def __init__(self, login):
global ldapcon
global basedn
if basedn == '':
basedn = cfg.get('ldap', 'basedn')
if ldapcon is None:
ldapuri = cfg.get('ldap', 'uri')
try:
ldapcon = self.ldap_obj = LDAP(ldapuri)
except Exception as e:
raise Exception('Failed to bind to Ldap: %s' % e)
try:
self._data = []
search = 'uid=%s' % str(login).strip()
self._data = ldapcon.search_s(basedn, ldap.SCOPE_SUBTREE, search)
except Exception as e:
raise self.LDAPException("Got error in search: '%s' '%s'<br/>%s" %
(basedn, search, e))
if not self._data:
raise self.LDAPException("%s not known" % login)
self.index = 0
self.iterator = False
self.unique = len(self._data) == 1
def linklint_handler(req, query, messages):
path = req.filename
req.filename = content_dir + "validator/linklint.pyl"
"""new version of py_handler() to work with PSP templates"""
# get string from our customized handler()
fakereq = DummyPage(req)
f = open(req.filename, "r")
filepath, filename = os.path.split(req.filename)
oldpath = sys.path
sys.path.append(filepath) #Allow files to import from their directory - bad?
module = imp.load_source(os.path.splitext(filename)[0], req.filename, f)
sys.path = oldpath
module.handler(fakereq, query, path)
user_content = fakereq.get_page()
title = ("%s | Linklint Check - " % cfg.get('core', 'rg_abbrev').upper()) + path
breadcrumbs = get_breadcrumbs(req)
data = dict(cfg=cfg, psp=psp, breadcrumbs=breadcrumbs, title=title,
user_content=user_content, nav_urls=nav_urls, messages=messages)
template = psp.PSP(req, filename=cfg.get('core', 'template_file'))
template.run(data)
f.close()
return apache.OK
def edit_handler(req, query, messages):
req.content_type = "text/html"
if not os.path.isfile(req.filename) and \
not (req.filename.endswith('_') and os.path.isfile(req.filename[:-1])):
req.send_http_header()
return apache.HTTP_NOT_FOUND
if req.filename.endswith('_'):
req.send_http_header()
return psp.handler(req)
user_content = ''
if not os.path.isfile(req.filename) or os.path.islink(req.filename) or \
not (req.filename.endswith('.pml') or req.filename.endswith('.html')):
sorry_text = """
<html><head><title>Can't edit this file | %s </title></head><body><h1>Can't edit this content</h1>
<p>To edit underlay or python code, ssh to the site</p>
</body></html>
""" % cfg.get('core', 'rg_abbrev').upper()
req.write(sorry_text)
return apache.HTTP_FORBIDDEN
f = open(req.filename, 'r')
title_re = re.search(r"\<title\>(.*)\<\/title\>", f.read())
f.close()
if title_re:
title = "Edit %s | %s " % (title_re.groups()[0], cfg.get('core', 'rg_abbrev').upper())
else:
title = "Edit %s | %s " % (req.filename, cfg.get('core', 'rg_abbrev').upper())
# See if page is requesting custom banner
banner_img = None
f = open(req.filename, 'r')
banner_re = re.search(r"\<banner\>(.*)\<\/banner\>", f.read())
f.close()
if banner_re:
banner_img = banner_re.groups()[0]
breadcrumbs = get_breadcrumbs(req)
f = open(req.filename, "r")
user_content = f.read()
f.close()
data = dict(psp=psp, breadcrumbs=breadcrumbs, title=title, banner_img=banner_img,
user_head_content='',
user_content=user_content, cfg=cfg,
nav_urls=nav_urls, messages=messages)
template_file = cfg.get('core', 'template_file_edit')
template = psp.PSP(req, filename=template_file)
template.run(data)
return apache.OK
def get_news_elements(item_count=6, max_title_len=80, max_story_len=350):
# Get news file, surround it by parent tags, and then join together
basedir = cfg.get('core', "base_dir")
