C++ for Fortran Programmers

Features

• Cover Type: Paperback with 540 pages
• Published by: Addison Wesley Longman
• Edition: 1st Edition July 1997
• Written in: English
• ISBN 10 Number: 0201924838
• ISBN 13 Number: 978-0201924831
• Book Dimensions: 9.2 x 7.3 x 1.1 inches
• Weighs: 2 pounds

Book Description
Using your existing knowledge of Fortran, C++ for Fortran Programmers gets you up and running with C++ quickly. By learning how individual elements of a Fortran program compare and translate into C++, this book helps you make a smooth transition to C++ and object-oriented concepts. Best-selling author and C++ authority, Ira Pohl, uses his trademark "dissection" technique to illustrate the underlying structure of programs and help you understand design trade-offs. Scientific and engineering coding examples are featured throughout the text.

The book provides a smooth transition to C++ and object-oriented programming for programmers already familiar with Fortran, and includes C++ to Fortran equivalencies making it easy to move from one language to another. Engineering computations are featured throughout; important scientific types such as complex number, vector and polynomial are implemented. It also incorporates the proposed ANSI C++ Standard including bool, namespaces, and the STL library, and features a chapter on the use of STL and efficient generic programming. All program code is fully tested and available on the World Wide Web.

Book Info
Provides a smooth transition to C++ and objectoriented programming for programmers familiar with Fortran. Includes C++ to Fortran equivalencies and features engineering computations. Paper. DLC: C++ (Computer program language)

Reader Reviews
First, let me say that this does seem to be a perfectly respectable C++ book. One could do worse, and it does not seem that knowing Fortran is a prerequisite. It's the "... For Fortran Programmers" part that is disappointing. I'm a member of the ostensible target demographic -- someone who has programmed for decades off and on in Fortran and is looking to learn the C (/ C++) language -- and this book sounded like exactly what I was looking for. Unfortunately, it did not live up to this expectation. The Fortran related material occupies only the first 20% of the book (plus a brief appendix), essentially the plain-old-C chapters. The "for Fortran programmers" content is mostly concerned with those areas where there are direct equivalences between the two languages that can be summarized neatly in tables (e.g. Fortran's "double precision" is the same as C's "double"). This kind of info is useful, but an experienced Fortran programmer will easily pick it up anyway. When it comes to more complicated issues such as control structures (IF, etc.), the emphasis seems to be to present Fortran equivalents of C structures, rather than the reverse. I guess this could be seen as using the presumed Fortran background to explicate a C language description, but I'd think most Fortran programmers learning C will be thinking the reverse, "I know how to do <X> in Fortran, how do I achieve the same thing in C?" Compounding this, the assumption seems to be that the Fortran programmed in is Fortran 90, which is far from universally adopted (especially with the immense amount of legacy code in use). Of course, a lot of the C++ language simply has no direct equivalent in Fortran. But if this were really written from the start for Fortran programmers there should be specific discussion showing how to do things that a Fortran programmer finds clumsy (such as matrix manipulation) more smoothly with the "++" part of C++, and the book should use examples drawn from scientific programming (which has been the primary domain of Fortran programmers). I am not saying this book should be a simple translation manual, but it should start with what the Fortran programmer already knows and build from there, and deal with problems that the Fortran programmer is typically interested in. In summary, this is a decent C++ book, but the Fortran content seems to be a clumsy afterthought, grafted on for marketing reasons. Comment | | (Report this)