Reprint coming ...

I've been contacted recently a few times by (potential) readers complaining that they can't get hold of Imperfect C++, and wondering whether it's out of print. As I've previously reported, a reprint has been prepared and should, as I speak, be being used to create new gems from dead trees. ;-)

I'll drop a note to my editor, and ask how the production process is going ...

More Imperfect C++? More imperfections, anyway

For the rest of 2009, and probably through 2010, I'm planning to start looking at some more imperfections not covered in the book. Some of these will reflect reader feedback in the five years - is it that long!? - since Imperfect C++ was published; some will reflect my own further dissatisfactions with the language and its various implementations.

I don't have any firm plans yet to do a More Imperfect C++, but given the imminent expansion of the language with C++1x (formerly known as C++0x), it couldn't hurt to start thinking about it. Readers, and other interested parties, are invited to contact me with any ideas/imperfections they may have.

New Recommendation in Reprint: use of parent_class_type member typedef

A new recommendation has been added, in section 18.5.1 (on page 302):

Recommendation: Use parent_class_type for maintenance robustness.

The section discusses the utility of member types, and in particular why it is useful to have a member typedef for the (primary) parent type.

New Recommendation in Reprint: use of macros to define 64-bit integer literals

A new recommendation has been added, in section 15.4.2 (on page 240):

Recommendation: Use macros to abstract away differences for 64-bit integer literals.

This is due to the use of non-standard literal suffixes - L / LL, UL / ULL - in different compilers

New Recommendation in Reprint: use of bools in literal expressions

A new recommendation has been added, in section 15.3 (on page 237):

Recommendation: Do not use literal true in conditional expressions. Avoid using literal false in conditonal expressions.

This is due to the discussed issues regarding the possibility of multiple non-false values in C/C++.

Reprint pending

I'm just working through the final checks of the corrected manuscript that will be used in the upcoming reprint of Imperfect C++. All being well, the reprint should occur this month (Nov. 2009). It will incorporate all the feedback that has been submitted on the website's errata page between the publication and mid-2009, as well as that received via email and other sources.

A great many thanks go out to all readers who submitted feedback. I've done my best to ensure that every one of you is named in the acknowledgements in the reprinted version.

Matt

Welcome to the "Imperfect C++" blog!

Just a first "hello" to all readers of Imperfect C++, and of this blog. Even though the book is now over five years old, it still continues to draw references and the occasional controversy, so I've created this blog for three reasons:

• To announce news regarding the book (such as reprints, reviews, and so on) and other books (including the books I'm currently working on)
  
• To highlight, or discuss further, imperfections highlighted within the book
• To discuss new imperfections, libraries, techniques and tools that I think may be of interest to the readers of Imperfect C++

Thanks for listening

Matt