Write new applications, extend Nextcloud or integrate other software. With Nextcloud you can go all the way and enhance the server itself or apps from other authors with new API’s and core functionality.
This is wild: @hillai@twitter.com found that apparently in Azure picking "multi-tenant" means that _every Azure user worldwide_ can log in. And that Microsoft had some of Bing's accounts configured as such. And that with that he could manipulate Bing search results. And inject XSS payloads into Bing search results. And could use that XSS to steal office365 tokens.
How Wiz Research found a common misconfiguration in Azure Active Directory that compromised multiple Microsoft applications, including a Bing management portal
Is it a 100 year old phone? Or is it a @Raspberry_Pi hooked into all-new custom electronic internals interfacing with a reverse-engineered mechanical rotary dialler that initiates a set of Python scripts that play archival audio clips into the ear-piece depending what you dial? Yes.
@brecht a recent discussion around this pretty much landed in: this is not functionality suitable for curl to expose like this, so maybe better done separately... so here we are!
CELEBRATING TEN YEARS OF TEENAGE DREAM!!REMASTERED IN 4K!Get “California Gurls” from Katy Perry’s ‘Teenage Dream’: http://katy.to/TeenageDreamKaty Perry Comp...
@freddy this is using libcurl's URL parser so it will be restricted to what parsing powers libcurl provides. Which now is RFC3986 (with extensions) only.
I'd like a more fine-grained way to print several parts of the URL, for example like curl's -w option that allows using percent signs and variable names in braces for a slightly verbose, but printf-like DSL. Very common options like formatting '{scheme}://{host}' may even be given a long flag to print. Does that sound logical enough to submit an issue on GitHub?
for example like curl's -w option that allows using percent signs and variable names in braces for a slightly verbose, but printf-like DSL. Very common options like formatting {scheme}://{host} may...
I like the --redirect option! I'm not aware of any other existing command-line utility to convert base URL plus relative URL to an absolute URL, so I have to resort to a 5-line Python script every time I need to do it in Bash.
After seeing the examples, I'd appreciate slightly more consistency with parameters. Eg, how about --set-path and --get-path, rather than --path and --only-path
The one feature I'm missing most from shell is urlencode/urldecode. Most of the url splitting can be done in pure POSIX shell in just a few lines, and Bash's printf can easily do a urldecode of a properly encoded string (just replace % with \x), but I would love to have a tool that can do both encoding/decoding with any related validation.
Most of the time I use curl's --data-urlencode, ex with -G for url params, but some corner cases can be more difficult like a POST with url params.
🙏 I have wanted something like this a few times. Particularly for parsing environment-variable-supplied database urls, of the kind used by Heroku, in dev/maintenance scripts.
LibreOffice is developed by a worldwide community, and many members of this community live in German-speaking countries. From 24 to 26 March, some of them met at the Linuxhotel in Essen.
The biggest chemical producer in the world, German company BASF is a political powerhouse and has used its influence to push back against chemicals regulation in the EU.
Con 175 votos a favor, 158 en contra y 11 abstenciones, el Congreso de los Diputados ha aprobado hoy una enmienda transaccional de EH Bildu, PSOE y Unidas Podemos para la futura Ley de Industria.
I guess the --url argument would default to the contents of STDIN, applied once to each line? Does it even need to be a named argument? Could be applied to each unnamed argument in turn.
I'd definitely use this for some of my "this bash substitution works, but only if you don't breath on it" mission critical hacky Proof Of Concept scripts.
But perhaps the result can be a single `--output` parameter that takes one or more arguments, with an additional `--format` parameter that can take values such as `json`? That would make it incredibly powerful in various scripting environments.
