appy.gen: improved cleaning and formatting of XHTML content; appy.pod: added some default appy-related table styles for producing cells with text in bold/normal, aligned right/left, etc.

gen/__init__.py

@@ -11,8 +11,9 @@ from appy.gen.utils import GroupDescr, Keywords, getClassName, SomeObjects
 import appy.pod
 from appy.pod.renderer import Renderer
 from appy.shared.data import countries
+from appy.shared.xml_parser import XhtmlCleaner
 from appy.shared.utils import Traceback, getOsTempFolder, formatNumber, \
-                              XhtmlCleaner, FileWrapper, sequenceTypes
+                              FileWrapper, sequenceTypes
 
 # Default Appy permissions -----------------------------------------------------
 r, w, d = ('read', 'write', 'delete')
@@ -1238,8 +1239,7 @@ class String(Type):
             # When image upload is allowed, ckeditor inserts some "style" attrs
             # (ie for image size when images are resized). So in this case we
             # can't remove style-related information.
-            keepStyles = self.allowImageUpload or self.richText
+            value = XhtmlCleaner().clean(value, keepStyles=self.richText)
-            value = XhtmlCleaner.clean(value, keepStyles=keepStyles)
         Type.store(self, obj, value)
 
     def getFormattedValue(self, obj, value):

gen/ui/appy.css

@@ -1,10 +1,14 @@
 body { font: 75% Helvetica,Arial,sans-serif; background-color: #EAEAEA;
        margin-top: 18px}
 pre { font: 100% Helvetica,Arial,sans-serif; margin: 0}
-h1 { font-size: 11pt; margin:0;}
+h1 { font-size: 14pt; margin:0;}
-h2 { font-size: 10pt; margin:0; font-style: italic; font-weight: normal;
+h2 { font-size: 13pt; margin:0; font-style: italic; font-weight: normal;
      background-color: #d7dee4}
-h3 { font-size: 9pt; margin:0; font-weight: bold;}
+h3 { font-size: 12pt; margin:0; font-weight: bold;}
+h4 { font-size: 11pt; margin:0;}
+h5 { font-size: 10pt; margin:0; font-style: italic; font-weight: normal;
+     background-color: #d7dee4}
+h6 { font-size: 9pt; margin:0; font-weight: bold;}
 a { text-decoration: none; color: #503737;}
 a:visited { color: #503737;}
 table { font-size: 100%; border-spacing: 0px; border-collapse:collapse;}
@@ -34,10 +38,15 @@ label { font-weight: 600; font-style: italic; line-height: 1.4em;}
 legend { padding-bottom: 2px; padding-right: 3px; color: black;}
 ul { line-height: 1.2em; margin: 0 0 0.2em 0.6em;  padding: 0;
      list-style: none outside none;}
-li { margin: 0; background-image: url("ui/li.gif"); padding-left: 10px;
+ul li { margin: 0; background-image: url("ui/li.gif"); padding-left: 10px;
         background-repeat: no-repeat; background-position: 0 4px;}
 img {border: 0}
+
+/* Styles that apply when viewing content of XHTML fields, that mimic styles
+   that ckeditor uses for displaying XHTML content in the edit view. */
+.xhtml { margin-top: 10px }
 .xhtml img { margin-right: 5px }
+.xhtml p { margin: 3px 0 7px 0}
 
 .main { width: 900px; background-color: white; box-shadow: 3px 3px 3px #A9A9A9;
         border-style: solid; border-width: 1px; border-color: grey}

gen/ui/ckeditor/config.js

@@ -30,4 +30,8 @@ CKEDITOR.editorConfig = function( config )
   config.format_h2 = { element:'h2', attributes:{'style':'margin:0;padding:0'}};
   config.format_h3 = { element:'h3', attributes:{'style':'margin:0;padding:0'}};
   config.format_h4 = { element:'h4', attributes:{'style':'margin:0;padding:0'}};
+  config.entities = false;
+  config.entities_greek = false;
+  config.entities_latin = false;
+  config.fillEmptyBlocks = false;
 };

gen/ui/ckeditor/contents.css

@@ -12,3 +12,4 @@ ol,ul,dl {
   padding:0 40px;
 }
 img { margin-right: 5px}
+table { border-collapse: collapse; border-spacing: 0 }

gen/wrappers/ToolWrapper.py

@@ -149,7 +149,7 @@ class ToolWrapper(AbstractWrapper):
         '''Reindex all Appy objects. For some unknown reason, method
            catalog.refreshCatalog is not able to recatalog Appy objects.'''
         if not startObject:
-            # This is a global refresh. Clear the catallog completely, and then
+            # This is a global refresh. Clear the catalog completely, and then
             # reindex all Appy-managed objects, ie those in folders "config"
             # and "data".
             # First, clear the catalog.

