feat: smart-link resolve, direct downloads, and Bandcamp album track pick - #90
Conversation
Follow unknown purchase/description links to known gates, and allow pasting Dropbox/Drive/raw file URLs as a browserless download path.
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Plus Run ID: 📒 Files selected for processing (3)
🚧 Files skipped from review as they are similar to previous changes (3)
📝 WalkthroughWalkthroughThe change adds direct file links as gate sources and interactive Bandcamp album-track selection. It updates URL resolution, download routing, job state, browserless handling, progress reporting, the web UI, tests, and documentation. ChangesGate and download flows
Estimated code review effort: 4 (Complex) | ~60 minutes Sequence Diagram(s)sequenceDiagram
participant GateInput
participant GateResolver
participant DownloadServer
participant DirectDownloader
participant DownloadsDirectory
GateInput->>GateResolver: Submit gate URL
GateResolver->>DownloadServer: Return resolved provider and URL
DownloadServer->>DirectDownloader: Download direct provider URL
DirectDownloader->>DownloadsDirectory: Write downloaded file
DirectDownloader-->>DownloadServer: Return filename and progress
Possibly related PRs
🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✨ Finishing Touches 💡 1📝 Generate docstrings 💡
🧪 Generate unit tests (beta)
Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 5
🧹 Nitpick comments (3)
src/soundcloud.ts (1)
525-538: 📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick winNested ternary for provider labels is growing hard to read.
The provider-to-label mapping is now six levels of nested ternary. Each new
GateProvidervalue adds another level. Replace it with a lookup table for readability and to make future additions (new providers) a one-line change.♻️ Proposed refactor
+const GATE_PROVIDER_LABELS: Record<GateProvider, string> = { + droploud: 'Droploud', + gaterush: 'GateRush', + downloadgater: 'DownloadGater', + direct: 'direct download', + bandcamp: 'Bandcamp', + soundcloud: 'SoundCloud', + hypeddit: 'Hypeddit', +}; + async getGateURL(track: SoundcloudTrack) { const gate = await extractAndResolveGateUrl(track); if (!gate) { return null; } - const providerLabel = - gate.provider === 'droploud' - ? 'Droploud' - : gate.provider === 'gaterush' - ? 'GateRush' - : gate.provider === 'downloadgater' - ? 'DownloadGater' - : gate.provider === 'direct' - ? 'direct download' - : gate.provider === 'bandcamp' - ? 'Bandcamp' - : gate.provider === 'soundcloud' - ? 'SoundCloud' - : 'Hypeddit'; + const providerLabel = GATE_PROVIDER_LABELS[gate.provider];🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@src/soundcloud.ts` around lines 525 - 538, Replace the nested ternary assigned to providerLabel in the surrounding provider-handling flow with a provider-to-label lookup table keyed by gate.provider. Preserve all existing labels and the fallback for unknown providers, while making future provider additions a single mapping entry.src/utils.test.ts (1)
1-13: 🎯 Functional Correctness | 🔵 Trivial | ⚡ Quick winMissing test coverage for
resolveUnknownGateUrl.This file adds tests for
resolveGateProviderUrl(direct-URL matching) andfindKnownGateInHtml, butresolveUnknownGateUrl— the function that follows redirects and fetches HTML for unrecognized URLs — has no test coverage here. This is the function with the highest risk (network fetch, redirect following, HTML scanning, SoundCloud short-circuit). Add tests that mockfetchto cover: a redirect landing on a known gate, a page whose HTML embeds a known gate, a SoundCloud final URL (should returnnull), and a network failure (should returnnull).🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@src/utils.test.ts` around lines 1 - 13, Add tests in src/utils.test.ts for resolveUnknownGateUrl, mocking fetch to cover redirects to a known gate, HTML containing a known gate, SoundCloud final URLs returning null, and network failures returning null; import the function under test and preserve the existing test conventions and cleanup behavior.src/index.ts (1)
81-90: 🚀 Performance & Scalability | 🔵 Trivial | ⚡ Quick winDuplicate network resolution for the same manually entered URL.
resolveGateUrlOrFollow(value)runs once inside thevalidatecallback (Line 83) and again immediately after the prompt resolves (Line 90), on the identical raw input. For an unrecognized URL, both calls trigger a full redirect-following fetch plus HTML scan (resolveUnknownGateUrl), doubling latency and outbound requests for the same input. Cache the resolution fromvalidateand reuse it instead of callingresolveGateUrlOrFollowa second time.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@src/index.ts` around lines 81 - 90, Cache the result of resolveGateUrlOrFollow inside the prompt validate callback and reuse that cached resolution after the prompt completes, removing the second call for the same gateUrl. Preserve the existing validation rejection for missing results or soundcloud providers, and ensure the cached value is updated for each validation attempt.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@src/directDownload.ts`:
- Around line 77-119: Update downloadAudio to pass an AbortSignal.timeout(...)
signal to the fetch call, and enforce a maximum download size before or while
obtaining response.arrayBuffer(). Use the response Content-Length when available
and reject oversized files before buffering; preserve the existing successful
download and filename handling.
