- Pharo-users - lists.pharo.org

Re: Standalone html builder (a la seaside without seaside?)

by Tim Mackinnon

Lots of good tips in this thread to explore - thanks everyone, Iâll report back on what I find works bestâ¦ To Offray - with a template approach, I have often been tempted with that too but I like the idea of autocompletion and easier refactoring capabilities in Pharoâ¦ but you never know. Tim > On 28 Sep 2020, at 19:35, Franz Josef Konrad <franzl(a)fjkonrad.de> wrote: > > Am 28.09.2020 um 19:28 schrieb Tim Mackinnon: >> Hi - has anyone ever managed to extract the html builder out of seaside - or written something equivalent? >> I often find I want to build some HTML, but donât want the full seaside - and was wondering if anyone has managed to extract it, or have something similar? >> This combined with Renoir from BA-ST would give a good little light weight web potential to run with Zinc. >> Tim > > Some years ago I was looking for something for the same purpose and looked at this library <http://smalltalkhub.com/PierceNg/WaterMint-HTML/index.html> <http://forum.world.st/Standalone-HTML-Model-td4967128.html> . I don't know if this helps. I also just loaded WaterMint and looked at the code. Other than that I had no experience. > > Franz Josef

Sept. 28, 2020

Re: how to check for the statusCode

by Roelof Wobben

Sept. 28, 2020

Re: Standalone html builder (a la seaside without seaside?)

by Offray Vladimir Luna CÃ¡rdenas

Hi, I know this is not what you're asking for, but maybe some lateral thinking could help with your need. Jump to the end for the TL;DR suggestion or bear with me some paragraphs for a practical example. In our project about "IndieWeb with Brea", we use Pandoc and some external tools to produce the web site you see in [1] from Markdown files in [2] combining them with templates done is Mustache[3], as the one in [3a]. Everything is orchestratedÂ by the minimalist Pharo based Brea CMS[4] and produced using interactive Grafoscopio[5] notebooks and stored in Fossil[6] SCM, which proves timelines[6a], user management, permissions and wiki like views of each page (click on the top right corner Icons of each page to see it). It has been a pretty agile experience and kind of our "pocket infrastructures" approach to decoupled/headless CMS, even before we knew about them (and their rising). [1] https://mutabit.com/repos.fossil/indieweb/ [2] https://mutabit.com/repos.fossil/indieweb/tree?type=tree&name=docs/es [3] https://github.com/noha/mustache [3a] https://mutabit.com/repos.fossil/indieweb/doc/trunk/docs/es/pagina.mus.html [4] https://mutabit.com/repos.fossil/brea/ [5] https://mutabit.com/grafoscopio/en.html [6] https://fossil-scm.org/ [6a] https://mutabit.com/repos.fossil/indieweb/timeline [7] https://github.com/pillar-markup/MicroDown We use Pandoc because of historical reasons and because we need YAML metadata headers, multiformat support including PDF export with custom pre-processors (Pandoc filters). But, at some point I would like to use Microdown[7] as a pure Pharo native format when Pandoc is not available or as a replacement/complement for Pandoc filters. So, my lateral thinking suggestion is why not to avoid HTML altogether? Mustache and Microdown could take you far regarding light web deployment and development. Cheers, Offray On 28/09/20 12:28 p.Â m., Tim Mackinnon wrote: > Hi - has anyone ever managed to extract the html builder out of seaside - or written something equivalent? > > I often find I want to build some HTML, but donât want the full seaside - and was wondering if anyone has managed to extract it, or have something similar? > > This combined with Renoir from BA-ST would give a good little light weight web potential to run with Zinc. > > Tim

Sept. 28, 2020

Re: how to check for the statusCode

by Sven Van Caekenberghe

You *really* should read the HTTP chapters of http://books.pharo.org/enterprise-pharo/ in particular https://ci.inria.fr/pharo-contribution/job/EnterprisePharoBook/lastSuccessf… not just the Web App chapters. Also, any general introduction to HTTP (maybe even https://en.wikipedia.org/wiki/Hypertext_Transfer_Protocol) will help you. You can just say response statusLine You also really, really have to learn how to use the tools (the IDE) to help yourself. If you browse the ZnResponse class, you can see all this for yourself. I know there is a steep learning curve, and that it can be intimidating, but the whole idea behind Pharo is that you have this living object system that you can explore and learn from. The introduction texts (Pharo By Example, the MOOC, etc) all try to learn you that. Use spotter to find things, use browsers to look at code, use inspectors to look at objects, use senders and implementers, use class references, read class and method comments, study unit tests. Learn from the system. BTW, ZnEasy is just a simple class side facade, ZnClient is the real thing. > On 28 Sep 2020, at 20:34, Roelof Wobben via Pharo-users <pharo-users(a)lists.pharo.org> wrote: > > Hello, > > Sorry for asking so much questions but I lost the big picture right now. > > Im trying to get the response object back from a api call. > > That I can do with : > > ` response := ZnEasy get: url. ` > > and I see that it has the fields statusline which has the field code > > I thought I could use it like this : response at: #statusline ` > but that give me a error message that indexes needs to be integers. > > How do I get the code field again ? > > Roelof

