Two .pyc files under scripts/__pycache__ were committed at some point and have
been churning ever since — merely importing gdl-sync.py to check a config
rewrites them and dirties the tree, which is how they surfaced.
.gitignore had no Python entries at all, only Node ones. The files stay on
disk; this just untracks them.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01UXfdJu7QhSJLr47K7koTDF
Docker Build and Publish / build-and-push (push) Failing after 13s
Ships the gallery-dl sidecar work to the viewer. The visible change is
reel classification: official_artms' Reels tab drops from 781 items to
360, because the sidecars say the other 421 are ordinary feed videos the
clips endpoint returns via include_feed_video. Directory-based
classification counted them all as reels.
Also in this release: dates ranked by source rather than scan order, and
highlight items no longer appearing twice when the archive holds them
under both JDownloader naming conventions.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
JDownloader wrote story-shaped names for highlights during one period of
its life, so the same item exists on disk as both
0ct0ber19 - C5dQPEYpd9W.mp4
2024-04-07_0ct0ber19 - 01 - C5dQPEYpd9W.mp4
which parsed to the ids "C5dQPEYpd9W" and "01 - C5dQPEYpd9W" -- two posts
for one item. The leading ordinal is a position within a day's stories
and carries nothing the shortcode does not, so story and highlight ids
drop it. Post ids are untouched, since those are permalinks.
Measured on the two real files, same archive, cache cleared between:
without the fix the profile reads "Heestory - 2 items", with it
"Heestory - 1 item".
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Nothing in this tool had any memory: every invocation started from zero
and would happily re-enumerate a profile it had listed minutes earlier.
That is what suspended the account -- the listing passes, not the
downloads -- and an aborted run re-enumerating five profiles on restart
was a large part of the bill.
Three changes, in order of how much they save:
- Seeding is now a one-time bootstrap per source. After the first
successful sync the archive DB records everything gallery-dl has seen,
so the source is never probed again. A second full sync costs roughly
half what the first did.
- Stories never seed at all. A story cannot be in the archive before it
is fetched, so there is nothing to seed from, and probing would double
the cost of the cheapest surface we have.
- A source fetched within --min-interval (20h) is refused, and listing
results are cached for --probe-ttl (24h), so a restart mid-run is free
rather than a repeat. --force overrides both.
--only replaces --no-stories and takes any subset of the surfaces, which
is what makes a daily stories-only run possible: one source per profile,
no seeding, a handful of requests. Everything else stays monthly.
Tested with stdlib unittest -- no new dependencies, and it runs anywhere
the sync does. The cases include the aborted-restart scenario, which now
plans zero work.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The account was suspended on 2026-08-17 for "spam", during the session
that built this tooling. Both fetching docs were confidently wrong about
what the risk was, so both now carry the correction.
docs/jdownloader.md said the ban vector is instagram.com requests, which
is right, and implied that meant per-post metadata fetching, which is
only part of it. The suspension came from read-only verification:
automated browser scrolling to enumerate profile grids (~18 paginated
loads per profile, done twice on one after a selector bug), repeated
--simulate and -j passes over the same profiles, per-post /p/ fetches
while testing filename formats, and an aborted sync that re-ran every
listing pass before dying. None of that produced a file, and together it
rivalled the real sync for request count.
The rules that follow are in docs/gallery-dl.md: verify against the
archive rather than the live site, count read-only work against the same
budget, treat the first CDN 429 as the end of the session rather than a
pacing knob, and cache probe_live so a restart does not re-enumerate
everything. The warning order was CDN 429, then 400 on the highlights
tray, then suspension.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The previous commit had this backwards: the sidecar date was only
consulted when the existing date came from an mtime, so a filename date
silently outranked what Instagram itself reported.
The order is sidecar, then filename, then mtime -- metadata first,
mtime last, since mtime is when the file hit disk and says nothing about
when the post was made. Ties keep the incumbent so two equally
authoritative files cannot flip a post's date by scan order.
Extracted to src/lib/post-dates.ts rather than left inline, because the
rule is easy to state and easy to get wrong -- the tests include an
order-independence case that would have caught the original mistake.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The .json sidecars published with the ARTMS fetch were inert: the scanner
fed them through the Instaloader path, where `node.edge_media_to_caption`
and `checkIsStory`'s `product_type` are both absent, so nothing happened.