- Around line 60-119: Sanitize the Content-Disposition filename before
constructing the output path in DirectDownloader.downloadAudio. Apply basename()
or the existing sanitizeFilenamePart utility to fromHeader, while preserving the
current URL-derived fallback and default filename behavior, so
join('./downloads', filename) cannot escape the downloads directory.
In `@src/utils.ts`:
- Around line 395-435: Extract the shared direct-download host checks, extension
checks, and normalization logic used by matchDirectDownloadUrl,
isDirectDownloadUrl, and normalizeDirectDownloadUrl into a dependency-free
module such as directLinkRules.ts. Update both utils.ts and directDownload.ts to
import and reuse those shared symbols, removing the duplicated inline rules
while preserving existing URL matching and normalization behavior.
- Around line 462-500: Update resolveUnknownGateUrl to validate the initial
trimmed URL and each redirect destination before any network request, rejecting
non-public or otherwise unsafe addresses while preserving known-gate and
SoundCloud checks. Configure the fetch flow to inspect redirects and apply the
same validation to every destination before following it, so neither
extractAndResolveGateUrl nor the hypeddit route can trigger blind SSRF.
In `@webui/src/components/App.tsx`:
- Line 948: Update the placeholder in the relevant App component input to show
the Dropbox example as a complete valid URL, using the https:// scheme and ?dl=1
as the first query separator instead of &dl=1; leave the other supported URL
examples unchanged.
---
Nitpick comments:
In `@src/index.ts`:
- Around line 81-90: Cache the result of resolveGateUrlOrFollow inside the
prompt validate callback and reuse that cached resolution after the prompt
completes, removing the second call for the same gateUrl. Preserve the existing
validation rejection for missing results or soundcloud providers, and ensure the
cached value is updated for each validation attempt.
In `@src/soundcloud.ts`:
- Around line 525-538: Replace the nested ternary assigned to providerLabel in
the surrounding provider-handling flow with a provider-to-label lookup table
keyed by gate.provider. Preserve all existing labels and the fallback for
unknown providers, while making future provider additions a single mapping
entry.
In `@src/utils.test.ts`:
- Around line 1-13: Add tests in src/utils.test.ts for resolveUnknownGateUrl,
mocking fetch to cover redirects to a known gate, HTML containing a known gate,
SoundCloud final URLs returning null, and network failures returning null;
import the function under test and preserve the existing test conventions and
cleanup behavior.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro Plus
Run ID: 5d6fcea4-b738-40e0-807c-46f8e971f3a0
📒 Files selected for processing (9)
README.mdsrc/browserLaunch.tssrc/directDownload.tssrc/index.tssrc/server.tssrc/soundcloud.tssrc/utils.test.tssrc/utils.tswebui/src/components/App.tsx
D3SOX
changed the title
There was a problem hiding this comment.
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
webui/src/components/App.tsx (1)
990-990: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winFix the Dropbox query separator in the placeholder.
The example does not show an earlier query parameter, so use
?dl=1instead of&dl=1. Include thehttps://scheme so the example is a valid URL. This was already raised on a previous commit of this PR and remains unfixed.Proposed fix
- placeholder="https://hypeddit.com/... / droploud.com/gate/... / gaterush.me/... / downloadgater.com/g/... / artist.bandcamp.com/track/... / dropbox.com/...&dl=1" + placeholder="https://hypeddit.com/... / droploud.com/gate/... / gaterush.me/... / downloadgater.com/g/... / artist.bandcamp.com/track/... / https://www.dropbox.com/...?dl=1"🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@webui/src/components/App.tsx` at line 990, Update the placeholder in the App component to show the Dropbox example as a valid URL with the https:// scheme and ?dl=1 query separator, replacing the current &dl=1 form while leaving the other URL examples unchanged.
🧹 Nitpick comments (1)
src/jobStore.ts (1)
161-181: 📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick winAdd unit tests for the Bandcamp selection pause/resume/cancel interactions.
