Quote
That would somehow encourage mods not to interfere with each others data, but still doesn't enforce it. Adding the extra column won't be a problem, but then all of the ACS DB functions get an additional argument. What's the general opinion on this? Since a mod could just write "zanrun-time" (one argument, i.e. namespace) instead of "zanrun", "time" (two argument, i.e. namespace and section), we don't really get much out of the more complicated interface except of this slight encouragement to use a unique namespace. Another possibility would be renaming the namespace argument in the ACS functions to section and to force selecting a namespace via a new lump.

Quote from Watermelon

We also have to take into account hosting services like Best-Ever where people might maliciously try to write into other people's DB's.
How does one stop this?

Quote from AlexMax

However, part of me also agrees with you about the additional parameters to each DB command that would require. Perhaps the best thing we can do is document the use of these functions and ensure that the documentation has the mod-specific namespace as part of the documentation. That way, people who "learn" by copypasting will get this extra namepsacing "for free".

Quote from AlexMax

Are there any other instances in the ZDoom ACS API that global state is used...other than perhaps setting the currently in-use font?

Quote from AlexMax

Also, just an unrelated side note, I feel like providing constants along with the API like DB_ORDER_ASC/DB_ORDER_DESC would be a good idea, to make the function calls more readable by sight.

Quote from Jroc

I'd just like to say that it's definitely useful to have full control over both a keyname and namespace as I've seen some question whether we should be allowed to specify namespace as a parameter in each function.

#define CASE_DIFFERENCE ('a' - 'A')
int g_toLowerBuffer[1024];

function str toLower (str s)
{
    int i, c;

    for (i = 0; str[i] != '\0'; ++i)
    {
        c = GetChar (str, i);

        if (c >= 'A' && c <= 'Z')
            g_toLowerBuffer[i] = c + CASE_DIFFERENCE;
        else
            g_toLowerBuffer[i] = c;
    }

    g_toLowerBuffer[i] = '\0';
    return strparam (a:g_toLowerBuffer);
}

Quote from Jroc
Also, can I request the public beta on zandro.com be updated with the DB features?

Quote from Jroc

I think along with the sorting, something like GetDBSortedPosition will be needed. It's all very well to be able to sort and get the top x entries, but what if you want to find the position of a specific entry

Quote from StrikerMan780
So, the database system only uses SQLite? Or is it possible to connect to an external MySQL database server? I kind of need to be able to connect to an external server if I'm gonna do what I plan to do.

    -108:SetDBEntryInt(3),
    -109:GetDBEntryInt(2),
    -110:SetDBEntryString(3),
    -111:GetDBEntryString(2),
    -112:IncrementDBEntryInt(3),
    -113:PlayerIsLoggedIn(1),
    -114:GetPlayerAccountName(1),

Quote from StrikerMan780
How would I connect it to a MySQL server then? Any examples?

void DATABASE_SaveSetEntry ( const char *Namespace, const char *EntryName, const char *EntryValue );
void DATABASE_SaveSetEntryInt ( const char *Namespace, const char *EntryName, int EntryValue );
FString DATABASE_SaveGetEntry ( const char *Namespace, const char *EntryName );
void DATABASE_SaveIncrementEntryInt ( const char *Namespace, const char *EntryName, int Increment );

Quote from StrikerMan780

About threading... Hate always using it to compare with, but SourcePawn cannot use threading either. However, they get around that by using Callbacks.

Quote from Jroc
Player rankings will be one of the main uses of this database feature, there are already plans to add rankings for duel and ctf players.

Quote from StrikerMan780
With Zandronum on the other hand, Is there a way to get an external library like that to tell Zandro to execute an ACS Script? or is it not capable of that whatsoever?

Quote from StrikerMan780
Probably return blank query results, and parameter used to determine whether or not it failed, allowing the modder to choose how to react to the situation. The game would keep going as it's trying to connect. That's the whole idea of using a callback instead of just waiting for the function to return.

Quote from StrikerMan780

The only other thing I can think of... is some form of MySQL proxy. Having all stuff done via SQLite, and then having some sort of external system that synchronizes the databases periodically, merging changes/differences.

Quote from AlexMax
Keys are designed to be unique. Values are not. If you want a reverse mapping, just put the key/value pair in the database twice - once one way and once the other.

Quote
Alright, I added the internal function DATABASE_GetEntryRank and will expose this via ACS.

Quote from StrikerMan780
No way to get ACS Functions to call things that don't reside in the ACS System in and of itself to break free of whatever limitations ACS may have?

Quote from StrikerMan780
Perhaps there's a way to get a timestamp / last modified time of sorts? There must be some way to determine if the values in the database are outdated.

    int QuerySortedDBEntries ( string Namespace, int N, int Offset, bool Ascending ),
    int GetDBResultSize ( int Handle ),
    void ClearDBResultHandle ( int Handle ),
    string GetDBResultEntryNameString ( int Handle, int ResultsIndex ),
    string GetDBResultEntryValueString ( int Handle, int ResultsIndex ),
    int GetDBEntryRank ( string Namespace, string EntryName, bool Ascending ),

    -115:QuerySortedDBEntries(4),
    -116:GetDBResultSize(1),
    -117:ClearDBResultHandle(1),
    -118:GetDBResultEntryNameString(2),
    -119:GetDBResultEntryValueString(2),
    -120:GetDBEntryRank(3),

int handle = QuerySortedDBEntries ( "Random", 3, 0, 1 );
int numEntries = GetDBResultSize(handle);
Log (s:"Query returned entries: ", d:numEntries );
for ( int i = 0; i < numEntries; ++i )
{
    Log ( s:GetDBResultEntryNameString(handle, i), s:" ", s:GetDBResultEntryValueString(handle, i) );
}
ClearDBResultHandle ( handle );

Quote from Jroc
Thanks, this basically answers all of my questions, but why isn't there a function to return an entry value as an int rather than a string? True I can just use GetDBResultEntryNameString in GetDBEntryInt, but I can do that with GetDBEntryString too, so how come there is a function specifically to get the string?

Quote from Dusk
Looks like we could use an ACS constant for query sort order? That is, if I'm reading the QuerySortedDBEntries call correctly.

// Zandronum database additions ---------------------------------------------
#define DB_ORDER_DESC 0
#define DB_ORDER_ASC 1

Quote from Jroc

Could there be a flag that can enable or disable the database features, and can it be off by default?

    -115:SortDBEntries(4),
    -116:CountDBResults(1),
    -117:FreeDBResults(1),
    -118:GetDBResultKeyString(2),
    -119:GetDBResultValueString(2),
    -120:GetDBResultValue(2),
    -121:GetDBEntryRank(3),

#define DB_ORDER_ASC 0
#define DB_ORDER_DESC 1

Quote from Jroc

Perhaps some guidance on how many times Zandro can query the database in one tic before performance is affected would be helpful.