1
* [ ] https://github.com/GitoxideLabs/gitoxide instead of libgit2
2
3
* [ ] We might want to make an über-NIF including connecting to Comrak/ammonia/etc. ourselves (or a separate NIF for that), since overriding stuff in MDEx/Autumn is hard. See markdown.ex.
4
5
* [ ] show dag nav on mobile with popout
6
7
* [ ] use read replica where appropriate
8
* [ ] "hooks" also point HEAD to tags when received, oops
9
* [ ] autumn and mdex NIFs are both around 80MB; the latter includes the former. dedupe.
10
11
* [ ] relative URIs: ugc- prefix, #xyz shouldn't navigate away, etc.
12
* [ ] pressing browser forward/back should trigger fancy blob highlights, not
13
just the automatic :target ones
14
* [ ] some patch links really need to scroll back to top
15
16
* [ ] https://talya.nossa.ee ?
17
18
### short term
19
20
- [x] blob
21
- [x] images
22
- [ ] pdfs, etc.
23
- [ ] less polysaturated, leaner design
24
- [ ] do we use Binary explicitly to not have to pass Unk back as a linked list? @fun
25
- [ ] get our tests and coverage fixed again. @pls
26
- [ ] message to end-user when following a redirect (via git)
27
28
### features
29
30
- [ ] blob permalink
31
- [ ] responsive (mobile) styles
32
- [ ] override locale user option
33
- [ ] diff 2+ parent commit
34
- [ ] passkey support
35
- jfc https://shadowfacts.net/2023/phoenix-passkeys/
36
- [ ] default branch: choose from existing branches; non-UTF-8 support
37
38
### housekeeping
39
40
- [ ] tab indices on e.g. crear repositorio
41
- [ ] IPv6 check
42
- [ ] redirect to primary host if not on it
43
- [ ] reverse where the structs are defined (between Git2/Nossa.Storage*)
44
- [ ] rename RepositorySelf* to RepositorySettings*?
45
- [ ] extract CGI into its own application; stream request/response in/out
46
- [ ] consider erlexec, or write our own driver, so we can get stderr out of
47
the way
48
- [ ] gettext once-over in hard-to-reach spots
49
- [ ] verify typescript in compile step
50
- [ ] override elixir/erlang used in autumn/mdex to keep all together
51
- there's some wip of this in zmlqmwzq
52
53
### graceful degradation
54
55
- [ ] "method" links don't work with noscript since they don't do the right
56
thing
57
- [ ] can't dismiss flash
58
- [ ] javascriptless sidebar changes? (then they must go in the URL)
59
60
### production ops
61
62
- [ ] caching: nginx, varnish? fragment?, git object? how much does bandwidth
63
cost at ovh? bunny? integrate caching, backup into nix module.
64
- right now the pawiwo visibly reloads on every page
65
- [ ] DIY anubis? (for easy deploy to other Phoenix/etc. strength in diversity @fun
66
is a real consideration in this fight.)
67
- [ ] backup restore procedure/test
68
- [ ] feature that restores from backup as we make it. ez!
69
- [ ] admin diagnostics page (combine into/with live dashboard somehow?)
70
- [ ] per repo: audit log, pushes, etc. (git hooks?)
71
- [ ] LRU-ish app-level caching: memcached? redis? ets? mnesia?
72
73
### future
74
75
- [ ] avatar upload
76
- [ ] federation?
77
- identity: auth to other nossa instance
78
- auth another instance for avatars
79
- federated MR-type deal?
80
- don't want to redo a whole code review thing here, idt? check out gerrit
81
etc.
82
- see also simon tatham's thing on git; pull request is just "please check
83
again"
84
- opinionated about rebase/jj norms
85
- at the very least we could show interdiffs between pushes etc.
86
- [ ] merge queue? what do i need at work?
87
- [ ] Log in with GitHub/GitLab/Forgxejo? make it easy to work across, to not be
88
completely hecked by network effects?
89
90
91
## archivo
92
93
- [x] change "settings" top right to "account"
94
- [x] repository delete
95
- [x] repository storage
96
- [x] push keys
97
- [x] proper logging (insert some in e.g. storage creation failure)
98
- [x] some kind of log aggregation/interface on kala
99
- [x] test coverage back up
100
- [x] do some more!
