Wiki for the forum

[Utsuho Lazy]

It's reaaally easy.

Spoiler for Summary of adding fanhymns >_>:  [ Open ]

To add a new hymn, surround its name in ===, Eg. ===EXEC_PAJA/.===
To add the hymn contents, put the song under the name and on the line above the song, put <pre>
Eg. <pre>
Wee ki ra hymme

At the end of the song, on the line under it put </pre>
Eg. Was yea ra chs hymmnos mea
</pre>

When you want to add a person's name (like, Infel_phira or Vermillion Sky), you just surround the name in ==. Eg. ==Vermillion Sky==
And then you put all the songs under the name.

*Kinda gets it*

[ilutiern]

Hm, by the way, the fanhymn section is apparently already over 76 kb long... or big. >_< Should we just let it keep expanding? I think it might be a good idea to separate the sections either by each person having their own page for fanhymns (since many people have more than just a couple) or by dividing it between 'Standard Dialect', 'Pastalia Dialect' and 'Original Dialects'.

[wizzardx]

Hm, by the way, the fanhymn section is apparently already over 76 kb long... or big. >_< Should we just let it keep expanding? I think it might be a good idea to separate the sections either by each person having their own page for fanhymns (since many people have more than just a couple) or by dividing it between 'Standard Dialect', 'Pastalia Dialect' and 'Original Dialects'.

If you want to divide it into separate pages, I think the most natural division (in terms of Wiki structure) would be be by person. "Aquagon's Fanhymns", "ilutiern's Fanhymns", etc. I can split them off into separate articles for you if you'd like.

[Deciare]

To be fair, PHP is a far easier language to learn than most others I've used.

*Wonders if you are referring to Perl, the infamous write-only language ^^*

I'm referring to Python, too. XD Sorry, but I so like being able to do basically whatever without worrying much about includes, types, pointers/copies/references, or complications in how code is supposed to be structured.

Perl can be good at that, too, but PHP is less arcane about it. ^_^

[wizzardx]

I'm referring to Python, too. XD Sorry, but I so like being able to do basically whatever without worrying much about includes, types, pointers/copies/references, or complications in how code is supposed to be structured.

Those be fighting words ^_^.

From what little I've used of php, I would never want to use it unless I absolutely had to ^^;

Rant:

Spoiler:  [ Open ]

Badly structured libraries, inconsistent naming schemes and argument orders (verb noun, noun verb, etc), no exceptions, cryptic/non-existant error messages, no built-in logging functions, no parameterized db queries, different functions to call for each database, horrible things like magic_quotes (and having to work around it possibly or possibly/possibly not being setup on a given server), no unicode, too many global functions, no namespaces, ability to use non-assigned variables (and, the security problems that can cause), encouraging coders to mix html, logic, and SQL together in an unholy mess >_>.

(Yeah, I know a lot of these have been fixed in php 5. It still looks cryptic & perl-like with all the sigils, though ^^)

I'm sure that it is very convenient (especially since it is installed on just about every web server), but it's a big mess. I prefer using elegant programming languages ^^.

[Deciare]

Rant:

Wow, that read like an introduction to an article about how not to write programs. XD

inconsistent naming schemes and argument orders (verb noun, noun verb, etc),

That's what references are for. ^_^

no exceptions,

*doesn't use them anyway unless a function has no other way of communicating that it failed*

no parameterized db queries

I tried writing my own function for that with a project I was working on, but it turned out to be too unwieldly to use. Beyond some degree of complexity, it's just easier to sprintf() a plaintext query, with foreach loops to handle complex combinations of conditional statements.

different functions to call for each database

Write a custom DB abstraction class? phpBB3 has one that's compatible with lots of them.

no unicode

D: Thank goodness that isn't true any longer!

too many global functions

That's exactly one of the reasons I like it. XD Not needing to worry about what comes from what object or namespace.

ability to use non-assigned variables (and, the security problems that can cause)

Nothing's stopping me from initialising them myself, and I do.

encouraging coders to mix html, logic, and SQL together in an unholy mess

I use PHP if conditions to decide whether bits of HTML or JavaScript should be presented on the final HTML output. XD Why go through the trouble of learning how to generate HTML or SQL through an abstraction layer when I already have perfectly usable knowledge of HTML and SQL by themselves?

I'm sure that it is very convenient (especially since it is installed on just about every web server), but it's a big mess. I prefer using elegant programming languages ^^.

It is very convenient, yes. ^_^ No single structural paradigm is universally applicable to every project, and I find that the fewer constraints a language puts on my ability to structure my code how I believe is most efficient for its particular purpose, the more likely that language is to remain applicable for me.

I prefer to think, "I won't do it this way because it's sloppy or insecure", rather than, "I can't do it this way because someone else thinks it's sloppy or insecure".