@jimmysjolund In 2022 no less than 21 CVEs were registered in total for curl, which is the second busiest year so far. Because people are looking really hard to find the cracks. I blogged about it last year: daniel.haxx.se/blog/2022/08/22…
Coming up on Mosen at Large 221, register for our webinar about the future of Mosen at Large, the Be My Eyes Virtual Volunteer, and new speakers from Sonos mosen.org/malppreview0221/
Kia ora Mosen at Largers. Here’s what’s coming up in our next two-hour show that’s got the blind community talking. On 8 April US and UK time, I have an announcement to make about the future of thi…
Did you enjoy our CSUN wrap-up last In-process, and our coverage of David Cane's Task Aware Browsing Proposal? Or are you more into games? What did you think of Guy Barker's #accessible #sudoku game, Grid Game? How have you found it? Or, what is YOUR favourite game? Let us know! Otherwise, read all the details in last week's In-Process blog here: nvaccess.org/post/in-process-2… #CSUNATC23
We’ve got a big issue for you this week, let’s get into it! CSUN Last week, the CSUN Assistive Technology Conference was held in Anaheim, California. The NV Access team were there to share informat…
stormdragon2976 pushed changes to the testing branch of the audiogame-manager project Update action buttons to more closely reflect what they actually do according to the option selected. For example, the game launcher now says “Launch”, the -r options now shows “Remove”, etc.
Just set up a Sliding Sync proxy for my @matrix homeserver so I can use the next generation Element iOS client. Was super easy and wow the new client is not just amazingly fast, from loading content but also decrypting messages, but also has a nice UI and UX. It's under active development and available via TestFlight and boy it's fun to use already :)
Check out their YouTube to see what they've been up to and what's next. FOSDEM was packed with cool talks & demos. youtu.be/5iIs1zWuVOU
This video was recorded during FOSDEM 2023, and can also be found here: https://fosdem.org/2023/schedule/event/matrix_clients_as_good_as_youd_expect/We will ...
Coming up at 00:00 UTC (8:00 PM Eastern, 5:00 PM Pacific in the Americas) on The Global Voice Internet Radio: two hours of relaxing #newage, #ambient, and #meditationmusic on Northern Lights: The New Age Show with Kelly Sapergia. More information about the show is at ksapergia.net/northernlights/. Tune in either by visiting theglobalvoice.info and clicking on the Listen Live link, or put the following URL in your favorite media player: theglobalvoice.info:8000/broad… #TGVRadio #audio
Northern Lights is a weekly live and interactive show featuring the best mix of relaxing music from independent and mainstream artists. Genres you’ll hear include New Age, Ambient, and music …
Oh, the "protocols not platforms" problem again We observed this with other platforms already. It's often a variant of some kind of free-speech idea vs. freedom of expression, often manifested in US or EU centric backgrounds.
And while I kiindda get their point, I'm still for calling scams scams. And for disallowing nazi shit or similar. However, there's an argument to be made that there may come someone else (like Musk with Twitter), or many, who suddenly ban(s) crates by PoC or LGBTQIA*.
In principle the system for pulling dependencies should have no say in this; while at the same time promoting moderation through e.g., federated crate indices. It probably will pan out that the majority of users would "subscribe" to a cryptocurrencies-are-scam index (likely even andrew and jeremy), while technically allowing tech bros to have their own bullshit index. Moderation must be a social issue (discussing which crate index to subscribe to) and solved by social means rather than technical (making the only index "regulate"). Technology must simply enable social measures (e.g., through federation).
“Among the unexpected sources of support for the magazine was the #fediverse—the collection of decentralized Mastodon servers that soared in popularity following Elon Musk’s acquisition of Twitter. Many comments on the GoFundMe page mention that the donors learned about the Observer through #Mastodon—where the Texas Observer was one of the first journalistic outlets to set up its own server, @jamescanup said.”
Robots are being in used in religious rituals worldwide -- often because they're extremely reliable and precise, useful in traditions and rituals that prize proper execution ("orthopraxic")
The anthropologist Holly Waters wonders: What does it mean if robots can be better at religious observances than we humans are?
Welcome to the latest edition of the Linkfest! Thank you for being a subscriber; if you’re enjoying it, spread the word – it’s a pay-what-you can signup...
LfDI stellt Datenschutzverstoß beim Innenministerium fest. Innenministerium akzeptiert rechtliche Bewertung.
@lfdi: „Das Verfahren ist mit der Feststellung eines Datenschutzverstoßes für uns abgeschlossen. Wir gehen davon aus, dass das Innenministerium künftig den Personaldatenschutz einhält.“
Verfahren bezüglich der Übermittlung eines Anwaltsschreibens zu einem Disziplinarverfahren durch das Innenministerium an einen Pressevertreter: LfDI stellt Datenschutzverstoß fest und schließt Verfahren ab.
