[x3d-public] Minutes of JavaScript/JSON meeting 28 July 2021

[John Carlson]

I am a bit concerned with zjsonpatch tool, as I don't see a test case 
that takes 2 JSON files one wants to diff, as we would need in a 
roundtrip test.  That is, perhaps they can patch, but not diff.  I will 
dig deeper.

I'll look at the other tool in a bit.

John

On 6/29/21 10:05 AM, Don Brutzman wrote:
> Thanks for concise minutes John, and your patience while I connected 
> from high-background-noise environment of SFO baggage claim.
>
> Agreed that JSON diff will be important for us to pursue as we work to 
> achieve correct/consistent/repeatable X3D JSON schema validation.
>
> On 6/28/2021 3:11 PM, John Carlson wrote:
>> [...]
>> Somewhat useful tools for comparing output, but  there are online 
>> tools as well:
>>
>> X3DJSONLD/src/main/node/xmldiff.js
>> X3DJSONLD/src/main/node/jsondiff.js
>
> When I searched online, just about everything I found was based on 
> invocation online via web browsers.  This won't work for our build 
> processes.
>
> For example, there is a json-diff project that has colorized text 
> output, not sure if that is useful for console logs.
>
> * https://www.npmjs.com/package/json-diff
>
> Found another written in python:
>
> * https://pypi.org/project/json-diff/
>
> Found this helpful reference which lists 2 Java-based tools, 
> zjsonpatch and fge json-patch
>
> * 
> https://softwarerecs.stackexchange.com/questions/17812/java-library-that-diffs-json-and-generates-what-was-added-deleted-modified
>
> Recommendation was zjsonpatch which looks mature and last released a 
> year ago.  It refers to a standard for such diffing:
>
> * RFC 6902, JavaScript Object Notation (JSON) Patch
>   https://datatracker.ietf.org/doc/html/rfc6902
>
> Am hoping to keep prerequisites our giant suite of Quality Assurance 
> (QA) tools as simple as possible, thus using Java with Ant repeatably 
> has greatest utility.  Maintaining logs of outputs is further useful 
> because then version-control diffs of console output can tell us when 
> something slips sideways.
>
> all the best, Don