101
- [x] work out why jj fails to push!
102
- [x] fix up this /blah/!url nonsense
103
- [x] git store tests
104
- [x] git tests async: use per-runner store
105
- [x] last controller tests
106
- [x] don't rename storage: use only numeric ids in storage
107
- [x] figure out why our distribution metrics ("phoenix.*") don't make it to
108
grafana (or appear in /metrics at all)
109
- [x] try ETS
110
- [x] pawiwo icon
111
- [x] clean up RepositoryGitControllerTest
112
- [x] Anubis/Nepenthes/?
113
- See https://lobste.rs/s/dmuad3/mitigating_sourcehut_s_partial_outage --- we
114
will need to be prepared; robots.txt, etc., and above all usage stats/graphs
115
so we can see it happen.
116
- [x] https://git.sr.ht/robots.txt is a good start, though of course it isn't
117
saving them. Still, no reason to start from zero.
118
- [x] start of gettext support
119
- [x] main browsing interface, ui polish/consistency
120
- [x] navigate tree
121
- [x] zindex fix for icons overlaying nav bit
122
- [x] select commit
123
- need to resolve a rev instead of a branch now
124
- [x] include breadcrumbs in Diff view
125
- [x] wherever "path=" is sneaking into our URLs, get it out (breadcrumbs)
126
- [x] liveview for navigating within
127
- [x] extract inline classes into css (w/ @apply)
128
- [x] dark mode styles
129
- [x] actual colour system
130
- [x] pawiwo svg for colours
131
- [x] ok dark mode again
132
- [x] safari, chromium
133
- [x] repository redirects
134
- [x] tailwind v4
135
https://elixirforum.com/t/mix-tailwind-4-0-0-beta-1-support/67636/17
136
- [x] more post-liveview cleanup of repository controller; check over
137
auth-related stuff.
138
- [x] then back to (re)design
139
- [x] work out whether diff view will be within the same liveview, a different
140
one, etc.? - it may be overkill; navigating the tree responsively is nice,
141
but the huge diffs might make browsers sick. gotta test!
142
- [x] decide on the repository view design; do we keep the commits there? do
143
we make the sidebar collapsable a la ~~yuta~~ hexdocs etc.? something
144
else entirely?
145
- what do i want to see when i hit a root of a repo? what are the various
146
reasons i end up on one of those on github/srht/etc.?
147
* linked to as a project's "home page", searched and the first result (or
148
apparent best result for an official download or similar)
149
* directly navigated to because i want to browse the source of something
150
i don't have a local clone of (and i haven't yet hit the threshold
151
where i need to use tools to look thru; source aware views can sometimes
152
abrogate the need but often (e.g. github's views + semantic index)
153
they're an untenable degree of both meh design usability (in e.g., links
154
are all too small, everything bunched up on the side, often starting to
155
scroll horizontally by this point; not a priority) and flawed impl (in
156
e.g. elixir outlining. is this just from elixir-ls? tree-sitter?))
157
* often want a specific version (i.e. something i'm running/using),
158
sometimes just browse main/master if cbf/not relevant/more interested
159
in design than exact
160
* summary of this point: "whatever would be unnecessary if i always had
161
a clone of everything". i don't, but i'm in the position of improving
162
the situation for when you don't -- what actually helps?
163
* sometimes i do this even in the presence of a clone on disk, because
164
i'm *using* that one somehow, but i cbf creating a new worktree.
165
similar result. a browser window (and the way you can open lots of
166
them, to various paths/versions/etc) is a much smaller cognitive unit
167
to manage than a worktree.
168
* how often do i want to see the whole commit list straight up front like
169
this? (day separations would give this a lot more context, for what
170
it is.)
171
* almost always would prefer the *graph*; i almost never prefer the
172
straight log style locally. (think about jj log view.)
173
* there are three axes: commit (temporal), path (spatial), view (fn).
174
github exposes commit consistently, path inconsistently (compare blob
175
view to commit history restricted to a subpath), view inconsistently
176
("Commits" page vs 2.5 tree pages (root, non-root w/ README, non-root;
177
sidebar tree only visible in non-root) vs blob (very aligned with
178
non-root tree). root really is the odd one out, since it's a landing
179
page.)