It's been a while, so a number of things happened.
------
You've probably already seen it, since it was in the last TWIG, but @jamie added a way to create AdwAboutWindow from AppStream data. In particular, this should make release notes page far easier to work with - the release notes already support appstream markup and now it can just pull them from appdata directly rather than having to copy-paste them, and unlike application name, developer name etc they tend to be fairly long.
It doesn't unfortunately fill everything - for example, copyright line or credits still have to be specified manually since AppStream doesn't have that data in the first place. But it still considerably reduces the amount of copy-pasted information.
One part that is a bit awkward is that the appdata xml needs to be in gresource (the original plan was get the installed file by path, but it's pretty hard to do correctly in practice), and meson doesn't make easy to pull translated xmls into gresource - but it is doable.
And well, it means that libadwaita depends on libappstream now.
------
To reduce confusion a bit, @brainblasted deprecated adw_expander_row_add_action() and added adw_expander_row_add_suffix() instead - matching action rows instead. We really should have done it back in 1.0, but didn't for some reason - better late than never.
------
For sidebars, Chris also added a property that hides AdwHeaderBar title, incl. the generated one that's pulled from the window or browsing view. This should make it a lot easier to have a title in a sidebar only when folded, once the new sidebar widgets land.
------
A new contributor added AdwSwitchRow - a simple AdwActionRow subclass with a GtkSwitch. While this isn't too difficult to implement manually, it's a really common case and it gets annoying, so having a helper is nice. Another nicety is that it doesn't expose the GtkSwitch:state property, only :active - meaning it's much harder to misuse it.
------
@mks_h added a description property to AdwPreferencesPage - it looks same as preference group description, but shown at the very top of the page, making some of the Settings mockups easier to implement.
------
I finally finished and landed AdwToolbarView. We also did a few changes to toolbar styles, so you may notice the doc screenshots look a bit different now. ^^
This one I'm not announcing in TWIG yet, since the rest of the new widgetry is not there yet, so it's not as useful yet as it will be later, once the new sidebar widgets have landed.
A change that's sort of related to toolbar view, but also not really, is that the new header bar style with a shadow is now also used for GtkWindow titlebars. While we can't use it for any random header bars in a GtkBox, we can do it here. This does have a few side effects though:
- If an app is using a header bar as a window titlebar, but has, say, a GtkSearchBar within the window, the header bar shadow will be above the search bar. It's not a big problem in practice since very few apps do this (I checked every core app and every third party app I have installed and only found 2 instances), and it doesn't do much harm, just looks a bit weird - but still a thing to keep in mind.
- Late in the last cycle we changed GtkPopoverMenuBar style to use a header bar-like style as well, with the same colors and border, and a backdrop style. Well, it is quite common to use menu bars together with a header bar set as a GtkWindow titlebar as well, particularly with the automatic one GtkWindow adds by default. So, I just reverted that change and the menubars look plain and boring once again. If somebody wants them to look fancy and blend with the titlebar, AdwToolbarView does support them, so you can add a header bar and a menubar within one and it will look nice. There is even an example of that in libadwaita tests.
For adaptive states I extracted the code that makes them work from AdwWindow and into a reusable widget that's currently called AdwAdaptiveBin.
It works the same way as adaptive states on the window - you add one or more adaptive states, set their conditions add setters changing properties on the child widget, and they are safely applied on resize, with the tradeoff that the the widget has no minimum size and you must set it manually using the GtkWidget's width-request and height-request properties. If you don't, it will print a critical telling you that every time it's allocated, so hopefully people won't forget to do it.
Currently it's public in the branch, I'm not sure I'll keep it that way since a part of the idea for the new mechanism was to do it on the window so that it can all be done in one step - but at the very least it made the code a lot cleaner, so the widget is gonna stay even if it's gonna be private.
------
On GTK side I went ahead and tested gtk4-demo with all buttons being able to ellipsize. It... does not work well, it's a pretty big behavior change so it definitely needs to be an opt-in property. But that's fine.
