Re: X3: Alternative operator doesn't reduce?

--===============7641108187147454226==
Content-Type: multipart/alternative; boundary="000000000000671f2205aa1d79fb"

--000000000000671f2205aa1d79fb
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Fri, Jul 10, 2020 at 2:43 PM Michael Powell <[email protected]>
wrote:

>
>
> On Fri, Jul 10, 2020 at 5:38 PM Aaron Graham <[email protected]>
> wrote:
>
>> X3 compound attribute rules
>> <https://www.boost.org/doc/libs/1_73_0/libs/spirit/doc/x3/html/spirit_x3=
/quick_reference/compound_attribute_rules.html>
>> say this:
>>
>> a: A, b: A --> (a | b): A
>>
>>
>> However, when I write this simple bit of code (exposition purposes only)=
:
>>
>> #include <boost/spirit/home/x3.hpp>
>> int main() {
>>     using namespace boost::spirit::x3;
>>     auto const quoted_string =3D lexeme['"' >> *(char_ - '"') >> '"'];
>>     auto value =3D std::string{};
>>     auto assign_string =3D [&](auto& ctx){ value =3D _attr(ctx); };
>>     auto const any_string =3D (quoted_string | quoted_string)
>> [assign_string];
>>     auto val =3D std::string_view{"\"Test\""};
>>     phrase_parse(val.begin(), val.end(), any_string, space);
>> }
>>
>>
>> I get this error:
>>
>> error: no match for =E2=80=98operator=3D=E2=80=99 (operand types are =E2=
=80=98std::__cxx11::b
>> asic_string<char>=E2=80=99 and =E2=80=98
>> boost::variant<boost::detail::variant::over_sequence<boo
>> st::mpl::l_item<mpl_::long_<2>, std::__cxx11::basic_string<char>,
>> boost::mpl::l_
>> item<mpl_::long_<1>, std::__cxx11::basic_string<char>, boost::mpl::l_end=
>
>> > > >=E2=80=99
>> )
>>    7 |     auto assign_string =3D [&](auto& ctx){ value =3D _attr(ctx); =
};
>>      |
>>
>> Why is the attribute a variant instead of a string?
>>
>
> AFAIK, this is the way alternatives have always worked since even before
> X3.
>

I've written a lot of very complicated qi parsers and I don't remember ever
running across this problem. It's possible it just never came up, but I
doubt it.

But it doesn't matter. The documentation I linked in my previous email says=
:

a: A, b: A --> (a | b): A

and not:

a: A, b: A --> (a | b): variant<A, A>

But the error message from the compiler clearly expresses that that's what
I'm getting. I'll dig into the x3 code when I have some time and try to
figure it out.

Which being said, I'm not sure it makes a whole lot of sense, even in an
> academic sense, to alternative, (qs | qs), does it?
>

I knew someone would answer that way, which is why I had the disclaimer
"exposition purposes only". My real code looks a little different. But the
attribute type is the same. The point in this example is to present code
that's alternating between two clearly identical attribute types.

