Quick Links: Download Gideros Studio | Gideros Documentation | Gideros community chat | DONATE
Fixing broken links in Wiki (Articles Tutorials) - Page 7 — Gideros Forum

Fixing broken links in Wiki (Articles Tutorials)

123457»

Comments

  • hgy29hgy29 Maintainer
    antix said:

    @hgy29 oh, I thought the script was broken.

    It may still be broken btw ;)
  • @MoKaLux I was already able to scrape URLLoader.... I can scrape everything but it's just slow using my method because I basically scrape every page that is in the API Reference.

    The script @hgy29 supplied does not return Sprite or Bitmap, and some other stuff like (global) is also terribly broken.@hgy29 says it is because there is some issue on the wiki.

    I looked at the Bitmap and Sprite pages but have no idea what might be the issue. There doesn't seem to be any issues with those pages so there's nothing I can do to make my scraper more efficient.

    @SinisterSoft the changing from left panel to top bar is one of the media queries inside the CSS, and is to do with making the site work nicely on mobile devices . It is no longer an issue for me because I no longer have to go to the wiki to read the API Reference... I have it all scraped to my local machine.

    Sorry to annoy you all, I'll retreat to my corner now.

    Likes: MoKaLux

    +1 -1 (+1 / -0 )Share on Facebook
  • MoKaLuxMoKaLux Member
    edited December 2019
    @antix you do not annoy me (us?) at all and please don't retreat, we are all together trying to make THE BIG GIDEROS 2020 happen (I am excited) :)
    I also had a look at the wiki for the pages you mentioned and don't see where the issues could come from! This should be all we need unless formatting is somehow wrong?
    <!-- GIDEROSOBJ:Bitmap -->
    <!-- GIDEROSMTD:Bitmap.new(texture) creates a new Bitmap object -->
    ...
    :/ do spaces matter after and before -- ?
    my growING GIDEROS github repositories: https://github.com/mokalux?tab=repositories
  • hgy29hgy29 Maintainer
    edited December 2019
    @antix, I just had a look and Bitmap is returned by my script, so not sure what is wrong. There is an issue with globals though, they appear as if attached to bitmap, but that is definitely a bug in my script (object reference not reset probably). Sprite is there too
  • @hgy29, hmm interesting, it does now return Bitmap and Sprite. I wonder whats happening with (global)? It's not a problem anymore since Ive already got everything that I need, don't worry yourself with it.

    During my scraping I found inconsistent behavior with LiquidFun...

    Even though LiquidFun is a plugin, it is also actually an API on its own. as such it should be moved from the plugins section and added to the other page as an API.
  • Each of the plugins are technically APIs on their own though aren't they? In any event, seeing as it's not part of the main API and should be ticked as a plugin when exporting then it should be in the plugins for consistency otherwise it's pretty confusing.

    Likes: MoKaLux

    Coder, video game industry veteran (since the '80s, ❤'s assembler), arrested - never convicted hacker (in the '90s), dad of five, he/him (if that even matters!).
    https://deluxepixel.com
    +1 -1 (+1 / -0 )Share on Facebook
  • antixantix Member
    edited December 2019
    @SinisterSoft I don't agree because:

    "Main API" - Contains a bunch of namespaces
    "Lua API" contains a bunch of namespaces
    "Plugins" contains a bunch of namespaces

    The exception is when you open the "LiquidFun" namespace which contains... a bunch of namespaces. Technically this makes it an API, not a namespace.

    Anyway, to me this is not consistent because if you are systematically processing namespaces, and one of them is actually an API... then it's broken IMHO.

    Namespaces are supposed to contain Methods, Events, and Constants.... not other API's ;)

    I've just made a corner-case for my scraper so it doesnt bother trying to parse "LiquidFun" if it finds it when scraping namespaces. Corner-cases are generally created to accommodate cases where data is not consistent over a set of data.

    Each to his own I guess.
  • SinisterSoftSinisterSoft Maintainer
    edited December 2019
    I'm not sure what you mean, but the history of it is Box2D used to be part of the main API as the physics system.

    That was replaced at my request by @ar2rsawseen and @atilim for LiquidFun as it was a fully compatible replacement but had some optimisation and extra functionality with regard to particles (but I think this bit wasn't functioning until @hgy29 got involved).

    Later, @hgy29 was optimising the base size of Gideros and it was decided to make LiquidFun a plugin rather than part of the main API in order to save a large amount of space (this was needed to bring the base size down as a requirement for Facebook Instants - which we were given early access to, but that's a different story!).

    For a short time, LiquidFun was both in the main API (LF v1.0) and also as a plugin (LF v1.1).

