fnp / wolnelektury.git / commitdiff
? search:
summary | shortlog | log | commit | commitdiff | tree
raw | patch | inline | side by side (parent: a289712)
Added book2html and bookfragments utilities to repository.
authorMarek Stępniowski <marek@stepniowski.com>
Thu, 28 Aug 2008 12:28:07 +0000 (14:28 +0200)
committerMarek Stępniowski <marek@stepniowski.com>
Thu, 28 Aug 2008 12:28:07 +0000 (14:28 +0200)
bin/book2html.py [new file with mode: 0755] patch | blob
bin/book2html.xslt [new file with mode: 0644] patch | blob
bin/bookfragments.py [new file with mode: 0755] patch | blob

diff --git a/bin/book2html.py b/bin/book2html.py
new file mode 100755 (executable)
index 0000000..5bd2bb5
--- /dev/null
+++ b/bin/book2html.py
@@ -0,0 +1,55 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+import cStringIO
+import re
+import optparse
+import os
+import sys
+
+from lxml import etree
+
+
+# Parse args
+usage = """Usage: %prog [options] SOURCE [SOURCE...]
+Convert SOURCE files to HTML format."""
+
+parser = optparse.OptionParser(usage=usage)
+
+parser.add_option('-v', '--verbose', action='store_true', dest='verbose', default=False,
+    help='print status messages to stdout')
+
+options, input_filenames = parser.parse_args()
+
+if len(input_filenames) < 1:
+    parser.print_help()
+    exit(1)
+
+# Parse XSLT
+style = etree.parse('book2html.xslt')
+    
+# Do some real work
+for input_filename in input_filenames:
+    if options.verbose:
+        print input_filename
+    
+    output_filename = os.path.splitext(input_filename)[0] + '.html'
+    
+    # Transform
+    doc_file = cStringIO.StringIO()
+    expr = re.compile(r'/\s', re.MULTILINE | re.UNICODE);
+
+    f = open(input_filename, 'r')
+    for line in f:
+        line = line.decode('utf-8')
+        line = expr.sub(u'<br/>\n', line).replace(u'---', u'—').replace(u',,', u'„')
+        doc_file.write(line.encode('utf-8'))
+    f.close()
+
+    doc_file.seek(0);
+
+    parser = etree.XMLParser(remove_blank_text=True)
+    doc = etree.parse(doc_file, parser)
+
+    result = doc.xslt(style)
+    result.write(output_filename, xml_declaration=True, pretty_print=True, encoding='utf-8')
+
diff --git a/bin/book2html.xslt b/bin/book2html.xslt
new file mode 100644 (file)
index 0000000..6cba758
--- /dev/null
+++ b/bin/book2html.xslt
@@ -0,0 +1,150 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="2.0">
+
+<xsl:output method="xml" encoding="utf-8" doctype-public="-//W3C//DTD XHTML 1.1//EN" doctype-system="http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd" indent="yes" />
+
+<xsl:template match="text()" />
+
+<xsl:template match="extra|uwaga" />
+<xsl:template match="extra|uwaga" mode="inline" />
+
+<xsl:template match="/">
+    <html>
+        <head>
+            <title>book2html output</title>
+            <meta http-equiv="content-type" content="text/html;charset=utf-8"/>
+            <link rel="stylesheet" href="master.css" type="text/css" media="screen" charset="utf-8" />
+        </head>
+        <body>
+            <xsl:apply-templates />
+            <div id="footnotes">
+                <h3>Przypisy</h3>
+                <xsl:for-each select="descendant::*[self::pe or self::pa or self::pr or self::pt][not(parent::extra)]">
+                    <div>
+                        <a name="{concat('footnote-', generate-id(.))}" />
+                        <a href="{concat('#anchor-', generate-id(.))}" class="annotation">[<xsl:number value="count(preceding::*[self::pa or self::pe or self::pr or self::pt]) + 1" />]</a>
+                        <xsl:choose>
+                            <xsl:when test="count(akap|akap_cd|strofa) = 0">
+                                <p><xsl:apply-templates select="text()|*" mode="inline" /></p>
+                            </xsl:when>
+                            <xsl:otherwise>
+                                <xsl:apply-templates select="text()|*" mode="inline" />
+                            </xsl:otherwise>
+                        </xsl:choose>
+                    </div>
+                </xsl:for-each>
+            </div>
+        </body>
+    </html>
+</xsl:template>
+
+<xsl:template match="naglowek_akt|naglowek_czesc">
+    <h2><xsl:apply-templates mode="inline" /></h2>
+</xsl:template>
+
+<xsl:template match="naglowek_scena|naglowek_rozdzial">
+    <h3><xsl:apply-templates mode="inline" /></h3>
+</xsl:template>
+
+<xsl:template match="naglowek_osoba">
+    <h4><xsl:apply-templates mode="inline" /></h4>
+</xsl:template>
+
+<xsl:template match="kwestia">
+    <div class="kwestia">
+        <xsl:apply-templates select="strofa|akap" />
+    </div>
+</xsl:template>
+
+<xsl:template match="didaskalia">
+    <div class="didaskalia"><xsl:apply-templates mode="inline" /></div>
+</xsl:template>
+
+<xsl:template match="lista_osob">
+    <div class="person-list">
+        <h3><xsl:value-of select="naglowek_listy" /></h3>
+        <ol>
+            <xsl:apply-templates select="lista_osoba" />
+        </ol>
+    </div>
+</xsl:template>
+
+<xsl:template match="lista_osoba">
+    <li><xsl:apply-templates mode="inline" /></li>
+</xsl:template>
+
+<xsl:template match="begin" mode="inline">
+    <xsl:variable name="mnum" select="concat('m', substring(@id, 2))" />
+    <span class="theme-begin" fid="{substring(@id, 2)}">
+        <xsl:value-of select="string(following::motyw[@id=$mnum]/text())" />
+    </span>
+</xsl:template>
+
+<xsl:template match="end" mode="inline">
+    <span class="theme-end" fid="{substring(@id, 2)}"> </span>
+</xsl:template>
+
+<xsl:template match="begin|end">
+    <xsl:apply-templates select='.' mode="inline" />
+</xsl:template>
+
+<xsl:template name="verse">
+    <xsl:param name="line-content" />
+    <xsl:param name="line-number" />
+    <p>
+        <xsl:choose>
+            <xsl:when test="name($line-content) = 'wers_akap'">
+                <xsl:attribute name="style">indent: 1em</xsl:attribute>
+            </xsl:when>
+            <xsl:when test="name($line-content) = 'wers_wciety'">
+                <xsl:attribute name="style">indent: 2em</xsl:attribute>
+            </xsl:when>
+        </xsl:choose>
+        <xsl:apply-templates select="$line-content" mode="inline" />
+    </p>
+</xsl:template>
+
+<xsl:template match="pa|pe|pr|pt" mode="inline">
+    <a name="{concat('anchor-', generate-id(.))}" />
+    <a href="{concat('#footnote-', generate-id(.))}" class="annotation">[<xsl:number value="count(preceding::*[self::pa or self::pe or self::pr or self::pt]) + 1" />]</a>
+</xsl:template>
+
+<xsl:template match="strofa">
+    <div class="stanza">
+        <xsl:choose>
+            <xsl:when test="count(br) > 0">     
+                <xsl:call-template name="verse">
+                    <xsl:with-param name="line-content" select="br[1]/preceding-sibling::text() | br[1]/preceding-sibling::node()" />
+                    <xsl:with-param name="line-number" select="1" />
+                </xsl:call-template>    
+                <xsl:for-each select="br">             
+                               <!-- Każdy BR "zjada" to co jest za nim -->
+                    <xsl:variable name="lnum" select="count(preceding-sibling::br)" />
+                    <xsl:call-template name="verse">
+                        <xsl:with-param name="line-number" select="$lnum+2" />
+                        <xsl:with-param name="line-content" 
+                            select="following-sibling::text()[count(preceding-sibling::br) = $lnum+1] | following-sibling::node()[count(preceding-sibling::br) = $lnum+1]" />
+                    </xsl:call-template>
+                </xsl:for-each>
+            </xsl:when>
+            <xsl:otherwise>
+                <xsl:call-template name="verse">
+                    <xsl:with-param name="line-content" select="text() | node()" />
+                    <xsl:with-param name="line-number" select="1" />
+                 </xsl:call-template>           
+            </xsl:otherwise>
+        </xsl:choose>
+    </div>
+</xsl:template>
+
+<xsl:template match="akap|akap_dialog|akap_cd">
+    <p class="paragraph"><xsl:apply-templates mode="inline" /></p>
+</xsl:template>
+
+<xsl:template match="motyw" mode="inline" />
+
+<xsl:template match="dlugi_cytat">
+    <blockquote><xsl:apply-templates /></blockquote>
+</xsl:template>
+
+</xsl:stylesheet>
\ No newline at end of file
diff --git a/bin/bookfragments.py b/bin/bookfragments.py
new file mode 100755 (executable)
index 0000000..73d2711
--- /dev/null
+++ b/bin/bookfragments.py
@@ -0,0 +1,148 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+import optparse
+import os
+import copy
+
+from lxml import etree
+
+
+# Parse args
+usage = """Usage: %prog [options] SOURCE [SOURCE...]