The resolver map introduces new concurrency-sensitive behavior: superseding an existing waiter, resolving on cancel, and resolving on delete. Add tests that cover: a pending wait resolved by
resolveBandcampTrackSelection, a pending wait resolved bycancel(), a pending wait resolved bydelete(), and callingwaitForBandcampTrackSelectiontwice for the same job.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@src/jobStore.ts` around lines 161 - 181, Add unit tests covering the Bandcamp resolver lifecycle through waitForBandcampTrackSelection and resolveBandcampTrackSelection: verify a pending wait receives its track URL, cancel() resolves it with null, delete() resolves it with null, and a second wait for the same job resolves the first waiter with null while leaving the replacement waiter pending.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Outside diff comments:
In `@webui/src/components/App.tsx`:
- Line 990: Update the placeholder in the App component to show the Dropbox
example as a valid URL with the https:// scheme and ?dl=1 query separator,
replacing the current &dl=1 form while leaving the other URL examples unchanged.
---
Nitpick comments:
In `@src/jobStore.ts`:
- Around line 161-181: Add unit tests covering the Bandcamp resolver lifecycle
through waitForBandcampTrackSelection and resolveBandcampTrackSelection: verify
a pending wait receives its track URL, cancel() resolves it with null, delete()
resolves it with null, and a second wait for the same job resolves the first
waiter with null while leaving the replacement waiter pending.
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro Plus
Run ID: a164c4e4-4992-442c-8331-7d996e4e96c0
📒 Files selected for processing (8)
README.mdsrc/index.tssrc/jobStore.tssrc/server.tssrc/types.tssrc/ytdlp.tswebui/src/components/App.csswebui/src/components/App.tsx
🚧 Files skipped from review as they are similar to previous changes (2)
- README.md
- src/index.ts
Sanitize download filenames, bound fetch size/timeout, share link rules, block private resolve hops, rewrite Dropbox dl=0→dl=1, and test Bandcamp selection lifecycle.
There was a problem hiding this comment.
Actionable comments posted: 1
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
src/directDownload.ts (1)
64-94: 🔒 Security & Privacy | 🔴 Critical | 🏗️ Heavy liftBind SSRF validation to the connection used for every request.
- In
src/directDownload.ts#L73-L87, replace automatic redirects with a manual loop. Validate eachLocationURL before fetching it.- In
src/safeOutboundUrl.ts#L52-L82, do not validate one DNS result and then call hostname-based Bunfetch, which resolves separately. Use a client or connection path that uses the validated address while preserving the original hostname for TLS/SNI. Apply this tosrc/utils.ts#L468-L483as well; its manual redirect loop still has the DNS TOCTOU gap. Do not rely on an unsupported Bun custom dispatcher or connector.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@src/directDownload.ts` around lines 64 - 94, Bind SSRF validation to the actual connection for every request: in src/directDownload.ts lines 64-94, replace fetch’s automatic redirects with a manual redirect loop that validates each Location URL before the next fetch; in src/safeOutboundUrl.ts lines 52-82, change the validated request path to connect through the validated DNS address while preserving the original hostname for TLS/SNI, without using an unsupported Bun dispatcher or connector; also apply the same DNS-TOCTOU fix to the manual redirect loop in src/utils.ts lines 468-483.
🧹 Nitpick comments (2)
src/directLinkRules.ts (2)
34-45: 🎯 Functional Correctness | 🔵 Trivial | ⚡ Quick win
docs.google.comis a "known" host but has no normalization path.
isKnownDirectDownloadHost(lines 9-17) recognizesdocs.google.comas a direct-download host, butnormalizeDirectDownloadParsedUrlonly rewritesdropbox.comanddrive.google.comURLs. A plain Google Docs share link (e.g..../document/d/ID/edit) passed through unchanged will resolve to an HTML edit page, whichDirectDownloadercorrectly rejects with an HTML error — so the failure is graceful, but the feature silently doesn't do what the host allow-list implies it supports.Add an export-URL rewrite for
docs.google.com(mirroring the Driveuc?export=downloadpattern) or drop it fromisKnownDirectDownloadHostuntil normalization exists.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@src/directLinkRules.ts` around lines 34 - 45, Update normalizeDirectDownloadParsedUrl to handle docs.google.com URLs consistently with isKnownDirectDownloadHost: rewrite Google Docs document links containing a document ID to the appropriate export-download URL using the existing Drive normalization pattern, or remove docs.google.com from the allow-list if normalization is not being added. Preserve existing Dropbox and Drive behavior.