They are now recognised structurally -- flat, with post_shortcode and
type, and none of the markers the other two JSON shapes carry -- and used
for three things:
- `type` sets post.isReel, which post-tabs prefers over every fallback.
This is Instagram's own classification and it disagrees with ours a
lot: of 781 items in "official_artms - reels", the sidecars say only
360 are reels. The other 421 are feed videos the clips endpoint returns
via include_feed_video, and the directory-based rule counted them all.
- `description` fills the caption where no .txt exists.
- `date` dates a post whose filename could not.
Also fixes date precedence. Only JDownloader highlights lack a date in
the filename, so parseArchiveFilename now marks those as mtime-derived
and the scanner lets any real date replace them -- previously the date
depended on which file the scan reached first.
Verified against real published files: a directory of three type=post and
three type=reel renders 6 in the grid and exactly the 3 reels in the
Reels tab.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
A sync runs for hours and is normally watched through a redirected log,
where Python's block buffering withheld the per-source progress lines
until they happened to flush. The gallery-dl subprocesses write to the
same descriptor unbuffered, so the log also interleaved out of order.
Reconfiguring the streams in-process rather than relying on `python3 -u`
means it holds however the script is invoked.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Script comments should describe the script. Where a run has got to is
project state, so it belongs in docs/gallery-dl.md, which now records the
live withaseul publish, the file-ownership caveat and where the profile
list lives.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Adds --urls-file so a run can reference a hand-maintained list rather
than repeating --profile, which is how this actually gets used: the
ARTMS accounts now live in artms_account_links.txt at the archive root.
The parser takes what a person would paste. Full URLs, scheme-less URLs
and bare usernames all work; blank lines and # comments are ignored and
duplicates dropped, so the list can be appended to carelessly. Lines that
are not profiles are rejected loudly rather than silently syncing
nothing: an Instagram post URL yields the segment "p", which would
otherwise be treated as a username and create a directory called "p".
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
A dry-run publish against the real archive caught gdl-sync.config.json
being created in the archive root: it was written into the staging
directory, and staging is rsynced wholesale. It now lives as a sibling of
staging instead, with rsync excludes as a second line of defence.
The dry run is otherwise clean -- 322 files added, 0 deleted, no new
directories -- and confirms the property that matters most: of 74 new
media files, zero duplicate media already held under a different name.
The JD2 and gallery-dl naming really do converge.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Wires up the staging -> rsync step and replaces --archives with --index
(a listing source: local path or the viewer's API), --staging and
--publish, so the fetch host needs no copy of the archive.
rsync runs --ignore-existing with no --delete. That is a safety property
rather than an optimisation: the archive deliberately outlives Instagram,
so publishing must only ever add. It runs once at the end so a profile
that fails midway never reaches the archive half-written.
Exercised end-to-end against withaseul across all four surfaces,
publishing to a scratch directory. Seeding worked as designed (915 of 984
post items and 28 of 34 reel items already held), stories and highlights
returned no results cleanly, and the collab-reel case landed correctly:
"withaseul - reels" holds files owned by cher_ryppo, 0ct0ber19 and
official_artms, each with the owner in the filename and the crawl scope
as the directory.
The first run drew '429 Too Many Requests' from the CDN at 3M with 1-3s
sleeps and lost two videos. That is the tolerant surface complaining, so
the defaults are now 1M, 6-10s between requests, 3-6s between downloads,
sleep-429 of 120s and 8 retries. Re-running recovered both videos with
zero failures and zero 429s. Installing yt-dlp on the fetch host also
matters: without it DASH videos fall back to a progressive URL, which is
what the rate limiting hit hardest.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
gallery-dl skips already-held media either by file existence -- which
requires the archive mounted where it writes -- or by a sqlite
skip-archive, which requires nothing on disk. Using the latter lets the
fetch host write to local disk and rsync afterwards, avoiding tens of
thousands of small writes over CIFS and keeping a mid-sync failure from
leaving partial files on the live Resilio share.
The key is archive_prefix + archive_fmt: the literal "instagram" plus the
per-media numeric pk. Verified against a real run -- a 3-image carousel
produced 3 rows and a re-run skipped every media file.
media_id is absent from our filenames, so the DB cannot be built from
names alone, but the listing pass we already make maps every live item to
its media_id, and a file listing says which we hold. Seeding therefore
costs no extra Instagram requests and no archive content -- the listing
GET /api/archives/:name/files already serves is enough.
