| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | Made Registries work in EvalException (prepare, register, replace, cleanup) | pjenvey | 2006-12-07 | 1 | -53/+92 |
| | | | | | | | | | o Registries get their state restored from the restorer when EvalException begins executing, and their state changes are maintained by the restorer o StackedObjectProxy's _push/pop_object will no-op in that context o made get_saved_proxied_obj handle proxies having their objects de-registered within EvalException | ||||
| * | Fixed issue with pushing obj twice. | bbangert | 2006-12-04 | 1 | -1/+0 |
| | | |||||
| * | Added replace method to registry for replacing registered objects with a new ↵ | bbangert | 2006-12-04 | 1 | -0/+13 |
| | | | | | one. | ||||
| * | small refactor | pjenvey | 2006-12-02 | 1 | -2/+2 |
| | | |||||
| * | StackedObjectRestorer tests. fixed a couple edge cases: | pjenvey | 2006-12-01 | 1 | -20/+30 |
| | | | | | | | | | | o get_saved_proxied_obj needed to delve down to allow access to older (overlaid) RegistryManagers' objects (test_restorer_nested_middleman) o save_registry_state needed to save all state, so restoration can get to every overlaid RegistryManagers' objects (test_restorer_middlemen_nested_evalexception). cleaned it up to do all the saving at once, instead of piecemeal | ||||
| * | StackedObjectProxies now work within the interactive debugger. When | pjenvey | 2006-12-01 | 1 | -1/+136 |
| | | | | | | | | | | | | EvalException is enabled and RegistryManager and or EvalException detect an unexpected exception, they'll call StackedObjectRestorer to: o save the Registry state o change currently registered StackedObjectProxies to detect when they're being used within the interactive debugger via a threadlocal (resulting in a small performance hit). They'll get their proxied object from the restorer when they detect the presence of the debugger, otherwise they'll continue to work as normal | ||||
| * | use repr instead of __repr__ directly | pjenvey | 2006-11-30 | 1 | -1/+1 |
| | | |||||
| * | A big commit, primarily aesthetic/whitespace in nature. This is the result ↵ | ianb | 2006-10-20 | 1 | -1/+0 |
| | | | | | of running pylint over the codebase. Some minor/hard-to-reach typos were also picked up. | ||||
| * | Adding code-block directive to Python code samples. | bbangert | 2006-09-02 | 1 | -3/+9 |
| | | |||||
| * | exception message cleanup | pjenvey | 2006-08-18 | 1 | -6/+5 |
| | | |||||
| * | moved/deprecated StackedObjectProxy's function names to be preceded | pjenvey | 2006-08-16 | 1 | -19/+45 |
| | | | | | with underscores, so they're less likely to clash with the contained object's attributes | ||||
| * | RegistryManager was close()ing the app_iter, when it needs to simply ↵ | pjenvey | 2006-08-11 | 1 | -2/+0 |
| | | | | | propagate it up the chain | ||||
| * | Fixed issue with attempting to clean-up the Registry environment twice under ↵ | bbangert | 2006-08-10 | 1 | -8/+7 |
| | | | | | certain exception tossing conditions. | ||||
| * | Fixed issue with Registry still referring to the wrong place if an app down ↵ | bbangert | 2006-08-09 | 1 | -11/+7 |
| | | | | | the chain uses Registry and returns an iter. Apps returning iter's should hold their own ref to any objects that used a StackedObjectProxy. | ||||
| * | having StackedObjectProxy.__repr__ avoid raising an exception when it | pjenvey | 2006-08-03 | 1 | -1/+6 |
| | | | | | does not contain an internal object | ||||
| * | fix docstring | ianb | 2006-04-24 | 1 | -1/+1 |
| | | |||||
| * | More docs, passing __len__ into object | bbangert | 2006-03-27 | 1 | -0/+39 |
| | | |||||
| * | Added naming capability to StackedObject for better debugging when a object ↵ | bbangert | 2006-03-19 | 1 | -2/+4 |
| | | | | | isn't registered | ||||
| * | Fixed bug in setattr for stacked object | bbangert | 2006-03-19 | 1 | -1/+1 |
| | | |||||
| * | Added unit tests to ensure that TypeError is thrown if no default is ↵ | bbangert | 2006-03-12 | 1 | -5/+21 |
| | | | | | | | provided and no object is pushed to the Proxy. Added ability to define a default object for a StackedObjectProxy | ||||
| * | Adding copyright/license info | bbangert | 2006-03-10 | 1 | -0/+4 |
| | | |||||
| * | Doc change to make it clear that paste.registry is a Registry instance | bbangert | 2006-03-10 | 1 | -1/+1 |
| | | |||||
| * | Updating docs | bbangert | 2006-03-10 | 1 | -48/+90 |
| | | |||||
| * | Didn't need custom app_iter, as my unit test wasn't following WSGI spec by ↵ | bbangert | 2006-03-10 | 1 | -17/+2 |
| | | | | | calling close on the response. Unit tests for registry all passing | ||||
| * | Fixes iterable issue by using new IterWrap, likely need to break it into ↵ | bbangert | 2006-03-10 | 1 | -1/+15 |
| | | | | | wsgilib still, added test case for iterable app | ||||
| * | Adding registry unit tests, fixed registry to not use weakref's so any ↵ | bbangert | 2006-03-10 | 1 | -14/+3 |
| | | | | | object can be proxied (including dicts) | ||||
| * | Initial creation of registry and stacked object proxy, hasn't been tested ↵ | bbangert | 2006-03-10 | 1 | -0/+197 |
| yet but I'm feeling lucky | |||||
 |
index : delta/python-packages/paste.git | |
| bitbucket.org: Obsolete (use python-packages/paste-git) |
| summaryrefslogtreecommitdiff |