* generate ebook
* serve ebook file
* Update ebook.go not download same image twice anymore
* download ebook request api/ui part
* fix typo
* add stylesheet
* update hasEbook status
* download link update after ebook generate
update bookmark command in ui
* download ebook with bookmark title
* Apply suggestions from code review for better error handling
Co-authored-by: Felipe Martin <812088+fmartingr@users.noreply.github.com>
* Update internal/view/js/page/home.js fix typo
Co-authored-by: Felipe Martin <812088+fmartingr@users.noreply.github.com>
* import error lib and retuen missing error
* move ebook download action to update cache
* replace io/ioutil with io
* add missing error handling
* update Archive now always update ebook
* replace panic error with 404
* remove ebook with delete action
* add download ebook link to content page
* remove tags that not work correctly right now
* if file is pdf not generate ebook
* update style.css
* Revert "update style.css"
This reverts commit 519e10d6ce.
* remove download limit for api
* fix missing fmt.Errorf and change to errors.Wrap
* fix double panic
* return 404 if bookmark not exist
* change function name to GenerateEbook
* not isFatalErr anymore
* add unit test
* remove uneeded field for unit test
---------
Co-authored-by: Felipe Martin <812088+fmartingr@users.noreply.github.com>
This commit fixes URL malformation when trying to remove the utm social
query parameters from an URL, which upon finishing and reconstructing
would attach the equal symbol even if the original URL didn't have it.
This is a known Go "bug" [1] that isn't going to be "fixed". I quote
that because server side should behave the same for `?a=&b=1` and
`?a&b=1` for the `a` parameter, but sometimes that's not the case.
[1]: https://github.com/golang/go/issues/20820Fixes#409
* feat: async content download when creating via api
Invoking the content download code in a goroutine after saving the
bookmark, this way we can return a response to the user quickly while
the webpage is donwloaded and archived.
Cache api endpont (/api/cache) remains untouched until I understand
the logic behind it.
Also updated the API endpoint for the extension, though I'm unsure why
there's a difference between the "regular" API and the webext API,
they should be using the same APIs.
