David Abrahams wrote:

"Pavel Chikulaev" writes:

...

Hi David,

There is a strange symbols in /libs/parameter/doc/html/index.html in section 2.6.2 after second function code with modifier <span class="doublesize">. Absolutely the same in 2.7.2 and 2.7.3...

Looks fine to me:

http://boost-consulting.com/boost/libs/parameter/doc/html/index.html#forward...

What am I missing?

I think Pavel is referring to <span class="doublesize">⋮</span> that gets rendered as a fat asterisk under links2 here. HTH, m Send instant messages to your online friends http://au.messenger.yahoo.com

"David Abrahams" <dave@boost-consulting.com> wrote in message news:u4qa7vsxq.fsf@boost-consulting.com...

"Pavel Chikulaev" <pavel.chikulaev@gmail.com> writes:

...

Hi David,

There is a strange symbols in /libs/parameter/doc/html/index.html in section 2.6.2 after second function code with modifier <span class="doublesize">. Absolutely the same in 2.7.2 and 2.7.3...

Looks fine to me:

http://boost-consulting.com/boost/libs/parameter/doc/html/index.html#forward...

What am I missing?

There is something weird with in the code section right at the spot Pavel indicates. It is aligned over the "p" in "template" on the next line. On IE 6.0.2900 SP2, it shows up as a square about two characters wide. On Firefox 1.0.6 it looks like three dots or diamonds vertically aligned. It is hard to see without increasing the type size. While it isn't right, it is a nit rather than a serious worry. --Beman

Pavel Chikulaev wrote:

"Beman Dawes" wrote...

...

On IE 6.0.2900 SP2, it shows up as a square about two characters wide.

Same browser, same problem.

That's probably a browser bug, but...

Could that unicode character changed to ampersand-sequence? Moreover nowhere except these places unicode characters are used.

That doesn't help for all Browsers. Some of them simply don't support the vertical ellipsis, no matter how you spell it . They will instead display whatever they use for a missing glyph. I don't think we can't work around this portably without damaging the semantics of the text. I'm now in favour of ignoring this problem and letting the Browsers fix it. It's a rendering problem, after all. Regards, m Send instant messages to your online friends http://au.messenger.yahoo.com

"Pavel Chikulaev" <pavel.chikulaev@gmail.com> writes:

"David Abrahams" wrote...

...

Looks fine to me. What browser? IE 6.0.2900 SP2

...

I don't want a table; I want a sidebar. Table would solve that problem on all browsers... I mean table border=0 BTW, so no one will ever notice...

Show me how to get ReStructuredText to do that, and I'll consider it. -- Dave Abrahams Boost Consulting www.boost-consulting.com

"Pavel Chikulaev" <pavel.chikulaev@gmail.com> writes:

Hi David,

Please take a closer look at:

http://boost-consulting.com/boost/libs/parameter/doc/html/index.html#id15

[quote] template <class T> struct construct2 { typedef T result_type;

template <class A1, class A2> T operator() { return T(a1,a2); } };

and use Boost.Bind_ to generate the function object::

boost::bind(construct2<default_color_map>,num_vertices(g),i) [/quote]

That won't compile, and library called Boost.Bind (not Bind_) :).

Good catch.

That should be:

template <class T> struct construct2 { typedef T result_type;

template <class A1, class A2> T operator(A1 & a1, A2 & a2) { return T(a1,a2); } };

I'm afraid that won't compile either.

and use Boost.Bind to generate the function object::

boost::bind(construct2<default_color_map>(),num_vertices(g),i)

isn't it?

Something like that. Thanks, I'll make the fix. -- Dave Abrahams Boost Consulting www.boost-consulting.com

Andrey Melnikov <melnikov@simplexsoft.com> writes:

Pavel Chikulaev wrote:

...

Hi David,

There is a strange symbols in /libs/parameter/doc/html/index.html in section 2.6.2 after second function code with modifier <span class="doublesize">. Absolutely the same in 2.7.2 and 2.7.3...

In section 2.8.3 "Mnemonics" note can overlap code in that section when page less than 800 pixels wide. It's better to use tables IMO there.

I confirm that it works in Firefox/Windows and broken in IE6 exactly as Pavel describes.

I don't see any truly bad results in IE6. None of the code is obscured.

Also I found another bug there: "Mnemonics" is spelled incorrectly.

Thanks; I will fix that one.

And regarding tables: tables are considered an evil in modern XHTML/CSS2 world. Look at boost home page - you won't believe that it doesn't use tables.

I don't think that it's a good idea to fix a minor visual bug just because it appears in IE. I think we can afford having standard-compliant pages and ignore minor browser quirks. I think we should focus on getting all pages written in BoostBook and fixing the bugs in BoostBook XSLT styesheets then.

I see strange things in the code like <tt class="docutils literal"><span class="pre">. I think it's a fix for an old browser quirk because according to XHTML and CSS standards just <span> with correctly defined styles should be enough.

I'm going to contribute to this area after finishing with reviews of patches for BBv2 with Vladimir and Aleksey and HOWTO documentation improvements with Hendrik.

Fantastic; we need your contribution in all of these areas! Thank you! -- Dave Abrahams Boost Consulting www.boost-consulting.com