Re: Bad generated C-code ... - creation constraints.
- Further observations:
I have tried to compile with SmartEiffel, to see if it can find
anything od. This is quite hard work as the libraries are different
and the compiler needs tweaking to recognise the ISE syntax (e.g. SE
has a quite different syntax for free operators, inconsistent with
ETL First Edition, e.g. it rejects 'infix "#Whatnot"'). It is also
not as productive as EiffelStudio, as there is no browser.
It seemed for a moment that SmartEiffel had cracked it, as it
reported a lot of syntax errors in my generic classe: weird 'create'-
expressions where they ought not to be. But then I realised I had
put them there because EiffelStudio said the generic arguments
needed 'creation constraints', copying the syntax from the
libraries. Another tweak to SmartEiffel needed..
What is the status of creation constraints? Does the compiler fully
In fact, how do I get an authoritative definition of the language
without subscribing to all future editions of ETL?
> What is the status of creation constraints? Does the compilerI cannot talk for other Eiffel compilers. The Eiffel Software's Eiffel compiler
> fully support them?
full support creation constraints.
> In fact, how do I get an authoritative definition of theSoon we will have an ECMA standard for Eiffel that will be able to freely download
> language without subscribing to all future editions of ETL?
the standard and have the authoritative definition of Eiffel.
- For the record, there was indeed a problem that will be fixed in our next release.
Thanks for reporting it,
> -----Original Message-----
> From: prunesquelch [mailto:Patrick.Traill@...]
> Sent: Sunday, March 28, 2004 2:13 AM
> To: firstname.lastname@example.org
> Subject: [eiffel_software] Re: Bad generated C-code passes
> EIF_INTEGER_32 to EIF_REFERENCE (corrected)
> Thank-you very much for your reaction.
> I am in the process of reporting it as requested.
> I have spent quite a bit of time reducing the system, but
> reducing it too much makes the problem go away.
> I do not believe that I have tested it in finalized mode, but
> if asked to I shall be glad to do that.
> I was working from memory, not having access to my source.
> The operator may have been '>=' rather than '<='.
> The postcondition was (logically enough) defined in the base class.
> --- In email@example.com, "Emmanuel Stapf [ES]"
> <manus@e...> wrote:
> > Create the bug report ...
> > If possible try to reduce the size of the system as much as you can.
> > Also does this issue occurs only in workbench mode or also in
> > finalized mode?
> > Thanks,
> > Manu
> > > -----Original Message-----
> > > From: prunesquelch [mailto:Patrick.Traill@T...]
> > > Sent: Friday, March 26, 2004 8:03 AM
> > > To: firstname.lastname@example.org
> > > Subject: [eiffel_software] Re: Bad generated C-code passes
> > > EIF_INTEGER_32 to EIF_REFERENCE (corrected)
> ------------------------ Yahoo! Groups Sponsor
> ---------------------~--> Buy Ink Cartridges or Refill Kits
> for your HP, Epson, Canon or Lexmark Printer at MyInks.com.
> Free s/h on orders $50 or more to the US & Canada.
> Yahoo! Groups Links