shockrah
aa9c515b95
invites::valid_invite now uses proper db-lib function calls
2020-11-16 22:18:37 -08:00
shockrah
2448e1b200
get_online_members now uses db-lib for its back end
...
This means we've removed a ton of old raw sql code
2020-11-16 20:20:37 -08:00
shockrah
b966c61c20
Finally the change over to db::messages::Messages::send in userland code
...
- removed old helper function which is no longer needed
* cleaned up responses for empty,other, and _ in /channels/create endpoint handler
2020-11-12 13:18:43 -08:00
shockrah
d615a41c0c
Messages schema now contains foreign references to channel_id; not channel_name
2020-11-12 13:17:19 -08:00
shockrah
c78c8a5502
First pass build of messages::send functionality
2020-11-12 13:16:50 -08:00
shockrah
3925e2c57f
adding chrono to cargo.toml in db-lib
2020-11-12 13:15:24 -08:00
shockrah
a9c55bac90
New dependancy for db-lib chrono
...
! Used in messages module
2020-11-12 13:11:00 -08:00
shockrah
2330f47558
Merging in for the newly updated channels user module
...
Points of interest in this changelog are below:
* Finally using the db library with full integrations in userland channels
* Removed db_types altogether, now using db-lib's version of it
* Auth now also does queries using db-lib, which of course reduces eye strain when looking through that mess
* Userland channels is midly tested so I'm somewhat comfortable for now
* Userland invites still works but should now be trivial to changeover
* db-lib also includes tons more documentation which should make it easier to reason why it does certain things
New testing suite with python also makes writing unit tests stupid easy and is now qol update territory
Here's a contemporary list of things that at this point are simply nice to have:
* Better output from the python testing suite, green/red text and whatnot to more easily skim results
* Better logging overall on the test suite
* More robust error handling in db-lib however ALL the information regarding errors IS getting through to userland and IS handleable
This concern is more about code expressiveness and not really required at all (yet I hope)
Depdendancies:
* Using serde whereever possible since the old serialization methods were super sketchy
* No more .common module
* No more .db_types module
SQL:
Made it easier to purge channels and also fixed some _edgy_ features
like `unique` being inlined and not use at the end of a create table clause
Auth:
* Much cleaner managemetn of `if let` expressions and `matches`
They used to be all over the place but reduction of checks has fixed this mostly
* Now also uses the new db-lib which reduces a ridiculous amount of code
2020-11-07 21:44:38 -08:00
shockrah
5f8863e9df
updated /channel/delete test to pass in correct params, was always workign correctly however
2020-11-07 21:33:03 -08:00
shockrah
06a76b26df
Fixed content-type in channels::create + list response header
2020-11-07 21:01:09 -08:00
shockrah
c8a6aa204d
User API changes to now use the new behavior for adding channels
2020-11-07 18:27:14 -08:00
shockrah
4668ce7d0f
Channel add behavior now also returns the same new channel that was just added
...
! We do this because in order to `get` the channel later we need its id.
Some clients will be updating channel data periodically so this helps to make smaller queries possible
2020-11-07 18:26:19 -08:00
shockrah
700da3695f
Removing whats left of db_types references
...
Finally relying on db library crate for database types
2020-11-07 18:20:34 -08:00
shockrah
cb69d8b945
moving away from db_types in the members api module
2020-11-07 13:26:37 -08:00
shockrah
ec30e80ac1
Invites::filter 1st pass implementation
...
! untested
2020-11-04 00:33:35 -08:00
shockrah
84f04c20cd
integrating special response object to make pull
2020-11-04 00:25:55 -08:00
shockrah
4d85c094cc
removing errors for the sake of testing
2020-11-04 00:19:50 -08:00
shockrah
960abac7d9
DbInvites .filter method
...
! untested
2020-11-04 00:19:35 -08:00
shockrah
562377d6e2
1st pass implementation of the new .filter implementation
...
