From: David Abrahams Ferdinand Prantl wrote:
From: boost-bounces@lists.boost.org [mailto:boost-bounces@lists.boost.org] On Behalf Of David Abrahams
I just downloaded boost-1.32.0 and started to unzip it with the built-in facilities in Windows XP; the OS reported it was going to take 50 minutes and it was *crawling*. Using cygwin's unzip tool it was done in under a minute.
I'm thinking we should stop distributing raw .zip files without an extractor...
Hmm, I have never tried it. I always use Info-Zip's unzip or the
That's what I mean. I used the built-in unzipper in the Win XP file manager.
OK, I tried it with the WinXP extractor in the explorer (I used the copy&paste files method). It estimated at first 25 minutes, after a minute only 9 minutes and it finished after 20 minutes, wonderful... ;-) My machine is P4 2 GHz with 80 GB seagate / 2MB cache Norton antivirus and unbelievably messed up Windows installation ;-) Just to display the archive content (like a folder content) takes relatively long. I am not sure why but I would guess that it could be in the implementation of the shell namespace extension for archives in windows xp. It can be quite expensive to build the structure of shell item nodes for every file/folder in the archive, if the archive contains so many items. They should be created on demand, I faced the problem in my shell extension too. But without the source code I cannot guess. For example WinRAR shell extension "Extract Here" extracts the zip file quick - 2 minutes (it has no explore option, it is only a context menu shell extension) and I believe WinZIP or 7ZIP will have similar results. Btw., rd /q/s xxx is also faster that deleting using the explorer. I understand, that the zip distribution is primarily for the windows users and such slowness is painful. Nevertheless, a crappy implementation of some plugin in OS would annoy only the users who use it, the others can follow the recommendation from the boost download page, and download an alternative unzipping tool, for example from the info-zip site. There is no problem in the zipped file, it's just too big :-) An alternative would be a self-extracting archive, which can be produced by some free tools as well. Ferda