1 | """General helper functions for WAeUP. |
---|
2 | """ |
---|
3 | import os |
---|
4 | import re |
---|
5 | import sys |
---|
6 | import shutil |
---|
7 | import grok |
---|
8 | from cStringIO import StringIO |
---|
9 | from docutils.core import publish_string |
---|
10 | from zope.component.interfaces import IFactory |
---|
11 | from zope.interface import implementedBy |
---|
12 | from zope.schema import getFieldNames |
---|
13 | from zope.schema.fieldproperty import FieldProperty |
---|
14 | |
---|
15 | def removeFileOrDirectory(filepath): |
---|
16 | """Remove a file or directory. |
---|
17 | |
---|
18 | Different to :func:`shutil.rmtree` we also accept not existing |
---|
19 | paths (returning silently) and if a dir turns out to be a regular |
---|
20 | file, we remove that. |
---|
21 | """ |
---|
22 | filepath = os.path.abspath(filepath) |
---|
23 | if not os.path.exists(filepath): |
---|
24 | return |
---|
25 | if os.path.isdir(filepath): |
---|
26 | shutil.rmtree(filepath) |
---|
27 | else: |
---|
28 | os.unlink(filepath) |
---|
29 | return |
---|
30 | |
---|
31 | def copyFileSystemTree(src, dst, overwrite=False, del_old=False): |
---|
32 | """Copy contents of directory src to directory dst. |
---|
33 | |
---|
34 | Both directories must exists. |
---|
35 | |
---|
36 | If `overwrite` is true, any same named objects will be |
---|
37 | overwritten. Otherwise these files will not be touched. |
---|
38 | |
---|
39 | If `del_old` is true, copied files and directories will be removed |
---|
40 | from the src directory. |
---|
41 | |
---|
42 | This functions returns a list of non-copied files. |
---|
43 | |
---|
44 | Unix hidden files and directories (starting with '.') are not |
---|
45 | processed by this function. |
---|
46 | """ |
---|
47 | if not os.path.exists(src): |
---|
48 | raise ValueError('source path does not exist: %s' % src) |
---|
49 | if not os.path.exists(dst): |
---|
50 | raise ValueError('destination path does not exist: %s' % dst) |
---|
51 | if not os.path.isdir(src): |
---|
52 | raise ValueError('source path is not a directory: %s' % src) |
---|
53 | if not os.path.isdir(dst): |
---|
54 | raise ValueError('destination path is not a directory: %s' % dst) |
---|
55 | not_copied = [] |
---|
56 | for item in os.listdir(src): |
---|
57 | if item.startswith('.'): |
---|
58 | continue # We do not copy hidden stuff... |
---|
59 | itemsrc = os.path.join(src, item) |
---|
60 | itemdst = os.path.join(dst, item) |
---|
61 | |
---|
62 | if os.path.exists(itemdst): |
---|
63 | if overwrite is True: |
---|
64 | removeFileOrDirectory(itemdst) |
---|
65 | else: |
---|
66 | not_copied.append(item) |
---|
67 | continue |
---|
68 | |
---|
69 | if os.path.isdir(itemsrc): |
---|
70 | shutil.copytree(itemsrc, itemdst) |
---|
71 | else: |
---|
72 | shutil.copy2(itemsrc, itemdst) |
---|
73 | if del_old: |
---|
74 | removeFileOrDirectory(itemsrc) |
---|
75 | return not_copied |
---|
76 | |
---|
77 | |
---|
78 | def getInnerHTMLPart(html_code): |
---|
79 | """Return the 'inner' part of a complete HTML snippet. |
---|
80 | |
---|
81 | If there is a form part, get this. |
---|
82 | |
---|
83 | If there is no form part, try to return the body part contents. |
---|
84 | |
---|
85 | If there is no body, return as-is. |
---|
86 | |
---|
87 | Let's see how that works. If we deliver some doc with form, we |
---|
88 | will get that form only: |
---|
89 | |
---|
90 | >>> doc = '<html><form>My Form</form>Outside the form</html>' |
---|
91 | >>> getInnerHTMLPart(doc) |
---|