pod/renderer.py

@@ -78,7 +78,7 @@ CONTENT_POD_STYLES = f.read()
 f.close()
 
 # Default font added by pod in content.xml
-CONTENT_POD_FONTS = '<@style@:font-face style:name="PodStarSymbol" ' \
+CONTENT_POD_FONTS = '<@style@:font-face @style@:name="PodStarSymbol" ' \
                     '@svg@:font-family="StarSymbol"/>'
 
 # Default text styles added by pod in styles.xml
@@ -213,7 +213,8 @@ class Renderer:
                 nsUris={'style': pe.NS_STYLE, 'svg': pe.NS_SVG}),
             OdInsert(STYLES_POD_STYLES,
                 XmlElement('styles', nsUri=pe.NS_OFFICE),
-                nsUris={'style': pe.NS_STYLE, 'fo': pe.NS_FO}))
+                nsUris={'style': pe.NS_STYLE, 'fo': pe.NS_FO,
+                        'text': pe.NS_TEXT}))
         self.stylesParser = self.createPodParser('styles.xml', context,
                                                  stylesInserts)
         # Stores the styles mapping

pod/styles.in.content.xml

@@ -119,3 +119,25 @@
 <@style@:style @style@:name="podImageRight" @style@:family="graphic" @style@:parent-style-name="Graphics">
  <@style@:graphic-properties @style@:run-through="foreground" @style@:wrap="parallel" @style@:number-wrapped-paragraphs="no-limit" @style@:wrap-contour="false" @style@:vertical-pos="top" @style@:vertical-rel="paragraph" @style@:horizontal-pos="right" @style@:horizontal-rel="paragraph" @style@:mirror="none" @fo@:clip="rect(0cm, 0cm, 0cm, 0cm)" @fo@:margin-left="0.3cm" @fo@:margin-bottom="0.2cm"/>
 </@style@:style>
+<@style@:style @style@:name="podTablePara" @style@:family="paragraph" @style@:parent-style-name="Appy_Table_Content">
+ <@style@:text-properties @fo@:font-size="8pt" @fo@:font-weight="normal" @style@:font-weight-asian="normal" @style@:font-weight-complex="normal"/>
+</@style@:style>
+<@style@:style @style@:name="podTableParaBold" @style@:family="paragraph" @style@:parent-style-name="Appy_Table_Content">
+ <@style@:text-properties @fo@:font-size="8pt" @fo@:font-weight="bold" @style@:font-weight-asian="bold" @style@:font-weight-complex="bold"/>
+</@style@:style>
+<@style@:style @style@:name="podTableParaRight" @style@:family="paragraph" @style@:parent-style-name="Appy_Table_Content">
+ <@style@:paragraph-properties @fo@:text-align="end" @style@:justify-single-word="false"/>
+ <@style@:text-properties @fo@:font-size="8pt" @fo@:font-weight="normal" @style@:font-weight-asian="normal" @style@:font-weight-complex="normal"/>
+</@style@:style>
+<@style@:style @style@:name="podTableParaBoldRight" @style@:family="paragraph" @style@:parent-style-name="Appy_Table_Content">
+ <@style@:paragraph-properties @fo@:text-align="end" @style@:justify-single-word="false"/>
+ <@style@:text-properties @fo@:font-size="8pt" @fo@:font-weight="bold" @style@:font-weight-asian="bold" @style@:font-weight-complex="bold"/>
+</@style@:style>
+<@style@:style @style@:name="podTableCell" @style@:family="table-cell">
+ <@style@:table-cell-properties @fo@:padding="0.097cm" @fo@:border="0.018cm solid #000000"/>
+</@style@:style>
+<@style@:style @style@:name="podTableHeaderCell" @style@:family="table-cell">
+ <@style@:table-cell-properties @fo@:background-color="#e6e6e6" @fo@:padding="0.097cm" @fo@:border="0.018cm solid #000000">
+  <@style@:background-image/>
+ </@style@:table-cell-properties>
+</@style@:style>

