hello:
This can be a mis understanding but ..

from the manual...im using file_str_readto.
and :delimiter - STRING* or char*, used as end mark for the text, or NULL for no end mark.

So if i use null, there is no delimiter string, and i should get the full file load, (4000 chars max.. ok...)
STRING* rss_down="#400"
var filehandle_n;
filehandle_n=file_open_read ("rss_1.txt");
file_str_readto(filehandle_n,rss_down,NULL,4000);
file_close(filehandle_n);
error(rss_down); ->> check the output

the string contains only 1 word, after a space there is another word , but those are not shown , only the first.

replacing the delimiter for:
file_str_readto(filehandle_n,rss_down,"|",4000);

Does work. but .. i read NULL had no end mark , thus no limit to end?

Is this a missing detail? or i did something wrong?


---> last news:

in fact , with some time of studding this weird behavior... i found why that happen...

well,if the text is this:

Hello world , it will read, hello world if, the delimiter is set to null. (no end mark)

If the text is:
</tag> <rss version> : please note that i did a space between the tag and rss version, and in this case this 2 words output well and it works , and the string read matches.

But when i do without space: like this:

</tag><rss version> guess what is read? Answer: only </tag> the rest , that actually is part of the same word, since there is no space is missing...

So this is a bug somewhere....
it seams that this letters <>< can't be together this is what i call.. the fish bug, note the simbol.. <>< it looks a fish to me! ehhe.. and if the text contains <>< the tail is lost.

but this happens in a special condition...

this is the condition.. that i dont understand there must be some unix code or something hidden... i got this from a HTML CODE.
And its kind a funny..
So: try this 2.
Copy and past the bellow text into the .txt file that is loaded using the NULL delimiter! then on the txt file, select the fish tail (second"<" and cut the space between them with the backspace key so it will look a fish <>< then try to read.. and you will the read on the error output "<><" will become this "<>" only...

but if you erase everything and type by your own, <>< it will work.. but copy pasting this code of text wont.. which is weird.


this is the "magical" text you can copy and paste on a txt file:

<>

<



--> this looks like UNICODE UTF-8 Incompatible with GS... Which is bad:(

so for loading txt files with a html file, where you extracted from a URL, you need to USE the delimiter not NULL but "|" , since "|" its very rare to appear.


you can try this with the GS page for example...

Right click right now! on mozzila or other internet explorer, go to the view HTML source, save to txt file, UTF-8 (default for HTTP and web, or EMAIL communications), and let just the 3 lines just that.. and now past into the txt file, and try to read.. and you will notice only the first tag <bla bla> will appear.. there is a hidden character somewhere or some sort..

I dont quite get you need.

But IF you are just trying to read the ENTIRE file, try (tested)

I think using NULL in file_str_readto just means
"use the normal delimiters and delimit_str".

ANd im also getting errors with stupid unecessary things..

i have the file, that is encoded in UTF-8 or big endiens, or the unicode.. and the file has nothing on it, and the error outputs, weird simbols, and it wont recognize this : árvore and other words similar that use ^ ~ ´ ` acents...

Can Gs be able to read and decode UNICODE and UTF-8 properly? and not only AScii? because ...
im downloading html with a plugin and when i say to save the url to a xx.txt file, dont know why but it saves with UTF-8 coding. and then when i import that using Gs there is .. those little/big word simblos messing up the text..


from wikipedia:

which is exaclty what im recieving.. on the error message, even with no text..

I read also on wikipedia , there is a windows api to inter convert characters

hey, no its not NULL is NO delimiter.. i tryed my selft tons of ways and times.


Off-topic, i need to sleep its 6:14 AM jesus.

Long thread, short answer: Yes, when no delimiter is given, the default delimiter is used. Give "" when you want to read to the end of file. I'll change that so that NULL also reads to the end of file. UTF-8 is a web site coding and has nothing to do with text reading.

Jcl! sorry for the thread. But my file has nothing on it.. And engine is returning Simbols? i want it to convert the Reading from UTF-8 to ascii, becausse it can't Read words like this genética árvore acção .. understand me ? this is not right... That happens because the txt file that file-open-read is trying to read is not on ascii but utf-8 or something?

have you got any idea, how can i make it read text on a txt file that is encoded in utf-8 ?

would be nice to hear.

I guess the problem is some misunderstanding. The engine just reads your file, byte-wise. It does not convert UTF-8 or whatever coding you are using. You must do code conversion yourself. For this, build up a conversion table and a target font that contains the desired characters.