|
From: | David Bateman |
Subject: | Re: seekg calls in ls-mat5.cc |
Date: | Fri, 22 Apr 2005 09:40:55 +0200 |
User-agent: | Mozilla Thunderbird 0.8 (X11/20040923) |
John W. Eaton wrote:
The seekg's in ls-mat5.cc are all to seek to a particular byte position in a file, to either take into account the padding, to skip an unknown length header, or an element that could not be read. The padding might easily be handled since we know the length we read and can just read a block that we don't use. However, the other two cases are harder to see what might be done, and in fact I have no solution..There are a number of places in ls-mat5.cc where we have things like is.seekg (tmp_pos + static_cast<std::streamoff> (PAD (len))); This seems risky to me because I don't think there is any guarantee that a cast from an integer type to a std::streamoff type is guaranteed to work (std::streamoff might not be implemented with a simple integer data type). Is it really necessary to save and restore the file position when reading MAT files or is there another way that we could implement these functions to avoid all the calls to seekg? jwe
You'll note I removed all of the seekp's recently.. Regards David -- David Bateman address@hiddenMotorola Labs - Paris +33 1 69 35 48 04 (Ph) Parc Les Algorithmes, Commune de St Aubin +33 1 69 35 77 01 (Fax) 91193 Gif-Sur-Yvette FRANCE
The information contained in this communication has been classified as: [x] General Business Information [ ] Motorola Internal Use Only [ ] Motorola Confidential Proprietary
[Prev in Thread] | Current Thread | [Next in Thread] |