92 | '<form>My Form</form>' |
---|
93 | |
---|
94 | No form? Then seek for a body part and get the contents: |
---|
95 | |
---|
96 | >>> doc = '<html><body>My Body</body>Trailing Trash</html>' |
---|
97 | >>> getInnerHTMLPart(doc) |
---|
98 | 'My Body' |
---|
99 | |
---|
100 | If none of these is included, return what we got: |
---|
101 | |
---|
102 | >>> doc = '<html>without body nor form</html>' |
---|
103 | >>> getInnerHTMLPart(doc) |
---|
104 | '<html>without body nor form</html>' |
---|
105 | |
---|
106 | """ |
---|
107 | |
---|
108 | try: |
---|
109 | result = re.match('^.+(<form[^\>]*>.*</form>).+$', html_code, |
---|
110 | re.DOTALL).groups()[0] |
---|
111 | return result |
---|
112 | except AttributeError: |
---|
113 | # No <form> part included |
---|
114 | try: |
---|
115 | result = re.match('^.+<body[^\>]*>(.*)</body>.*$', html_code, |
---|
116 | re.DOTALL).groups()[0] |
---|
117 | return result |
---|
118 | except AttributeError: |
---|
119 | # No <form> and no <body> tag... |
---|
120 | pass |
---|
121 | return html_code |
---|
122 | |
---|
123 | class FactoryBase(grok.GlobalUtility): |
---|
124 | """A factory for things. |
---|
125 | |
---|
126 | This is a baseclass for easier creation of factories. Factories |
---|
127 | are utilities that are registered under a certain name and return |
---|
128 | instances of certain classes when called. |
---|
129 | |
---|
130 | In :mod:`waeup.sirp` we use factories extensively for |
---|
131 | batching. While processing a batch some importer looks up a |
---|
132 | factory to create real-world instances that then get filled with |
---|
133 | data from imported CSV files. |
---|
134 | |
---|
135 | To get rid of reimplementing the same stuff over and over again, |
---|
136 | most notably the methods defined here, we offer this base class |
---|
137 | (which will *not* be registered as a factory itself). |
---|
138 | |
---|
139 | Real factories can then be created like this: |
---|
140 | |
---|
141 | >>> import grok |
---|
142 | >>> from waeup.sirp.utils.helpers import FactoryBase |
---|
143 | >>> class MyObject(object): |
---|
144 | ... # Some class we want to get instances of. |
---|
145 | ... pass |
---|
146 | >>> class MyObjectFactory(FactoryBase): |
---|
147 | ... # This is the factory for MyObject instances |
---|
148 | ... grok.name(u'waeup.sirp.factory.MyObject') |
---|
149 | ... factory = MyObject |
---|
150 | |
---|
151 | That's it. It is essential to set the ``factory`` attribute, which |
---|
152 | will determine the class of which instances should be created when |
---|
153 | called. The given name must even be unique amongst all utilities |
---|
154 | registered during runtime. While you can pick any name you like |
---|
155 | you might want to prepend ``waeup.sirp.factory.`` to the name |
---|
156 | string to make sure it does not clash with names of other |
---|
157 | utilities one day. |
---|
158 | |
---|
159 | Before all this works we have to grok the baseclass once and our |
---|
160 | freshly defined factory. This executes all the component |
---|
161 | registration stuff we don't want to do ourselves. In daily use |
---|
162 | this is done automatically on startup of a :mod:`waeup.sirp` |
---|
163 | system. |
---|
164 | |
---|
165 | >>> grok.testing.grok('waeup.sirp.utils.helpers') |
---|
166 | >>> grok.testing.grok_component( |
---|
167 | ... 'MyObjectFactory', MyObjectFactory |
---|
168 | ... ) |
---|
169 | True |
---|
170 | |
---|
171 | After grokking we (and importers) can create objects without |
