1
|
|
|
########################################################################## |
2
|
|
|
# MET v2 Metadate Explorer Tool |
3
|
|
|
# |
4
|
|
|
# This Software is Open Source. See License: https://github.com/TERENA/met/blob/master/LICENSE.md |
5
|
|
|
# Copyright (c) 2012, TERENA All rights reserved. |
6
|
|
|
# |
7
|
|
|
# This Software is based on MET v1 developed for TERENA by Yaco Sistemas, http://www.yaco.es/ |
8
|
|
|
# MET v2 was developed for TERENA by Tamim Ziai, DAASI International GmbH, http://www.daasi.de |
9
|
|
|
# Current version of MET has been revised for performance improvements by Andrea Biancini, |
10
|
|
|
# Consortium GARR, http://www.garr.it |
11
|
|
|
########################################################################## |
12
|
|
|
|
13
|
|
|
from os import path |
14
|
|
|
from lxml import etree |
15
|
|
|
import simplejson as json |
16
|
|
|
|
17
|
|
|
from django.db import models |
18
|
|
|
from django.contrib.auth.models import User |
19
|
|
|
from django.core import validators |
20
|
|
|
from django.core.files.base import ContentFile |
21
|
|
|
from django.utils.translation import ugettext_lazy as _ |
22
|
|
|
|
23
|
|
|
from pyff.mdrepo import MDRepository |
24
|
|
|
from pyff.pipes import Plumbing |
25
|
|
|
|
26
|
|
|
from met.metadataparser.xmlparser import MetadataParser |
27
|
|
|
from met.metadataparser.utils import compare_filecontents |
28
|
|
|
|
29
|
|
|
|
30
|
|
|
class JSONField(models.CharField): |
31
|
|
|
""" |
32
|
|
|
JSONField is a generic textfield that neatly serializes/unserializes |
33
|
|
|
JSON objects seamlessly |
34
|
|
|
|
35
|
|
|
The json spec claims you must use a collection type at the top level of |
36
|
|
|
the data structure. However the simplesjon decoder and Firefox both encode |
37
|
|
|
and decode non collection types that do not exist inside a collection. |
38
|
|
|
The to_python method relies on the value being an instance of basestring |
39
|
|
|
to ensure that it is encoded. If a string is the sole value at the |
40
|
|
|
point the field is instanced, to_python attempts to decode the sting because |
41
|
|
|
it is derived from basestring but cannot be encodeded and throws the |
42
|
|
|
exception ValueError: No JSON object could be decoded. |
43
|
|
|
""" |
44
|
|
|
|
45
|
|
|
# Used so to_python() is called |
46
|
|
|
__metaclass__ = models.SubfieldBase |
47
|
|
|
description = _("JSON object") |
48
|
|
|
|
49
|
|
|
def __init__(self, *args, **kwargs): |
50
|
|
|
super(JSONField, self).__init__(*args, **kwargs) |
51
|
|
|
self.validators.append(validators.MaxLengthValidator(self.max_length)) |
52
|
|
|
|
53
|
|
|
@classmethod |
54
|
|
|
def get_internal_type(cls): |
55
|
|
|
return "TextField" |
56
|
|
|
|
57
|
|
|
@classmethod |
58
|
|
|
def to_python(cls, value): |
59
|
|
|
"""Convert our string value to JSON after we load it from the DB""" |
60
|
|
|
if value == "": |
61
|
|
|
return None |
62
|
|
|
|
63
|
|
|
try: |
64
|
|
|
if isinstance(value, basestring): |
65
|
|
|
return json.loads(value) |
66
|
|
|
except ValueError: |
67
|
|
|
return value |
68
|
|
|
|
69
|
|
|
return value |
70
|
|
|
|
71
|
|
|
def get_prep_value(self, value): |
72
|
|
|
"""Convert our JSON object to a string before we save""" |
73
|
|
|
|
74
|
|
|
if not value or value == "": |
75
|
|
|
return None |
76
|
|
|
|
77
|
|
|
db_value = json.dumps(value) |
78
|
|
|
return super(JSONField, self).get_prep_value(db_value) |
79
|
|
|
|
80
|
|
|
def get_db_prep_value(self, value, connection, prepared=False): |
81
|
|
|
"""Convert our JSON object to a string before we save""" |
82
|
|
|
|
83
|
|
|
if not value or value == "": |
84
|
|
|
return None |
85
|
|
|
|
86
|
|
|
db_value = json.dumps(value) |
87
|
|
|
return super(JSONField, self).get_db_prep_value(db_value, connection, prepared) |
88
|
|
|
|
89
|
|
|
|
90
|
|
|
class Base(models.Model): |
91
|
|
|
""" |
92
|
|
|
Class describing an entity that can be updated from metadata file. |
93
|
|
|
Each object parsed from the XML extends this base class that contains shared methods. |
94
|
|
|
""" |
95
|
|
|
|
96
|
