March 2017 - cdi-dev - Jboss List Archives

[JBoss JIRA] (CDI-690) Automatically start a request context for async observer methods

by John Ament (JIRA)

[ https://issues.jboss.org/browse/CDI-690?page=com.atlassian.jira.plugin.sy... ] John Ament commented on CDI-690: -------------------------------- Yes, I raised this for the SE use case. Martin is correct, this already happens in the RI for SE. > Automatically start a request context for async observer methods > ---------------------------------------------------------------- > > Key: CDI-690 > URL: https://issues.jboss.org/browse/CDI-690 > Project: CDI Specification Issues > Issue Type: Feature Request > Reporter: John Ament > > In many other places we start an implicit request context for method invocations (post constructs for instance). We should have an implicit request context for async observer methods. -- This message was sent by Atlassian JIRA (v7.2.3#72005)

7 years, 9 months

1
0
0 / 0

[JBoss JIRA] (CDI-692) isResolvable missing from 5.6.1

by John Ament (JIRA)

[ https://issues.jboss.org/browse/CDI-692?page=com.atlassian.jira.plugin.sy... ] John Ament commented on CDI-692: -------------------------------- Did you point me to PRD because the line is supposed to be there? Or some other reason? > isResolvable missing from 5.6.1 > ------------------------------- > > Key: CDI-692 > URL: https://issues.jboss.org/browse/CDI-692 > Project: CDI Specification Issues > Issue Type: Bug > Reporter: John Ament > > Section 5.6.1 is missing the new isResolvable method http://docs.jboss.org/cdi/spec/2.0.Beta1/cdi-spec.html#dynamic_lookup -- This message was sent by Atlassian JIRA (v7.2.3#72005)

7 years, 9 months

1
0
0 / 0

[JBoss JIRA] (CDI-692) isResolvable missing from 5.6.1

by Antoine Sabot-Durand (JIRA)

[ https://issues.jboss.org/browse/CDI-692?page=com.atlassian.jira.plugin.sy... ] Antoine Sabot-Durand commented on CDI-692: ------------------------------------------ Sorry for that, bad regression. It's now ok. Note that the correct URL is http://docs.jboss.org/cdi/spec/2.0-PRD/cdi-spec.html since we are in public review. > isResolvable missing from 5.6.1 > ------------------------------- > > Key: CDI-692 > URL: https://issues.jboss.org/browse/CDI-692 > Project: CDI Specification Issues > Issue Type: Bug > Reporter: John Ament > > Section 5.6.1 is missing the new isResolvable method http://docs.jboss.org/cdi/spec/2.0.Beta1/cdi-spec.html#dynamic_lookup -- This message was sent by Atlassian JIRA (v7.2.3#72005)

7 years, 9 months

1
0
0 / 0

[JBoss JIRA] (CDI-689) Typo or unclear message in 24.1.2

by Antoine Sabot-Durand (JIRA)

[ https://issues.jboss.org/browse/CDI-689?page=com.atlassian.jira.plugin.sy... ] Antoine Sabot-Durand updated CDI-689: ------------------------------------- Fix Version/s: 2.0 .Final > Typo or unclear message in 24.1.2 > --------------------------------- > > Key: CDI-689 > URL: https://issues.jboss.org/browse/CDI-689 > Project: CDI Specification Issues > Issue Type: Bug > Reporter: John Ament > Priority: Minor > Fix For: 2.0 .Final > > > Section 24.1.2 includes the following text: > {quote} > When Running in Java EE, the container must extend the rules defined in Observer method invocation context and must also ensure that > All kinds of observers are called in the same client security context as the invocation of Event.fire() or Event.fireAsync() or BeanManager.fireEvent(). > {quote} > The break after "that" in the first line seems weird, its not clear if its supposed to continue into the second line, or if the sentence is intended to continue. -- This message was sent by Atlassian JIRA (v7.2.3#72005)

7 years, 9 months

1
0
0 / 0

[JBoss JIRA] (CDI-688) Contextual bean stores

by Antoine Sabot-Durand (JIRA)

[ https://issues.jboss.org/browse/CDI-688?page=com.atlassian.jira.plugin.sy... ] Antoine Sabot-Durand commented on CDI-688: ------------------------------------------ +1 to [~mkouba] comment. Right now, nothing prevents you to create a custom scope with your own store. > Contextual bean stores > ---------------------- > > Key: CDI-688 > URL: https://issues.jboss.org/browse/CDI-688 > Project: CDI Specification Issues > Issue Type: Feature Request > Reporter: John Ament > > Right now, the notion of how beans are stored are ambiguous. The implementations back them by maps. This may work well, but there's some amount of plugability required to make passivation capable beans work right. A single request makes sense to store in a map. Sessions are stored in the HTTPSession. We should allow these bean stores to be more pluggable, allowing users or implementations to override how their beans are stored. -- This message was sent by Atlassian JIRA (v7.2.3#72005)