+Extract theme fragments from SOURCE."""
+
+parser = optparse.OptionParser(usage=usage)
+
+parser.add_option('-v', '--verbose', action='store_true', dest='verbose', default=False,
+    help='print status messages to stdout')
+
+options, input_filenames = parser.parse_args()
+
+if len(input_filenames) < 1:
+    parser.print_help()
+    exit(1)
+
+
+class Fragment(object):
+    def __init__(self, id, themes):
+        super(Fragment, self).__init__()
+        self.id = id
+        self.themes = themes
+        self.events = []
+        
+    def append(self, event, element):
+        self.events.append((event, element))
+    
+    def closed_events(self):
+        stack = []
+        for event, element in self.events:
+            if event == 'start':
+                stack.append(('end', element))
+            elif event == 'end':
+                try:
+                    stack.pop()
+                except IndexError:
+                    print 'CLOSED NON-OPEN TAG:', element
+        
+        stack.reverse()
+        return self.events + stack
+    
+    def to_string(self):
+        result = []
+        for event, element in self.closed_events():
+            if event == 'start':
+                result.append(u'<%s %s>' % (element.tag, ' '.join('%s="%s"' % (k, v) for k, v in element.attrib.items())))
+                if element.text:
+                    result.append(element.text)
+            elif event == 'end':
+                result.append(u'</%s>' % element.tag)
+                if element.tail:
+                    result.append(element.tail)
+            else:
+                result.append(element)
+        
+        return ''.join(result)
+    
+    def __unicode__(self):
+        return self.to_string()
+
+
+# Do some real work
+for input_filename in input_filenames:
+    if options.verbose:
+        print input_filename
+    
+    output_filename = os.path.splitext(input_filename)[0] + '.fragments.html'
+    
+    open_fragments = {}
+    closed_fragments = {}
+    lost_text = []
+    
+    for event, element in etree.iterparse(input_filename, events=('start', 'end')):
+        
+        # Process begin and end elements
+        if element.tag == 'span' and element.get('class', '') in ('theme-begin', 'theme-end'):
+            if not event == 'end': continue # Process elements only once, on end event
+            
+            # Open new fragment
+            if element.get('class', '') == 'theme-begin':
+                fragment = Fragment(id=element.get('fid'), themes=element.text)
+                
+                # Append parents
+                if element.getparent().tag != 'body':
+                    parents = [element.getparent()]
+                    while parents[-1].getparent().tag != 'body':
+                        parents.append(parents[-1].getparent())
+                    
+                    parents.reverse()
+                    for parent in parents:
+                        fragment.append('start', parent)
+                
+                open_fragments[fragment.id] = fragment
+                    
+            # Close existing fragment
+            else:
+                try:
+                    fragment = open_fragments[element.get('fid')]
+                except KeyError:
+                    print '%s:closed not open fragment #%s' % (input_filename, element.get('fid'))
+                else:
+                    closed_fragments[fragment.id] = fragment
+                    del open_fragments[fragment.id]
+            
+            # Append element tail to lost_text (we don't want to lose any text)
+            if element.tail:
+                for fragment_id in open_fragments:
+                    open_fragments[fragment_id].append('text', element.tail)
+        
+        
+        # Process all elements except begin and end
+        else:
+            # Omit annotation tags
+            if len(element.get('name', '')) or element.get('class', '') == 'annotation':
+                if event == 'end' and element.tail:
+                    for fragment_id in open_fragments:
+                        open_fragments[fragment_id].append('text', element.tail)
+            else:
+                for fragment_id in open_fragments:
+                    open_fragments[fragment_id].append(event, copy.copy(element))
+
+
+    for fragment_id in open_fragments:
+        print '%s:warning:unclosed fragment #%s' % (input_filename, fragment_id)
+
+    output_file = open(output_filename, 'w')
+    output_file.write("""
+        <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">
+        <html><head>
+            <title>bookfragments output</title>
+            <meta http-equiv="content-type" content="text/html;charset=utf-8"/>
+            <link rel="stylesheet" href="master.css" type="text/css" media="screen" charset="utf-8" />
+        </head>
+        <body>""")
+    for fragment in closed_fragments.values():
+        html = u'<div class="fragment"><h3>[#%s] %s</h3>%s</div>' % (fragment.id, fragment.themes, fragment)
+        output_file.write(html.encode('utf-8'))
+    output_file.write('</body></html>')
+    output_file.close()
+