Add --force-page-break-before-attr

src/libprs500/__init__.py

@@ -13,7 +13,7 @@
 ##    with this program; if not, write to the Free Software Foundation, Inc.,
 ##    51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
 ''' E-book management software'''
-__version__   = "0.3.93"
+__version__   = "0.3.94"
 __docformat__ = "epytext"
 __author__    = "Kovid Goyal <kovid@kovidgoyal.net>"
 __appname__   = 'libprs500'

src/libprs500/ebooks/lrf/__init__.py

@@ -153,6 +153,8 @@ def option_parser(usage):
                       '''if the current page has only a few elements.''')
     chapter.add_option('--force-page-break-before', dest='force_page_break',
                        default='$', help='Like --page-break-before, but page breaks are forced.')
+    chapter.add_option('--force-page-break-before-attr', dest='force_page_break_attr',
+                       default='$,,$', help='Force a page break before an element having the specified attribute. The format for this option is tagname regexp,attribute name,attribute value regexp. For example to match all heading tags that have the attribute class="chapter" you would use "h\d,class,chapter". Default is %default''')
     prepro = parser.add_option_group('PREPROCESSING OPTIONS')
     prepro.add_option('--baen', action='store_true', default=False, dest='baen',
                       help='''Preprocess Baen HTML files to improve generated LRF.''')

src/libprs500/ebooks/lrf/html/convert_from.py

@@ -35,11 +35,11 @@ except ImportError:
 from libprs500.ebooks.BeautifulSoup import BeautifulSoup, BeautifulStoneSoup, \
                 Comment, Tag, NavigableString, Declaration, ProcessingInstruction
 from libprs500.ebooks.lrf.pylrs.pylrs import Paragraph, CR, Italic, ImageStream, \
-                TextBlock, ImageBlock, JumpButton, CharButton, Bold,\
+                TextBlock, ImageBlock, JumpButton, CharButton, \
                 Plot, Image, BlockSpace, RuledLine, BookSetting, Canvas, DropCaps, \
                 LrsError
 from libprs500.ebooks.lrf.pylrs.pylrs import Span as _Span
-from libprs500.ebooks.lrf import Book, PRS500_PROFILE
+from libprs500.ebooks.lrf import Book
 from libprs500.ebooks.lrf import option_parser as lrf_option_parser
 from libprs500.ebooks import ConversionError
 from libprs500.ebooks.lrf.html.table import Table 
@@ -262,7 +262,7 @@ class HTMLConverter(object):
     SELECTOR_PAT   = re.compile(r"([A-Za-z0-9\-\_\:\.]+[A-Za-z0-9\-\_\:\.\s\,]*)\s*\{([^\}]*)\}")
     PAGE_BREAK_PAT = re.compile(r'page-break-(?:after|before)\s*:\s*(\w+)', re.IGNORECASE)
     IGNORED_TAGS   = (Comment, Declaration, ProcessingInstruction)
-    # Fix <a /> elements 
+     
     MARKUP_MASSAGE   = [
                         # Close <a /> tags
                         (re.compile("(<a\s+.*?)/>|<a/>", re.IGNORECASE), 
@@ -926,6 +926,29 @@ class HTMLConverter(object):
                                        blockwidth=pwidth, blockheight=pheight),
                             left, 0)
     
+    def process_page_breaks(self, tag, tagname, tag_css):
+        if 'page-break-before' in tag_css.keys():
+            if tag_css['page-break-before'].lower() != 'avoid':
+                self.end_page()
+            tag_css.pop('page-break-before')
+        end_page = False
+        if 'page-break-after' in tag_css.keys() and \
+           tag_css['page-break-after'].lower() != 'avoid':
+            end_page = True
+            tag_css.pop('page-break-after')
+        if (self.force_page_break_attr[0].match(tagname) and \
+           tag.has_key(self.force_page_break_attr[1]) and \
+           self.force_page_break_attr[2].match(tag[self.force_page_break_attr[1]])) or \
+           self.force_page_break.match(tagname):
+            self.end_page()
+            self.page_break_found = True
+        if not self.page_break_found and self.page_break.match(tagname):
+            if len(self.current_page.contents) > 3:
+                self.end_page()
+                if self.verbose:
+                    print 'Forcing page break at', tagname
+        return end_page
+    
     def parse_tag(self, tag, parent_css):
         try:
             tagname = tag.name.lower()
@@ -940,23 +963,8 @@ class HTMLConverter(object):
                 return
         except KeyError:
             pass
-        if 'page-break-before' in tag_css.keys():
+        end_page = self.process_page_breaks(tag, tagname, tag_css)
-            if tag_css['page-break-before'].lower() != 'avoid':
+        
-                self.end_page()
-            tag_css.pop('page-break-before')
-        end_page = False
-        if 'page-break-after' in tag_css.keys() and \
-           tag_css['page-break-after'].lower() != 'avoid':
-            end_page = True
-            tag_css.pop('page-break-after')
-        if self.force_page_break.match(tagname):
-            self.end_page()
-            self.page_break_found = True
-        if not self.page_break_found and self.page_break.match(tagname):
-            if len(self.current_page.contents) > 3:
-                self.end_page()
-                if self.verbose:
-                    print 'Forcing page break at', tagname
         if tagname in ["title", "script", "meta", 'del', 'frameset']:            
             pass
         elif tagname == 'a' and self.link_levels >= 0:
@@ -991,13 +999,17 @@ class HTMLConverter(object):
                 previous = self.current_block
                 self.process_children(tag, tag_css)
                 target = None
-                if self.current_block == previous:
+                if self.current_block == previous:                    
-                    self.current_para.append_to(self.current_block)
+                    if self.current_para.has_text():
-                    self.current_para = Paragraph()
+                        self.current_para.append_to(self.current_block)
-                    if self.current_block.has_text():
+                        self.current_para = Paragraph()
-                        target = self.current_block                        
+                        target = self.current_block                 
-                    else:
+                    else: # Empty <a> element
-                        target = BlockSpace()
+                        self.current_page.append(self.current_block)                        
+                        self.current_block = self.book.create_text_block(
+                                textStyle=self.current_block.textStyle,
+                                blockStyle=self.current_block.blockStyle)
+                        target = self.book.create_text_block()
                         self.current_page.append(target)
                 else:
                     found = False
@@ -1330,6 +1342,11 @@ def process_file(path, options):
         options.link_exclude = le
         options.page_break = pb
         options.chapter_regex = re.compile(options.chapter_regex, re.IGNORECASE)
+        fpba = options.force_page_break_attr.split(',')
+        if len(fpba) != 3:
+            fpba = ['$', '', '$']
+        options.force_page_break_attr = [re.compile(fpba[0], re.IGNORECASE), fpba[1],
+                                         re.compile(fpba[2], re.IGNORECASE)]
         conv = HTMLConverter(book, fonts, path, options)
         conv.process_links()
         oname = options.output