Re: [xml] Error in the Tutorial

[Raymond Wiker <Raymond Wiker fast no>]

Igor Zlatkovic writes:
> On 27.05.2004 02:54, John Fleck wrote:
>
> > Thanks, I'll fix that.
>
> No, John, leave it. The code is working and I have no idea why would a 
> compiler give an error here. If you want to fix it, then better do a
>
>    xmlChar *xpath = (xmlChar*) "//keyword";
>
> Duplicating the string here is misleading.
>
> By the way, I cannot reproduce the error with any of the three compilers 
> I have here:
>
> Microsoft's Visual C/C++ says it isn't a big deal:
>
>    C:\Temp>cl -Ic:\opt\fsw\include -W4 -c xpathex.c
>    Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 13.10.3077
>     for 80x86
>    Copyright (C) Microsoft Corporation 1984-2002. All rights reserved.
>    test.c
>    test.c(37) : warning C4057: 'initializing' : 'xmlChar *' differs
>     in indirection to slightly different base type from 'char [10]'
>    C:\Temp>
>
>
> GNU C compiler doesn't say a thing. It just compiles.
>
>    C:\Temp>gcc -Ic:\opt\fsw\include -Wall -c xpathex.c
>    C:\Temp>
>
>
> Intel's C/C++ compiler is very conversational, but it doesn't mention 
> our line 37:
>
>    C:\Temp>icl -Ic:\opt\fsw\include -W4 -c xpathex.c
>    Intel(R) C++ Compiler for 32-bit applications, Version 8.0
>     Build 20031017Z
>    Copyright (C) 1985-2003 Intel Corporation.  All rights reserved.
>    test.c
>    test.c(5): remark #1418: external definition with no prior declaration
>      getdoc (char *docname) {
>      ^
>    test.c(18): remark #1418: external definition with no prior
>      declaration getnodeset (xmlDocPtr doc, xmlChar *xpath){
>                  ^
>    test.c(55): remark #181: argument is incompatible with corresponding
>      format string conversion printf("keyword: %s\n", keyword);
>                                                       ^
>    C:\Temp>
>
>
> Note that all three compilers produced a valid object file. Also note 
> that in all three cases I had to turn all possible warnings on in order 
> to get any message on the screen. Finally note that only one of the 
> three had mentioned the line 37 at all.
>
> Concusion: Don't trust t-onliners. They lie. :-)

        I think you'd need a combination of high(ish) warning level
and compiler/system that treats chars as signed to get problems. In
any case, the correct solution is (as you say) to cast to xmlChar *. I
have definitely had to add such casts to placate the compiler on one
or more platforms.