180
* big commit *graph* that stashes in side: file listing-style tree view
181
that then collapses into expanding tree sidebar: blob.
182
* then you can shift around points in the graph while retaining
183
visibility over the directory tree, no reloads/scrolls breaking
184
"immersion".
185
* does the graph stash over to a plain log? i should make sure while i'm
186
developing that i work with the browser the sizes i want to actually
187
be using it at. (i.e. push back against default to 2/3rds code, unless
188
i'm developing a usecase where i'd have the browser at 1/3rd).
189
* sometimes you do want to view just the log of a file or subtree.
190
kinda wanna apply that filter *to* the far-left graph, as a
191
saved-in-navigation thing. then you can nav to a commit (still
192
centred on the tree or file you started the filter from), see it in
193
context, move around, and then at some point you back up or clear
194
the 'path filter' on the DAG view.
195
* note that we want the current path to remain in the tree view
196
when navigating the DAG view even when we don't have a path filter
197
active. same with current blob, if any.
198
* this kind of removes that fake third axis, exposes the other two as
199
proper orthogonal ones. github flattening the DAG with no way to
200
recover is criminal.
201
* and then there's commit view, which is a whole other thing: viewing a
202
commit message & diff.
203
- [x] let's fix our URIs in gen i guess, ?path= is ugly
204
- [x] support submodules in tree
205
- [x] be explicit in which params are UTF-8 encoded & which aren't
206
- [x] global support for non-UTF-8; no error messages
207
- [x] navigate commits on left; focus current commit as navigating
208
- [x] compile-time the storage layer choice
209
- [x] vyxos push fails!
210
- [x] "last commit modified" links are incorrect for subtrees, there's a TODO in
211
git2.rs somewhere
212
- [x] decompose git2's monster "resolve_commit" into somewhat orthogonal ops
213
- [x] syntax highlighting
214
- [x] little live demo with channels of push response
215
- [x] reaffirm backup situation for kala
216
- [x] nix syntax highlighting egumi??
217
- [x] wow! nossa in docker/k8s!
218
- [x] ES: "Clave de empujar creada exitosamente" when creating a new repo!?
219
- [x] new repo => brocken
220
- [x] remove register links if register is disabled
221
- [x] install our own git hooks in repositories, set HEAD on push if unset
222
- [x] assign default branch on first push
223
- [x] this causes clone failures, the weird nix flake stuff we've been seeing, etc.
224
- [x] floating windows don't preserve state on livereload
225
- [x] let's fix the first thing: navigating back to a subtree shouldn't close an
226
open blob. we can keep our existing assigns, or just stream.
227
- [x] open new windows in specified order
228
- [x] default orientation based on aspect ratio
229
- [x] splith(3): need to have up/down too; splitv(3) needs to have left/right.
230
it's too hard to just do that with merges and unmerges.
231
- [x] skip floating windows when finding clientInDirection; similarly,
232
moveClient/mergeClient; their behaviour is very puzzling if it's not clear
233
what's happening
234
- [x] "hooks" are breaking on current live because they're symlinks into old
235
versions of nossa, since GC'd from the nix store!
236
* [x] shynet: add javascript call on live patch
237
* [x] what's nom? https://github.com/NixOS/nixpkgs/pull/430275#pullrequestreview-3089109039
238
* [x] set inline-block for repo breadcrumbs so the underline doesn't over-extend
239
* [x] NixOS tests to ensure the module stays good even after we move to deploying on k8s
240
* [x] per-repo deploy keys: R/O, R/W
241
* i want this for flux
242
* [x] everything except the UI
243
* [x] the UI
244
* [x] update the unique index to be per-repo since we've changed auth anyway
245
* [x] tree view isn't getting the right commits for files in subdirectories
246
since we're back to one-directory-at-a-time and need to redo how we search
247
for the last changed commit.
248
* [x] actually show commit details at top of tree/blob views
249
- [x] branch selector disappears on safari with javascript on
250
* [x] address our lack of sendmail in container; send mail to status.hostIP:25 (no auth)
251
- [x] dag infinite scroll/load to context
252
* [x] visual confirmation on "copy to clipboard" shouldn't be a modal, just temporarily change the icon
253