Measured on 0ct0ber19: 2275 live items, 2248 seeded, 27 left to fetch --
exactly the media of the two posts added since the last crawl.
The trap worth the comment it carries: posts and reels are filed under
post_shortcode, while stories and highlights use the per-item shortcode
(post_shortcode there is the containing reel's id, shared by every item).
Matching on the wrong field seeded 5 of 2275 rather than failing loudly.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Every claim in docs/gallery-dl.md was measured against the live site and
the archive rather than taken from documentation, because two of the
assumptions turned out to be wrong.
The safety model is the reason the config looks the way it does.
gallery-dl has two API backends: the graphql one issues a request PER
POST for every video and carousel -- the pattern that got this account
banned via Instaloader -- while the default rest one paginates listings
at 30-50 items and carries carousel_media, video_versions and
product_type inline. A 300-post profile costs ~10 requests.
Findings worth recording:
- JD2 stamped filenames in desktop LOCAL time (US Eastern), not UTC.
Across 212 comparable posts: UTC 19 mismatches, UTC-5 10, UTC-4 zero.
{date:Olocal/%Y-%m-%d} reproduces it; the trailing separator must be
omitted or it lands in the strftime format.
- A profile's reels tab returns collab reels owned by OTHER accounts, so
the directory must be forced with -D. JD2 did the same: chuuo3o and
official_artms filenames sit inside "0ct0ber19 - reels".
- Stories and highlights need per-item {shortcode}; {post_shortcode} is
the reel's id and is shared by every item. {date} is per-item, verified
on a 154-item highlight with distinct times.
- gallery-dl reproduces JD2's caption .txt exactly, including writing
nothing for an empty caption and omitting the trailing newline.
- The json sidecar needs `include`, not `fields`; `fields` silently does
nothing in mode:json and leaks audio_user blobs. It yields `type`
(post/reel) -- Instagram's own flag, which can retire the lone-video
heuristic once the scanner reads it.
Naming differences between the two tools are cosmetic: EXPORT_RE already
makes the index optional and parseInt normalises zero-padding, so a mixed
archive parses identically. Tests pin that down.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Docker Build and Publish / build-and-push (push) Failing after 10s
The header rendered allPosts.length, which is pre-dedupe — 0ct0ber19
showed "303 posts" over a 300-tile grid. Instagram's counter equals its
grid, so count the grid.
Also correct CLAUDE.md. v1.7.0 claimed the grid holds everything "as on
Instagram"; Instagram actually includes a reel in the grid only when the
creator shared it to feed, per post. Measured live: official_artms has
21 reels in its first 34 grid tiles, 0ct0ber19 has 1 in 214. Archives
carry no such flag, so showing everything approximates the behaviour
rather than reproducing it.
Records the DOM trap that caused the wrong reading in the first place:
grid reels link to /reel/<code>/, not /<user>/p/<code>/.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Docker Build and Publish / build-and-push (push) Failing after 10s
The Posts tab filtered reels out, so the grid was not the archive — it
was the archive minus its videos. For `for.heejin` that hid 533 of 1225
posts; for `loonatheworld`, 1100 of 3813. On Instagram the grid holds
everything and the Reels tab is a filtered view of that same set.
Extract the tab logic to src/lib/post-tabs.ts so the reel heuristic is
testable outside the component, and add dedupePostCopies: the jd2 flow
crawls the profile URL and the /reels URL separately because the profile
page misses some reels, so the two overlap and a reel can land on disk
twice. Those are two posts with distinct directory-scoped ids, which the
grid would now render side by side; the reels-source copy wins so the
survivor is still recognised as a reel.
Deciding what *is* a reel stays a guess for most archives. Instagram
marks it with product_type ("clips" vs "feed" vs "igtv" — all three are
GraphVideo, and aspect ratio does not separate them), but only newer
Instaloader captures carry it: 1101 of gibiofficial's 5919 sidecars, and
only 2 marked clips. JDownloader archives carry none, so those still
fall back to treating a lone video as a reel.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
ergosteurandClaude Opus 5