This approach is neat for observability, but it's worth noticing that it essentially quantises all of your samples down to the vertical resolution of your graph. If you somehow introduced a bug that caused an error that was smaller than the step size then these tests wouldn't catch it.
(e.g. if you somehow managed to introduce a constant DC-offset of +0.05, with the shown step size of 0.2, these tests would probably never pick it up, modulo rounding.)
That said, these tests are great for asserting that specific functionality does broadly what it says on the tin, and making it easy to understand why not if they fail. We'll likely start using this technique at Fourier Audio (shameless plug) as a more observable functionality smoke test to augment finer-grained analytic tests that assert properties of the output waveform samples directly.
That's true that it quantizes (aka bins) the samples, so it isn't right for tests that need to be 100% sample-perfect, at least vertically speaking. I suppose it is a compromise between a few tradeoffs - easy readability just from looking at the code itself (you could do images, but then there's a separate file you have to keep track of, or you're looking at binary data as a float) vs strict correctness. The evaluation of these tradeoffs would definitely depend on what you're doing, and in my case, most of the potential bugs are going to relate to horizontal time resolution, not vertical sample depth resolution.
If the precise values of these floats is important in your domain (which it very well may be), a combination of approaches would probably be good!
Would love to hear how well this approach works for you guys. Keep me updated :)
I'm not sure it makes sense to separate "vertical" correctness from "horizontal" correctness when it comes to "did the feature behave" though; to extend the example in TFA, if your fade progress went from 0->0.99 but then stopped before it actually reached 1 for some reason, you might find that you still had a (small, but still present) signal on the output, which, if the peak-peak amplitude was < 0.1, the test wouldn't catch.
Obviously any time you're working with floating-point sample data the precise values of floats will almost always not be bit-accurate against what your model predicts (sometimes even if that model is a previous run of the same system with the same inputs as in this case); it's about defining an acceptable deviation. I guess what I'm saying is that for audio software, a peak-peak error of 0.1 equates to a signal at -20 dBFS (ref DBFS@1.0) (which of course is quite a large amount of error for an audio signal), so perhaps using higher-resolution graphs would be a good idea.
Fair points here. Unfortunately adding more vertical resolution starts to get a little unwieldy to navigate through. Maybe it could start using different characters to multiply the resolution to something sufficiently less forgiving of errors. If it could choose between even 3 chars, for example, it would effectively squash 3 possible values into one line, tripling the resolution.
I think more resolution may give you more false negatives, which might not be helpful. We’ve used similar tools for integration testing at work and the smallest usually irrelevant change can bust the reference cases, due to the high-detail in the reference, which means going through all the changed tests and then seeing that everything is still fine.
For this, just thinking about sound, I wonder if you could invert the reference wave form and add it to the test to see how well it cancels? Then instead of just knowing there was a diff, you could get measurements of the degree of the diff.
A more accurate and only slightly more complex process for this is to generate numerical text representations of the desired test waveforms and then feed them through sox to get actual wave files. The numerical text representations are likely even easier to generate programmatically than the ascii->audio transformation.
Imagine if we had terminals that could handle graphical data. We wouldn't have to do weird kludges like this, we could just plot the waveforms in the output of our tools.
But it's 2021, and not only is this not possible, there is not even a path forward to a world where this would be possible. It's just not an option. Nobody is working on this, nobody is trying to make this happen. We're just sitting here with our text terminals, and we can't even for a second imagine that there could be anything else.
I would point out that sixels exist. There is a nice library, libsixel for working with it, which includes bindings into many languages. If the author of sixel-tmux is to be believed, the relative lack of adoption is a result of unwillingness on the part of maintainers of some popular open source terminal libraries to implement sixel support.
I can't comment on that directly, but I will say, it's pretty damn cool to see GnuPlot generating output right into one's terminal. lsix is also pretty handy as well.
But yeah, I agree, I'm not a fan of all the work that has gone into "terminal graphics" that are based on unicode. It's a dead-end, as was clear to DEC even back in '87 (and that's setting aside that the VT220 had it's own drawing capabilities, though they were more limited). Maybe sixel isn't the best possible way of handling this, but it does have the benefit of 34 years of backwards-compatibility, and with the right software, you can already use it _now_.
> If the author of sixel-tmux is to be believed, the relative lack of adoption is a result of unwillingness on the part of maintainers of some popular open source terminal libraries to implement sixel support.
If you have any doubt, look no further than this thread: the sixel format is attacked not for any technical reasons, but for its age, RIGHT HERE ON HN:
>> "That's a protocol that's a good forty years old, and even that is not supported. And I can see why, why on earth would you want to be adding support for that in 2021? What a ridiculous state of affairs."
What's ridiculous is, with so many examples and quotes, some people still thing I must be "emotional" (I had a long discussion here... https://news.ycombinator.com/item?id=28761043 ) or that a few million colors is not sufficient for the terminal (!)