! Totally untested so far and WILL have to go through a battery of tests before I'm confident with this
2020-11-03 23:37:14 -08:00
shockrah
8ce88faa78
new filtertype parameter is being given for our .filter method
2020-11-03 23:36:13 -08:00
shockrah
f9bc6b3dc9
channels modules have newly updated list endpoints
...
/channels/create requires a rework however
2020-11-03 23:16:24 -08:00
shockrah
188184460f
Adding serde to db libraries list of deps
2020-11-03 21:44:23 -08:00
shockrah
e71f960ee1
For now the test flag just dumps _everything_ to stdout instead of for the sake of usability
2020-11-03 21:42:40 -08:00
shockrah
1431d62ba5
ignoring diesel's schema.rs since we're straight up not using it
...
Only usign diesel for things like database instantiation in development
Production builds will likely use a different more custom solution
2020-11-02 00:13:13 -08:00
shockrah
7a093c1fbc
new delete_channel endpoint handler
...
* to be tested
2020-11-02 00:12:16 -08:00
shockrah
3824546bba
removing unique restriction from channel name (for now i hope)
2020-11-02 00:11:43 -08:00
shockrah
06c81ce2f2
Removal of unused code or things that were trivial to switch to the library like db types
2020-11-02 00:11:15 -08:00
shockrah
543feef330
Foreign keys in messages now delete when parent key is deleted
2020-11-01 22:33:25 -08:00
shockrah
4cb8f578ed
new test for deleting channels
...
-x- this test will fail but the patch is required for testing self hosting our code base
2020-10-28 22:31:52 -07:00
shockrah
7023ce2b7a
new passing test of sending a message
2020-10-21 23:00:56 -07:00
shockrah
b008a0d3e1
* Removed 'unused import' warning
...
+ Added library crate level docs to channels module
Should hopefully make things easier
2020-10-21 22:39:16 -07:00
shockrah
35dac99d88
new helper method to build json body in requests
2020-10-21 22:32:59 -07:00
shockrah
33234f8e18
build.sh now shows server output in stdout
2020-10-21 22:32:43 -07:00
shockrah
ec732dfd34
changed auth calls to use the new db api
2020-10-21 21:40:55 -07:00
shockrah
28dbbc4132
vim somtimes gets prompts stuck in the code somehow yea idk either
2020-10-21 20:24:02 -07:00
shockrah
c5db9bae51
tfw build script failed before because of random spaces
2020-10-21 20:22:33 -07:00
shockrah
75a9bbe316
new build + test scripst for ci/cd pipelines
2020-10-21 19:36:47 -07:00
shockrah
90b131c60a
channels delete is now based on http deletes
2020-10-10 23:08:00 -07:00
shockrah
75eac74f3f
forgot to expose modules for use as external crate
2020-10-10 19:53:33 -07:00
shockrah
dfc9f88e66
first sampling of what the lib migration interactions will look like
2020-10-10 19:53:13 -07:00
shockrah
41c28cc845
dummy commit before mass library migration
2020-10-10 14:44:00 -07:00
shockrah
9f3e734b0b
moar warning removals
2020-10-02 13:24:45 -07:00
shockrah
b897e07273
removing frivolous sql select(why was this even here tho)
2020-10-02 13:24:23 -07:00
shockrah
f8ef9ef6eb
nothing of substance in this readme if ill be honest
2020-10-02 08:59:13 -07:00
shockrah
e58d4abb28
removed some more warnings until these vars are used
2020-10-01 20:14:56 -07:00
shockrah
04dca141b5
base for querying online users
2020-10-01 19:50:57 -07:00
shockrah
d2f74b563b
but why was this there tho
2020-10-01 18:55:50 -07:00
shockrah
fa04d2d722
removed dead code
2020-09-30 22:42:28 -07:00
shockrah
65532def98
added delete method for messages
2020-09-17 21:54:54 -07:00
shockrah
2642fdb8b3
adding wrapper for messages
2020-09-17 21:51:23 -07:00