Commit Graph
6 Commits
Author SHA1 Message Date
Adam MonsenandAdam Saghy 3f2079566c FINERACT-2219: capture v1.11.0 lessons learned in asciidoc and fineract release plugin
* don't bother making MD5 files, they aren't necessary
* step12 changed, stub it out and reference FINERACT-1817 for future improvements (looks like svn mv is not yet supported)
* use latest vote message template, based on https://lists.apache.org/thread/cs5t0ho39ktg2jxr74dqd7fg9nyfcz3k
* use more explicit form of `git push` to push release tag, see https://stackoverflow.com/questions/5195859/how-do-you-push-a-tag-to-a-remote-repository-using-git
* use consistent language re: tarball, file, artifact, release candidate
* fix some source listings that shouldn't be using macro expansion: this messes up URLs (hides scheme)
* fix asciidoc link to FINERACT-1229
* clean up copyright years: use variable properly and remove unnecessary extra copyright in preface (it appears directly above, in the license)
* de-indent URL in apache license -- this makes the link clickable in rendered HTML and PDF asciidoc outputs
2025-03-24 14:30:21 +01:00
Adam MonsenandAdam Saghy 34cdf4d01f FINERACT-2216: add github project metadata
1. Link to project homepage at https://fineract.apache.org
2. Add some tags/labels/topics

See also:

* https://github.com/apache/infrastructure-asfyaml#repo_meta
* https://docs.github.com/en/repositories/managing-your-repositorys-settings-and-features/customizing-your-repository/classifying-your-repository-with-topics
2025-03-19 12:13:45 +01:00
Adam MonsenandGitHub 8b80f5a934 FINERACT-2170: improve asciidoc: clean URIs, more (#4271)
* disable broken fineract-client API asciidoc include
  * this has a side effect of removing ~1,200 pages from the PDF
* remove duplicated/outdated text, reduce sources of truth
  * link to current FAQ
  * introduction is all references now... it was way out of date and also needed formatting fixes
* unset `show-link-uri` to improve PDF rendering
* remove unnecessary/unused `.asciidoctorconfig` file
2025-01-29 12:41:34 -06:00
Adam MonsenandAdam Saghy 4956bdd574 FINERACT-2149: disable liquibase phone home
At some point, Liquibase added an analytics/telemetry "phone home" to gather user data.

I get how companies and FOSS projects greatly benefit from knowing more about their users, but I think this should only be done with proper consent. I would not assume/expect any libraries to send analytics home. The default behavior of trackers like these should always be to *not* phone home without a user first consenting / opting in.

---

I'm not sure this is the right change, but it works (feedback welcome). My naïve test/check to know this works is visually inspecting the output of the log message I added in TenantDatabaseUpgradeService to see if `liquibase.analytics.enabled` is set, and to what. I'm happy to automate this, but it seemed unnecessary (again, feedback welcome).

We could also disable the phone home with an environment variable ( LIQUIBASE_ANALYTICS_ENABLED=false , see https://docs.liquibase.com/analytics/home.html ). I'm not sure where that should be set if we did. Likely several places (for build/dev/test/ci/qa/etc.).

I originally noticed the phone home because I block requests from my computer to 3rd party tracking/analytics services, and I was getting this error in my console:

    java.net.ConnectException: Connection refused at liquibase.analytics.LiquibaseAnalyticsListener

At first I misunderstood this as a database connection error, but it was really just liquibase throwing an error because it couldn't connect to its own analytics server.

If we were to create a configurable property, I think the ideal place to set the value would be in fineract-provider/src/main/resources/application.properties, (default to false, allow override) but our version of spring boot doesn't support it yet ( see https://github.com/spring-projects/spring-boot/issues/43067 ) so we'd have to add the usual scaffolding required for a Fineract setting. I'm not sure we want/need that, because I can't imagine anyone *wanting* to set the flag to `true`, and my naïve hardcoded global change is less code to maintain than providing a configurable setting. If there's demand to make it configurable later, we could of course do that (especially once it becomes easier to do so). I feel like how we let Liquibase behave is up to us, though. An under the hood / upstream decision we get to make.

See also:

* https://lists.apache.org/thread/5s7zjsxdcqk1qvjnc1wbhqqv66shpl54
* Analytics is enabled by default in OSS version · Issue #6503 · liquibase/liquibase · GitHub https://github.com/liquibase/liquibase/issues/6503
* Add a method for Spring Boot to configure the new analytics feature · Issue #6501 · liquibase/liquibase · GitHub https://github.com/liquibase/liquibase/issues/6501
2024-11-27 17:13:23 +01:00
Adam MonsenandArnold Gálovics bb18aadb74 FINERACT-2147: fix extra swagger UI endpoint paths
I noticed an extra `v1/` while visiting https://localhost:8443/fineract-provider/swagger-ui/index.html

All fineract-provider (unit) tests passed.

I also manually tested out hitting a few /api/v1/ endpoints via the Swagger UI web page at the URL above and those worked as well.
2024-11-19 16:54:58 +01:00
Adam MonsenandAdam Saghy b053c90009 bump MariaDB version
EOL for 11.2 is coming up in a few weeks, see: https://mariadb.org/about/#maintenance-policy

I upgraded my local MariaDB dev database to 11.4 and performed a few cursory GETs and PUTs via the Fineract API--just what I've learned to do so far. Nothing broke. I didn't run all the tests though.

I didn't see any obvious red flags in https://mariadb.com/kb/en/changes-improvements-in-mariadb-11-4/
2024-10-29 21:53:21 +01:00