[MERGE] forward port of branch saas-2 up to revid 4965 chs@openerp.com-20131025103939-5gta1eifjhx1tsmi

bzr revid: chs@openerp.com-20131025104326-4k5hayl0bwdgt0mm
This commit is contained in:
Christophe Simonis 2013-10-25 12:43:26 +02:00
commit d91819cb43
5 changed files with 111 additions and 56 deletions

View File

@ -34,6 +34,7 @@
<field name="lang"/>
</group>
<group>
<field name="module" group="base.group_no_one"/>
<field name="type"/>
<field name="res_id"/>
</group>

View File

@ -47,33 +47,8 @@ class base_module_import(osv.osv_memory):
}
def importzip(self, cr, uid, ids, context):
(data,) = self.browse(cr, uid, ids , context=context)
module_data = data.module_file
zip_data = base64.decodestring(module_data)
fp = StringIO()
fp.write(zip_data)
try:
file_data = zipfile.ZipFile(fp, 'r')
except zipfile.BadZipfile:
raise osv.except_osv(_('Error!'), _('File is not a zip file!'))
init_file_name = sorted(file_data.namelist())[0]
module_name = os.path.split(init_file_name)[0]
file_path = os.path.join(ADDONS_PATH, '%s.zip' % module_name)
try:
zip_file = open(file_path, 'wb')
except IOError:
raise osv.except_osv(_('Error!'),
_('Can not create the module file: %s!') % \
(file_path,) )
zip_file.write(zip_data)
zip_file.close()
self.pool.get('ir.module.module').update_list(cr, uid,
{'module_name': module_name,})
self.write(cr, uid, ids, {'state':'done', 'module_name': module_name},
context)
return False
#TODO: drop this model and the corresponding view/action in trunk
raise NotImplementedError('This feature is not available')
def action_module_open(self, cr, uid, ids, context):
(data,) = self.browse(cr, uid, ids , context=context)

View File

@ -116,8 +116,7 @@ class _column(object):
self.groups = False # CSV list of ext IDs of groups that can access this field
self.deprecated = False # Optional deprecation warning
for a in args:
if args[a]:
setattr(self, a, args[a])
setattr(self, a, args[a])
def restart(self):
pass

View File

@ -149,6 +149,59 @@ class TestCleaner(unittest2.TestCase):
for text in out_lst:
self.assertNotIn(text, new_html, 'html_email_cleaner did not remove unwanted content')
def test_05_shorten(self):
# TEST: shorten length
test_str = '''<div>
<span>
</span>
<p>Hello, <span>Raoul</span>
<bold>You</bold> are
pretty</p>
<span>Really</span>
</div>
'''
# shorten at 'H' of Hello -> should shorten after Hello,
html = html_email_clean(test_str, shorten=True, max_length=1, remove=True)
self.assertIn('Hello,', html, 'html_email_cleaner: shorten error or too short')
self.assertNotIn('Raoul', html, 'html_email_cleaner: shorten error or too long')
self.assertIn('read more', html, 'html_email_cleaner: shorten error about read more inclusion')
# shorten at 'are' -> should shorten after are
html = html_email_clean(test_str, shorten=True, max_length=17, remove=True)
self.assertIn('Hello,', html, 'html_email_cleaner: shorten error or too short')
self.assertIn('Raoul', html, 'html_email_cleaner: shorten error or too short')
self.assertIn('are', html, 'html_email_cleaner: shorten error or too short')
self.assertNotIn('pretty', html, 'html_email_cleaner: shorten error or too long')
self.assertNotIn('Really', html, 'html_email_cleaner: shorten error or too long')
self.assertIn('read more', html, 'html_email_cleaner: shorten error about read more inclusion')
# TEST: shorten in quote
test_str = '''<div> Blahble
bluih blouh
<blockquote>This is a quote
<span>And this is quite a long quote, after all.</span>
</blockquote>
</div>'''
# shorten in the quote
html = html_email_clean(test_str, shorten=True, max_length=25, remove=True)
self.assertIn('Blahble', html, 'html_email_cleaner: shorten error or too short')
self.assertIn('bluih', html, 'html_email_cleaner: shorten error or too short')
self.assertIn('blouh', html, 'html_email_cleaner: shorten error or too short')
self.assertNotIn('quote', html, 'html_email_cleaner: shorten error or too long')
self.assertIn('read more', html, 'html_email_cleaner: shorten error about read more inclusion')
# shorten in second word
html = html_email_clean(test_str, shorten=True, max_length=9, remove=True)
self.assertIn('Blahble', html, 'html_email_cleaner: shorten error or too short')
self.assertIn('bluih', html, 'html_email_cleaner: shorten error or too short')
self.assertNotIn('blouh', html, 'html_email_cleaner: shorten error or too short')
self.assertNotIn('quote', html, 'html_email_cleaner: shorten error or too long')
self.assertIn('read more', html, 'html_email_cleaner: shorten error about read more inclusion')
# shorten waaay too large
html = html_email_clean(test_str, shorten=True, max_length=900, remove=True)
self.assertIn('Blahble', html, 'html_email_cleaner: shorten error or too short')
self.assertIn('bluih', html, 'html_email_cleaner: shorten error or too short')
self.assertIn('blouh', html, 'html_email_cleaner: shorten error or too short')
self.assertNotIn('quote', html, 'html_email_cleaner: shorten error or too long')
def test_10_email_text(self):
""" html_email_clean test for text-based emails """
new_html = html_email_clean(test_mail_examples.TEXT_1, remove=True)
@ -230,7 +283,7 @@ class TestCleaner(unittest2.TestCase):
for ext in test_mail_examples.BUG_1_OUT:
self.assertNotIn(ext, new_html, 'html_email_cleaner did not removed invalid content')
new_html = html_email_clean(test_mail_examples.BUG2, remove=True, shorten=True, max_length=4000)
new_html = html_email_clean(test_mail_examples.BUG2, remove=True, shorten=True, max_length=250)
for ext in test_mail_examples.BUG_2_IN:
self.assertIn(ext, new_html, 'html_email_cleaner wrongly removed valid content')
for ext in test_mail_examples.BUG_2_OUT:

