Added html support for motion text.

Use html parser 'beautifulsoup' to replace html tags which not supported by reportlab pdf.
'pip install bs4' required.

Fixed nested ol lists and unicode warning.

New requirement: beautifulsoup4

HTML support for reportlab.

Move parse html code into new function. Added missing html support in pdf for motion reason.

Added version number for bleach and beautifulsoup4. Moved to core requirements block.

Added bs4, bleach and html5lib requirements to portable build script. Updated portable readme for 1.4.

extras/win32-portable/create_portable.txt

@@ -4,7 +4,7 @@ How to create a new portable Windows distribution of OpenSlides:
 1.) Follow the OpenSlides installation instructions for windows,
     but add the option "-Z" when executing easy_install, e.g.:
 
-      easy_install -Z django django-mptt reportlab pil
+      easy_install -Z django django-mptt beautifulsoup4 bleach pillow qrcode reportlab tornado
 
 2.) Run in the main directory of the OpenSlides checkout:
 
@@ -14,5 +14,5 @@ How to create a new portable Windows distribution of OpenSlides:
     in the 'dist' directory
 
 
-NOTE: Creating the portable Windows distribution of OpenSlides is not possible,
+NOTE: Creating the portable Windows distribution of OpenSlides is not possible
-if Python is installed in 64-bit version. 
+if Python is installed in 64-bit(!) version.

extras/win32-portable/prepare_portable.py

@@ -90,6 +90,15 @@ SITE_PACKAGES = {
     },
     "qrcode": {
         "copy": ["qrcode"],
+    },
+    "beautifulsoup4": {
+        "copy": ["bs4"],
+    },
+    "bleach": {
+        "copy": ["bleach"],
+    },
+    "html5lib": {
+        "copy": ["html5lib"],
     }
 }
 

openslides/motion/pdf.py

@@ -10,6 +10,9 @@
     :license: GNU GPL, see LICENSE for more details.
 """
 
+import random
+from bs4 import BeautifulSoup
+
 from reportlab.lib import colors
 from reportlab.lib.units import cm
 from reportlab.platypus import (
@@ -147,14 +150,89 @@ def motion_to_pdf(pdf, motion):
     pdf.append(table)
     pdf.append(Spacer(0, 1 * cm))
 
+    # motion title
     pdf.append(Paragraph(motion.title, stylesheet['Heading3']))
-    pdf.append(Paragraph(motion.text.replace('\r\n', '<br/>'), stylesheet['Paragraph']))
+
+    # motion text
+    convert_html_to_reportlab(pdf, motion.text)
+    pdf.append(Spacer(0, 1 * cm))
+
+    # motion reason
     if motion.reason:
         pdf.append(Paragraph(_("Reason:"), stylesheet['Heading3']))
-        pdf.append(Paragraph(motion.reason.replace('\r\n',  '<br/>'), stylesheet['Paragraph']))
+        convert_html_to_reportlab(pdf, motion.reason)
     return pdf
 