Sept. 28, 2020

Re: Standalone html builder (a la seaside without seaside?)

by Esteban Maringolo

I've been playing with something like that, but still using Seaside. I was calling it "WAStatelessApplication", which is nothing but a way to route requests to components, and have these components not using callbacks nor continuations. I'm really comfortable with the Seaside canvas way of doing HTML, so I'll eventually end up building it, but currently it is just another abandoned project. For a very few specific things, I used a RESTful handler and a WABuilder to generate the HTML output, using regular components to render it. Regards! Esteban A. Maringolo On Mon, Sep 28, 2020 at 2:28 PM Tim Mackinnon <tim(a)testit.works> wrote: > > Hi - has anyone ever managed to extract the html builder out of seaside - or written something equivalent? > > I often find I want to build some HTML, but donât want the full seaside - and was wondering if anyone has managed to extract it, or have something similar? > > This combined with Renoir from BA-ST would give a good little light weight web potential to run with Zinc. > > Tim

Sept. 28, 2020

Re: Standalone html builder (a la seaside without seaside?)

by Franz Josef Konrad

Am 28.09.2020 um 19:28 schrieb Tim Mackinnon: > Hi - has anyone ever managed to extract the html builder out of seaside - or written something equivalent? > > I often find I want to build some HTML, but donât want the full seaside - and was wondering if anyone has managed to extract it, or have something similar? > > This combined with Renoir from BA-ST would give a good little light weight web potential to run with Zinc. > > Tim > Some years ago I was looking for something for the same purpose and looked at this library <http://smalltalkhub.com/PierceNg/WaterMint-HTML/index.html> <http://forum.world.st/Standalone-HTML-Model-td4967128.html> . I don't know if this helps. I also just loaded WaterMint and looked at the code. Other than that I had no experience. Franz Josef

Sept. 28, 2020

how to check for the statusCode

by Roelof Wobben

Hello, Sorry for asking so much questions but I lost the big picture right now. Im trying to get the response object back from a api call. That I can do with : ` response := ZnEasy get: url. ` and I see that it has the fields statusline which has the field code I thought I could use it like this :Â response at: #statusline ` but that give me a error message thatÂ indexes needs to be integers. How do I get the code field again ? Roelof

Sept. 28, 2020

Re: Standalone html builder (a la seaside without seaside?)

by Eric Gade

> On 28 Sep 2020, at 19:28, Tim Mackinnon <tim(a)testit.works> wrote: > > > I often find I want to build some HTML, but donât want the full seaside > - and was wondering if anyone has managed to extract it, or have something > similar? > I had a similar thought a few months back when I was looking for libraries to build a lightweight static site generator for personal use (I gave up due to time constraints) > Hi - has anyone ever managed to extract the html builder out of seaside - > or written something equivalent? > You might want to dip into the code for Amber Smalltalk. From what I recall, that system was able to extract the TagBrush and related classes -- Eric

Sept. 28, 2020

Re: Standalone html builder (a la seaside without seaside?)

by Sven Van Caekenberghe

Hi Tim, > On 28 Sep 2020, at 19:28, Tim Mackinnon <tim(a)testit.works> wrote: > > Hi - has anyone ever managed to extract the html builder out of seaside - or written something equivalent? > > I often find I want to build some HTML, but donât want the full seaside - and was wondering if anyone has managed to extract it, or have something similar? > > This combined with Renoir from BA-ST would give a good little light weight web potential to run with Zinc. > > Tim There is a minimal HTML builder to be found in ZnHtmlOutputStream. It is used for a couple of examples and default functionality of ZnServer. There are even unit tests. It is not the same as the Seaside HTML approach though. Sven

Sept. 28, 2020

Standalone html builder (a la seaside without seaside?)

by Tim Mackinnon

Hi - has anyone ever managed to extract the html builder out of seaside - or written something equivalent? I often find I want to build some HTML, but donât want the full seaside - and was wondering if anyone has managed to extract it, or have something similar? This combined with Renoir from BA-ST would give a good little light weight web potential to run with Zinc. Tim

Sept. 28, 2020