20-28: 🎯 Functional Correctness | 🔵 Trivial | ⚡ Quick winTighten the
dl/rawquery-param heuristic.
urlLooksLikeDirectDownloadreturnstruefor anyhttps?://URL that has adlorrawquery parameter (lines 25-26), regardless of host. Many unrelated sites userawordlfor other purposes (CDN cache-busting, image sizing, tracking). When such a URL is actually a legitimate gate page, this heuristic misclassifies it asdirect, bypassesresolveUnknownGateUrl's redirect/HTML-scanning logic, and forces the user through a hard failure inDirectDownloader("returned HTML instead of a file") instead of correctly finding the real gate.Consider requiring the extension match or a known host in combination with the
dl/rawsignal, rather than accepting either alone on an arbitrary host.♻️ Suggested tightening
export function urlLooksLikeDirectDownload(url: URL): boolean { if (!/^https?:$/i.test(url.protocol)) return false; if (isKnownDirectDownloadHost(url.hostname)) return true; if (AUDIO_OR_ARCHIVE_EXT_RE.test(url.pathname)) return true; - // Dropbox preview links often use dl=0; treat any dl= as a download intent. - if (url.searchParams.has('dl')) return true; - if (url.searchParams.has('raw')) return true; + // Dropbox preview links often use dl=0; only treat dl=/raw= as a download + // intent when the path also looks like a file (avoids false positives on + // unrelated HTML pages that happen to use these common query names). + if ( + (url.searchParams.has('dl') || url.searchParams.has('raw')) && + /\.[a-z0-9]{2,4}$/i.test(url.pathname) + ) { + return true; + } return false; }🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@src/directLinkRules.ts` around lines 20 - 28, Update urlLooksLikeDirectDownload so dl/raw query parameters only qualify a URL when combined with an AUDIO_OR_ARCHIVE_EXT_RE pathname match or a known direct-download host. Preserve the existing protocol, host, and extension checks, but remove the unconditional dl/raw acceptance for arbitrary hosts.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@src/directDownload.ts`:
- Around line 96-122: Update the download flow around response.arrayBuffer() to
stream response.body through a byte-counting limit enforced during writes,
aborting when the accumulated bytes exceed MAX_DOWNLOAD_BYTES. Preserve the
existing filename selection and size-error behavior, and avoid relying on
content-length or post-buffer validation as the primary limit.
---
Outside diff comments:
In `@src/directDownload.ts`:
- Around line 64-94: Bind SSRF validation to the actual connection for every
request: in src/directDownload.ts lines 64-94, replace fetch’s automatic
redirects with a manual redirect loop that validates each Location URL before
the next fetch; in src/safeOutboundUrl.ts lines 52-82, change the validated
request path to connect through the validated DNS address while preserving the
original hostname for TLS/SNI, without using an unsupported Bun dispatcher or
connector; also apply the same DNS-TOCTOU fix to the manual redirect loop in
src/utils.ts lines 468-483.
---
Nitpick comments:
In `@src/directLinkRules.ts`:
- Around line 34-45: Update normalizeDirectDownloadParsedUrl to handle
docs.google.com URLs consistently with isKnownDirectDownloadHost: rewrite Google
Docs document links containing a document ID to the appropriate export-download
URL using the existing Drive normalization pattern, or remove docs.google.com
from the allow-list if normalization is not being added. Preserve existing
Dropbox and Drive behavior.
- Around line 20-28: Update urlLooksLikeDirectDownload so dl/raw query
parameters only qualify a URL when combined with an AUDIO_OR_ARCHIVE_EXT_RE
pathname match or a known direct-download host. Preserve the existing protocol,
host, and extension checks, but remove the unconditional dl/raw acceptance for
arbitrary hosts.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro Plus
Run ID: b96f643d-2eab-4058-bcd7-712e9a1baf48
📒 Files selected for processing (7)
src/directDownload.tssrc/directLinkRules.tssrc/jobStore.test.tssrc/safeOutboundUrl.tssrc/utils.test.tssrc/utils.tswebui/src/components/App.tsx
🚧 Files skipped from review as they are similar to previous changes (2)
- webui/src/components/App.tsx
- src/utils.ts
Connect via DNS-validated public addresses with Host/SNI, follow redirects manually, enforce size limits while streaming, and tighten direct-link heuristics.
|
Addressed the latest CodeRabbit round in 7e10bc8:
|
|
@coderabbitai review |
✅ Action performedReview finished.
|
There was a problem hiding this comment.