news = ["<xml>"]
news.extend(readfile(os.path.join(basedir, "content/news/home.pml")))
news.append("</xml>")
news = "\n".join(news)
# Preprocess string for ultra-conservative parser
news = re.sub(r"&\S*;", " ", news)
news = news.replace("<br>", "<br />")
parser = etree.HTMLParser()
tree = etree.parse(StringIO(news), parser)
news_items = []
for entry in tree.iter('tr'): # Go through each row
for title in entry.iter('th'): # Titles
txt = "".join(list(title.itertext()))
txt = txt[:max_title_len-3] + "..." if len(txt) > max_title_len else txt
for body in entry.iter('td'): # Stories
txt = "".join(list(body.itertext()))
txt = txt[:max_story_len-3] + "..." if len(txt) > max_story_len else txt
news_items.append(txt.strip().replace('\n', ' '))
if len(news_items) >= item_count * 2:
break
news_elements = [news_items[i:i+2] for i in range(0, len(news_items), 2)]
return news_elements
def save_handler(req, query, messages):
path = req.filename
content_dir = cfg.get('core', 'content_dir')
length = int(req.headers_in['content-length'])
postvars = parse_qs(req.read(length), keep_blank_values=1)
if b'content' not in postvars:
return apache.HTTP_NOT_FOUND
# Editor adds <p> </p> if file starts with a blank line or a comment
user_content = postvars[b'content'].pop().decode('utf-8').replace(r'<p> </p>', '', 1)
tidy = Tidy()
OPTIONS = {
'wrap': 72,
'hide-comments' : 'no',
'preserve-entities': 'yes',
'char-encoding' : 'utf8',
'indent-attributes' : 'yes',
}
doc = '<!DOCTYPE html><head><title>foo</title></head><body>' + user_content + '</body></html>'
doc, errs = tidy.tidy_document(doc, options=OPTIONS)
if errs is not '':
req.write("<strong>Errors found</strong><br/><pre>")
req.write(cgi.escape(errs))
req.write("</pre>")
return apache.OK
doc = re.sub(r'^.*<body>(.*)</body>.*', r'\1', doc, flags=re.DOTALL)
doc = re.sub(r'^ ', '', doc, flags=re.MULTILINE)
if path.startswith(content_dir):
filename = path + ".new"
f = open(filename, "w")
f.write(doc)
f.close()
os.rename(filename, path)
os.chmod(path, 0o664)
else:
req.write("<strong>Attempt to update a file not in the web directory</strong>")
return apache.HTTP_FORBIDDEN
if 'save' in query:
del query['save']
req.args = req.args.replace('?save', '')
return psp_handler(req, query, messages)
def req_to_title(req):
basedir = cfg.get('core', "base_dir")
# Convert a request to a guess at what the title should be
parse_filename = req.filename[len(basedir):-4]
if parse_filename.endswith("home"):
parse_filename = parse_filename[:-5]
parts = parse_filename.split("/")
title = ""
if len(parts) > 1:
title = "%s - %s -" % (parts[-2], parts[-1])
elif len(parts) > 0:
title = "%s -" % parts[0]
if title == " -":
title = ""
if title == " - -":
title = ""
return title
def search_handler(req, query, messages):
path = os.path.dirname(req.filename)
fakereq = DummyPage(req)
f = open(req.filename, "r")
filepath, filename = os.path.split(req.filename)
oldpath = sys.path
sys.path.append(filepath) #Allow files to import from their directory - bad?
module = imp.load_source(os.path.splitext(filename)[0], req.filename, f)
sys.path = oldpath
module.handler(fakereq, query, path)
user_content = fakereq.get_page()
title = module.get_title(req, query, path)
breadcrumbs = get_breadcrumbs(req)
data = dict(cfg=cfg, psp=psp, breadcrumbs=breadcrumbs, title=title,
user_content=user_content, nav_urls=nav_urls, messages=messages)
template = psp.PSP(req, filename=cfg.get('core', 'template_file'))
template.run(data)
