News for hackers. Stuff that matters.

jandecaluwe

Re:Show HN: Urubu – A micro CMS for static websites on Wednesday February 17, 2016 @05:52AM (#11116933)
A showstopper feature for me is that it should be straightforward to use the tool to write its own manual. Surprisingly few static generators, including Pelican, do that however.

But Urubu's manual is written using Urubu: http://urubu.jandecaluwe.com/manual/

Re:Show HN: Urubu – A micro CMS for static websites on Wednesday February 17, 2016 @04:55AM (#11116683)
It is correct that there is "one main content" per page, available in the 'body' attribute.

However, to me the distinctive feature of a blog is not that, but the way entries are ordered (i.e. reverse order by date). For a manual, faq, feature requests,... I need a different ordering scheme, but I am happy with the "one main content" concept.

That having said, there is of course a need combine content in a single page. A layout (aka template) can combine content as needed. It can use other attributes of the page, specified in the front matter. Or it can use the content from other pages. One could put subpages in a subdirectory, and use some filename convention so that a layout can access them.

Urubu has support for pages that are only present for content, and not for being rendered themselves: you can specify a 'null' layout in the front matter.

Re:Show HN: Urubu – A micro CMS for static websites on Wednesday February 17, 2016 @04:27AM (#11116508)
The sites are typically in a git repo, so content updates are through commits or pull requests.

For an example of feature requests that are ordered by status and number, see http://dev.myhdl.org/meps/

Re:Show HN: Urubu – A micro CMS for static websites on Wednesday February 17, 2016 @03:02AM (#11116131)
My main use case are technical-oriented websites.

So I need to support content like a manual, a faq, news, feature requests, essays, and yes, also a blog. I also want good navigation support, and an easy way to refer to other project pages. Furthermore, I want programmatorial flexibility, which because of my background means Python and Jinja2 templates.

I wrote this because I didn't find an existing solution that met those requirements. I am not claiming it is "better" in a general sense than anything else - I think it all depends on your requirements and background.

Re:Show HN: Urubu – A micro CMS for static websites on Wednesday February 17, 2016 @02:47AM (#11116078)
No multilanguage support.

I do use autopulling for some websites, works great indeed.

Re:Show HN: Urubu – A micro CMS for static websites on Wednesday February 17, 2016 @02:43AM (#11116058)
It is heavily inspired by some Jekyll features, and I have tried to add recognition for this in the docs. However, it is not blog-oriented and it is in Python ;-).

The distinguishing feature is that the the website structure matches the source structure exactly, which enables a number of other features such as flexible ordering and wiki-links.

Re:Show HN: Urubu – A micro CMS for static websites on Wednesday February 17, 2016 @02:40AM (#11116041)
Urubu is for Python 3 also, from a single codebase. The regression suite currently tests for 2.7 and 3.4.

Update: I see where the question is coming from - the Quickstart site suggest 2.7 only. Thanks - will update.

Re:Show HN: Urubu – A micro CMS for static websites on Tuesday February 16, 2016 @05:24PM (#11113600)
Compared to others, I believe configuration is minimal. Also, there is a companion site to get started immediately, and update as you go:

http://urubu-quickstart.jandecaluwe.com

Re:Show HN: Urubu – A micro CMS for static websites on Tuesday February 16, 2016 @05:13PM (#11113539)
Hello:

I created this project because I needed a generator for a static website - not just a blog.

To accomplish this, the website structure matches the source structure exactly. In a subdirectory, you use the index.md file to specify the content, either explicitly by listing it, or implicitly by specifying the ordering key and direction. A blog would be reverse order by date, but any other scheme is possible.

In addition, this structure makes it possible to refer to other pages in the project using a wiki-like syntax. Basically, Urubu resolves the Markdown syntax for reference links over the project, instead of just within a page. This is a unique feature afaik.

Furthermore, there is support for tags, search, and templating constructs within pages. And it works well with Bootstrap.

It works great for a number of sites that I manage.

Interested to hear your feedback!

You have it completely backwards. The best "real engineers" do not "think directly" on how to solve the problem with registers and combinational logic. They think about how to solve the problem functionally and let synthesis take care of the logic as much as possible. MyHDL is ideal for that.

That is how ASICs/FPGAs have been designed by the best teams for the last 2 decades. And it is much more productive than the textual schematic entry you are describing. There is simply no comparison. Synthesis works.

As for your word-playing attempt: "description" stands for a number of things, among those "describing behavior". Just check the Verilog/VHDL LRM or MyHDL manual.

Complete nonsense. First, a synthesizable MyHDL description is at exactly the same abstraction level as a synthesizable Verilog/VHDL description. This paradigm is the basis for the standard, successful industrial flow. This is what real hardware designers do and it is not HLS.

In sharp contrast to what you suggest, logic generation is not the main hardware design problem. And the fact that you think you can beat RTL synthesis by "generating" or "constructing" the logic itself proves one thing: you don't have the slightest idea of the real power of RTL synthesis.

The main problem with hardware design is verification. VERIFICATION. And for that reason, HDLs should not be limited a "fully synthesizable" subuset, but support powerful modeling concepts in the first place. And so far, nothing beats the event-driven paradigm for that purpopse.

Stop the misrepresentation and do your homework first please. MyHDL is as cycle accurate as Verilog & VHDL. It's based on the same event-driven paradigm - the paradigm that real industrial designers have been using to design real products for two decades. And you do not need conversion to Verilog/VHDL to achieve this - the core of MyHDL is a simulator.

The real difference is that in event-driven languages, clock events are explicit, instead of implicit like in Chisel and the whole array of dead HDLs that preceded it. So if history is any guide, Chisel is dead upon arrival.

Re:Megaprocessor on Tuesday June 23, 2015 @03:45AM (#9763169)
Please. Of course it is, and that is made very clear on the home page. http://www.myhdl.org/
Re:MyHDL – Design hardware with Python – Examples on Wednesday May 07, 2014 @08:29AM (#7709746)
MyHDL is implemented as a minimalistic pure Python library, which basically means that you can use Python features for modeling - pretty powerful. In addition, it supports conversion of a language subset, so that you can also use it as a synthesizable RTL language. The latter feature is absent from SystemC.
Re:MyHDL – Design hardware with Python – Examples on Wednesday May 07, 2014 @01:50AM (#7708790)
MyHDL has been and is being used for production designs.

For parametrizability, you have the full Python power at your disposal. This is also true for conversion, because conversion happens after elaboration of the design.