Matthew Tretter
80008aee11
Merge pull request #215 from saulshanabrook/patch-5
...
pass in upscale keyword to thumbnail template tag
2013-05-20 16:09:05 -07:00
Matthew Tretter
32522114db
Improve default cache backend handling
2013-05-13 21:28:41 -04:00
Saul Shanabrook
6023e9216a
pass in upscale keyword to thumbnail template tag
2013-05-11 09:12:29 -03:00
Matthew Tretter
34739819aa
Fix None checks
...
Files can be falsy
2013-05-10 04:43:01 -04:00
Matthew Tretter
c89b18aa95
Split before_access into two signals
...
Differentiating between when the generated file content is required and
when the generated file is just required to exist gives us more
flexibility with strategies.
2013-05-10 04:39:46 -04:00
Matthew Tretter
6db082bca2
Reset content pointer
2013-05-10 03:08:43 -04:00
Matthew Tretter
906fbbd463
Reduce calls to backend
...
Reading from an `ImageCacheFile`, will result in accessing its `file`
attribute repeatedly which would result in the `before_access` signal
being dispatched, which in turn would result in many unnecessary calls
to the image cache backend. With this change, we don't send
`before_access` if the file has already been created.
Similarly, we don't need to try to generate the image if we know for
certain that it's already been generated (because we have a reference
to it).
2013-05-10 02:56:44 -04:00
Matthew Tretter
bc49f9cf8b
Store the file after we generate it
...
This means reading ImageCacheFiles won't result in an additional storage
operation.
2013-05-10 02:42:28 -04:00
Matthew Tretter
06dd238993
Store cachefile strategy on file; not generator
...
The file can appeal to the generator for the value, but it shouldn't
require it; it just does that as a convenience.
2013-05-10 00:51:47 -04:00
Matthew Tretter
d5d5fc0550
Add cachefile backend fallback
...
The generator shouldn't have to implement `cachefile_backend`
2013-05-10 00:26:46 -04:00
Matthew Tretter
4efa05099d
Use pilkit's process_image utility
2013-05-09 22:02:03 -04:00
Matthew Tretter
d94c4bb0a9
Merge branch 'canon-pickle' into develop
2013-04-30 20:50:29 -04:00
Matthew Tretter
f173861b53
Fix invalid char range
2013-04-30 09:32:47 -04:00
Matthew Tretter
c704db2da5
Sanitize cache keys for memcached
...
Closes #208
2013-04-30 09:32:47 -04:00
Matthew Tretter
f5b171979b
Canonicalize sets and dicts before hashing
2013-04-25 02:30:37 -04:00
Matthew Tretter
3e019f5dac
Merge pull request #204 from danxshap/close_file_fix
...
Re-open source file on ValueError in ImageSpec.generate()
2013-04-23 10:59:14 -07:00
Matthew Tretter
0a98eb0e79
Use DEFAULT_FILE_STORAGE by default
...
As raised in #203 , if `IMAGEKIT_DEFAULT_FILE_STORAGE` isn't set,
`DEFAULT_FILE_STORAGE` should be used.
2013-04-22 09:05:49 -04:00
danxshap
13c92db760
Re-open source file on ValueError in ImageSpec.generate()
2013-04-21 16:06:13 -04:00
Eric Brelsford
7946fe54b5
processors in PILKit: module
...
Help those looking in the code find processors.
2013-04-10 14:24:02 -04:00
Matthew Tretter
261f164127
Bump version to 3.0b1
2013-04-05 16:43:26 -04:00
Matthew Tretter
c4431fe296
Don't require generator to unregister
2013-04-05 16:41:53 -04:00
Matthew Tretter
6f9f99e86c
Fields shouldn't cause AlreadyRegistered exceptions
2013-04-05 15:28:36 -04:00
Matthew Tretter
091b2137d0
Always call set_spec_id
2013-04-05 14:57:41 -04:00
Matthew Tretter
41ab2c0fa3
Use different name for different signature
2013-04-05 14:57:11 -04:00
Matthew Tretter
dafebc9a4d
Don't call register without spec
2013-04-05 14:38:29 -04:00
Matthew Tretter
9891314b8e
Don't error if same generator is registered
2013-04-05 12:25:02 -04:00
Matthew Tretter
681527fa9d
Remove implicit source logic from descriptor
...
This needs to be known earlier, when registering the source group, so
we do it in `contribute_to_class` instead.
Closes #188 .
2013-04-03 00:21:04 -04:00
Matthew Tretter
2b10e85813
Clarify "force" kwarg
2013-04-02 22:55:44 -04:00
Matthew Tretter
3ca8c53698
Also defer existance check
2013-04-02 22:37:52 -04:00
Matthew Tretter
728368abf6
Add async cachefile backend
2013-04-02 22:29:53 -04:00
Matthew Tretter
86cd23e906
Remove actions module
2013-03-15 01:06:28 -04:00
Matthew Tretter
3d6a9de9aa
Add existence check timeout
2013-03-15 01:05:53 -04:00
Matthew Tretter
1e129c5b70
Convert to boolean
2013-03-15 00:49:24 -04:00
Matthew Tretter
70ff6dc788
Store state when force-generated
2013-03-15 00:30:58 -04:00
Matthew Tretter
c48c720f8a
Add __nonzero__ method that will work for async
2013-03-15 00:27:23 -04:00
Matthew Tretter
fc87c0497c
Add abstract cachefile backend for illustration
2013-03-15 00:20:15 -04:00
Matthew Tretter
54dda25adb
Generation goes through backend; use states
...
We might as well account for asynchronous generation in our
CachedFileBackend.
2013-03-15 00:19:19 -04:00
Matthew Tretter
ac92b77709
Fix loopy existence check
...
Sometimes two wrongs do make a right I guess. But better to have two
rights.
2013-03-14 23:04:30 -04:00
Matthew Tretter
f181d30008
Rename file_exists to exists
2013-03-14 22:58:28 -04:00
Matthew Tretter
044a3625f9
Correct error message
2013-03-14 22:55:40 -04:00
Matthew Tretter
c64eee40b3
Bump version to 3.0a5
2013-02-25 22:43:37 -05:00
Matthew Tretter
92a3c2688c
Error when attempting to generate image w/o source
2013-02-25 22:42:47 -05:00
Matthew Tretter
61aa1c32e7
Bump version to 3.0a4
2013-02-25 22:22:27 -05:00
Matthew Tretter
b33869a9f8
Sourceless specs are falsy
...
Passes test added in 190153d#187
2013-02-25 22:18:33 -05:00
Matthew Tretter
dff0182310
Remove another old util
2013-02-12 21:40:15 -05:00
Matthew Tretter
f5a078b68f
Remove old util
2013-02-12 21:40:15 -05:00
Matthew Tretter
5c944c5efb
Bump version to 3.0a3
2013-02-10 16:05:21 -05:00
Matthew Tretter
b53b7c3cae
Clarify ImageFieldSourceGroup docstrings
2013-02-10 16:01:50 -05:00
Matthew Tretter
90ae9e8339
Fix get_nonabstract_descendants
...
Previously, function was only returning the first non-abstract
descendant. (It should also return concrete descendants of concrete
descendants, which it now does.)
Also, converted to generator.
2013-02-10 15:58:22 -05:00
Matthew Tretter
df41459e65
Fix signals for abstract models
...
Includes a fix for undispatched signals, as well as signals being
handled twice.
A regression of #126
Related: #185
2013-02-10 15:42:10 -05:00
