roxen.lists.pike.general

Subject Author Date
Re: FakeFile() and to assign() or not to assign() Stephen R. van den Berg <srb[at]cuci[dot]nl> 12-02-2009
Martin Stjernholm wrote:
>"Mirar @ Pike  importm??te f??r mailinglistan" <<6341[at]lyskom.lysator.liu.se>>
wrote:

>> I thought that was because Stdio.FILE were allowed to buffer, while
>> Stdio.File is supposed to be lowlevel and not buffer. (Like fd->write
>> and FILE*->fwrite in C.)

>True, the contents of a FakeFile is a string, which you might call
>buffer. But implementing unread() and ungets() would require another
>buffer on top of it, so you'd have to add much of the Stdio.FILE
>implementation on top of FakeFile too. I think a good solution would
>be if Stdio.FILE and FakeFILE shared the same implementation.

Take a look at the implementation I provided.  It's lightweight.
In fact, check all of my submissions.  The series is done, for the moment.
Special points to review:
- Documentation for Parser.Tabular; I tried to be complete, it's not
  perfect yet (formatting autodoc has never been my strong suite).
- The test I added for Parser.Tabular, is this going to work and/or
  prudent?
-- 
Sincerely,
           Stephen R. van den Berg.
The first 90% of code accounts for the first 90% of development time.
The remaining 10% of code accounts for the other 90% of development time.