What's less fine is that GtkCenterLayout prioritizes the center child over start/end. In practice it means that if you have, say, a dialog with the following header bar:
[ Cancel ] Add a New World Clock [ Add ]
and the title doesn't completely fit, the buttons will ellipsize first, and you will end up with this:
[ … ] Add a New World Clock [ … ]
and only then it will start ellipsizing the title:
[ … ] Add a New… [ … ]
That's not great, it should do the opposite here, first fully ellipsizing the title and only then ellipsizing buttons if it doesn't fit even then. I guess GtkCenterLayout needs a way to control this, because there are situations where prioritizing the center child makes sense. But that's something for another week.
------
The :pop-content-on-fold property on AdwDualPaneView is actually implemented now, so there is an easy way to make it immediately show sidebar instead of content when folding, same as how when visible-child is set to the sidebar right now.
------
As I mentioned in neurodifferent.me/@alexm/10998…, I started working on a widget that can replace AdwFlap for the utility pane case, but didn't have a good name for it. Well, I still don't have a good name and suggestions are still welcome but the widget is now in the merge request as well, and AdwHeaderBar has learned to manage window buttons on it just like it does with AdwDualPaneView.
One thing it doesn't do for these widgets is handle nesting. It probably should implement it though.
Some signatories of the open letter to pause AI training walk back their positions, others turned out to be fake, and many experts disagree with its proposal (Chloe Xiang/VICE)
La #InteligenciaArtificial como tal no existe 🤖 Es un concepto publicitario. Son herramientas máquinas y técnicas desarrolladas y manejadas por humanos. Y es el resultado del trabajo mal pago e invisible de millones de personas en todo el mundo
Recent updates to Mutter (included in #GNOME44!) deliver improved performance for even the most demanding apps, including games and interactive 3D apps.
It will be included in the upcoming Fedora 38 release (which should be sometime in April) as will Ubuntu 23.04. And GNOME 44 builds already landed in openSUSE's Tumbleweed and MicroOS
it is curious to me to follow from a distance the way in which audio and video handling (on linux, and more broadly) have been both converging yet not quite arriving in the same place.
It's interesting to thnk about the differences: more computational load for visual display, but more sensitivity to timing errors for audio (due to human sensory mechanisms).
read the blog post and couldn't stop wondering "why don't they just double buffer as simply as every audio API on every platform?"
@PaulDavisTheFirst Because adding more buffers means keeping the GPU and CPU at a higher clock setting, which consumes more power. You can do it for some things, like an application, but not for everything, otherwise your power consumption and battery life will just be miserable
@ebassi no, not adding more buffers. Just accepting that you only ever have two, and the one not being used by the video hardware MUST be ready by the next vsync (or its equivalent). No games, just deadlines.
since i work on production software, power&battery issues tend to be of less consequence in the way i think about this stuff.
One piece of slang that has long embodied the short attention span Internet age is TL;DR, short for “too long; didn’t read.” With the explosion of generative AI tools, we’re rapidly entering the age of TL;DW: “too long, didn’t write.
Klare Absage zur #Vorratsdatenspeicherung 🎉 Ein Generalverdacht gegenüber allen Menschen und damit ein Einschnitt in ihre Privatsphäre und ihr Recht auf anonyme Internetnutzung ist und bleibt grundrechtswidrig. #PIRATEN #fckvds
"Wenn ein Smartphone durch einen Akkudefekt auf dem Müll landet, sollte die Bezeichnung „Smart“ hinterfragt werden."
Wo soll ich unterschreiben? Besser kann man es kaum sagen. Vollste Zustimmung!
Passend dazu freue ich mich schon auf das SHIFTphone8. Bei dem ich den (Reserve-) Akku des SHIFT6mq gleich weiter verwenden kann. Ist zwar genau umgekehrt dann (Akku noch heile; 6mq aber auch), aber ebenso nachhaltig!
@mat im Sinne der Nachhaltigkeit möchte ich nur so viele bestellen, wie ich auch wirklich benötige. Meine Bestellung ist schon bei @shiftphones angekommen und die Summe beglichen.