pod/styles.in.styles.xml

@@ -4,3 +4,18 @@
                         @style@:font-name-asian="PodStarSymbol" @style@:font-size-asian="9pt"
                         @style@:font-name-complex="PodStarSymbol" @style@:font-size-complex="9pt"/>
 </@style@:style>
+<@style@:style style:name="AppyStandard" style:family="paragraph" style:class="text" style:master-page-name="">
+ <@style@:paragraph-properties fo:margin-left="0cm" fo:margin-right="0cm" fo:margin-top="0.101cm" fo:margin-bottom="0.169cm" fo:text-indent="0cm" style:auto-text-indent="false" style:page-number="auto"/>
+ <@style@:text-properties style:font-name="DejaVu Sans" fo:font-size="10pt"/>
+</@style@:style>
+<@style@:style @style@:name="Appy_Table_Content" @style@:display-name="Appy Table Contents" @style@:family="paragraph"
+             @style@:parent-style-name="AppyStandard" @style@:class="extra">
+ <@style@:paragraph-properties @fo@:margin-top="0cm" @fo@:margin-bottom="0cm" @text@:number-lines="false" @text@:line-number="0"/>
+ <@style@:text-properties @fo@:font-size="8pt"/>
+</@style@:style>
+<@style@:style @style@:name="Appy_Table_Heading" @style@:display-name="Appy Table Heading" @style@:family="paragraph"
+             @style@:parent-style-name="Appy_Table_Contents" @style@:class="extra">
+ <@style@:paragraph-properties @fo@:text-align="center" @style@:justify-single-word="false" @text@:number-lines="false"
+                             @text@:line-number="0"/>
+ <@style@:text-properties @fo@:font-weight="bold" @style@:font-weight-asian="bold" @style@:font-weight-complex="bold"/>
+</@style@:style>

shared/packaging.py

@@ -106,7 +106,8 @@ class Debianizer:
 
     def __init__(self, app, out, appVersion='0.1.0',
                  pythonVersions=('2.6',), zopePort=8080,
-                 depends=('openoffice.org', 'imagemagick'), sign=False):
+                 depends=('zope2.12', 'openoffice.org', 'imagemagick'),
+                 sign=False):
         # app is the path to the Python package to Debianize.
         self.app = app
         self.appName = os.path.basename(app)
@@ -261,10 +262,6 @@ class Debianizer:
         # Create postinst, a script that will:
         # - bytecompile Python files after the Debian install
         # - change ownership of some files if required
-        # - [in the case of a app-package] execute:
-        #   apt-get -t squeeze-backports install zope2.12
-        #   (if zope2.12 is defined as a simple dependency in field "Depends:"
-        #   it will fail because it will not be searched in squeeze-backports).
         # - [in the case of an app-package] call update-rc.d for starting it at
         #   boot time.
         f = file('postinst', 'w')
@@ -276,8 +273,6 @@ class Debianizer:
                                                                   self.appName)
             content += 'if [ -e %s ]\nthen\n%sfi\n' % (bin, cmds)
         if self.appName != 'appy':
-            # Install zope2.12 from squeeze-backports
-            content += 'apt-get -t squeeze-backports install zope2.12\n'
             # Allow user "zope", that runs the Zope instance, to write the
             # database and log files.
             content += 'chown -R zope:root /var/lib/%s\n' % self.appNameLower

shared/utils.py

@@ -263,35 +263,6 @@ def formatNumber(n, sep=',', precision=2, tsep=' '):
             res += sep + splitted[1]
     return res
 
-# ------------------------------------------------------------------------------
-class XhtmlCleaner:
-    # Regular expressions used for cleaning.
-    classAttr = re.compile('class\s*=\s*".*?"')
-    comment = re.compile('<!--.*?-->', re.S)
-
-    '''This class has 2 objectives:
-
-       1. The main objective is to format XHTML p_s to be storable in the ZODB
-          according to Appy rules.
-          a. Every <p> or <li> must be on a single line (ending with a carriage
-             return); else, appy.shared.diff will not be able to compute XHTML
-             diffs;
-          b. Optimize size: HTML comments are removed.
-
-       2. If p_keepStyles (or m_clean) is False, some style-related information
-          will be removed, in order to get a standardized content that can be
-          dumped in an elegant and systematic manner into a POD template.
-    '''
-    @classmethod
-    def clean(klass, s, keepStyles=False):
-        '''Returns the cleaned variant of p_s.'''
-        if not keepStyles:
-            # Format p_s according to objective 2.
-            s = klass.classAttr.sub('', s)
-        # Format p_s according to objective 1.
-        s = klass.comment.sub('', s)
-        return s
-
 # ------------------------------------------------------------------------------
 def lower(s):
     '''French-accents-aware variant of string.lower.'''

shared/xml_parser.py

@@ -18,7 +18,7 @@
 # Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301,USA.
 