[ilutiern]

Hm, by the way, the fanhymn section is apparently already over 76 kb long... or big. >_< Should we just let it keep expanding? I think it might be a good idea to separate the sections either by each person having their own page for fanhymns (since many people have more than just a couple) or by dividing it between 'Standard Dialect', 'Pastalia Dialect' and 'Original Dialects'.

If you want to divide it into separate pages, I think the most natural division (in terms of Wiki structure) would be be by person. "Aquagon's Fanhymns", "ilutiern's Fanhymns", etc. I can split them off into separate articles for you if you'd like.

That'd be very helpful D: Could you also tell me how to do it?

[Eleinia]

Hm, by the way, the fanhymn section is apparently already over 76 kb long... or big. >_< Should we just let it keep expanding? I think it might be a good idea to separate the sections either by each person having their own page for fanhymns (since many people have more than just a couple) or by dividing it between 'Standard Dialect', 'Pastalia Dialect' and 'Original Dialects'.

If you want to divide it into separate pages, I think the most natural division (in terms of Wiki structure) would be be by person. "Aquagon's Fanhymns", "ilutiern's Fanhymns", etc. I can split them off into separate articles for you if you'd like.

That'd be very helpful D: Could you also tell me how to do it?

I think you just need to create new pages, like how Wizzard said in this earlier post. If he's not around, I can try to help you. Otherwise, he might >.> me again.

[wizzardx]

Wow, that read like an introduction to an article about how not to write programs. XD

I had to let off some steam ^_^. I had to do some PHP code maintenance at work (add error checking, logging, etc). It was a .... disturbing experience, coming from Python land >_>.

inconsistent naming schemes and argument orders (verb noun, noun verb, etc),

That's what references are for. ^_^

Reference books, or reference variables? Either one sounds like a nasty work-around >.<.

no exceptions,

*doesn't use them anyway unless a function has no other way of communicating that it failed*

Lack of exceptions means you need to either check for errors after every function call, or have silent errors that get ignored (the bane of PHP apps ^^;). Exceptions mean that if there are problems, your program logic *will* respond to them in some way, and you can do something sensible, without adding "if" blocks around every function call.

Similarly, for bash scripts, I always add 'set -e' to the top, to stop immediately on errors (and then add specific handler logic or ignore errors as appropriate)

no parameterized db queries

I tried writing my own function for that with a project I was working on, but it turned out to be too unwieldly to use. Beyond some degree of complexity, it's just easier to sprintf() a plaintext query, with foreach loops to handle complex combinations of conditional statements.

Yes, I tried that to in a C++ app (to avoid having to build up vectors (dynamic arrays) one parameter at a time), but it became a huge mess of weird macros. It helps a lot of your programming language/library supports that kind of thing.

Compare these two: (pseudo-code):

Code:  [ Select all ]

# Some initial code:
a = "some string 'with quotes'"
b = 123456

# Non-parametized (pseudo-php):

mysql_exec("SELECT * FROM foo WHERE field_a = '" . mysql_escape(a) . "' AND field_b = $b")

# Parameterized (pseudo-python):

curs.execute("SELECT * FROM foo WHERE field_a = %s AND field_b = %s", [a, b])



Parameterised has these advantages:
- Don't need to escape/quote strings. The libraries handle that
- Don't need database-specific functions (for executing, quoting, etc). The libraries handle that
- Query string looks a lot simpler
- Query and parameters are separate, so you can easily swap them out etc

different functions to call for each database

Write a custom DB abstraction class? phpBB3 has one that's compatible with lots of them.

It's nice if the language's standard library comes with a standardised, but simple DB API, rather than re-inventing the wheel in each project >_>. Although you also get more advanced (and complex) 3rd-party libraries like SQLAlchemy for projects that want object relational mapping & automatic query generation.

no unicode

D: Thank goodness that isn't true any longer!

I don't use unicode much myself, but I understand the importance ^^. It came in useful a while back when I was making trainer programs to help myself study the Kana ^^. In Python 3, unicode strings are the standard string type, so you can't use them as byte arrays any more >_>. You need to use a special 'bytes' type for that. Which makes sense I guess ^^.

too many global functions

That's exactly one of the reasons I like it. XD Not needing to worry about what comes from what object or namespace.

Which results in code where every function name has a set of library prefixes at the start, like "mysql_foo", "pgsql_foo", etc. Namespaces (at least how Python does them) can make your code a lot more readable ^^. For instance:

Code:  [ Select all ]

   
from pgsql import Connection as Con

a = Con('<db connection string 1>')
b = Con('<db connection string 2>')


As opposed to using something like pgsql_connect() You can use the exact same code later with a different database, by changing the import line, or assigning "Con" to something else.