There is none so blind as those who will not see...
Sixels are fun, but I was disappointed by libsixel. It’s not really a general–purpose library; most of it is there only to implement various command–line arguments of img2sixel. Most of the functions determine what to do by parsing strings taken from the command–line arguments, so reusing it is super annoying.
When implementing a program that outputs sixels, you are better off looking elsewhere. SDL1.2-SIXEL is a good choice in general, if you are writing C or don’t mind using the C bindings for your preferred language.
I don’t see why sixels couldn’t work. You’d probably want a tool to decode them, diff the images, and then output another sixel image. I’m admittedly not sure of such a tool existing off the shelf though.
I’m not aware of text editors supporting sixels, which could make preparing the tests a challenge. Certainly, you could imagine a text editor supporting them, but I’m not aware of one that does personally.
I will concede that for your specific use case, an off the shelf ASCII plotting library probably involves less custom tooling.
The problem is NOT THE FORMAT, the problem is the lack of tooling: links and w3m are among the rare text browsers that can display images in the console.
It's just a matter of the browser sending the image to the terminal in some format it can understand, but if that hasn't be thought about as a possibility (say, for text reflow issues) it's going to be far more complicated than just adding a new format, as you will have to work both on say the text reflow issues (ex: how do you select the size of the placeholder, when expressed in characters?), on top of the picture display issues.
Said differently, it would be easier to have console IDE that supported graphics if any format whatsoever (sixel, kitty...) was supported by a console IDE; we could then argue about the ideal format.
Arguing about the ideal format BEFORE letting the ecosystem grow using whatever solution there is only results in a negative loop.
It's like if a startup was arguing about the ideal technological stack even before trying to find a product market fit!!
Personally, I do not care much about sixels, kitty or iterm format - all I want is to see some kind of support for a format that's popular enough for tools using it to emerge.
Yes, it would be better if that supported format was the option that had the greatest chance of succeeding, but right now, that is a very remote concern: first we need tools, then if in the worst case they are for a "bad" format, we can write transcoders to whatever format people prefer!
Right now, there is rarely any "input" to transcode (how many console tools support say iTerm format?), so we have a much bigger bootstrapping problem.
> an off the shelf ASCII plotting library probably involves less custom tooling
With a terminal like msys2 or xterm, no custom tooling is required: just use the regular gnuplot after doing the export for the desired resolution, font, and font size.
gnuplot is far more standard than plotting library that often require special Unicode fonts on top of requiring you to use their specific format.
That's a protocol that's a good forty years old, and even that is not supported. And I can see why, why on earth would you want to be adding support for that in 2021? What a ridiculous state of affairs.
In truth, it's because text is quite easy to handle. It's easy to make a program that handles text, too.
And so we have a lot of text editors, diff tools, efficient compression, tools like sort and uniq: the whole unix ecosystem.
So if you transform sound to text, you can then use text tools to compare the output to catch differences. A simple serialization of numerical sample values would have caught the bug, but I agree that having a way of visualizing the output is nice.
Command line input, programming, etc. is also still mostly done with text, because it's easy to transform. Of course, you can imagine working at a higher level with objects (like powershell does IIRC), mimetypes, etc.
The venerable xterm and a lot of later physical terminals (those things with CRTs) can emulate Tektronix (Tektronix, that today makes instruments, also made computer terminals with fancy storage CRTs that were kind of e-paper-like, but green - and sometimes yellow - screen) graphics. iTerm2 and some others, as pointed out, can do Sixel graphics (a format designed originally for DEC dot-matrix printers that some DEC terminals also implement).
VTE and, with it, almost every Linux distro, will get Sixel support soon. I volunteered to add Tektronix graphics to it too, but this is neither a dire need, nor something I have done before, so it'll take some time.
Because things that existed 40 years ago are useful, already have software written for it, are compatible in sometimes unforeseen ways (a DEC dot-matrix graph can be printed as is on a Sixel-compatible terminal!) and have been battle tested for ages.
There is a reason the Unix way of bytestream-based shell and pipes is still useful and present these days to the point that That Other OS is now embedding Linux in it.
Also, these ancient terminals often had some interesting typography options that are encoded in the ANSI standard that most modern terminals don't bother (line attributes that generate wider and taller cells are one such example).
These formats may be more desirable than more modern and complete ones such as PostScript for other reasons. I wouldn't advise implementing a terminal capable of rendering PostScript graphics because it's one more way to infiltrate malware in your computer by rendering untrusted inputs (There are a lot of RCE opportunities in exploiting vulnerable decoders).
In TempleOS you can mix text, images, hyperlinks, and 3d models in the terminal. This is true for the whole system: you could literally have a spinning 3d model of a tank as a comment in a source file. That's right, it took a literal schizophrenic to make an OS with a feature that should have been standard decades ago.
