Re: [cc65] Common macros for dirent.d_type

From: <>
Date: 2012-07-07 11:48:33
On 2012-07-07, at 05:13, Groepaz wrote:

> On Friday 06 July 2012, you wrote:
>> CBM DOS filetypes are similar in function to beloved msdos filename
>> extensions, aren't they?
> not at ALL.

Why? They are both part of metadata of the file. They both suggest what the content may be. They both suggest what you might do with the file. They both impose some restrictions on how they are handled by default... 

>> Yet you don't insist on using FOO.EXE for data files instead of e.g.
>> FOO.DAT, do you?
> for the most part i find it utterly stupid to draw conclusions from the 
> filename to its content.

We are on the very same page! That's why I wrote "beloved" ;-) Yet this utter stupidity is with us for long and not only it seems to be there to stay but moreover it already infected or is further infecting systems, which were previously immune to this form of stupidity. Mac, Linux.. 

> the one and only data file i have ever used on pc for my own stuff was named 
> "config.sys", btw =)

I see... not :-P

>> Why then examples in the books about programming data access using
>> sequential files uses SEQ files? Do you think the authors were stupid?
> they were making examples for TEXT files.

Which might be the most common / easiest to explain type of use. Yet I don't recall any of them saying "don't use SEQ for anything else than textual data". It is rather just plain natural to use SEQ when it comes to sequential DATA. Which is BTW the same what CBM DOS does when opening file for writing with appropriate secondary address! I admit that for a long moment I was fooled by your argument that "BASIC creates PRG by default". This is not true as Andre pointed out. It happens only when you use channel# 1 (aka sec addr 1), using which is against the rules set in 1541 manual for example:

   OPEN file#, device#, channel#, "0:name,type,direction"
[...] The channel# is a data channel, number 2 through 14 [...]

Which IMHO takes away the only valid point I thought you had but - OK - seems that we keep having different opinions on this matter.


To unsubscribe from the list send mail to with
the string "unsubscribe cc65" in the body(!) of the mail.
Received on Sat Jul 7 11:48:48 2012

This archive was generated by hypermail 2.1.8 : 2012-07-07 11:48:52 CEST