Our frontend requests the `/me` pathname in order to determine whether
it's logged in or not. Due to the fact that the sameSite attribute of
the session cookie was set to `strict` in a previous commit, the session
token was no longer sent along with HTTP calls initiated by JS. This is
due to the RFCs definition of "safe" HTTP calls in RFC7231.
The bug triggers the UI to show up like an unauthenticated user, even
after a successful login. In order to debug it a look into the send
cookies to the `/me` turned out to be very enlightening.
The fix this patch implements is rather simple, it replaces the sameSite
attribute to `lax` which enables the cookies for those requests again.
Some older and mobile clients were unaffected by this due to the lack of
implementations of sameSite policies.
References:
https://tools.ietf.org/html/rfc7231#section-4.2.1https://tools.ietf.org/html/draft-ietf-httpbis-rfc6265bis-05#section-5.3.7.1https://developer.mozilla.org/en-US/docs/Web/HTTP/Headers/Set-Cookie/SameSitee77e7b165a
Signed-off-by: Sheogorath <sheogorath@shivering-isles.com>
We enabled the `secure` flag for various cookies in previous commits.
This caused setups behind reverse proxies to drop cookies as the nodejs
instance wasn't aware of the fact that it was able to hand out secure
commits using an insecure connection (between the codimd instance and
the reverse proxy).
This patch makes express, the webserver framework we use, aware of
proxies and this way re-enabled the handing out of cookies. Not only the
cookie monster will enjoy, but also functionality like authentication
and real-time editing will return as intended.
References:
https://www.npmjs.com/package/express-session#cookiesecure383d791a50
Signed-off-by: Sheogorath <sheogorath@shivering-isles.com>
The socket.io cookie doesn't really have any purpose as it's no longer
user in modern socket.io versions. This patch disables it.
References:
https://github.com/socketio/socket.io/issues/2276
Signed-off-by: Sheogorath <sheogorath@shivering-isles.com>
While HSTS should take care of most of this, setting cookies to be
secure, and only applied on same site helps to improve situations where
for whatever reason, downgrade attacks are still a thing.
This patch adds the `sameSite` and `secure` to the session cookie and
this way prevent all accidents where a browser may doesn't support HSTS
or HSTS is intentionally dropped.
Reference:
https://www.npmjs.com/package/express-session#cookiesecure
Signed-off-by: Sheogorath <sheogorath@shivering-isles.com>
According to https://github.com/socketio/socket.io/issues/2276 this cookie is not used for anything. To avoid browser warnings about the sameSite attribute, we disable it here.
Signed-off-by: David Mehren <dmehren1@gmail.com>
Modern browsers do not support (or will stop supporting) sameSite: none (or no sameSite attribute) without the Secure flag. As we don't want everyone to be able to make requests with our cookies anyway, this commit sets sameSite to strict. See https://developer.mozilla.org/de/docs/Web/HTTP/Headers/Set-Cookie/SameSite
Signed-off-by: David Mehren <dmehren1@gmail.com>
Because of circular import problems, this commit also moves the error messages from response.js to errors.js
Signed-off-by: David Mehren <dmehren1@gmail.com>
Thanks to our great translators that made it to translate the major
parts of CodiMD into Arabic!
Signed-off-by: Sheogorath <sheogorath@shivering-isles.com>
There was some awesome work by Hồng in the recent days who translated
CodiMD completely into Vietnamese language! This patch provides this
awesome contributions.
Signed-off-by: Sheogorath <sheogorath@shivering-isles.com>
Thanks for the work of the translator Vladan we got a serbian
translation added! Those few changes will add serbian language support
for future CodiMD releases.
Signed-off-by: Sheogorath <sheogorath@shivering-isles.com>
It seems like the inital work on the hsts module expected milliseconds.
This has either changed or was never true. Either way, it caused that
the current defaults resulted in theory in a 1000 year HSTS policy.
Luckily helmet was smart enough to not go higher than 1 year.
Anyway, this patch fixes the multiplication of the configured size with
1000 by removing this multiplication.