--000000000000671f2205aa1d79fb
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div dir=3D"ltr"><br></div><br><div class=3D"gmail_quote">=
<div dir=3D"ltr" class=3D"gmail_attr">On Fri, Jul 10, 2020 at 2:43 PM Micha=
el Powell &lt;<a href=3D"mailto:[email protected]">[email protected]=
m</a>&gt; wrote:<br></div><blockquote class=3D"gmail_quote" style=3D"margin=
:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"=
><div dir=3D"ltr"><div dir=3D"ltr"><br></div><br><div class=3D"gmail_quote"=
><div dir=3D"ltr" class=3D"gmail_attr">On Fri, Jul 10, 2020 at 5:38 PM Aaro=
n Graham &lt;<a href=3D"mailto:[email protected]" target=3D"_blank">aar=
[email protected]</a>&gt; wrote:<br></div><blockquote class=3D"gmail_quote=
" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);=
padding-left:1ex"><div dir=3D"ltr">X3 <a href=3D"https://www.boost.org/doc/=
libs/1_73_0/libs/spirit/doc/x3/html/spirit_x3/quick_reference/compound_attr=
ibute_rules.html" target=3D"_blank">compound attribute rules</a> say this:<=
blockquote style=3D"margin:0px 0px 0px 40px;border:none;padding:0px"><div><=
span style=3D"color:rgb(0,0,0);font-size:12px">a</span><span style=3D"font-=
size:12px;color:rgb(112,112,112)">:</span><span style=3D"color:rgb(0,0,0);f=
ont-size:12px"> </span><span style=3D"color:rgb(0,0,0);font-size:12px">A</s=
pan><span style=3D"font-size:12px;color:rgb(112,112,112)">,</span><span sty=
le=3D"color:rgb(0,0,0);font-size:12px"> </span><span style=3D"color:rgb(0,0=
,0);font-size:12px">b</span><span style=3D"font-size:12px;color:rgb(112,112=
,112)">:</span><span style=3D"color:rgb(0,0,0);font-size:12px"> </span><spa=
n style=3D"color:rgb(0,0,0);font-size:12px">A</span><span style=3D"color:rg=
b(0,0,0);font-size:12px"> </span><span style=3D"font-size:12px;color:rgb(11=
2,112,112)">--&gt;</span><span style=3D"color:rgb(0,0,0);font-size:12px"> <=
/span><span style=3D"font-size:12px;color:rgb(112,112,112)">(</span><span s=
tyle=3D"color:rgb(0,0,0);font-size:12px">a</span><span style=3D"color:rgb(0=
,0,0);font-size:12px"> </span><span style=3D"font-size:12px;color:rgb(112,1=
12,112)">|</span><span style=3D"color:rgb(0,0,0);font-size:12px"> </span><s=
pan style=3D"color:rgb(0,0,0);font-size:12px">b</span><span style=3D"font-s=
ize:12px;color:rgb(112,112,112)">):</span><span style=3D"color:rgb(0,0,0);f=
ont-size:12px"> </span><span style=3D"color:rgb(0,0,0);font-size:12px">A</s=
pan></div></blockquote><div><br></div><div>However, when I write this simpl=
e bit of code (exposition purposes only):</div><blockquote style=3D"margin:=
0px 0px 0px 40px;border:none;padding:0px"><div>#include &lt;boost/spirit/ho=
me/x3.hpp&gt;</div><div>int main() {</div><div>=C2=A0 =C2=A0 using namespac=
e boost::spirit::x3;</div><div>=C2=A0 =C2=A0 auto const quoted_string =3D l=
exeme[&#39;&quot;&#39; &gt;&gt; *(char_ - &#39;&quot;&#39;) &gt;&gt; &#39;&=
quot;&#39;];</div><div>=C2=A0 =C2=A0 auto value =3D std::string{};</div><di=
v>=C2=A0 =C2=A0 auto assign_string =3D [&amp;](auto&amp; ctx){ value =3D _a=
ttr(ctx); };</div><div>=C2=A0 =C2=A0 auto const any_string =3D (quoted_stri=
ng | quoted_string) [assign_string];</div><div>=C2=A0 =C2=A0 auto val =3D s=
td::string_view{&quot;\&quot;Test\&quot;&quot;};</div><div>=C2=A0 =C2=A0 ph=
rase_parse(val.begin(), val.end(), any_string, space);</div><div>}</div></b=
lockquote><div><br></div><div>I get this error:</div><blockquote style=3D"m=
argin:0px 0px 0px 40px;border:none;padding:0px"><div><span style=3D"font-fa=
mily:monospace;font-weight:bold;color:rgb(255,84,84)">error: </span><span s=
tyle=3D"font-family:monospace;color:rgb(0,0,0)">no match for =E2=80=98</spa=
n><span style=3D"font-family:monospace;font-weight:bold;color:rgb(0,0,0)">o=
perator=3D</span><span style=3D"font-family:monospace;color:rgb(0,0,0)">=E2=
=80=99 (operand types are =E2=80=98</span><span style=3D"font-family:monosp=
ace;font-weight:bold;color:rgb(0,0,0)">std::__cxx11::b</span></div><div><sp=
an style=3D"font-family:monospace">asic_string&lt;char&gt;<span style=3D"co=
lor:rgb(0,0,0)">=E2=80=99 and =E2=80=98</span><span style=3D"font-weight:bo=
ld;color:rgb(0,0,0)">boost::variant&lt;boost::detail::variant::over_sequenc=
e&lt;boo</span></span></div><div><span style=3D"font-family:monospace">st::=
mpl::l_item&lt;mpl_::long_&lt;2&gt;, std::__cxx11::basic_string&lt;char&gt;=
, boost::mpl::l_</span></div><div><span style=3D"font-family:monospace">ite=
m&lt;mpl_::long_&lt;1&gt;, std::__cxx11::basic_string&lt;char&gt;, boost::m=