The example was very simple, but when code uses a lot of modules (or sub-modules and sub-sub-modules ^^;), namespaces and aliasing like the above can make your code a lot easier to read.

I guess that's how you use references in PHP?

ability to use non-assigned variables (and, the security problems that can cause)

Nothing's stopping me from initialising them myself, and I do.

Nothing's stopping you from making hard-to-detect typos in variable names either >_>. For instance, code like this:

Code:  [ Select all ]

   
db_execute('DELETE FROM customers WHERE name LIKE '*$customr*');


(The above is contrived, obviously, but the basic idea holds for more complicated code)

Where your code declared '$customer', but you made a typo in the query. Most programming languages will flag an error for lines (with equivalent logic), because of a non-existant variable, but PHP will go and delete everything from the table because the non-existant $customr variable evaluates to an empty string. At least PHP, iirc, gives you a way to log use of unknown variables, but you have to go find that option and enable it >_>.

encouraging coders to mix html, logic, and SQL together in an unholy mess

I use PHP if conditions to decide whether bits of HTML or JavaScript should be presented on the final HTML output. XD Why go through the trouble of learning how to generate HTML or SQL through an abstraction layer when I already have perfectly usable knowledge of HTML and SQL by themselves?

For small projects that's fine. But in larger projects, combining presentation and backend logic can lead to a lot of problems. For instance when it comes to making alternative user interfaces/themes. That's why web frameworks usually make such a big deal about "Master-view-controller".

It is very convenient, yes. ^_^ No single structural paradigm is universally applicable to every project, and I find that the fewer constraints a language puts on my ability to structure my code how I believe is most efficient for its particular purpose, the more likely that language is to remain applicable for me.

That's true. I love Python to bits, but sadly you can't use it for everything ^^. For instance, the significant whitespace causes a lot of problems if you try to use it in web templating, and you end up with hideous pseudo-Python templating languages like this:

Code:  [ Select all ]

{% extends "base_generic.html" %}

{% block title %}{{ section.title }}{% endblock %}

{% block content %}
<h1>{{ section.title }}</h1>

{% for story in story_list %}
<h2>
  <a href="{{ story.get_absolute_url }}">
    {{ story.headline|upper }}
  </a>
</h2>
<p>{{ story.tease|truncatewords:"100" }}</p>
{% endfor %}
{% endblock %}


That's a real example, from the Django documentation. It's a nice hack for a templating language, but it's incredibly ugly! PHP looks far better by comparison >_>. Code aesthetic is very important to me, which is one reason why I like Python compared to most other languages, most of the time ^^.

(You do get better-looking templating languages, but the above is from the most popular Python-based web programming framework)

[wizzardx]

Double posting, because the above reply was a wall of text, mainly for Deci ^^.

Hm, by the way, the fanhymn section is apparently already over 76 kb long... or big. >_< Should we just let it keep expanding? I think it might be a good idea to separate the sections either by each person having their own page for fanhymns (since many people have more than just a couple) or by dividing it between 'Standard Dialect', 'Pastalia Dialect' and 'Original Dialects'.

If you want to divide it into separate pages, I think the most natural division (in terms of Wiki structure) would be be by person. "Aquagon's Fanhymns", "ilutiern's Fanhymns", etc. I can split them off into separate articles for you if you'd like.

That'd be very helpful D: Could you also tell me how to do it?

@ Eleinia: >_>



Yes, like Eleinia said, you just make a new article. The method I would use:

1) Edit one of the longer sections (like yours ^_^)

2) Add this text to your section:

Code:  [ Select all ]

ilutiern has written several Fanhymns. You can read them [[ilutiern's Fanhymns|here]].


3) Preview the page. You now get a broken link in your preview.

4) Open the broken link in a new tab (opening up an "edit new article" page)

5) Cut and paste your Hymns over from your original section, to the new page.

6) Change "===" to "==" in your new article (since your sub-sections are now main sections. Unless you want to split by dialect or whatever later).

7) Add some kind of description to the top of your new page (so that people visiting your page randomly can know what the stuff below is about).

8) Save both the articles.

@Lazy & ilutiern: You might find this page useful, for understanding how sections (and wiki formatting in general) work.

Update: I've moved ilutiern and Aquagon's Fanhymns into separate articles. They are linked from the main Fanhymn article.

[Utsuho Lazy]

Umm, I was wondering, what are the font (beside Hymmnos), font size and color tags for the Wiki? And do those tags work within <pre> tags? I like to format my Hymns with different sizes and text colors, if I do, I mustn't use <pre> tags since it's similar to BBCode [code] tags right?

[wizzardx]

Umm, I was wondering, what are the font (beside Hymmnos), font size and color tags for the Wiki? And do those tags work within <pre> tags? I like to format my Hymns with different sizes and text colors, if I do, I mustn't use <pre> tags since it's similar to BBCode [ code ] tags right?