View File

@ -175,26 +175,45 @@ def html_email_clean(html, remove=False, shorten=False, max_length=300):
iteration += 1
new_node = _insert_new_node(node, -1, new_node_tag, text[idx:] + (cur_node.tail or ''), None, {})
def _truncate_node(node, position, find_first_blank=True):
def _truncate_node(node, position, simplify_whitespaces=True):
""" Truncate a node text at a given position. This algorithm will shorten
at the end of the word whose ending character exceeds position.
:param bool simplify_whitespaces: whether to try to count all successive
whitespaces as one character. This
option should not be True when trying
to keep 'pre' consistency.
"""
if node.text is None:
node.text = ''
# truncate text
end_position = position if len(node.text) >= position else len(node.text)
innertext = node.text[0:end_position]
outertext = node.text[end_position:]
if find_first_blank:
stop_idx = outertext.find(' ')
if stop_idx == -1:
stop_idx = len(outertext)
truncate_idx = -1
if simplify_whitespaces:
cur_char_nbr = 0
word = None
node_words = node.text.strip(' \t\r\n').split()
for word in node_words:
cur_char_nbr += len(word)
if cur_char_nbr >= position:
break
if word:
truncate_idx = node.text.find(word) + len(word)
else:
stop_idx = 0
node.text = innertext + outertext[0:stop_idx]
truncate_idx = position
if truncate_idx == -1 or truncate_idx > len(node.text):
truncate_idx = len(node.text)
# compose new text bits
innertext = node.text[0:truncate_idx]
outertext = node.text[truncate_idx:]
node.text = innertext
# create <span> ... <a href="#">read more</a></span> node
read_more_node = _create_node('span', ' ... ', None, {'class': 'oe_mail_expand'})
read_more_link_node = _create_node('a', 'read more', None, {'href': '#', 'class': 'oe_mail_expand'})
read_more_node.append(read_more_link_node)
# create outertext node
overtext_node = _create_node('span', outertext[stop_idx:])
overtext_node = _create_node('span', outertext)
# tag node
overtext_node.set('in_overlength', '1')
# add newly created nodes in dom
@ -223,17 +242,16 @@ def html_email_clean(html, remove=False, shorten=False, max_length=300):
html = '<div>%s</div>' % html
root = lxml.html.fromstring(html)
# remove all tails and replace them by a span element, because managing text and tails can be a pain in the ass
for node in root.getiterator():
quote_tags = re.compile(r'(\n(>)+[^\n\r]*)')
signature = re.compile(r'([-]{2,}[\s]?[\r\n]{1,2}[\s\S]+)')
for node in root.iter():
# remove all tails and replace them by a span element, because managing text and tails can be a pain in the ass
if node.tail:
tail_node = _create_node('span', node.tail)
node.tail = None
node.addnext(tail_node)
# form node and tag text-based quotes and signature
quote_tags = re.compile(r'(\n(>)+[^\n\r]*)')
signature = re.compile(r'([-]{2,}[\s]?[\r\n]{1,2}[\s\S]+)')
for node in root.getiterator():
# form node and tag text-based quotes and signature
_tag_matching_regex_in_text(quote_tags, node, 'span', {'text_quote': '1'})
_tag_matching_regex_in_text(signature, node, 'span', {'text_signature': '1'})
