the csv exports for mastodon had some formatting issues that was making
importing problematic, so those issues have been fixed so importing the
doc is easier
The admin area needed to be reorganized as it was a bit all over the
place to accomodate new features.
And edit link was added to location detail pages so authorized members
can edit when necessary.
Form elements were loosened up a bit to give them a bit more breathing
room
The plumbing for editing location info has been updated, so that data
can be changed by authorized memebers. Also added a new data point for
locations to store archive links
part 2 will focus on setting up permissions and authorizations as well
as smoothing out adding new members and member roles. an edit link will
be added to locations, which will be visible for members with the
correct permissions
finally found the correct exception that captures a connect error when a
source url cannot be loaded, so it logs the error and just keep
executing the script.
Now automation can be enabled since the system knows how to handle
errors. Fuck. Yes.
The script for collection blocklist info was getting a little heavy, so
it's been split in two to make it more accessible and easier to
maintain. The first part collects data from current sources and stores
it for the second part of the script, which compiles all the data and
updated the database.
The Source model was expanded to track votes so block and silence counts
can be easily verified on the backend.
changed the function that retrieves a single location to decide what to
get based on the variable type, so it doesn't need to be two seperate
functions
The Location Controller was getting too heavy so an Update and
Maintenance service class was created to offload most of it's
functionality. Location upating was moved to LocationRepository
There was also a small issue with responsive list links not adapting
properly in Safari that was fixed
addded some preliminary url validation to make sure a source url is
still active.
the consolidation process is starting to get top heavy, so it's a stop
gap until a more effecient process can be created.
Adding total actions count to location data has made it possible to
simplify backend data queries, so the custom pagination was optimized
and location repositories have been cleaned up.
the front end controller has been cleaned up as well, which has resulted
in the appropriate template being polished up as well.
Added a new field to the locations table for total actions which will be
used to set the active state of location. this will result in data being
parsed and sorted easier instead of doing those calculations on the back
end, so DB queries will be simpler
also fixed actions counts for csv imports
Seperated data logic for locations and put it into its own repository
class for better organization and readability of controller classes.
Data methods are still simple so no need for an interface class just
yet, but will probably implement at a later date
Changed the appeal process so that each request is tracked in the
database to make reviewing and time limits easier to manage.
An email is still sent but it's just a notifcation to let the admin know
an appeal has been filed.
Hit the major friction points in the responsive UI. Still have some
polishing to do but there shouldn't be any show stoppers at this points.
Also moved some variable to the env so they can be changed easily when
necessary
Responsive styles are non-existent so it was time to get that sorted.
This first pass was just getting a feel for what can be done with
list items since that's one of the main components of the site.
Second pass will clean this up as well as the majority of text styling
so it all smoothly adapts.
Ha, or that's the plan anyway.
Put together a quick test to get internal messaging working for appeals
and joining current sources requests. Now that it works, forms that will
leverage messaging will be created.
Updated the design for pages displaying information about specifiic
locations to clearly display heat rating, silence and suspeneded
statistics.
Also inlcude a descriptor for heat rating.
Added some preliminary Bad Spade stats to the index to set up a space
for additional information to be shared in the future. Starting with an
accurate display of locations tracked, Current Sources count and
last update date
Search methodolgy has been tweaked and the correspending result links
have been updated, as well as links in the Listing page to display both
silence and suspend counts and the overall heat rating.
Also plugged in new silence and suspend icons provideb by
https://rage.love/@puf. Big thanks for that, man.
Started the front end refresh by starting to plug in the new design on
the index page. The search methodology was still not consistent, so that
process was cleaned up to display what is actually being tracked based
on the two action criteria, then updating the design for the location
links to be clear and display the heat rating.
Added a repo link to the About section.
https://tenforward.social/@noracodes reported an issue with the CSV not
rendering properly because of the mix of single and double quotes, so
that's been cleaned up.
Also fixed action sorting which was reversed.
There are some carriage returns in some of the descriptions which was
throwing off the CSV cell assignment, so added a quick patch that
replaces said returns with a simple space to keep formatting intact
Reactivated CSV exports based on Heat Rating, which is the percentage of
Current Sources the have taken action, a suspend or a silence, against
an instanct. The higher the Heat Rating, the more Sources that have
taken actions agaisnt it, so they should be viewed with caution
Actions taken against and instance of have been separted into their
respective buckets so they can be display properly rather than simply
grouped together and mislabled.
This gives a better sense of the severity of response per instance.
The count data is now separated by specific action rather than grouping everything together.
This gives a clearer picture of the severity of a response by current sources
There was a small syntax bug that was causing the sytem to error out
when resetting rating terms.
Also plugged a bug that was returning locations with ony one block
count, which is below the criteria for instances that should be listed
and shown in search results
Finally moved over the public search API from the old version and
updated the about page to show the new data structure
Also tweaked the location update script to change 'defederate' to
'suspend' for the sake of consistency
Tweaked the about page to include some langauge talking about where the
data is coming from and how it's being used and linking to the
respective sources
Switched from using external CSV exports to having the system talk to
instances directly to build and update the database.
NOTE: Added a new table called Source with the corresponding model to
access it
The archived build errors out when it finds the images json field is
null, so this will replace a null value with an empty array instead.
Should probalby just make images non null and an empty array the default
Plugged in noted counts that detail how many times an instance as been
suspended or silenced by members of the Trusted Sources.
https://writer.oliphant.social/oliphant/trusted-source-membership
Also added some additional meta data to links displaying instances lists
so status and counts can be identified at a glance