    Then it was fully removed from the internal API when we had done enough testing to show there would be no compatibility errors using the plugin.

    ------------------------------------------------------------------

    My opinion is that if it's optional and in the plugins folder then it's a plugin and it's documentation should therefore also be in plugins.
    Coder, video game industry veteran (since the '80s, ❤'s assembler), arrested - never convicted hacker (in the '90s), dad of five, he/him (if that even matters!).
    https://deluxepixel.com
  • @SinisterSoft what I mean is that if an API can be defined as a collection of namespaces... and LiquidFun is a collection of namespaces... then it's an API... not a namespace (which it currently is presented as).

    Basically if it looks like a duck, waddles like a duck, and quacks like a duck.. it is generally a duck ;)
  • @MoKaLux yes, that link opens a page with a bunch of XML inside containing fbinstant stuff.

    The XML contains fbinstant but a load of other stuff also, which isn't overly relative to fbinstant.
  • @hgy29 is there any way your script could be fixed and export basically the entire API Reference contents so that it could be processed offline?
  • I discovered that this page can export wikimedia pages...

    https://wiki.giderosmobile.com/index.php/Special:Export

    Whilst you can stuff page names into the URL, it becomes too long and cannot be processed.

    That leaves the undesirable task of having to populate the list manually so there is seemingly no way to programmatically use this API to download all of the API Reference pages. It does however quickly export all API Reference pages in XML format, and that is under 2MB in size.

    Currently my reader looks like this...

    It accepts URL stuff so you can tell it to start at a specified page. Personally I feel it looks nicer than the wiki, and is much easier to navigate.

    That is NOT to take away from the fantastic work my @MoKaLux and others to improve the wiki, they are doing a great job.

    I strongly think that there should be a mechanism in place so that a user can download the entire API Reference to either print it in hard-copy, or to present it in a different manner.

    @hgy29 hi if you don't want to fix your script can I get some access so I can fix/modify it myself please?
    image.png
    1060 x 1040 - 114K

    Likes: plicatibu

    +1 -1 (+1 / -0 )Share on Facebook
  • MoKaLuxMoKaLux Member
    edited December 2019
    @antix you can do it another way:
    -go to this page: https://wiki.giderosmobile.com/index.php/Special:AllPages or this one: https://wiki.giderosmobile.com/index.php/All_Classes_and_more
    -with the mouse select the pages you want
    -use the page you mentioned: https://wiki.giderosmobile.com/index.php/Special:Export
    -in the box below paste the pages you previously selected
    -click export

    export.png
    727 x 627 - 32K

    Likes: plicatibu

    my growING GIDEROS github repositories: https://github.com/mokalux?tab=repositories
    +1 -1 (+1 / -0 )Share on Facebook
  • antixantix Member
    edited December 2019
    @MoKaLux yes, I know that. What I'm wanting is a single-click method to download the entire API Reference because doing it manually is just too time consuming :)
  • MoKaLuxMoKaLux Member
    edited December 2019
    @antix time consuming?
    I used this page https://wiki.giderosmobile.com/index.php/All_Classes_and_more copied everything from top to bottom and got this result in 5secs!
    You don't need to do any editing (spaces, section titles) just copy and paste.
    PS: i changed from xml to txt for forum.
    PS2: i created this page just for you :) Tell me if some links are missing.
    txt
    txt
    GiderosMobile-20191215060156.txt
    2M
    my growING GIDEROS github repositories: https://github.com/mokalux?tab=repositories
  • hgy29hgy29 Maintainer
    antix said:


    @hgy29 hi if you don't want to fix your script can I get some access so I can fix/modify it myself please?

    Sorry, I forgot to fix it, being busy with releasing gideros, I’ll do it tomorrow and send you the script, however I can’t let you access the server files without setting up some form of isolation beforehand, since this server also stores private business data.
  • @hgy29 ahh I didn't know you had other sensitive stuff there.. I understand :)

    If you could fix the script that would be good. If it could just dump the entire API reference that would be even better.

    No great rush either... I'm now back working on my UI editor and updating my apps to accommodate googles constant demands regarding adverts ;)
  • hgy29hgy29 Maintainer
    @antix, I fixed the empty namespace issue (no longer tied to Bitmap object), and cleanup a bit of FBInstant mess (some objects were declared as methods on the wiki).
    My script isn't really fit for dumping the entire API, I would need to rewrite it as it looks more lire spaghetti code currently...
  • @hgy29, thanks! I'll have a look tomorrow. I suppose I can just use your script to then scrape the entire thing... I'm not sure how I'll use it yet :)
Sign In or Register to comment.