OpenEuphoria

Re: Resources and Euphoria (Opening a new thread)

• Posted by "Graeme." <hmi at POWERUP.COM.AU> Jun 05, 1998
• 784 views

At 09:13 PM 6/3/98 -0700, you wrote:
>---------------------- Information from the mail header -----------------------
>Sender:       Euphoria Programming for MS-DOS <EUPHORIA at LISTSERV.MUOHIO.EDU>
>Poster:       Hawke <mdeland at NWINFO.NET>
>Subject:      Re: Resources and Euphoria (Opening a new thread)
>-------------------------------------------------------------------------------
>
>> Jesus wrote:
>>
>> >So I was proposing to have a single file with a dual contents. At the
>> >beginning, the normal .EXE. After the .EXE, a kind of "resource" file that
>> >would consist on an array (or sequence) of headers that would describe the
>> >later contents. A header should have a couple of pointers -where the
>> >resource begins and its length-, a name -to identify it-, a descriptor -to
>> >know what kind of resource is-, etc.
>>
>
>hrmmmm....
>If I read this right, you want data appended to a bound program.
>Since binding _already_ appends 'data' (ergo: include files) and since
>there is _already_ a program that converts data (bmp's I believe) into
>'includable' files...(can't remember the filename for the bmp->include
>converter...anyone?)...why not modify that program for your other data
>types as well? the waves and such can all be turned into ahem 'ascii'
>(after all, isn't all data ascii anyway :) )
>
>Here's the benefits/detrements......

The include file is one of mine, its called incbm.e
It already contains a routine called incseq() for including
data sequences, it is definatly the easiest way to include
small to medium sized amounts of data with a bound file.

The system does, however have one major draw-back; it requires
all included data to be declared as constant (loaded into memory
and possibly swapped out) at program execution. The system
Jesus and David Cuny are talking about would allow rescources
to be loaded and freed at will (highly desirable).

Graeme

BTW: One post of incbm.e had a couple of bugs
in it (like a 64K limit on data sequences). If
anyone is interested in using it you can get the
final version here:





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