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (2)
src/safeOutboundUrl.ts (1)
73-83: 🔒 Security & Privacy | 🟠 Major | ⚡ Quick winNormalize IPv6 literals before validation.
url.hostnameincludes brackets, whilenet.isIP()returns0for bracketed IPv6. The code then callsdns.lookup()and rejects every IPv6 literal withENOTFOUND. Strip the brackets once and use the unbracketed value forisBlockedHostname(),isIP(),isPrivateOrLocalIp(), andaddress.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@src/safeOutboundUrl.ts` around lines 73 - 83, Normalize url.hostname by removing IPv6 brackets once before the validation flow, then use the normalized host for isBlockedHostname(), isIP(), isPrivateOrLocalIp(), and the returned address. Preserve existing rejection behavior for blocked or private/local hosts while allowing valid IPv6 literals to bypass DNS lookup.src/directLinkRules.ts (1)
18-41: 🔒 Security & Privacy | 🟡 Minor | ⚡ Quick winReject malformed Google Drive share links before direct download. Valid
/file/d/ID/viewlinks convert correctly, and Dropbox usesURLSearchParams.set('dl', '1'). However, malformed Drive paths are classified as direct and may pass through unchanged. Validate the file ID before normalization.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@src/directLinkRules.ts` around lines 18 - 41, Update normalizeDirectDownloadParsedUrl to validate the Google Drive file ID extracted by the driveFile pattern before converting it to the uc download URL. Reject malformed Drive share paths rather than treating them as direct downloads, while preserving conversion for valid /file/d/ID/view links and existing Dropbox normalization.
🧹 Nitpick comments (1)
src/safeOutboundUrl.ts (1)
59-95: 🔒 Security & Privacy | 🔵 Trivial | ⚡ Quick winNormalize IPv6 brackets in
isBlockedHostname.URL.hostnameretains brackets for IPv6 literals, so unbracketed blocklist entries do not match.isPrivateOrLocalIpcatches these values later, but the blocklist check should normalize brackets directly.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@src/safeOutboundUrl.ts` around lines 59 - 95, Normalize IPv6 brackets from the host value before passing it to isBlockedHostname in resolveSafeConnectTarget, while preserving the original host for subsequent IP validation and error reporting as needed.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Outside diff comments:
In `@src/directLinkRules.ts`:
- Around line 18-41: Update normalizeDirectDownloadParsedUrl to validate the
Google Drive file ID extracted by the driveFile pattern before converting it to
the uc download URL. Reject malformed Drive share paths rather than treating
them as direct downloads, while preserving conversion for valid /file/d/ID/view
links and existing Dropbox normalization.
In `@src/safeOutboundUrl.ts`:
- Around line 73-83: Normalize url.hostname by removing IPv6 brackets once
before the validation flow, then use the normalized host for
isBlockedHostname(), isIP(), isPrivateOrLocalIp(), and the returned address.
Preserve existing rejection behavior for blocked or private/local hosts while
allowing valid IPv6 literals to bypass DNS lookup.
---
Nitpick comments:
In `@src/safeOutboundUrl.ts`:
- Around line 59-95: Normalize IPv6 brackets from the host value before passing
it to isBlockedHostname in resolveSafeConnectTarget, while preserving the
original host for subsequent IP validation and error reporting as needed.
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro Plus
Run ID: f1c66e80-eadb-4c24-8aff-2e1e281df2d0
📒 Files selected for processing (5)
src/directDownload.tssrc/directLinkRules.tssrc/safeOutboundUrl.tssrc/soundcloud.tssrc/utils.ts
🚧 Files skipped from review as they are similar to previous changes (3)
- src/soundcloud.ts
- src/directDownload.ts
- src/utils.ts
Strip IPv6 brackets before IP/blocklist checks, and only treat Google Drive URLs with a valid file id as direct downloads.
|
Addressed the latest CodeRabbit outside-diff notes in this push:
|
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@src/directLinkRules.ts`:
- Around line 58-65: Update the Google Drive normalization branch around
extractDriveFileId to read any input resourcekey via url.searchParams and
preserve it in the returned download URL’s query parameters. Keep the existing
URL format and validation unchanged, and add a regression test covering a link
containing resourcekey.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro Plus
Run ID: 046694f1-2f6d-40ed-bc57-3fc2b80d3a55
📒 Files selected for processing (3)
src/directLinkRules.tssrc/safeOutboundUrl.tssrc/utils.test.ts
🚧 Files skipped from review as they are similar to previous changes (2)
- src/utils.test.ts
- src/safeOutboundUrl.ts
Link-shared Drive files can require resourcekey; keep it when rewriting to uc?export=download.
1 participant
Summary
dl=1, Google Drive, raw audio/archive links) as a browserless download path when pasting a gate URLdirectprovider through CLI, server, and Web UI copyTest plan
dl=1as the manual gate URL and confirm the file lands in./downloadspurchase_urlis an lnk.to/smart-link that lands on Hypeddit and confirm auto-resolutionbun test ./src/utils.test.tsandbun run lintSummary by CodeRabbit
New Features
Documentation
Bug Fixes