Board index » cppbuilder » Re: Static linked library in Package not allowed

2007-08-10 11:11:50 AM
cppbuilder77
it is a dynamic loaded library instead of a standalone executable. It does
NOT contain object files.
. Ed

Valence wrote:
or higher level programming language. When advocating precise usage of
words, please do.
.a
p.s. the code in import library doesn't do anything. It's linker and OS
loader that do things.
 

"Alex Bakaev [TeamB]" < XXXX@XXXXX.COM >wrote in message
.a
 

{smallsort}

Valence wrote:
it.
--
Rudy Velthuis [TeamB] rvelthuis.de
"Forgive your enemies, but never forget their names."
-- John F. Kennedy (1917-1963)
 

"Ed Mulroy [TeamB]" < XXXX@XXXXX.COM >wrote in message
Thanks Ed. I can always count on you for exacting and precise information. I
love all you guys, and find these exchanges extremely fun, and informative.
I remember the good ole' days, and expected that you would eventually find
this thread irresistable. It seemed right down your ally.
Where's Chris to rebut? I haven't seen you guys spar in a good while!
<\greetings>
<retoric>
I think that the BCB 6 documenter (god rest his soul) was trying to be
general enough to be "generally" understood. He wasn't communicating to the
ultra-geeks or the prudes. At least not to level of exactitude that is, of
course, necessary in this thread.
<\retoric>
 

"Rudy Velthuis [TeamB]" < XXXX@XXXXX.COM >wrote in message
 

"Ed Mulroy [TeamB]" < XXXX@XXXXX.COM >wrote in message
rebuttal. Do know that I have updated my vocabulary, with support for you
guys and google, and accept the definition of "implicit linking". With
reference to a "static" import library. I will simply say, sorry I didn't
quote it as I tried to quote everything else like this.
linker per that library is dynamic, how so? And if not dynamic, is it
static.
But again, I forget the thread I'm in. I'd better add that I agree that an
import library is not a static library, though it might be obvious
elsewhere. Any attempt to be general here is certainly folly.
 

Valence wrote:
--
Rudy Velthuis [TeamB] rvelthuis.de
"Gentleman: Knows how to play the bagpipes, but doesn't."
 

Valence wrote:
using, just to win your argument. That is plain silly. <shrug>
So to be clear: import libraries do not contain executable code. "And
thats not up for debate" (sic).
--
Rudy Velthuis [TeamB] rvelthuis.de
"If we knew what it was we were doing, it would not be called
research, would it?" -- Albert Einstein
 

Rudy Velthuis [TeamB] < XXXX@XXXXX.COM >wrote:
ask something that you then answer yourself
later?
Schobi
--
XXXX@XXXXX.COM is never read
I'm HSchober at gmx dot de
"If there were some arcane way to remove the heads of every
newsgroup troll on the planet, I think it would elevate
humans to a whole new level of intelligence."
Rocky Frisco
 

Hendrik Schober wrote:
interested in your take.
--
Rudy Velthuis [TeamB] rvelthuis.de
"Maybe this world is another planet's Hell."
-- Aldous Huxley (1894-1963)
 

Rudy Velthuis [TeamB] < XXXX@XXXXX.COM >wrote:
"explicit linking" doesn't seem to fit as well to me.
But that doesn't change the fact that I keep hearing
"static/dynamic linking" referring to DLLs and that
most people know what it refers to. With even both MS
(who, after all, invented DLLs) and Borland/Codegear
using these terms, I don't see how one can call it
stupid to talk about "statically linking a DLL". (This
was, if you care to remember, where I came into this
discussion.) The term is used that way whether one
likes it or not. And there are far worse terms used
throughout our field (like RAII, which isn't about
initialization at all).
Schobi
--
XXXX@XXXXX.COM is never read
I'm HSchober at gmx dot de
"If there were some arcane way to remove the heads of every
newsgroup troll on the planet, I think it would elevate
humans to a whole new level of intelligence."
Rocky Frisco
 

"Hendrik Schober" < XXXX@XXXXX.COM >wrote:
initialisation that is meant - you're initialising the object with the
allocated resource.
(Bjarne has been reported as saying that he apologises for that
terminology. But we look to be stuck with it.)
Alan Bellingham
--
Team Browns
ACCU Conference 2008: 2-5 April 2008 - Oxford (probably), UK
 

Valence wrote:
discussion about software development. Every profession has a vocabulary
that is unique to it. When speaking within a profession, you are
expected to understand and use that vocabulary appropriately. In the
software development community, code means machine code, not the born on
date code that you see on bottle of Budweiser.
A PE executable file consists of various segments. Some segments contain
code (ie machine code), and some contain data. The import table resides
in a data segment called idata. If you start claiming that the import
data segment contains 'code' from the import library, then you are just
going to confuse people.
You wrote this:
<quote>
The import library used to "statically" link to a dll does contain the
appropriate code to dynamically link to the dll and bind your names to it.
</quote>
Every sane developer would read that and think that you are talking
about machine code in the import library. Not letters, numbers and symbols.
to mean machine instructions that an x86 CPU can process and execute.
The code is a jump to an address. However, that code only ends up in the
code segment of the executable if the DLL author is stuck in the year
1994. For most DLLs, the code in the import library is discarded, and
the linker inserts a more efficient call.
See this article by Matt Pietrek, specifically the section on importing
functions. Pietrek is a very smart guy. It is well worth it to read just
about everything he writes.
msdn.microsoft.com/msdnmag/issues/02/02/PE/
He also has a follow up article here:
msdn.microsoft.com/msdnmag/issues/02/03/PE2/default.aspx
H^2
 

Valence wrote:
not know everything, but one thing I have learned from being a citizen
of this country is that being in the majority doesn't make you right.
H^2