thegeneralist01/archivr
1
Fork 0
mirror of https://github.com/thegeneralist01/archivr synced 2026-07-22 03:05:32 +02:00
Code Issues Activity

feat(core): add via_freedium to CaptureConfig; route WebPage captures through Freedium mirror

Browse source 
When via_freedium is true and the locator is not already a Freedium URL,
perform_capture passes https://freedium-mirror.cfd/<original-url> to
singlefile::save() so paywalled articles are fetched via the mirror.
The original locator is kept for requested_locator and canonical_locator
in the DB entry. An empty cookie map is used for the mirror fetch to
prevent original-domain credentials from being sent to freedium-mirror.cfd.
This commit is contained in:
TheGeneralist 2026-07-19 12:09:30 +02:00
parent 289037235c
commit a926d7cf2d
Signed by: thegeneralist01
SSH key fingerprint: SHA256:pp9qddbCNmVNoSjevdvQvM5z0DHN7LTa8qBMbcMq/R4
1 changed files with 14 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

15
crates/archivr-core/src/capture.rs
View file

@ -87,6 +87,9 @@ pub struct CaptureConfig {
pub reader_mode: bool, pub reader_mode: bool,
/// Override for modal-closer browser-script behavior during WebPage captures. /// Override for modal-closer browser-script behavior during WebPage captures.
pub modal_closer_enabled: Option<bool>, pub modal_closer_enabled: Option<bool>,
/// Route WebPage captures through the Freedium mirror to bypass paywalls.
/// The original locator is still recorded in the DB; only the fetch URL changes.
pub via_freedium: bool,
} }
/// Resolves which cookies apply to `url` by evaluating all rules in ordinal order. /// Resolves which cookies apply to `url` by evaluating all rules in ordinal order.
@ -1038,7 +1041,17 @@ pub fn perform_capture(
// Source: web page — archive as a self-contained HTML snapshot via single-file-cli // Source: web page — archive as a self-contained HTML snapshot via single-file-cli
if source == Source::WebPage { if source == Source::WebPage {
match downloader::singlefile::save(locator, store_path, &timestamp, &cookies, config.ublock_enabled, config.cookie_ext_enabled, config.reader_mode, config.modal_closer_enabled) { // When via_freedium is enabled and the URL is not already a freedium mirror,
// fetch through the mirror to bypass paywalls. Store the original locator in DB.
// Use an empty cookie jar for the mirror URL: cookies resolved for the original
// domain (e.g. NYT, Medium) must not be sent to freedium-mirror.cfd.
let (fetch_url, fetch_cookies): (String, HashMap<String, String>) =
if config.via_freedium && !locator.starts_with("https://freedium-mirror.cfd/") {
(format!("https://freedium-mirror.cfd/{}", locator), HashMap::new())
} else {
(locator.to_string(), cookies.clone())
};
match downloader::singlefile::save(&fetch_url, store_path, &timestamp, &fetch_cookies, config.ublock_enabled, config.cookie_ext_enabled, config.reader_mode, config.modal_closer_enabled) {
Ok(result) => { Ok(result) => {
let file_extension = ".html".to_string(); let file_extension = ".html".to_string();
let temp_html = store_path let temp_html = store_path