28 replies to this topic

[Ghostkeeper]

SimplXML
Version 1.0.2, by Ghostkeeper

SimplXML is an XML (eXtensible Markup Language) Parser for Game Maker (registered) completely within 24kb of GML code. It reads an XML file and puts it in a tree-like structure, making it easy to use for the average programmer.
I created this set of scripts to let game makers be able to use the oh-so-handy XML files in their games and applications. I noticed there was a reasonably large demand, but only one tryout about 1,5 years ago.
Meanwhile, I've gotten some competition! If you're having trouble with this XML parser, take a look at this script by Major Victory. It's more line based than SimplXML, it uses the SAX method in stead of the DOM method to read it.
Coding this baby wasn't easy. It took me a full 2 weeks coding just 1 script (the loading script obviously). Yet still, not all functions of XML as stated by w3c are implented (non-declaring parameters, language identification, element type declarations and parsed entities, document type declaration, I didn't think it would be used much / easy enough to program). The functions that are supported and a full documentation of all supported rules is in the readme file.

In the zip file is an example of how to extract information, as well as information on how to use SimplXML and what structure is the supported one.

xml_findvalue() is made by Carlos Ramos, I am grateful to him to make me that script, and you should be too!

Download: (15kb, zipped, version 1.0.2)
ghostkeeper.nl

Version History:

SimplXML 1.0.2 - Updated Carlos Ramos' xml_findvalue() script, he just posted an update. Thanks, mate!SimplXML 1.0.1 - Fixed a small problem with xml_findvalue() and xml_findposition() that would give an error. - Added the .gme file (GM7's beta extension file) to the pack. This will be removed when GM7 goes to alpha mode.SimplXML 1.0.0 - Got enough reactions and did enough tests to release the project in alpha. - Implented 3 new scripts: xml_findvalue(), xml_findposition() and xml_close().SimplXML 0.3.3 - Implented <![cdata[ ... ]]> tag support. - Made the code check for if the file exists before trying to read it.SimplXML 0.3.2 - Bug fixed where a filtered entity gets placed at the wrong position.SimplXML 0.3.1 - Bug fixed where a space at the end of a line caused the program to freeze.SimplXML 0.3.0 - Initial release. 0.1.0 and 0.2.0 were tryouts but failed. Eventually on the third complete redo I succeed.

Have fun with it!

Edited by Ghostkeeper, 04 February 2007 - 10:27 AM.

[takua108]

Wow. That's pretty awesome. I'm still looking into its functionality, but it looks good to me so far. I just wanna parse a configuration file into my game, and this looks like it'll do the trick.

I'll check back in once I've worked further with this. Again, this look sweet so far!

[Killamancer]

what is xml and parse?

[Ghostkeeper]

what is xml

XML is a language that is used to store data. It's often used by professional programs too. Take a look at the example .xml file I included with the download.

and parse?

To parse XML is letting a program read it (as far as I understood, English is not my native language).

I'll look into some more examples, like how to actually make a game with this (the current example only reads it and shows the contents in a tree-form).
Uses for XML are (among others):
- News pages (RSS feeds)
- Configuration files.
- Databases.
- File system.
- Menu structure.
- Even an HTML browser! (hard, but this is the first step)

Edited by Ghostkeeper, 20 October 2006 - 01:48 PM.

[takua108]

It doesn't look at DTD at all, does it? I just figured out how to use DTD today and it would be nice (but not necessary at all) if it could validate the XML document using DTD.

And this was posted at school, so again, I'll verify that everything works once I get home. I'm sure it will ^^

EDIT: Okay...I'm stumped. My XML file looks like this:
[codebox]<!-- ============================================= -->
<!-- = The TOPDOWN Engine Configuration XML File = -->
<!-- ============================================= -->
<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<!DOCTYPE configuration SYSTEM "config.dtd">

<configuration>
<graphics>
<config name="resolution_w" value="1024" />
<config name="resolution_h" value="768" />
<config name="vsync" value="0" />
</graphics>

<performance>

<config name="water_reflect_objects" value="0" />
<config name="water_reflect_sky" value="0" />
<config name="shells" value="0" />
<config name="particles" value="0" />
</performance>

<violence>
<config name="blood_cloud" value="0" />
<config name="blood_splatter" value="0" />

<config name="blood_stay" value="0" />
<config name="gibs" value="0" />
<config name="gibs_spray" value="0" />
</violence>

<player>
<config name="name" value="takua108" />
<config name="color_primary" value="000000" />
<config name="color_secondary" value="FFFFFF" />

</player>

<statistics>
<config name="enabled" value="0" />
</statistics>

<messenger>
<config name="enabled" value="0" />
<config name="status_text" value="music" />
</messenger>

<autoupdate>
<config name="enabled" value="0" />
<config name="remind_check" value="0" />
<config name="remind_new" value="0" />
</autoupdate>

</configuration>[/codebox]
How do I read the attributes of each "<config />," filtered by their parent tag (i.e. configuration, performance, violence, etc.)?