return apache.OK
# Run a link checker over the live website.
# Send email on error
#
from ertos_config import cfg
import smtplib
import os
from email.mime.text import MIMEText
from subprocess import run
import sys
sys.path.append(
os.path.join(os.path.dirname(os.path.abspath(__file__)), "../code"))
cfg.read('config.cfg')
mailhost = cfg.get('hosts', 'mailhost')
RECIPIENTS = ['webmaster@%s' % cfg.get('hosts', 'maildomain')]
CONTENTROOT = cfg.get('core', 'content_dir')
OUTPUTDIR = '%s/linklint' % CONTENTROOT
URL = "https://%s/" % cfg.get('hosts', 'live')
URL1 = "https://docs.%s/" % cfg.get('hosts', 'live')
linklint = run([
'/usr/bin/linkchecker', '-F',
'html/%s/index.html' % OUTPUTDIR, '--no-status', '--ignore-url="print$"'
'--ignore-url="/blob/"'
'--check-extern', URL, URL1
],
encoding='utf-8',
capture_output=True)
def handler(req):
import time
start = time.time()
if not os.path.isfile(req.filename):
return apache.HTTP_NOT_FOUND
# reads config and set some global variables
global base_dir, content_dir, nav_urls
#cfg.read(req.get_options()['site_config'])
base_dir = cfg.get('core', 'base_dir') + "/"
content_dir = cfg.get('core', 'content_dir') + "/"
in_production = cfg.getboolean('core', 'production')
d = os.path.dirname(req.filename)
nav_urls = dir2list(content_dir)
req.content_type = "text/html"
if req.args is not None:
query = cgi.parse_qs(req.args)
else:
query = {}
# TODO: Make "messages" append to an error log instead
messages = []
try:
res = apache.OK
if not in_production and req.args is not None:
if 'linklint' in req.args:
return linklint_handler(req, query, messages)
if 'validate' in req.args:
return validate_handler(req, query, messages)
if 'save' in req.args:
return save_handler(req, query, messages)
if 'edit' in req.args:
return edit_handler(req, query, messages)
if req.filename.endswith('search/home.pyl'):
res = search_handler(req, query, messages)
elif req.filename.endswith(".pml") or req.filename.endswith('.psp'):
res = psp_handler(req, query, messages)
elif req.filename.endswith(".pyl"):
res = pyl_handler(req, query, messages)
elif req.filename.endswith(".pyx"):
res = pyx_handler(req, query, messages)
else:
res = apache.HTTP_NOT_FOUND
if req.content_type == "text/html":
req.write("<!-- page generated in %s seconds -->" % str(time.time() - start))
return res
except Exception as e:
# cgitb should NEVER be used on the live public site
# it can leak internal data like usernames and passwords
formatted_lines = traceback.format_exc().splitlines()
tb = ""
for line in formatted_lines:
tb += line +"\n"
if not in_production:
import cgitb
#cgitb.enable()
cgitb.Hook(file=req).handle()
sorry_text = """<html><head><title>404 Not Found | %s | Data61</title></head><body><h1>404 Not Found</h1>
<p>Some content in this page is currently unavailable or no longer exists.<br/>
The webmaster has been notified. Please try again later.</p></body></html>
""" % cfg.get('core', 'rg_abbrev').upper()
req.write(sorry_text)
if not in_production:
req.write("<p>Traceback:</p>")
req.write("<p>"+tb.replace("\n", "<br />")+"</p>")
if in_production:
# send webmasters a mail with the trace
import smtplib, time
from email.mime.text import MIMEText
text = """
Problem with %s webpage at:
%s:%s.
Please check...
Trace follows:
%s
Arguments (if any):
%s
mod-python hard working at %s
""" % (cfg.get('core', 'rg_abbrev').upper(),
req.server.server_hostname,
req.filename, tb, str(query),
time.strftime('%x %X %Z'))
msg = MIMEText(text)
md = cfg.get('hosts', 'maildomain')
mx = cfg.get('hosts', 'mailhost')
webmasters = ['webmaster@' + md]
msg['Subject'] = '<%s Web> - Exception caught' % cfg.get('core', 'rg_abbrev').upper()
msg['From'] = 'webmaster@' + md
msg['To'] = 'webmaster@' + md
s = smtplib.SMTP()
s.connect(mx)
s.sendmail(webmasters[0], webmasters, msg.as_string())
s.close()
return apache.OK
def psp_handler(req, query, messages):
req.content_type = "text/html; charset=utf-8"
if not os.path.isfile(req.filename) and \
not (req.filename.endswith('_') and os.path.isfile(req.filename[:-1])):
req.send_http_header()