Having dipped my toe in making Generative AI, I can say both are equally likely. Generative AI simply picks the next most common thing to say based on the context of the prompt and the prior words generated. It's prone to not stating the truth. But Google is also prone to lying and using peoples' data without their explicit consent of the use case.
If you love the #Firefox Translations add-on as much as we do and find yourself thinking "this should REALLY come to Thunderbird," then this bug is worth keeping an eye on:
We observed this with other platforms already. It's often a variant of some kind of free-speech idea vs. freedom of expression, often manifested in US or EU centric backgrounds.
but the widget is now in the merge request as well, and AdwHeaderBar has learned to manage window buttons on it just like it does with AdwDualPaneView.
Brecht Savelkoul
in reply to daniel:// stenberg:// • • •daniel:// stenberg://
in reply to Brecht Savelkoul • • •Brecht Savelkoul
in reply to daniel:// stenberg:// • • •Natanael ⚠️
in reply to daniel:// stenberg:// • • •Jan Ainali
in reply to daniel:// stenberg:// • • •daniel:// stenberg://
in reply to daniel:// stenberg:// • • •A better name? · Issue #1 · curl/urler
GitHubTomas Ekeli
in reply to daniel:// stenberg:// • • •ijk64✅
in reply to daniel:// stenberg:// • • •V Martín
in reply to daniel:// stenberg:// • • •Rasmus Malver
in reply to daniel:// stenberg:// • • •Katy Perry - California Gurls (Official Music Video) ft. Snoop Dogg
YouTubeSander Steffann
in reply to daniel:// stenberg:// • • •moved to @amanjeev@hachyderm.io
in reply to daniel:// stenberg:// • • •“purl”.
long story - curl brings “earl” to mind but its the worker so a peasant so maybe “p” from that and “url”.
sorry.
Sean McArthur
in reply to daniel:// stenberg:// • • •Frederik Braun �
in reply to daniel:// stenberg:// • • •daniel:// stenberg://
in reply to Frederik Braun � • • •Tunapunk
in reply to daniel:// stenberg:// • • •daniel:// stenberg://
in reply to Tunapunk • • •daniel:// stenberg://
in reply to daniel:// stenberg:// • • •more fine-grained output option · Issue #10 · curl/urler
GitHubJérémy Lecour
in reply to daniel:// stenberg:// • • •Alexey Shpakovsky
in reply to daniel:// stenberg:// • • •--redirectoption! I'm not aware of any other existing command-line utility to convert base URL plus relative URL to an absolute URL, so I have to resort to a 5-line Python script every time I need to do it in Bash.Sjlver
in reply to daniel:// stenberg:// • • •looks useful!
After seeing the examples, I'd appreciate slightly more consistency with parameters. Eg, how about --set-path and --get-path, rather than --path and --only-path
Thanks for your work!
Thomas Guyot-Sionnest
in reply to daniel:// stenberg:// • • •The one feature I'm missing most from shell is urlencode/urldecode. Most of the url splitting can be done in pure POSIX shell in just a few lines, and Bash's printf can easily do a urldecode of a properly encoded string (just replace % with \x), but I would love to have a tool that can do both encoding/decoding with any related validation.
Most of the time I use curl's --data-urlencode, ex with -G for url params, but some corner cases can be more difficult like a POST with url params.
Ben Cotton (he/him)
in reply to daniel:// stenberg:// • • •Chris Bandy
in reply to daniel:// stenberg:// • • •I thought about the interface for a similar tool some time ago. I'll port some issues over later today.
github.com/cbandy/uri
GitHub - cbandy/uri: Create URIs in a CLI
GitHubdaniel:// stenberg://
in reply to Chris Bandy • • •heyda.tech
in reply to daniel:// stenberg:// • • •daniel:// stenberg://
Unknown parent • • •Michal Vyskočil
in reply to daniel:// stenberg:// • • •David JONES
in reply to daniel:// stenberg:// • • •Brad
in reply to daniel:// stenberg:// • • •MrCopilot
in reply to daniel:// stenberg:// • • •Can I nominate:
mniu
MyNameIsUrl
Luci Callous Thumb
in reply to daniel:// stenberg:// • • •