The wiki supports several HTML tags, including <font> and <span>. You can read more about the <font> tag here.

(Technically, the <font> tag is deprecated and web developers are supposed to use <span> and CSS formatting for presentation, but you don't have to worry about that for the wiki)

And yes, the wiki supports fonts inside pre-formatted sections. Instead of using <pre>, you put a single space in front of each of the lines.

Here is an example:

Code:  [ Select all ]

<b><hymmnos>Was yea ra chs hymmnos mea</hymmnos></b>
<i><font color="blue">(I will be glad to turn myself into a song)</font></i>


(There is a single space in front, even though it's hard to see. Try quoting my reply and then editing the text above)

You can see how the above looks in the Wiki sandbox.

[Utsuho Lazy]

Ah, okay, thank you so much~

[wizzardx]

I've added another example to the sandbox, which shows how to use the (more technically correct) "span" tag, for italics, fonts, colors, sizes, etc.

[Utsuho Lazy]

I've added another example to the sandbox, which shows how to use the (more technically correct) "span" tag, for italics, fonts, colors, sizes, etc.

...Looks amazingly complex and hard to remember ;_;

Oh well I'll try. Gambatte mea~

[wizzardx]

...Looks amazingly complex and hard to remember ;_;

Oh well I'll try. Gambatte mea~

If there's any commonly-used fonts/styles/etc, I can add more tags, like with <hymmnos>

That example was pretty complicated, most of the time you can just use these:
- <hymmnos>text</hymmnos> - text in hymmnos font
- ''text'' - Text in italics (2 single quotes on both sides)
- '''text''' - Text in bold (3 single quotes on both sides)

And for other fonts (besides Hymmnos), you can use either the span tag (recommended):

Code:  [ Select all ]

<span style="font-family: times">text in times new roman</span>


Or the font tag (deprecated):

Code:  [ Select all ]

<font face="times">text in times new roman</font>


You don't have to remember the exact syntax, you can just copy & paste the tags above, and change the parts that look interesting ^^;.

Update: Just added an example to the bottom of the sandbox.

[Utsuho Lazy]

After about 45 minutes, I've given up with Arabic text. There's absolutely no way I can do something simple as to add a space to the beginning of each Arabic line, no matter what...

[wizzardx]

After about 45 minutes, I've given up with Arabic text. There's absolutely no way I can do something simple as to add a space to the beginning of each Arabic line, no matter what...

What is the hard part? If you need help adding spaces to lots of lines, most text editors (more advanced than notepad) let you indent text. You will probably need to:

1. Get a decent text editor (eg: Notepad++)
2. Set it's ident to 1 space, instead of the default (usually 8)
3. Copy your text into the editor
4. Select all the text
5. Press tab, to indent everything by one space
6. Copy text over to the wiki edit box.

Update: Alternatively, you don't need to use pre-formatted text. You can also put line breaks on the end of each line (<br>), although that's kind of ugly. I've added an example to the bottom of the wiki sandbox.

[Utsuho Lazy]

I mean, Arabic is supposed to be encoded right-to-left, but somehow the Wiki (and our forums too, at least in the post making page) forces it to be left-to-right. As such, whenever I want to add a space to the beginning of the Arabic line, it is actually added at the end of the line. And if I add the space to the end of the line, it also is added at the end! There's no way I can add a space (or any normal character, for same reason) at the beginning of an Arabic line.

For example, could you try with this:

شسیتبمسیتب یسمابمسی مسیتبمسنی سیب
یسشبنسیتب سیبشسب
شیسبشسیبشبسیب

I can add space to the beginning of the first and second line, but not third (it always jump to the end of the line).

Is there another way that can allow me to both line break and font edit? I don't like Courrier font for Arabic, and I also want to customize the size/color.

And Notepad++ doesn't even support Arabic and Japanese texts ;_;

[wizzardx]

I think I got it working, although it was a pain to edit >.<. I think that Firefox recognizes after the first line or so, that editing direction is meant to go backwards, so things start going pear-shaped >_>.

Check the bottom of the Sandbox page, is that correct?

How I did it was, for simple indented Arabic text:

1. Put a space and an "x" on the start of each line
2. Copy the arabic text, line by line, after each of the <space><x> on each line
3. Remove the x, line by line

During (3) the spaces jumped around confusingly (to the end of the line), but when I previewed the page, the text was in a "preformatted" section.

I think it should be possible in Firefox to edit the arabic and get those spaces in the "front", but what you'll be seeing on the screen during editing is spaces on the end.

For the bottom example (with fonts) it was actually a lot simpler. I put those formatting instructions in first, and then copied and pasted in the Arabic text last, line by line, and it seemed to work.

Does that help?