return apache.HTTP_NOT_FOUND
# Debugging output. Needs PythonDebug in the apache conf.
if req.filename.endswith('_'):
req.send_http_header()
return psp.handler(req)
user_content = ''
# trick PSP to give us a string for the user page
# and pass it to the template
fakereq = DummyPage(req)
user_page = psp.PSP(fakereq, filename=req.filename)
data = dict(cfg=cfg)
try:
user_page.run(data)
except apache.SERVER_RETURN:
req.page = fakereq.page
req.status = fakereq.status
return apache.OK
user_content = fakereq.get_page()
# some variables needed in our templates
breadcrumbs = get_breadcrumbs(req)
f = open(req.filename, "r")
title_re = re.search(r"\<title\>(.*)\<\/title\>", f.read())
f.close()
# If the user content was a tuple, it means we specified extra data for the <head>
user_head_content = ""
if type(user_content) is tuple:
user_head_content = user_content[1]
user_content = user_content[0]
if cfg.has_option('core', 'titletemplate'):
title = cfg.get('core', 'titletemplate')
else:
title = "%s | CSIRO's Data61" % cfg.get('core', 'rg_abbrev').upper()
if title_re:
title_end = title_re.end()
title = "%s | %s" % (title_re.groups()[0], title)
else:
# Default to the ugly URL
cur_url = req.filename[len(cfg.get('core', 'content_dir')):].strip('/')
live_url = '%s' % cfg.get('hosts', 'url')+'/'+cur_url
live_url = cur_url
# Set the title
title = "%s | %s" % (title, live_url)
# See if page is requesting custom banner
banner_img = None
f = open(req.filename, "r")
banner_re = re.search(r"\<banner\>(.*)\<\/banner\>", f.read())
f.close()
if banner_re:
banner_img = banner_re.groups()[0]
data = dict(psp=psp, breadcrumbs=breadcrumbs, title=title,
banner_img=banner_img, user_head_content=user_head_content,
user_content=user_content, cfg=cfg, nav_urls=nav_urls,
messages=messages)
template_file = cfg.get('core', 'template_file')
if 'media' in query:
if query['media'][0] == 'print':
template_file = cfg.get('core', 'template_file_print')
try:
template = psp.PSP(req, filename=template_file)
except Exception as e:
req.send_http_header()
req.write("Template prep failed, $s\n", str(e))
traceback.print_exc()
return apache.OK
req.send_http_header()
try:
template.run(data)
except Exception as e:
req.write("template run failed, %s\n" % str(e))
traceback.print_exc()
return apache.OK
def pyl_handler(req, query, messages):
"""new version of py_handler() to work with PSP templates"""
# get string from our customized handler()
fakereq = DummyPage(req)
fakereq.user = req.user
try:
f = open(req.filename, "r")
filepath, filename = os.path.split(req.filename)
oldpath = sys.path
sys.path.append(filepath) #Allow files to import from their directory - bad?
module = imp.load_source(os.path.splitext(filename)[0], req.filename, f)
except:
req.status = apache.HTTP_NOT_FOUND
return apache.OK
sys.path = oldpath
status = module.handler(fakereq, query)
user_content = fakereq.get_page()
# if this is a file download, then don't spit out template
if "Content-Disposition" in req.headers_out:
return apache.OK
if cfg.has_option('core', 'titletemplate'):
title = cfg.get('core', 'titletemplate')
else:
title = "%s | seL4" % cfg.get('core', 'rg_abbrev').upper()
if "get_title" in dir(module):
title = module.get_title(req) + " | " + title
usr_head_content = ""
if "get_head" in dir(module):
usr_head_content = module.get_head(req)
breadcrumbs = get_breadcrumbs(req)
data = dict(cfg=cfg, psp=psp, breadcrumbs=breadcrumbs, title=title,
user_content=user_content,
user_head_content=usr_head_content,
nav_urls=nav_urls, messages=messages)
template_file = cfg.get('core', 'template_file')
if 'media' in query:
if query['media'][0] == 'print':
template_file = cfg.get('core', 'template_file_print')
if status != 0:
req.status = status
try:
template = psp.PSP(req, filename=template_file)
except Exception as e:
req.write("psp failed: %s" % str(e))
return apache.OK
req.content_type = "text/html; charset=utf-8"
try:
template.run(data)
except apache.SERVER_RETURN:
req.page = fakereq.page
req.status = fakereq.status
return apache.OK
except Exception as e:
req.write("template run failed: %s" % str(e))
return apache.OK