@ -245,7 +263,10 @@ def html_email_clean(html, remove=False, shorten=False, max_length=300):
quote_begin = False
overlength = False
cur_char_nbr = 0
for node in root.getiterator():
for node in root.iter():
# do not take into account multiple spaces that are displayed as max 1 space in html
node_text = ' '.join((node.text and node.text.strip(' \t\r\n') or '').split())
# root: try to tag the client used to write the html
if 'WordSection1' in node.get('class', '') or 'MsoNormal' in node.get('class', ''):
root.set('msoffice', '1')
@ -277,27 +298,30 @@ def html_email_clean(html, remove=False, shorten=False, max_length=300):
# 1/ truncate the text at the next available space
# 2/ create a 'read more' node, next to current node
# 3/ add the truncated text in a new node, next to 'read more' node
if shorten and not overlength and cur_char_nbr + len(node.text or '') > max_length:
if shorten and not overlength and cur_char_nbr + len(node_text) > max_length:
node_to_truncate = node
while node_to_truncate.get('in_quote') and node_to_truncate.getparent() is not None:
node_to_truncate = node_to_truncate.getparent()
overlength = True
node_to_truncate.set('truncate', '1')
node_to_truncate.set('truncate_position', str(max_length - cur_char_nbr))
cur_char_nbr += len(node.text or '')
if node_to_truncate == node:
node_to_truncate.set('truncate_position', str(max_length - cur_char_nbr))
else:
node_to_truncate.set('truncate_position', str(len(node.text or '')))
cur_char_nbr += len(node_text)
# Tree modification
# ------------------------------------------------------------
for node in root.iter():
if node.get('truncate'):
_truncate_node(node, int(node.get('truncate_position', '0')))
_truncate_node(node, int(node.get('truncate_position', '0')), node.tag != 'pre')
# Post processing
# ------------------------------------------------------------
to_remove = []
for node in root.getiterator():
for node in root.iter():
if node.get('in_quote') or node.get('in_overlength'):
# copy the node tail into parent text
if node.tail and not node.get('tail_remove'):
@ -306,17 +330,20 @@ def html_email_clean(html, remove=False, shorten=False, max_length=300):
to_remove.append(node)
if node.get('tail_remove'):
node.tail = ''
# clean node
for attribute_name in ['in_quote', 'tail_remove', 'in_overlength', 'msoffice', 'hotmail', 'truncate', 'truncate_position']:
node.attrib.pop(attribute_name, None)
for node in to_remove:
if remove:
node.getparent().remove(node)
else:
if not 'oe_mail_expand' in node.get('class', ''): # trick: read more link should be displayed even if it's in overlength
node_class = node.get('class', '') + ' ' + 'oe_mail_cleaned'
node_class = node.get('class', '') + ' oe_mail_cleaned'
node.set('class', node_class)
# html: \n that were tail of elements have been encapsulated into <span> -> back to \n
html = etree.tostring(root, pretty_print=False)
linebreaks = re.compile(r'<span>([\s]*[\r\n]+[\s]*)<\/span>', re.IGNORECASE | re.DOTALL)
linebreaks = re.compile(r'<span[^>]*>([\s]*[\r\n]+[\s]*)<\/span>', re.IGNORECASE | re.DOTALL)
html = _replace_matching_regex(linebreaks, html, '\n')
return html