>%s</dt></div><dd>%s</dd>' % (id, id, pml_to_html(pml))
-    for rule in FOOTNOTE_HTML_RULES:
+    return html 
-        html = rule[0].sub(rule[1], html)
-        
-    html = pml_to_html(html)
-        
-    return html
-    
-def sidebar_to_html(sidebars):
-    html = sidebars
-    for rule in FOOTNOTE_HTML_RULES:
-        html = rule[0].sub(rule[1], html)
-        
-    html = pml_to_html(html)
-        
-    return html
 
 def html_to_pml(html):
-    pml = html
+    pml = ''
-    for rule in HTML_PML_RULES:
+    
-        pml = rule[0].sub(rule[1], pml)
+    for dom_tree in BeautifulSoup(html).findAll('body'):
+        body = unicode(dom_tree.pretty_print())
+
+        for rule in HTML_PML_RULES:
+            body = rule[0].sub(rule[1], pml)
+            
+        pml += body
 
     # Replace symbols outside of cp1512 wtih \Uxxxx
 

src/calibre/ebooks/pdb/ereader/reader.py

@@ -8,7 +8,7 @@ __license__   = 'GPL v3'
 __copyright__ = '2009, John Schember <john@nachtimwald.com>'
 __docformat__ = 'restructuredtext en'
 
-import os, sys, struct, zlib
+import os, re, sys, struct, zlib
 
 from calibre import CurrentDir
 from calibre.ebooks import DRMError
@@ -16,7 +16,7 @@ from calibre.ebooks.metadata import MetaInformation
 from calibre.ebooks.pdb.formatreader import FormatReader
 from calibre.ebooks.pdb.ereader import EreaderError
 from calibre.ebooks.pdb.ereader.pmlconverter import pml_to_html, \
-    footnote_to_html, sidebar_to_html 
+    footnote_sidebar_to_html 
 from calibre.ebooks.mobi.palmdoc import decompress_doc
 from calibre.ebooks.metadata.opf2 import OPFCreator
 
@@ -42,14 +42,6 @@ class HeaderRecord(object):
         
         self.num_text_pages = self.non_text_offset -1
         self.num_image_pages = self.metadata_offset - self.image_data_offset
-
-        # Can't tell which is sidebar and footnote if they have same offset.
-        # They don't exist if offset is larget than last_record.
-        # Todo: Determine if the subtraction is necessary and find out
-        # what _rec means.
-        end_footnote_offset = self.sidebar_offset if self.sidebar_offset != self.footnote_offset else self.last_data_offset
-        self.num_footnote_pages = end_footnote_offset - self.footnote_offset if self.footnote_offset < self.last_data_offset else 0 
-        self.num_sidebar_pages = self.sidebar_offset - self.last_data_offset if self.footnote_offset < self.last_data_offset else 0
         
 
 class Reader(FormatReader):
@@ -94,44 +86,10 @@ class Reader(FormatReader):
         assumed to be encoded as Windows-1252. The encoding is part of
         the eReader file spec and should always be this encoding.
         '''
-        if number not in range(1, self.header_record.num_text_pages):
+        if number not in range(1, self.header_record.num_text_pages + 1):
             return ''
             
         return self.decompress_text(number)
-            
-    def get_footnote_page(self, number):
-        if number not in range(self.header_record.footnote_offset, self.header_record.footnote_offset + self.header_record.num_footnote_pages):
-            return ''
-            
-        return self.decompress_text(number)
-        
-    def get_sidebar_page(self, number):
-        if number not in range(self.header_record.sidebar_offset, self.header_record.sidebar_offset + self.header_record.num_sidebar_pages - 1):
-            return ''
-            
-        return self.decompress_text(number)
-
-    def has_footnotes(self):
-        if self.header_record.num_footnote_pages > 1:
-            try:
-                content = self.decompress_text(self.header_record.footnote_offset)
-                
-                if content.contains('</footnote>'):
-                    return True
-            except:
-                pass
-        return False
-        
-    def has_sidebar(self):
-        if self.header_record.num_sidebar_pages > 1:
-            try:
-                content = self.decompress_text(self.header_record.sidebar_offset)
-                
-                if content.contains('</sidebar>'):
-                    return True
-            except:
-                pass
-        return False
 
     def extract_content(self, output_dir):
         output_dir = os.path.abspath(output_dir)
@@ -144,22 +102,25 @@ class Reader(FormatReader):
         for i in range(1, self.header_record.num_text_pages + 1):
             self.log.debug('Extracting text page %i' % i)
             html += pml_to_html(self.get_text_page(i))
-
+        
-        # Untested: The num_.._pages variable may not be correct!
+        if self.header_record.footnote_rec > 0:
-        # Possibly use .._rec instead?
-        '''
-        if has_footnotes():
             html += '<br /><h1>%s</h1>' % _('Footnotes')
-            for i in range(self.header_record.footnote_offset, self.header_record.num_footnote_pages):
+            footnoteids = re.findall('\w+(?=\x00)', self.section_data(self.header_record.footnote_offset).decode('cp1252' if self.encoding is None else self.encoding))
+            for fid, i in enumerate(range(self.header_record.footnote_offset + 1, self.header_record.footnote_offset + self.header_record.footnote_rec)):
                 self.log.debug('Extracting footnote page %i' % i)
-                html += footnote_to_html(self.get_footnote_page(i))
+                html += '<dl>'
+                html += footnote_sidebar_to_html(footnoteids[fid], self.decompress_text(i))
+                html += '</dl>'
                 
-        if has_sidebar():
+        
+        if self.header_record.sidebar_rec > 0:
             html += '<br /><h1>%s</h1>' % _('Sidebar')
-            for i in range(self.header_record.sidebar_offset, self.header_record.num_sidebar_pages):
+            sidebarids = re.findall('\w+(?=\x00)', self.section_data(self.header_record.sidebar_offset).decode('cp1252' if self.encoding is None else self.encoding))
+            for sid, i in enumerate(range(self.header_record.sidebar_offset + 1, self.header_record.sidebar_offset + self.header_record.sidebar_rec)):
                 self.log.debug('Extracting sidebar page %i' % i)
-                html += sidebar_to_html(self.get_sidebar_page(i))
+                html += '<dl>'
-        '''
+                html += footnote_sidebar_to_html(sidebarids[sid], self.decompress_text(i))
+                html += '</dl>'
         
         html += '</body></html>'
         

src/calibre/ebooks/pdb/header.py

@@ -63,12 +63,12 @@ class PdbHeaderReader(object):
 class PdbHeaderWriter(object):
 
     def __init__(self, identity, title):
-        self.identity = identity[:8]
+        self.identity = identity.ljust(3, '\x00')[:8]
         self.title = title.ljust(32, '\x00')[:32]
-
+        
-    def build_header(self, sections):
+    def build_header(self, offsets):
         '''
-        Sections is a list of section offsets
+        Offsets is a list of section offsets
         '''
 
 

src/calibre/ebooks/pdb/input.py

@@ -8,6 +8,7 @@ __docformat__ = 'restructuredtext en'
 import os
 
 from calibre.customize.conversion import InputFormatPlugin
+from calibre.ebooks.pdb.header import PdbHeaderReader
 from calibre.ebooks.pdb import PDBError, IDENTITY_TO_NAME, get_reader
 
 class PDBInput(InputFormatPlugin):