During yesterday's stream, it was brought to my attention that Heroku
deployments with the default blank value set for custom CSS causes a
bizarre appearance (all black and white with missing UI elements).
Setting the custom css variable to the default seems to fix this
problem.
If the alt for a particular service is blank, the original source is
used instead.
Example:
1. Site alts enabled in config
2. User wants wikipedia links, not wikiless
3. WHOOGLE_ALT_WIKI set to ""
4. All available alt links redirected to farside, except wikipedia
Fixes#704
Bibliogram uses a slightly different URL format than Instagram, and
requires a "u/" before the username when replacing Instagram links. This
was already implemented everywhere else except the Dockerfile.
Old iOS 12 devices will pass the Referer HTTP header to the site user clicks.
Websites will know those traffic come from Whoogle search.
Adding "nofollow noopener noreferrer" solves the issue.
The time filter (past day/hour/month/etc) was using the result element
background color instead of the page background color, which wasn't
providing enough contrast with the default text color.
After an uptime of 109 days, the usage of /var/lib/tor was still 10.9 MB. A
reply in issue #648 reported a higher usage, which was fixed by setting the
size a bit higher (12MB instead of 11MB).
Wikipedia, imgur, and translate alternatives were all still using
hardcoded URLs when replaced with their respective alternative frontend.
This updates them to use farside instead.
Links that were directed at G domains were previously removed
universally, when really they only needed to be removed from the footer
to reduce possible confusion caused by mixed Whoogle and G links.
Fixes#656
Bang searches without an actual query (i.e. just searching "!gh") will
now redirect to the home page. I guess people do this for some reason
and don't like that it redirects to the correct bang result URL, but
without an actual search term.
Fixes#595
Recent changes to ads in search results caused Whoogle to display ads
for certain searches. In particular, ads recently started appearing
grouped into one div, as opposed to a singular ad per div. This was
accompanied by the div label "ads" (instead of just "ad"), which threw
off the existing ad filter. The ad keyword blacklist has been updated
accordingly, and has been enhanced to only check against alpha chars for
each label.
This only seems to have affected English language searches, and only for
very specific searches.
If the config dir already exists, setting the mode (`-m 777`) doesn't
actually work as it should. This change splits the command into two
separate commands for directory creation and enabling the directory to
be writable by all.
Fixes#658
Currency amounts returned by google seem to randomly include unicode
chars ('\xa0' noted in #642) which broke the currency calculator
included in the project. This ensures that only strings that can be
converted to float are ever used in the conversion.
Fixes#642
Rather than only checking for an available update on app init, the check
for updates now performs the check once every 24 hours on the first
request sent after that period.
This also now catches the requests.exceptions.ConnectionError that is
thrown if the app is initialized without an active internet connection.
Fixes#649
Removes dependency on class names for creating the "my ip" info card in
the results list for searches pertaining to the user's public IP.
Adds test to prevent this from happening again.
Note to anyone reading this and looking to contribute: please avoid
using hardcoded class names at all costs. This approach of
creating/removing content just results in issues if/when Google decides
to introduce/remove class names from the result page.
Fixes#657
Also removes fosshost instance from readme
From @benbusby:
I'm unable to get in touch with fosshost support about the whoogle
instance being unavailable, and am no longer interested in
maintaining the instance due to the lack of communication.
Introduces a header for switching between result types (i.e. "All", "News",
etc) that is consistent between the different result types. Previously, image
results had a tab header that was formatted in a drastically different manner,
which was jarring when switching from a different result page to the Images
page.
Created a G class enum to reference class names returned in search
results. As noted in the class doc, this should only be used/updated as
a last resort, as class names change frequently. For some instances,
such as replacing the tbm tab, it's a lot easier to just replace by
header name than attempting to replace it based on how the element is
structured.
Also updated a few styles to revert the latest styling changes being
applied by Google.
Co-authored-by: jacr13 <ramos.joao@protonmail.com>
Co-authored-by: Ben Busby <contact@benbusby.com>
Google's latest formatting changes broke the modifications made when enabling
`WHOOGLE_MINIMAL`. This updates the result filtering to work with the new
changes.
Fixes#634
min-width was previously set to 736px for all screen sizes, which forced
content off screen for smaller devices such as mobile phones. This
modifies the search stylesheet to only apply a min-width style to
devices > 800px wide.
There have been some recent formatting changes made by Google for search
results that do not look good (especially for dark themes). This
mostly overrides those styles to resemble the original Whoogle
result formatting.
The fosshost team decommissioned the region that Whoogle was hosted in,
but hasn't provided an option to transfer the domain record to the new VM. Until
that is fixed, the instance is inaccessible.
Introduces a new 'scan' workflow for scanning the main branch container for
vulnerabilities nightly. By default, this will fail for any 'medium' or higher
vulnerability.
Fixes#613
Depending on bash wasn't strictly necessary, as the two minimal scripts
in the repo were both nearly POSIX anyways.
Aside from simplifying the repo's dependencies a little bit, this also
helps reduce the overall Docker image size as an added bonus.