---|
172 | knowing about the location of the real class definition, just by |
---|
173 | the factory name: |
---|
174 | |
---|
175 | >>> from zope.component import createObject |
---|
176 | >>> obj = createObject('waeup.sirp.factory.MyObject') |
---|
177 | >>> isinstance(obj, MyObject) |
---|
178 | True |
---|
179 | |
---|
180 | We can also use the regular utility lookups to find our new |
---|
181 | factory: |
---|
182 | |
---|
183 | >>> from zope.component import getUtility |
---|
184 | >>> from zope.component.interfaces import IFactory |
---|
185 | >>> factory = getUtility( |
---|
186 | ... IFactory, name='waeup.sirp.factory.MyObject' |
---|
187 | ... ) |
---|
188 | >>> isinstance(factory, MyObjectFactory) |
---|
189 | True |
---|
190 | |
---|
191 | And this factory generates `MyObject` instances: |
---|
192 | |
---|
193 | >>> obj = factory() |
---|
194 | >>> isinstance(obj, MyObject) |
---|
195 | True |
---|
196 | |
---|
197 | """ |
---|
198 | grok.baseclass() # Do not grok this class, do not register us. |
---|
199 | grok.implements(IFactory) |
---|
200 | # You can override any of the following attributes in derived |
---|
201 | # classes. The `grok.name` setting *must* even be set to some |
---|
202 | # unique value. |
---|
203 | grok.name(u'waeup.Factory') |
---|
204 | title = u"Create instances of ``factory``.", |
---|
205 | description = u"This factory instantiates new applicant instances." |
---|
206 | factory = None |
---|
207 | |
---|
208 | def __call__(self, *args, **kw): |
---|
209 | """The main factory function. |
---|
210 | |
---|
211 | Returns an instance of the requested object. |
---|
212 | """ |
---|
213 | return self.factory() |
---|
214 | |
---|
215 | def getInterfaces(self): |
---|
216 | # Required by IFactory |
---|
217 | return implementedBy(self.factory) |
---|
218 | |
---|
219 | def ReST2HTML_w_warnings(source_string): |
---|
220 | """Convert a reStructuredText string to HTML preserving warnings. |
---|
221 | |
---|
222 | Returns a tuple ``(<HTML_CODE>, <WARNINGS>)``, both being |
---|
223 | strings. Where ``<HTML_CODE>`` is the HTML code generated from the |
---|
224 | source string (in unicode), ``<WARNINGS>`` is a string containing |
---|
225 | any warning messages or ``None``. |
---|
226 | |
---|
227 | Regular multi-line ReStructuredText strings will be returned as |
---|
228 | HTML code: |
---|
229 | |
---|
230 | >>> from waeup.sirp.utils.helpers import ReST2HTML |
---|
231 | >>> source = ''' |
---|
232 | ... Headline |
---|
233 | ... ======== |
---|
234 | ... |
---|
235 | ... - A list item |
---|
236 | ... - Another item |
---|
237 | ... |
---|
238 | ... Thanks for watching! |
---|
239 | ... ''' |
---|
240 | >>> html, warnings = ReST2HTML_w_warnings(source) |
---|
241 | >>> print html |
---|
242 | <div class="document" id="headline"> |
---|
243 | <h1 class="title">Headline</h1> |
---|
244 | <BLANKLINE> |
---|
245 | <ul class="simple"> |
---|
246 | <li>A list item</li> |
---|
247 | <li>Another item</li> |
---|
248 | </ul> |
---|
249 | <p>Thanks for watching!</p> |
---|
250 | </div> |
---|
251 | |
---|
252 | Here no warnings happened, so the `warnings` are ``None``: |
---|
253 | |
---|
254 | >>> warnings is None |
---|
255 | True |
---|
256 | |
---|
257 | If warnings happen then they can be retrieved in the returned |
---|
258 | ``warnings``. We try to render an erraneous document: |
---|
259 | |
---|
260 | >>> source = ''' |
---|
261 | ... Headline |
---|
262 | ... ====== |
---|
263 | ... |
---|
264 | ... Thanks for watching! |