Thanks a million, though; the XML was parsed perfectly!

Edited by takua108, 21 September 2006 - 02:55 AM.

[Ghostkeeper]

Okay...I'm stumped. My XML file looks like this:

<!--            =============================================            --><!--            = The TOPDOWN Engine Configuration XML File =            --><!--            =============================================            --><?xml version="1.0" encoding="UTF-8" standalone="no"?><!DOCTYPE configuration SYSTEM "config.dtd"><configuration>    <graphics>        <config name="resolution_w" value="1024" />        <config name="resolution_h" value="768" />        <config name="vsync" value="0" />    </graphics>    <performance>        <config name="water_reflect_objects" value="0" />        <config name="water_reflect_sky" value="0" />        <config name="shells" value="0" />        <config name="particles" value="0" />    </performance>    <violence>        <config name="blood_cloud" value="0" />        <config name="blood_splatter" value="0" />        <config name="blood_stay" value="0" />        <config name="gibs" value="0" />        <config name="gibs_spray" value="0" />    </violence>    <player>        <config name="name" value="takua108" />        <config name="color_primary" value="000000" />        <config name="color_secondary" value="FFFFFF" />    </player>    <statistics>        <config name="enabled" value="0" />    </statistics>    <messenger>        <config name="enabled" value="0" />        <config name="status_text" value="music" />    </messenger>    <autoupdate>        <config name="enabled" value="0" />        <config name="remind_check" value="0" />        <config name="remind_new" value="0" />    </autoupdate></configuration>

How do I read the attributes of each "<config />," filtered by their parent tag (i.e. configuration, performance, violence, etc.)?

<{POST_SNAPBACK}>

I created this script for you:

xmlfile = xml_open(working_directory+"/YourFile.xml");currentgroup = 0;grouplevel = 0;currentoption = 0;i = 0;repeat(xml_getlength(xmlfile))    {    switch(xml_getvalue(xmlfile,i))        {        case "graphics":            {            currentgroup = 1;            grouplevel = xml_getlevel(xmlfile,i);            j = 0;            break;            };        case "performance":            {            currentgroup = 2;            grouplevel = xml_getlevel(xmlfile,i);            j = 0;            break;            };        case "violence":            {            currentgroup = 3;            grouplevel = xml_getlevel(xmlfile,i);            j = 0;            break;            };        case "player":            {            currentgroup = 4;            grouplevel = xml_getlevel(xmlfile,i);            j = 0;            break;            };        case "statistics":            {            currentgroup = 5;            grouplevel = xml_getlevel(xmlfile,i);            j = 0;            break;            };        case "messenger":            {            currentgroup = 6;            grouplevel = xml_getlevel(xmlfile,i);            j = 0;            break;            };        case "autoupdate":            {            currentgroup = 7;            grouplevel = xml_getlevel(xmlfile,i);            j = 0;            break;            };        default:            {            if(xml_getlevel(xmlfile,i) == grouplevel+3)                {                switch(xml_getvalue(xmlfile,i))                    {                    case "resolution_w": global.resolution_w = real(xml_getvalue(xmlfile,i+2)); break;                    case "resolution_h": global.resolution_h = real(xml_getvalue(xmlfile,i+2)); break;                    case "vsync": global.vsync = real(xml_getvalue(xmlfile,i+2)); break;                    case "water_reflect_objects": global.water_reflect_objects = real(xml_getvalue(xmlfile,i+2)); break;                    case "water_reflect_sky": global.water_reflect_sky = real(xml_getvalue(xmlfile,i+2)); break;                    case "shells": global.shells = real(xml_getvalue(xmlfile,i+2)); break;                    case "particles": global.particles = real(xml_getvalue(xmlfile,i+2)); break;                    case "blood_cloud": global.blood_cloud = real(xml_getvalue(xmlfile,i+2)); break;                    case "blood_splatter": global.blood_splatter = real(xml_getvalue(xmlfile,i+2)); break;                    case "blood_stay": global.blood_stay = real(xml_getvalue(xmlfile,i+2)); break;                    case "gibs": global.gibs = real(xml_getvalue(xmlfile,i+2)); break;                    case "gibs_spray": global.gibs_spray = real(xml_getvalue(xmlfile,i+2)); break;                    case "name": global.name = xml_getvalue(xmlfile,i+2); break;                    case "color_primary": global.color_primary = xml_getvalue(xmlfile,i+2); break;                    case "color_secondary": global.color_secondary = xml_getvalue(xmlfile,i+2); break;                    case "status_text": global.status_text = xml_getvalue(xmlfile,i+2); break;                    case "remind_check": global.remind_check = real(xml_getvalue(xmlfile,i+2)); break;                    case "remind_new": global.remind_new = real(xml_getvalue(xmlfile,i+2)); break;                    case "enabled":                        {                        if(currentgroup == 5){global.statistics_enabled = real(xml_getvalue(xmlfile,i+2));};                        if(currentgroup == 6){global.messenger_enabled = real(xml_getvalue(xmlfile,i+2));};                        if(currentgroup == 7){global.autoupdate_enabled = real(xml_getvalue(xmlfile,i+2));};                        };                    };                };            };        };    i += 1;    };

