Add Radost website proposal to trognotes ideas collection

This commit is contained in:
Wookey 2020-04-23 23:04:29 +01:00
parent 56c0db8c05
commit 6669939bf6

View File

@ -91,7 +91,7 @@ Most people will not want to read this at all. This is for speleosoftwarearcheol
<a href="https://gitorious.org/stroggle/stroggle-gitorious-wiki.git/">stroggle-gitorious-wiki</a></br>
but note that this domain has an expired ertificate so https:// complains.
h3 id="automation">Automation on expo.survex.com</h3>
<h3 id="automation">Automation on expo.survex.com</h3>
<p>Ths section is entirely out of date (June 2014), and moved here for historic interest</p>.
@ -266,7 +266,7 @@ If I could find some people that agree with this, then we could try to reach a
compromise on:
(1) how do we store our data in a file system,
(2) how do we use this XML (let's do a common spec, but keep it simple)
(3) how do we aim not to high and not end up dead like CaveXML :)
(3) how do we aim not too high and not end up dead like CaveXML :)
After we do that, everyone goes away to do their own projects and write their
own code. Or maybe we have some degree of co-operation in actually writing the
@ -381,12 +381,78 @@ Backend datafiles:
(all static info)
Storage:
non-html or > 200K go in 'files' (PDF, PNG, JPEG, DOC, ODF, SVG)
convert small 800x600 version into website by default. (matching structure?
non-html or > 280K go in 'files' (PDF, PNG, JPEG, DOC, ODF, SVG)
convert small 1024x768 version into website by default. (matching structure?
</pre>
<hr />
<h3 Radost's proposal</h3>
<p>Radost Waszkiewicz (CUCC member 2016-2019) proposed a plan for superceding troggle</p>
<pre>
Hey,
on the design sesh we've looked a bit on the way data is organised in the
loser repo and how to access it via troggle.
A proposal arised that all this database shannaingans is essentially
unnecessary - we have about 200 caves, about 250 entrances, about 200
people and couple dozen expos. We don't need efficient lookups at all. We
can write something which will be 'slow' and make only things we actually
care about. Similarly I see little gain from doing the html - python himera
template pages. These contain mainly nested for loops which could just as
well be written in e.g. python.
I'd advocate following solution:
- give up on fixing things in troggle
- have a small number of .py scripts which generate static webpages with
content we want
Why do this:
- we don't have multiple intermediate layers all of which are difficult to
maintain
- anyone can run this on their machine without need for local vm's
- to change something there is one thing you change - script that generates
pages belonging to particular pattern. currently you need to take care of
parsers, view script, url script, and template
- it runs offline (just like that as pages are just blocks of literal text
+ some javascript)
- we never end up with no working copy during expo - run scripts once
before expedition and put results in /backup and just leave it there till
next year. Something goes wrong on the daily updated version - we have a
very easy fallback
- just one (or very close to it) programming language
- change to python3 to parse silly chars
- wayyyy fewer lines of code and substantially better code to comment ratio
(if this is us writing this now)
- compatible with all the already existing static pages
- we can host this on a potato hosting as we're not running anything fancy
anymore
- get's rid of the horrifying url rewrites that correspond to no files that
exist
How much work would this actually take:
- most likely one script per page type, so far page types that are
obviously needed:
--cave index
--individual cave descriptions
--logbooks
- more than half of the parsers are already rewriten by me and can be
changed to do this instead of modifying SQL database with minimal effort
- html/css side of things already exists, but it would be nice to go for a
more modern look there as well
Things this solution doesn't solve:
- no webpage-based 'wizzard' for creating caves and such (did people use it
much? do we care?) -> maybe 'send an email to this address' is the ultimate
solution to this
- uploading photos is still difficult (in the sense that they don't get
minified automatically)
Rad
</pre>
</body>
</html>