Also to simplify the reading of the defaults, we split them into their
components, 60 times 60 seconds so we get one hour. 24 of those hours so
we get a day and finally 365 days to get our original wanted default of
one year.
Reference:
d69d65ea74
Signed-off-by: Sheogorath <sheogorath@shivering-isles.com>
During the upgrade of winston in
c3584770f2 a the class extension for
streaming was removed.
This caused silent crashes. Somehow winston simply called
`process.exit(1)` whenever `logger.write()` was called. This is really
bad and only easy to debug because of the testing right after upgrading.
However, reimplementing the stream interface as it was, didn't work, due
to the fact that `logger.write()` is already implemented and causes the
mentioned problem. So we extent the object with an `stream` object that
implements `write()` for streams and pass that to morgan.
So this patch fixes unexpected exiting for streaming towards our logging
module.
References:
https://www.digitalocean.com/community/tutorials/how-to-use-winston-to-log-node-js-applicationsc3584770f2https://stackoverflow.com/a/28824464
Signed-off-by: Sheogorath <sheogorath@shivering-isles.com>
Currently we only provide the version from `package.json`. This means
that during updates of instances, e.g. the demo instance, which runs
latest master instead of a stable release, changes are not reflected to
the webclient.
This patch adds a fullversion string that contains the current commit
and this way makes that clients are notified about changes.
Signed-off-by: Sheogorath <sheogorath@shivering-isles.com>
`uws` was deprecated by its maintainer and starts to cause more and more
problems and issue reports. So it's time to replace it and use a
maintained project instead. Lucky us, `uws` and `ws` can be used in an
identical way, without problems. To provide better performance, we
install the optional packages as well.
Signed-off-by: Sheogorath <sheogorath@shivering-isles.com>
A little minor change, by moving the CodiMD version header in its own
middleware. Should simplify to determine the version number of the
Backend in future.
Signed-off-by: Sheogorath <sheogorath@shivering-isles.com>
Since static path is providing with a high expiration data, we provide
configs via API. This shouldn't add any noticeable load while making it
uncached and this way working again.
Signed-off-by: Sheogorath <sheogorath@shivering-isles.com>
Since the original idea of using a symlink didn't work anyway, we should
remove the zh.json symlink from the repo. It doesn't provide any
benefit but alters the repo on start of HackMD.
Signed-off-by: Sheogorath <sheogorath@shivering-isles.com>
This translation was contributed via POEditor by the user Basix.
Thanks a lot for your work!
Signed-off-by: Sheogorath <sheogorath@shivering-isles.com>
This commit should prevent the i18n module from adding missing
translations to the local files in setups that are not for development.
This way we keep the directory clean and idempotent.
Signed-off-by: Sheogorath <sheogorath@shivering-isles.com>
It's sad but it's not working. For multiple releases this should be
already broken which shows how often it's used.
As there is also a security issue related to that, it's better to
remove the feature completely. Whoever wants to rewrite it, feel free to
go.
This commit removes the Google Drive integration from HackMD's Frontend
editor and this way removes the need to provide any API key and Client
ID in the frontend.
Signed-off-by: Sheogorath <sheogorath@shivering-isles.com>
As recently discovered we send the clientSecret to the webclient which
is potentionally dangerous. This patch should fix the problem and
replace the clientSecret with the originally intended and correct way to
implement it using the API key.
Signed-off-by: Sheogorath <sheogorath@shivering-isles.com>
This refactors the configs a bit to now use camel case everywhere.
This change should help to clean up the config interface and make it
better understandable.
Signed-off-by: Sheogorath <sheogorath@shivering-isles.com>
This commit adds a referrer policy to all requests.
The usage of `same-origin` allows HackMD to still interpret all requests
and this way not break anything. But it prevents 3rd party scripts,
pictures and more to get informations that may lead to not secured note.
It has to be mentioned that this maybe breaks some features of the
Google Analytics embedding. This has to be tested.
Fixes#724
Signed-off-by: Sheogorath <sheogorath@shivering-isles.com>