Be sure to update your filename. Working with this kind of data is hard and you shouldn't be shy to use long code with many switch and if statements. I'd also be glad to help people on their interpretation of XML files like Takua108's. By the way, Takua, I checked out your W.I.P. topic in your signature, looks promising .

And no, it doesn't look at the DTD at all. Main reason: I don't understand it myself! Could you give me some explaination? I've taken a look at it at the w3c page I found (the page in the first post), and it seems quite hard to program too, next to the entities. Another reason: After you said about the DTD, I scanned my C: drive for XML files. I picked several out, but none of them had a <!DOCTYPE> tag.

Last thing: I updated to SimplXML 0.3.2, with another bug filtered. I'll add a version history to the first post.

[takua108]

Wow. I didn't check this topic for awhile, and I was surprised to see that you responded with a fully-functional script. Looking at your code, I understand how your system works now.

Thanks a billion for this system and that script, they're lifesavers!

[the_mysterious_gamer]

This is great, I could see this in use as an alternative for INIs. But as a parser/compiler freak, I do have to tell you something: its coded terribly. Your entire parser is contained in one script, I cannot see any error handling functions and it isnt structured... at all! PM me and I could help you with this.

[Ghostkeeper]

Haha thanks for reanimating my topic!

The terrible coding is because things went wrong all time. Somewhere half way I realized that the string_copy() function returns a string shifted than I thought. I often just tested it, and modified the function (implented +1 or -1 or +2 or something) to make it work right. Also, when testing, sometimes string_copy() returned an empty string, and I used string_delete() in stead which worked right most of the time.
I know now I was just using it wrong probably. However, the code is quite complicated and I don't feel to dig in it again to make it more consistent.

About the entire parser being in one script, I think that's the best way to be, it takes less memory and is easier to overview. There is however (you probably noticed it) a frequent call to the xml_filter() script. This is the only one that I used so often that I found it handy to make a seperate script.

I removed all comment before I published it. If you look carefully though, you can see that there are still empty lines between every action that is performed.

I've learnt very much about string handling functions in GM while coding this. During the code, my coding has adapted and improved. That's why it's so messy.

And nope, there are no error handling functions in there. Thank you for the request though.

[the_mysterious_gamer]

You may think its easier now to keep it in an unstructured way, but I bet you could half, if not quarter (I can make a parser for GML in 300 lines), the lines of code, by making it in a top down structured way. In the long run it will be easier, to look at, and to add to. If you first write some E/BNF of it and create a function for every rule updating will just come naturally, and little to no editting is needed for new features.

EDIT

Haha thanks for reanimating my topic!

I never quite understood that comment until I read the dates...

Edited by the_mysterious_gamer, 22 November 2006 - 08:02 PM.

[Ghostkeeper]

I just posted an update.
I'm sorry to be so slow. This is not due to the messy code The election results of our government are on TV and I'm following it.

[Carlos_Ramos]

I know it's almost been a month, but thanks for these scripts
This is a clever way to store the XML info, I would never have figured this out.

[EDIT]
I'm writing an add-on script to search for the value of a specific tag via the tags name to help you out

Edited by Carlos_Ramos, 14 December 2006 - 02:40 AM.

[Jugado]

Keep working Ghostkeeper. Nice job!!!
It would be nice to see in a future GM working with a XML database, for example this one: dbxml

[Carlos_Ramos]

Ok, here's a script to find the value of a tag. (You'll need to create and rename a script, then copy the following code into it)

////xml_findvalue//////////////////////////////////////////////////
// This function finds the value of specific tag based on its name
//
// (argument0: (real) ID of the XML file, as returned by xml_open()
// (argument1: (string) tag name
// (argument2: (real) place to start looking
// return: (string) The value of the tag
// return: -1 if no value found
///////////////////////////////////////////Carlos Ramos////////////

//Search through the entire file
repeat(xml_getlength(argument0))
{
    value = xml_getvalue(argument0,argument2);  //Grabs the value at location i

    //Checks if the next positions layer is greater than this one, if so, then return the value
    if(string(value) == string(argument1) && xml_getlevel(argument0, argument2+1) > xml_getlevel(argument0, argument2))
    {
        return xml_getvalue(argument0,argument2+1); //Return the string
    }

    argument2+=1;   //Next position
}

return -1;  //No values found, end function

The way it works is by checking the current level, and returning the value that is one level less.