Nobody tries to make actually interesting new operating systems anymore. OS research today is just "let's implement unix with $security_feature", nobody is actually trying to make computers more powerful or fun to use, or design a system based off of a first-principles understanding of what a computer should be.
God I wish I was born in the lisp machine timeline
The features you describe belong to the app ecosystem, not to the OS - IMHO the OS is about hardware and drivers, and what kind of graphics is supported by your terminal and source file editor is orthogonal to the OS and could be done in any of the current OS'es; but that would require a rewrite/redesign/reimagining of the whole standard application package which seems a much larger project than "merely" an OS.
"There are more things in heaven and earth, Horatio, than are dreamt of in your philosophy"
An OS facilitates communication between programs running on a computer. Unix lets those programs communicate by sending characters of text to each other. You could just as easily imagine an OS that lets them communicate by sending images, audio/video, 3d models, etc. An OS can be way more than what you think it is. To detox your brain from this unix worldview, spend some time in a VM and play around with amigaOS or opengenera. Those were actual coherent OSes with an actual view of what a computer should be and how it should behave. Unix isn't.
> reimagining of the whole standard application package which seems a much larger project than "merely" an OS.
By OS, I don't mean kernel. I mean the base set of software that lets you interact with your computer and do interesting stuff with it.
The line between app platform and OS is a blurry one. The Amiga OS, for instance, has libraries for specific file types that expose standardized entry points. This way, if you install the library for Photoshop files, all graphics programs that adhere to that protocol will be able to read and write Photoshop PSD files. Microsoft had DDE and, later, OLE, for embedding objects from one program into data from another in a standard way all programs were supposed to share. It was a pain.
This blurry line is present in other environments as well. In the Apple Lisa, installing a program resulted in new templates in the Stationery folder. In Smalltalk, installing a program adds its class definitions to the system as independent entities you could use in your own programs.
Not all operating systems are the children of Unix and VMS.
the downside of rich terminal output is that media formats become the system's responsibility. applications can't output media in formats that aren't provided by the system, because then the terminal wouldn't know how to display it and interop with other applications (e.g. piping) wouldn't work either.
You could let a program create an API for manipulating a new type of data and inform the system about it so that other programs could use it. This is more or less what AmigaOS did; you installed a datatype for e.g. a PSD file, then all your programs that worked with images could read PSD files. I think it's a nice idea.
This isn't a graphical problem. All that's required is storing arrays of validation data and a diff tool to check for mismatches. Visualizing the results is useful for failure analysis but not a core requirement. That can readily be done with free tools like matplotlib. We live in that world today.
Once you've got the waveforms as arrays, what do you need the ASCII rendering for?
Instead of diffing ASCII-rendered waveforms, save the arrays and diff the arrays (and then use any kind of numerical metric on the residual). Scientist programmers have all sorts of techniques for testing and debugging software that processes sampled signals.
If we go beyond ASCII, Unicode specifies 2x2 mosaics since ever (they were present in DEC terminals) and 2x3 mosaics (from Teletext and the TRS-80) since version 13. Some more enlightened terminals (such as VTE) implement those symbols without the need of font support.
Or you can use Braille to get 2x4 mosaics, but they usually look terrible.
For audio you might also consider U+2581—U+2588, LOWER ONE EIGHTH BLOCK through FULL BLOCK. And then if you really want to go all out there are sixels, but that’s basically just an image; you probably lose the easy ability to compare them. On the other hand they’re not available in every terminal.
This is such a great idea! I've really struggled with how to test real-time audio code in the live looper I've been working on . Most of my tests use either very small, hand-constructed arrays, or arrays generated by some function.
This is both tedious and makes it very hard to debug test failures (especially with cases like crossfades, pan laws, and looping). I love the idea of having a visual representation that lets me see what's going wrong in the test output, and I'm definitely going to try to implement some similar tests.
I'm also curious what the state-of-the-art is for these sorts of tests. Does anyone have insight into what e.g., ableton's test suite looks like?
> I'm also curious what the state-of-the-art is for these sorts of tests. Does anyone have insight into what e.g., appleton's test suite looks like?
I don't know, but if I were to make an educated guess, maybe rendering stuff to actual audio files is a common approach? That way when something goes wrong, they can inspect it in a standard waveform editor?
Nice! I became obsessed with rendering sparkline representations of chunks of audio for the same reason: to inspect failures when writing tests / refactoring. I wrote a JUCE module (C++) and integration with lldb to make it quick to inspect chunks of audio in the IDE: https://github.com/sudara/melatonin_audio_sparklines
In short, because text is much easier to deal with than bitmaps, and there is much more tooling that "just works" for text than actual graphics, like Expecto's textual diffing in assertations. @MayeulC said it well: https://news.ycombinator.com/item?id=28856884
Not yet, but it certainly could be. Would it be useful to publish the helper classes that render the waves out to ASCII? That's really the guts of the thing. After that, you just use whatever testing framework you want to do the actual diffing (in my case Expecto for F#).