pl::l_end&gt; &gt; &gt; &gt;<span style=3D"color:rgb(0,0,0)">=E2=80=99</spa=
n></span></div><div><span style=3D"font-family:monospace">)
</span></div><div><span style=3D"font-family:monospace"> =C2=A0=C2=A0=C2=A0=
7 | =C2=A0=C2=A0=C2=A0=C2=A0auto assign_string =3D [&amp;](auto&amp; ctx){ =
<span style=3D"font-weight:bold;color:rgb(255,84,84)">value =3D _attr(ctx)<=
/span><span style=3D"color:rgb(0,0,0)">; };
</span></span></div><div><span style=3D"font-family:monospace"> =C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0|</span></div><div><br></div></blockquote><div>Why is =
the attribute a variant instead of a string?</div></div></blockquote><div><=
br></div><div>AFAIK, this is the way alternatives have always worked since =
even before X3.</div></div></div></blockquote><div><br></div><div>I&#39;ve =
written a lot of very complicated qi parsers and I don&#39;t remember ever =
running across this problem. It&#39;s possible it just never came up, but I=
 doubt it.</div><div><br></div><div>But it doesn&#39;t matter. The document=
ation I linked in my previous email says:</div></div><blockquote style=3D"m=
argin:0 0 0 40px;border:none;padding:0px"><div class=3D"gmail_quote"><div><=
pre class=3D"gmail-table-programlisting" style=3D"color:rgb(0,0,0);font-siz=
e:12px"><span class=3D"gmail-identifier">a: A</span><span class=3D"gmail-sp=
ecial" style=3D"color:rgb(112,112,112)">,</span> <span class=3D"gmail-ident=
ifier">b</span><span class=3D"gmail-special" style=3D"color:rgb(112,112,112=
)">:</span> <span class=3D"gmail-identifier">A</span> <span class=3D"gmail-=
special" style=3D"color:rgb(112,112,112)">--&gt;</span> <span class=3D"gmai=
l-special" style=3D"color:rgb(112,112,112)">(</span><span class=3D"gmail-id=
entifier">a</span> <span class=3D"gmail-special" style=3D"color:rgb(112,112=
,112)">|</span> <span class=3D"gmail-identifier">b</span><span class=3D"gma=
il-special" style=3D"color:rgb(112,112,112)">):</span> <span class=3D"gmail=
-identifier">A</span></pre></div></div></blockquote><div class=3D"gmail_quo=
te"><div>and not:</div></div><blockquote style=3D"margin:0 0 0 40px;border:=
none;padding:0px"><div class=3D"gmail_quote"><div><pre class=3D"gmail-table=
-programlisting" style=3D"color:rgb(0,0,0);font-size:12px"><span class=3D"g=
mail-identifier">a: A</span><span class=3D"gmail-special" style=3D"color:rg=
b(112,112,112)">,</span> <span class=3D"gmail-identifier">b</span><span cla=
ss=3D"gmail-special" style=3D"color:rgb(112,112,112)">:</span> A <span clas=
s=3D"gmail-special" style=3D"color:rgb(112,112,112)">--&gt;</span> <span cl=
ass=3D"gmail-special" style=3D"color:rgb(112,112,112)">(</span><span class=
=3D"gmail-identifier">a</span> <span class=3D"gmail-special" style=3D"color=
:rgb(112,112,112)">|</span> <span class=3D"gmail-identifier">b</span><span =
class=3D"gmail-special" style=3D"color:rgb(112,112,112)">):</span> <span cl=
ass=3D"gmail-identifier">variant</span><span class=3D"gmail-special" style=
=3D"color:rgb(112,112,112)">&lt;</span><span class=3D"gmail-identifier">A</=
span><span class=3D"gmail-special" style=3D"color:rgb(112,112,112)">,</span=
> A<span class=3D"gmail-special" style=3D"color:rgb(112,112,112)">&gt;</spa=
n></pre></div></div></blockquote><div class=3D"gmail_quote"><div>But the er=
ror message from the compiler clearly expresses that that&#39;s what I&#39;=
m getting. I&#39;ll dig into the x3 code when I have some time and try to f=
igure it out.</div><div><br></div><blockquote class=3D"gmail_quote" style=
=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding=
-left:1ex"><div dir=3D"ltr"><div class=3D"gmail_quote"><div>Which being sai=
d, I&#39;m not sure it makes a whole lot of sense, even in an academic sens=
e, to alternative, (qs | qs), does it?=C2=A0<br></div></div></div></blockqu=
ote><div><br></div><div>I knew someone would answer that way, which is why =
I had the disclaimer &quot;exposition purposes only&quot;. My real code loo=
ks a little different. But the attribute type is the same. The point in thi=
s example is to present code that&#39;s alternating between two clearly ide=
ntical attribute types.</div></div></div>

--000000000000671f2205aa1d79fb--


--===============7641108187147454226==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline


--===============7641108187147454226==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

_______________________________________________
Spirit-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/spirit-general

--===============7641108187147454226==--