Next, I'll make it so you can enter in a tags pathname, for more specific searches.

[Carlos_Ramos]

Ok, I've got it. Replace the last script with this one, this script works with both pathnames, and single tag names:

//////////////////////////xml_findvalue////////////////////////////////
// This function looks for the value of a tag at with a specific pathname
//
// (argument0: (real) ID of the XML file, as returned by xml_open()
// (argument1: (string) tag url
// (argument2: (real) place to start looking
// return: (string) The value of the tag
// return: -1 if no value found
/////////////////////////////////////////////////////Carlos Ramos/////////

//Setup Variables
num = 0;    //Number of strings
str[0] = argument1; //Stores the individual sections of a string, index 0 = the string containin '/'
parent_level = 0; //The parents level
cur_tag = 0;    //The current tag we are searching for

//Split up the url into separate strings
for(i = 1; i <= string_count("\",str[0]); i += 1)
{
    str[i] = string_copy(str[0],0,string_pos("\",str[0]));   //Grab a section
    str[0] = string_copy(str[0],string_pos("\",str[0])+1,string_length(str[0])-1); //Remove the grabbed section from the url string
    i-=1;   //Because we took one out, we need to decrease the current i
}

//Search through the entire file
repeat(xml_getlength(argument0))
{
    value = xml_getvalue(argument0,argument2);  //Grabs the value at location i

    if(cur_tag >= num){cur_tag = 0;}    //str[0] actually contains the last tag, so look for this last
    
    //show_message(str[cur_tag] + " :: " + value);
    //If we find the wanted tag, store it's info
    if(str[cur_tag] == value)
    {
        parent_level = xml_getlevel(argument0,argument2);   //Store this level
        
        //No longer searching here, so clear the parent variables    
        if(xml_getlevel(argument0,argument2+1) <= parent_level){parent_level = 0; cur_tag = 1;}

        if(cur_tag == 0){return xml_getvalue(argument0,argument2+1);}   //Return the value
        cur_tag += 1;
    }

    //No longer searching here, so clear the parent variables    
    if(xml_getlevel(argument0,argument2) <= parent_level){parent_level = 0; cur_tag = 1;}
    
    argument2+=1;   //Next position
}

return -1;  //No values found, end function

This function allows you to search for the value of a tag's pathname. Lets take the following:

<a>
   <b>
   <e>
         <d>  </d></e>
   <f>
      <c>
         <d> Hello World </d></f></c></b></a>

To extract "Hello World", you would look for the tag named d. However, the first d contains no useful information. So, to extract the d we need, we use

xml_findvalue(xml_file, "a\f\d");

Hope this helps out.

[Ghostkeeper]

Sure does!

I've completed version 1.0.0 and I've included (a slightly modified version of) your script in this, Carlos.

Also, be sure to check out the new function to get the position of an item with a certain path.

[Ansgar]

This might be quite handy. I hope I'll find a use for it.
One little glitchie: The readme file still says it's version 0.3.1

Cheers
-Ansgar aka Zombone-

EDIT: It's fixed now. Thumbs up.

Edited by Ansgar, 16 December 2006 - 11:11 AM.

[Ankerheegaard]

Hi there,

Looks like a great way to extend Game Maker, but I have this simple XML file like this;

<?xml version="1.0" encoding="ANSI" standalone="no"?>
    <Areas>	
	<Area ID="1" Object="100068" Name="Fiona" Country="Denmark" />
	<Area ID="2" Object="100070" Name="Holstein" Country="Denmark" />
   </Areas>

Ill try to extract the values by ID but I keep getting only the first value "Fiona" when i use the following code;

area_name=xml_findvalue(XML_FILE_NAME,"Areas\ID\2\Name");

What am I doing wrong?

Thank you in advance.

[tangibleLime]

You forgot my script set (based off of databot's): http://www.technosof...ipts/GM-XML.zip

[Ankerheegaard]

You forgot my script set (based off of databot's): http://www.technosof...ipts/GM-XML.zip

<{POST_SNAPBACK}>

Thank you for your quick reply, but I think I got the same problem with your scripts. I still only get the first area (Fiona).