 
+def convert_html_to_reportlab(pdf, text):
+    # parsing and replacing not supported html tags for reportlab...
+    soup = BeautifulSoup(text)
+    # read all list elements...
+    for element in soup.find_all('li'):
+        try:
+            # ... and replace ul list elements with <para><bullet>&bull;</bullet>...<para>
+            if element.parent.name == "ul":
+                if element.ul:
+                    # for nested ul lists use simple spaces (pragmatic solution)
+                    element.li.insert(0,'&nbsp;&nbsp;&nbsp;&nbsp;')
+                    element.insert_before(element.find_all('li'))
+                    element.clear()
+                else:
+                    element.name = "para"
+                    bullet_tag = soup.new_tag("bullet")
+                    bullet_tag.string = "&bull;"
+                    element.insert(0, bullet_tag)
+            # ... and replace ol list elements with <para><bullet><seq id="%id"></seq>.</bullet>...</para>
+            if element.parent.name == "ol":
+                # set list id if element is the first of numbered list
+                if not element.find_previous_sibling():
+                    id = random.randrange(0, 101)
+                if element.ol:
+                    # nested ol list
+                    element.li.insert(0,'&nbsp;&nbsp;&nbsp;&nbsp;')
+                    element.insert_before(element.find_all('li'))
+                    element.clear()
+                else:
+                    element.name = "para"
+                    element.insert(0, soup.new_tag("bullet"))
+                    element.bullet.insert(0, soup.new_tag("seq"))
+                    element.bullet.seq['id'] = id
+                    element.bullet.insert(1, ".")
+        except AttributeError:
+            pass
+    # remove tags which are not supported by reportlab (replace tags with their children tags)
+    for tag in soup.find_all('ul'):
+        tag.unwrap()
+    for tag in soup.find_all('ol'):
+        tag.unwrap()
+    for tag in soup.find_all('li'):
+        tag.unwrap()
+    # print paragraphs with numbers
+    text = soup.body.contents
+    paragraph_number = 1
+    for paragraph in text:
+        paragraph = str(paragraph)
+        # ignore empty paragraphs (created by newlines/tabs of ckeditor)
+        if paragraph == '\n' or paragraph == '\n\n' or paragraph == '\n\t':
+            continue
+        if "<pre>" in paragraph:
+            pdf.append(Paragraph(paragraph.replace('\n', '<br/>'), stylesheet['InnerMonotypeParagraph'], str(paragraph_number)))
+            paragraph_number += 1
+        elif "<para>" in paragraph:
+            pdf.append(Paragraph(paragraph, stylesheet['InnerListParagraph']))
+        elif "<seqreset" in paragraph:
+            pass
+        elif "<h1>" in paragraph:
+            pdf.append(Paragraph(paragraph, stylesheet['InnerH1Paragraph']))
+        elif "<h2>" in paragraph:
+            pdf.append(Paragraph(paragraph, stylesheet['InnerH2Paragraph']))
+        elif "<h3>" in paragraph:
+            pdf.append(Paragraph(paragraph, stylesheet['InnerH3Paragraph']))
+        else:
+            pdf.append(Paragraph(str(paragraph), stylesheet['InnerParagraph'], str(paragraph_number)))
+            paragraph_number += 1
+
+
 def all_motion_cover(pdf, motions):
     """Create a coverpage for all motions."""
     pdf.append(Paragraph(config["motion_pdf_title"], stylesheet['Heading1']))

openslides/utils/pdf.py

@@ -13,6 +13,7 @@
 from datetime import datetime
 from os.path import join as path_join
 
+from reportlab.lib import colors
 from reportlab.lib.styles import StyleSheet1, ParagraphStyle
 from reportlab.lib.units import cm
 from reportlab.pdfbase import pdfmetrics
@@ -44,7 +45,7 @@ PAGE_WIDTH = defaultPageSize[0]
 stylesheet = StyleSheet1()
 stylesheet.add(ParagraphStyle(
     name='Normal',
-    fontName='Ubuntu',
+    #fontName='Ubuntu',
     fontSize=10,
     leading=12,
 ))
@@ -54,6 +55,51 @@ stylesheet.add(ParagraphStyle(
     leading=14,
     spaceAfter=15
 ))
+stylesheet.add(ParagraphStyle(
+    name='InnerParagraph',
+    parent=stylesheet['Normal'],
+    leading=14,
+    spaceBefore=5,
+    spaceAfter=5,
+    bulletIndent=-15,
+    bulletFontSize=8,
+    bulletColor=colors.grey
+))
+stylesheet.add(ParagraphStyle(
+    name='InnerListParagraph',
+    parent=stylesheet['InnerParagraph'],
+    bulletIndent=10,
+    bulletFontSize=10,
+    bulletColor=colors.black,
+    leftIndent = 30
+))
+stylesheet.add(ParagraphStyle(
+    name='InnerMonotypeParagraph',
+    parent=stylesheet['InnerParagraph'],
+    fontName='Courier',
+))
+stylesheet.add(ParagraphStyle(
+    name='InnerH1Paragraph',
+    parent=stylesheet['InnerParagraph'],
+    fontName='Ubuntu-Bold',
+    fontSize=16,
+    spaceBefore=20,
+    spaceAfter=10,
+))
+stylesheet.add(ParagraphStyle(
+    name='InnerH2Paragraph',
+    parent=stylesheet['InnerH1Paragraph'],
+    fontSize=12,
+    spaceBefore=20,
+    spaceAfter=10,
+))
+stylesheet.add(ParagraphStyle(
+    name='InnerH3Paragraph',
+    parent=stylesheet['InnerH2Paragraph'],
+    fontSize=10,
+    spaceBefore=15,
+    spaceAfter=5,
+))
 stylesheet.add(ParagraphStyle(
     name='Small',
     parent=stylesheet['Normal'],

requirements.txt

@@ -1,5 +1,8 @@
+# Requirements for OpenSlides Core
 Django==1.5.1
 django-mptt==0.5.5
+beautifulsoup4==4.1.3
+bleach==1.2.1
 pillow==2.0.0
 qrcode==2.7
 reportlab==2.7
@@ -10,4 +13,3 @@ Fabric==1.6.0
 coverage==3.6
 django-discover-runner==0.3
 pep8==1.4.5
-bleach