7 years, 9 months

1
0
0 / 0

[JBoss JIRA] (CDI-687) Programmatic activation of an existing conversation by conversation id

by Martin Kouba (JIRA)

[ https://issues.jboss.org/browse/CDI-687?page=com.atlassian.jira.plugin.sy... ] Martin Kouba commented on CDI-687: ---------------------------------- Do you mean to allow to change the way a long-running conversation id is extracted from an HTTP request (e.g. to use an HTTP request header) or to "detach" the long-running conversations from a ServletRequest? > Programmatic activation of an existing conversation by conversation id > ---------------------------------------------------------------------- > > Key: CDI-687 > URL: https://issues.jboss.org/browse/CDI-687 > Project: CDI Specification Issues > Issue Type: Feature Request > Reporter: John Ament > > Currently, conversation activation can only work if you have an HTTP request with cid query param passed in. We should expand this to allow any arbitrary lookup of conversation by id. -- This message was sent by Atlassian JIRA (v7.2.3#72005)

7 years, 9 months

1
0
0 / 0

[JBoss JIRA] (CDI-693) Introduce a Instance#ifResolved(Consumer<T>) method

by Martin Kouba (JIRA)

[ https://issues.jboss.org/browse/CDI-693?page=com.atlassian.jira.plugin.sy... ] Martin Kouba updated CDI-693: ----------------------------- Fix Version/s: 2.1 (Discussion) > Introduce a Instance#ifResolved(Consumer<T>) method > --------------------------------------------------- > > Key: CDI-693 > URL: https://issues.jboss.org/browse/CDI-693 > Project: CDI Specification Issues > Issue Type: Feature Request > Reporter: John Ament > Fix For: 2.1 (Discussion) > > > in many cases, the logic around isResolvable/isUnsatisfied/isAmbiguous could be replaced by a consumer of the resolved instance. It would be good to introduce a ifResolved(Consumer<T>) method that consumes the resolved instance, not invoking the consumer if the instance was not resolvable. -- This message was sent by Atlassian JIRA (v7.2.3#72005)

7 years, 9 months

1
0
0 / 0

[JBoss JIRA] (CDI-688) Contextual bean stores

by Martin Kouba (JIRA)

[ https://issues.jboss.org/browse/CDI-688?page=com.atlassian.jira.plugin.sy... ] Martin Kouba commented on CDI-688: ---------------------------------- Right now, this is undefined, i.e. it's completely up to the implementations (assuming the requirements for normal scopes are met: "6.3. Normal scopes and pseudo-scopes"). What kind of pluggability do you have in mind? > Contextual bean stores > ---------------------- > > Key: CDI-688 > URL: https://issues.jboss.org/browse/CDI-688 > Project: CDI Specification Issues > Issue Type: Feature Request > Reporter: John Ament > > Right now, the notion of how beans are stored are ambiguous. The implementations back them by maps. This may work well, but there's some amount of plugability required to make passivation capable beans work right. A single request makes sense to store in a map. Sessions are stored in the HTTPSession. We should allow these bean stores to be more pluggable, allowing users or implementations to override how their beans are stored. -- This message was sent by Atlassian JIRA (v7.2.3#72005)

7 years, 9 months

1
0
0 / 0

[JBoss JIRA] (CDI-691) Fire initialized/destroyed for conversation and session scoped

by Martin Kouba (JIRA)

[ https://issues.jboss.org/browse/CDI-691?page=com.atlassian.jira.plugin.sy... ] Martin Kouba commented on CDI-691: ---------------------------------- Tomas is right, {{@ConversationScoped}} and {{@SessionScoped}} are not defined for Java SE. > Fire initialized/destroyed for conversation and session scoped > -------------------------------------------------------------- > > Key: CDI-691 > URL: https://issues.jboss.org/browse/CDI-691 > Project: CDI Specification Issues > Issue Type: Feature Request > Reporter: John Ament > > Section 6.7 indicates that initialized/destroyed are called for request and application scopes. need to also fire for conversation and session. -- This message was sent by Atlassian JIRA (v7.2.3#72005)

7 years, 9 months

1
0
0 / 0

[JBoss JIRA] (CDI-690) Automatically start a request context for async observer methods

by Martin Kouba (JIRA)

[ https://issues.jboss.org/browse/CDI-690?page=com.atlassian.jira.plugin.sy... ] Martin Kouba commented on CDI-690: ---------------------------------- [~tremes] You're right but what about Java SE? It should work there as well (and I believe it does in Weld). Also info about async observers is missing in the "The request context is destroyed:..." part. > Automatically start a request context for async observer methods > ---------------------------------------------------------------- > > Key: CDI-690 > URL: https://issues.jboss.org/browse/CDI-690 > Project: CDI Specification Issues > Issue Type: Feature Request > Reporter: John Ament > > In many other places we start an implicit request context for method invocations (post constructs for instance). We should have an implicit request context for async observer methods. -- This message was sent by Atlassian JIRA (v7.2.3#72005)

7 years, 9 months

1
0
0 / 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

cdi-dev March 2017