pyramid_raven integrates the raven python client and raven-js with a Pyramid web application. It provides a configured raven client at request.raven and a pyramid_layout panel called raven-js.
Raven has built in support for Pyramid applications via a Paste filter and logging handler. pyramid_raven is an alternative integration, useful when:
you handle exceptions within your application – i.e.: you register a catch all Exception view that renders an error page within your application, thus preventing errors reaching the WSGI pipeline
you also want to log javascript errors
Setup
Install using pip:
pip install pyramid_raven
Configure Raven DSN address in the INI configuration of your application:
raven.dsn = https://xxx:yyy@sentry.example.com/1
… or provide a SENTRY_DSN environment variable:
SENTRY_DSN=http://public:secret@example.com/1
Configure your application to include the package:
config.include('pyramid_raven')
Configure
Any raven.* namespaced settings in your .ini configuration file will be passed to the raven client constructor – although it’s your responsibility to coerce them to the right type, e.g.:
raven.timeout=3
You can augment the (already fairly comprehensive set of) request attributes sent with the logged server side exceptions by adding space seperated request property or method names (called with no args) to:
pyramid_raven.additional_request_properties=foo bar pyramid_raven.additional_request_methods=baz bam
You can also override the panel template:
pyramid_raven.panel_tmpl=mypkg:templates/foo.tmpl
Use
You can use it to record server side errors:
@view_config(context=Exception)
def system_error_view(context, request):
"""Example catch all exception handler."""
# Notify sentry.
request.raven.captureException()
# XXX E.g.: render error page.
# ...
Client side JavaScript errors
Client side error tracking requires that you have pyramid_layout configured before pyramid_raven.
And client side errors:
<!-- in production, in the head of your main layout, above all other js -->
${panel('raven-js')}
E.g.: then in any of your subsequently loaded scripts:
throw new Error('This javascript error will be logged.')
Note that if you load your scripts from an external domain (e.g.: from a CDN) then errors will not be logged by many browsers, due to cross origin security (not leaking information across domains). You can workaround this using CORS but browser support is, at the time of writing, wobbly to say the least.
Tests
To run the tests, pip install nose coverage mock and e.g.:
$ nosetests --with-doctest --with-coverage --cover-tests --cover-package pyramid_raven pyramid_raven ...... Name Stmts Miss Cover Missing ------------------------------------------------------- pyramid_raven 9 0 100% pyramid_raven.client 69 0 100% pyramid_raven.constants 5 0 100% pyramid_raven.panel 39 0 100% ------------------------------------------------------- TOTAL 122 0 100% ---------------------------------------------------------------------- Ran 6 tests in 0.143s OK
Metadata
Release files for pyramid_raven 0.2.2
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| pyramid_raven-0.2.2.tar.gz | 8.5 kB | Details |
Release files / pyramid_raven-0.2.2.tar.gz
| Download URL | pyramid_raven-0.2.2.tar.gz |
|---|---|
| Size | 8.5 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
9355728207d436979bd42db12c9b5cddb3c053afda61e87abb054275e5d249db
|
|
BLAKE2b-256 checksum How to use checksums |
ef2895bd5200aa4fd8cb667b98f315e68af732afc0b4e0ccf0b7a72712e6d943
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |