Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
    alaisslx
    @alaisslx
    All Servers are HTTP, but the client is HTTPS.
    Please use Chrome and allow insecure content for that URL (small Shield icon in the address bar)
    alaisslx
    @alaisslx
    for Firefox this seems to be the setting security.mixed_content.block_active_content accessible in about:config
    Spacebel sa (Christophe Noel)
    @spacebel
    @bpross-52n , I have started to commit corrections/misses in a dedicated branch :https://github.com/opengeospatial/wps-rest-binding/commits/spacebel-dev . If you wish, you'll be able to pick commits of interest.
    Benjamin Pross
    @bpross-52n
    @spacebel Great, thanks!
    Spacebel sa (Christophe Noel)
    @spacebel
    @bpross-52n But I realised your version does not use the same schema definition, so my contributions are not correct.
    Spacebel sa (Christophe Noel)
    @spacebel
    Any discussing pending on the notification/webHook mechanism on asynchronous execution ? opengeospatial/wps-rest-binding#31 . Any comment on the proposed schema ?
    Marian Neagul
    @mneagul
    @spacebel Maybe we can brainstorm this tomorrow so we can work on a PoC
    Spacebel sa (Christophe Noel)
    @spacebel
    +1
    Gérald Fenoy
    @gfenoy
    @spacebel, @mneagul any way to participate to this brainstorming session ?
    theoesque
    @theoesque
    We've implemented an asynchronous (light weight observer pattern) and a synchronous (using sse) return method for WPS outputs. As part of the Routing Pilot work. Happy to chat about!
    Brad Hards
    @bradh
    @alaisslx there is a client page on the wiki if you want to add your client there https://github.com/opengeospatial/OGC-API-Hackathon-2019/wiki/Clients
    Brad Hards
    @bradh
    Something for tomorrow, but this is a feeling a lot more complex than features and maps. I get that there is irreducable complexity in processing, but there are too many "knobs" for a typical web coder to get right. Maybe doing "WPS 2.0 in REST" isn't the right approach. For example, could we drop synchronous operation completely? Any consumer than needs it can convert async into sync.
    What would a clean sheet processing design for the web look like?
    Spacebel sa (Christophe Noel)
    @spacebel
    I personally believe that synchronous mode is an old think of the past, and that only async (with notification) makes sense. But I understand that some use it for chaining processes that have a single result and short processing (while I also believe than using workflow/dag is the right approach). Does a "good standard" force users for the "right" approach, or shall it be open and flexible ?
    Brad Hards
    @bradh
    open and flexible doesn't imply support for everything. If there is a "mostly right" approach, its probably bad to include the "mostly wrong" approaches too :-)
    Otherwise, well, GML...
    Matthias Mohr
    @m-mohr
    How would chaining without synchronous execution work?
    Brad Hards
    @bradh
    The async notification would invoke the next step in the chain?
    Matthias Mohr
    @m-mohr
    Once notifications are included, yes.
    Francesco Bartoli
    @francbartoli

    How would chaining without synchronous execution work?

    once you submit a DAG/workflow then your machinery should know which is the state and the next step

    Spacebel sa (Christophe Noel)
    @spacebel
    It depends what you call "chaining". I believe any serious workflow/business process shall be handled by a workflow engine/ dag engine/ script engine (allowing parallel executions etc.). We have prototyped chaining using CWL engine (TB14), BPMN engine (ESE), BPEL engine (SSEGrid). When using simple scripting, the WPS client used in the script simply needs to support either polling strategy, either notification (callback).
    Marian Neagul
    @mneagul
    @spacebel I agree, for current serious scenarios we need a Workflow engine. Allowing triggers would open the opportunity for future scenarios and coupling WPS Servers or Processes without requiring a separate middleware.
    Spacebel sa (Christophe Noel)
    @spacebel
    For those interested by sync mode (and raw single output), are you not interested in having the HTTP Get KVP parameters from XML spec for submitting jobs. Using that strategy (which I personally have not used), you can submit to a process an input by ref, with a HREF which itself generates a processing output (chaining). I believe this is a pro of sync mode. Note also that we can make use of CONFORMANCE: some implementation may support sync mode, others async or both (I think it is reported in the capabilities in XML)
    Marian Neagul
    @mneagul
    Also triggers would allow different scenarios not related to triggering other jobs
    Brad Hards
    @bradh
    I don't want the core of OGC API for Processes to be as complex as current WPS.
    Francesco Bartoli
    @francbartoli

    I don't want the core of OGC API for Processes to be as complex as current WPS.

    Totally agree

    Brad Hards
    @bradh
    So what do we absolutely need before it just won't work at all?
    Spacebel sa (Christophe Noel)
    @spacebel
    Agreed. But is the WPS REST spec a new version of WPS (3.0) or another story ? I don't understand the lifecycle of the existing spec (which required change request and stuff for any change). This is question :)
    Brad Hards
    @bradh
    If you model WFS 2.0 -> WFS 3.0, its another story.
    Semantic versioning, major version breaks backwards compatibility.
    If you just map WPS 2 into JSON, we haven't moved forwards.
    Francesco Bartoli
    @francbartoli
    Honestly I didn’t see WFS3 backwards compatible with WFS2 so the question is why we should make this for OGC API - Processes?
    Spacebel sa (Christophe Noel)
    @spacebel
    Sure, but my understanding (at least in TB14) was that the proposed REST API is a new binding of WPS 2.0 spec, and not a new specification 2.0.1, 2.1, 3.0 or whatever.
    Brad Hards
    @bradh
    If that is what everyone else wants, that's OK. I don't see that as the OGC API philosophy though. I think we should be trying for something that is really developer friendly, meets the common case user needs, and allows for extensions in the future. WPS 2.0 binding should be a non-goal.
    Of course that is easier to say than to do.
    Francesco Bartoli
    @francbartoli
    That’s exactly my feeling too
    Spacebel sa (Christophe Noel)
    @spacebel
    @bradh I'm strictly in favor of your approach. But I would like OGC to clarify this.
    Benjamin Pross
    @bpross-52n
    Right now, I think we are in between a REST binding for WPS 2.0 and WPS 3.0/OGC API Processes. We need to discuss in what direction we want to move forward in the Standards Working Group.
    The advantage of a REST binding for WPS 2.0 is that the functionality is limited to what WPS 2.0 specifies, i.e. it could be moved forward rather quickly from now. But limiting it to WPS 2.0 could also be a disadvantage..
    Matthias Mohr
    @m-mohr
    alaisslx
    @alaisslx
    @bradh added my client to the wiki.
    Spacebel sa (Christophe Noel)
    @spacebel
    Following the idea to move Process Description format to JSON Schema, I made a draft and faced some critical issues: opengeospatial/wps-rest-binding#45
    sima-qian
    @sima-qian
    Hi all, I’ve been spending the hackathon trying to build a client to act as an interactive GUI for the WPS APIs. I’ve documented some issues I came across while interacting with the various WPS implementations: opengeospatial/wps-rest-binding#46 — take from it what you will
    Spacebel sa (Christophe Noel)
    @spacebel
    Submitted the support of callback relying on openAPI callbacks in the API document.
    opengeospatial/wps-rest-binding@8cc2bda
    alaisslx
    @alaisslx
    @sima-qian I can confirm these observations from my client.
    There were quite a few things learned regarding CORS when using such APIs in the browsers as well during this Hackathon.
    Spacebel sa (Christophe Noel)
    @spacebel
    @francbartoli yes :)