---|
265 | ... ''' |
---|
266 | >>> html, warnings = ReST2HTML_w_warnings(source) |
---|
267 | >>> print html |
---|
268 | <div class="document" id="headline"> |
---|
269 | <h1 class="title">Headline</h1> |
---|
270 | <BLANKLINE> |
---|
271 | <p>Thanks for watching!</p> |
---|
272 | </div> |
---|
273 | |
---|
274 | >>> print warnings |
---|
275 | <string>:3: (WARNING/2) Title underline too short. |
---|
276 | <BLANKLINE> |
---|
277 | Headline |
---|
278 | ====== |
---|
279 | <BLANKLINE> |
---|
280 | |
---|
281 | As you can see, the warnings are not displayed inline the document |
---|
282 | but can be retrieved from the returned warnings, which is a string |
---|
283 | or ``None``. |
---|
284 | """ |
---|
285 | warnings = StringIO() |
---|
286 | fulldoc = publish_string( |
---|
287 | source_string, writer_name='html4css1', |
---|
288 | settings_overrides={ |
---|
289 | 'report_level': 0, |
---|
290 | 'warning_stream': warnings, |
---|
291 | }) |
---|
292 | warnings.seek(0) |
---|
293 | warning_msgs = warnings.read() |
---|
294 | if warning_msgs: |
---|
295 | # Render again, this time with no warnings inline... |
---|
296 | fulldoc = publish_string( |
---|
297 | source_string, writer_name='html4css1', |
---|
298 | settings_overrides={ |
---|
299 | 'report_level': 10000, |
---|
300 | 'halt_level': 10000, |
---|
301 | 'warning_stream': warnings, |
---|
302 | }) |
---|
303 | if warning_msgs == '': |
---|
304 | warning_msgs = None |
---|
305 | result = getInnerHTMLPart(fulldoc).strip() |
---|
306 | if not isinstance(result, unicode): |
---|
307 | result = result.decode('utf-8') |
---|
308 | return result, warning_msgs |
---|
309 | |
---|
310 | def ReST2HTML(source_string): |
---|
311 | """Render a string containing ReStructuredText to HTML. |
---|
312 | |
---|
313 | Any warnings about too short headings, etc. are silently |
---|
314 | discarded. Use :func:`ReST2HTML_w_warnings` if you want to get any |
---|
315 | warnings. |
---|
316 | |
---|
317 | The returned string will be unicode. |
---|
318 | |
---|
319 | A regular document will be rendered like this: |
---|
320 | |
---|
321 | >>> source = ''' |
---|
322 | ... Headline |
---|
323 | ... ======== |
---|
324 | ... |
---|
325 | ... Thanks for watching! |
---|
326 | ... ''' |
---|
327 | >>> html = ReST2HTML(source) |
---|
328 | >>> print html |
---|
329 | <div class="document" id="headline"> |
---|
330 | <h1 class="title">Headline</h1> |
---|
331 | <BLANKLINE> |
---|
332 | <p>Thanks for watching!</p> |
---|
333 | </div> |
---|
334 | |
---|
335 | A document with markup problems (here: the underline is too short) |
---|
336 | will look similar: |
---|
337 | |
---|
338 | >>> source = ''' |
---|
339 | ... Headline |
---|
340 | ... ====== |
---|
341 | ... |
---|
342 | ... Thanks for watching! |
---|
343 | ... ''' |
---|
344 | >>> html = ReST2HTML(source) |
---|
345 | >>> print html |
---|
346 | <div class="document" id="headline"> |
---|
347 | <h1 class="title">Headline</h1> |
---|
348 | <BLANKLINE> |
---|
349 | <p>Thanks for watching!</p> |
---|
350 | </div> |
---|
351 | |
---|
352 | """ |
---|
353 | html, warnings = ReST2HTML_w_warnings(source_string) |
---|
354 | return html |
---|
355 | |
---|
356 | def attrs_to_fields(cls): |
---|
357 | """Turn the attributes of a class into FieldProperty instances. |
---|
358 | """ |
---|
359 | iface = list(implementedBy(cls))[0] |
---|
360 | for field_name in getFieldNames(iface): |
---|
361 | setattr(cls, field_name, FieldProperty(iface[field_name])) |
---|
362 | return cls |
---|