|
|
file_url = models.CharField(verbose_name='Metadata url', |
97
|
|
|
max_length=1000, |
98
|
|
|
blank=True, null=True, |
99
|
|
|
help_text=_(u'Url to fetch metadata file')) |
100
|
|
|
file = models.FileField(upload_to='metadata', blank=True, null=True, |
101
|
|
|
verbose_name=_(u'metadata xml file'), |
102
|
|
|
help_text=_("if url is set, metadata url will be " |
103
|
|
|
"fetched and replace file value")) |
104
|
|
|
file_id = models.CharField(blank=True, null=True, max_length=500, |
105
|
|
|
verbose_name=_(u'File ID')) |
106
|
|
|
|
107
|
|
|
registration_authority = models.CharField(verbose_name=_('Registration Authority'), |
108
|
|
|
max_length=200, blank=True, null=True) |
109
|
|
|
|
110
|
|
|
editor_users = models.ManyToManyField(User, null=True, blank=True, |
111
|
|
|
verbose_name=_('editor users')) |
112
|
|
|
|
113
|
|
|
class Meta(object): |
114
|
|
|
abstract = True |
115
|
|
|
|
116
|
|
|
class XmlError(Exception): |
117
|
|
|
pass |
118
|
|
|
|
119
|
|
|
def __unicode__(self): |
120
|
|
|
return self.url or u"Metadata %s" % self.id |
121
|
|
|
|
122
|
|
|
def load_file(self): |
123
|
|
|
if not hasattr(self, '_loaded_file'): |
124
|
|
|
# Only load file and parse it, don't create/update any objects |
125
|
|
|
if not self.file: |
126
|
|
|
return None |
127
|
|
|
self._loaded_file = MetadataParser(filename=self.file.path) |
128
|
|
|
return self._loaded_file |
129
|
|
|
|
130
|
|
|
def _get_metadata_stream(self, load_streams): |
131
|
|
|
try: |
132
|
|
|
load = [] |
133
|
|
|
select = [] |
134
|
|
|
|
135
|
|
|
count = 1 |
136
|
|
|
for stream in load_streams: |
137
|
|
|
curid = "%s%d" % (self.slug, count) |
138
|
|
|
load.append("%s as %s" % (stream[0], curid)) |
139
|
|
|
if stream[1] == 'SP' or stream[1] == 'IDP': |
140
|
|
|
select.append( |
141
|
|
|
"%s!//md:EntityDescriptor[md:%sSSODescriptor]" % (curid, stream[1])) |
142
|
|
|
else: |
143
|
|
|
select.append("%s" % curid) |
144
|
|
|
count = count + 1 |
145
|
|
|
|
146
|
|
|
if len(select) > 0: |
147
|
|
|
pipeline = [{'load': load}, {'select': select}] |
148
|
|
|
else: |
149
|
|
|
pipeline = [{'load': load}, 'select'] |
150
|
|
|
|
151
|
|
|
md = MDRepository() |
152
|
|
|
entities = Plumbing(pipeline=pipeline, id=self.slug).process( |
153
|
|
|
md, state={'batch': True, 'stats': {}}) |
154
|
|
|
return etree.tostring(entities) |
155
|
|
|
except Exception, e: |
156
|
|
|
raise Exception( |
157
|
|
|
'Getting metadata from %s failed.\nError: %s' % (load_streams, e)) |
158
|
|
|
|
159
|
|
|
def fetch_metadata_file(self, file_name): |
160
|
|
|
file_url = self.file_url |
161
|
|
|
if not file_url or file_url == '': |
162
|
|
|
return |
163
|
|
|
|
164
|
|
|
metadata_files = [] |
165
|
|
|
files = file_url.split("|") |
166
|
|
|
for curfile in files: |
167
|
|
|
cursource = curfile.split(";") |
168
|
|
|
if len(cursource) == 1: |
169
|
|
|
cursource.append("All") |
170
|
|
|
metadata_files.append(cursource) |
171
|
|
|
|
172
|
|
|
req = self._get_metadata_stream(metadata_files) |
173
|
|
|
|
174
|
|
|
try: |
175
|
|
|
self.file.seek(0) |
176
|
|
|
original_file_content = self.file.read() |
177
|
|
|
if compare_filecontents(original_file_content, req): |
178
|
|
|
return False |
179
|
|
|
except Exception: |
180
|
|
|
pass |
181
|
|
|
|
182
|
|
|
filename = path.basename("%s-metadata.xml" % file_name) |
183
|
|
|
self.file.delete(save=False) |
184
|
|
|
self.file.save(filename, ContentFile(req), save=False) |
185
|
|
|
return True |
186
|
|
|
|
187
|
|
|
@classmethod |
188
|
|
|
def process_metadata(cls): |
189
|
|
|
""" |
190
|
|
|
Method that process the metadata file and updates attributes accordingly. |
191
|
|
|
""" |
192
|
|
|
raise NotImplementedError() |
193
|
|
|
|
194
|
|
|
|
195
|
|
|
class XmlDescriptionError(Exception): |
196
|
|
|
""" |
197
|
|
|
Class representing an error in the XML file. |
198
|
|
|
""" |
199
|
|
|
pass |
200
|
|
|
|
201
|
|
|
|
202
|
|
|
class Dummy(models.Model): |
203
|
|
|
""" |
204
|
|
|
Dummy object necessary to thest Django funcionalities. |
205
|
|
|
""" |
206
|
|
|
pass |
207
|
|
|
|