 # ------------------------------------------------------------------------------
-import xml.sax, difflib, types
+import xml.sax, difflib, types, cgi
 from xml.sax.handler import ContentHandler, ErrorHandler, feature_external_ges,\
                             property_interning_dict
 from xml.sax.xmlreader import InputSource
@@ -887,4 +887,127 @@ class XmlComparator:
             else:
                 lastLinePrinted = False
         return not atLeastOneDiff
+
+# ------------------------------------------------------------------------------
+class XhtmlCleaner(XmlParser):
+
+    # Tags that will not be in the result, content included, if keepStyles is
+    # False.
+    tagsToIgnoreWithContent = ('style', 'colgroup')
+    # Tags that will be removed from the result, but whose content will be kept,
+    # if keepStyles is False.
+    tagsToIgnoreKeepContent= ('x', 'font')
+    # All tags to ignore
+    tagsToIgnore = tagsToIgnoreWithContent + tagsToIgnoreKeepContent
+    # Attributes to ignore, if keepStyles if False.
+    attrsToIgnore = ('align', 'valign', 'cellpadding', 'cellspacing', 'width',
+                     'height', 'bgcolor', 'lang', 'border', 'class')
+    # Attrs to add, if not present, to ensure good formatting, be it at the web
+    # or ODT levels.
+    attrsToAdd = {'table': {'cellspacing':'0', 'cellpadding':'6', 'border':'1'},
+                  'tr':    {'valign': 'top'}}
+
+    # Tags that required a line break to be inserted after them.
+    lineBreakTags = ('p', 'li', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'td')
+    '''This class has 2 objectives:
+
+       1. The main objective is to format XHTML p_s to be storable in the ZODB
+          according to Appy rules.
+          a. Every <p> or <li> must be on a single line (ending with a carriage
+             return); else, appy.shared.diff will not be able to compute XHTML
+             diffs;
+          b. Optimize size: HTML comments are removed.
+
+       2. If p_keepStyles (or m_clean) is False, some style-related information
+          will be removed, in order to get a standardized content that can be
+          dumped in an elegant and systematic manner into a POD template.
+    '''
+    def clean(self, s, keepStyles=True):
+        # Must we keep style-related information or not?
+        self.env.keepStyles = keepStyles
+        self.env.currentContent = ''
+        # The stack of currently parsed elements (will contain only ignored
+        # ones).
+        self.env.currentElems = []
+        # 'ignoreTag' is True if we must ignore the currently walked tag.
+        self.env.ignoreTag = False
+        # 'ignoreContent' is True if, within the currently ignored tag, we must
+        # also ignore its content.
+        self.env.ignoreContent = False
+        return self.parse('<x>%s</x>' % s)
+
+    def startDocument(self):
+        # The result will be cleaned XHTML, joined from self.res.
+        self.res = []
+
+    def endDocument(self):
+        self.res = ''.join(self.res)
+
+    def startElement(self, elem, attrs):
+        e = self.env
+        # Dump any previously gathered content if any
+        if e.currentContent:
+            self.res.append(e.currentContent)
+            e.currentContent = ''
+        if e.ignoreTag and e.ignoreContent: return
+        if not e.keepStyles and (elem in self.tagsToIgnore):
+            e.ignoreTag = True
+            if elem in self.tagsToIgnoreWithContent:
+                e.ignoreContent = True
+            else:
+                e.ignoreContent = False
+            e.currentElems.append( (elem, e.ignoreContent) )
+            return
+        # Add a line break before the start tag if required (ie: xhtml differ
+        # needs to get paragraphs and other elements on separate lines).
+        if (elem in self.lineBreakTags) and self.res and \
+           (self.res[-1][-1] != '\n'):
+            prefix = '\n'
+        else:
+            prefix = ''
+        res = '%s<%s' % (prefix, elem)
+        # Include the found attributes, excepted those that must be ignored.
+        for name, value in attrs.items():
+            if not e.keepStyles and (name in self.attrsToIgnore): continue
+            res += ' %s="%s"' % (name, value)
+        # Include additional attributes if required.
+        if elem in self.attrsToAdd:
+            for name, value in self.attrsToAdd[elem].iteritems():
+                res += ' %s="%s"' % (name, value)
+        self.res.append('%s>' % res)
+
+    def endElement(self, elem):
+        e = self.env
+        if e.ignoreTag and (elem in self.tagsToIgnore):
+            # Pop the currently ignored tag
+            e.currentElems.pop()
+            if e.currentElems:
+                # Keep ignoring tags.
+                e.ignoreContent = e.currentElems[-1][1]
+            else:
+                # Stop ignoring elems
+                e.ignoreTag = e.ignoreContent = False
+        elif e.ignoreTag and e.ignoreContent:
+            # This is the end of a sub-tag within a region that we must ignore.
+            pass
+        else:
+            self.res.append(self.env.currentContent)
+            # Add a line break after the end tag if required (ie: xhtml differ
+            # needs to get paragraphs and other elements on separate lines).
+            if elem in self.lineBreakTags:
+                suffix = '\n'
+            else:
+                suffix = ''
+            self.res.append('</%s>%s' % (elem, suffix))
+            self.env.currentContent = ''
+
+    def characters(self, content):
+        if self.env.ignoreContent: return
+        # Remove blanks that ckeditor may add just after a start tag
+        if not self.env.currentContent or (self.env.currentContent == ' '):
+            toAdd = ' ' + content.lstrip()
+        else:
+            toAdd = content
+        # Re-transform XML special chars to entities.
+        self.env.currentContent += cgi.escape(content)
 # ------------------------------------------------------------------------------