On Thu, May 24, 2007 at 11:55:35AM -0400, Richard Heck wrote: > Georg Baum wrote: > > Richard Heck wrote: > > > > > >> That's the approach I'm now taking, more or less, as Enrico found yet > >> further problems. He also suggested a simpler way, namely: note the time > >> when we start the conversion; then look at the modification times of the > >> files after the conversion to see which ones got generated. There could > >> be an issue if the converter runs really fast, > >> > > I don't think so. Modification times are stored with a very fine precision > > (at least on ext3, but I don't believe that it is special here). > > > How do I get at these highly precise times? The boost file_write_time() > function returns a time_t, which seems to be in seconds. (As you say, > this may not work anyway.)
I think that it is so because posix requires that file timestamps are given in seconds (http://www.opengroup.org/onlinepubs/009695399/basedefs/sys/stat.h.html) > So I'm thinking the right flag is something like "multifile", which > signals that the converter generates multiple files. The associated > behavior would then be copying the whole temporary directory over to a > subdirectory of the buffer's file directory. At least then we don't > litter /home/rgheck/files/ (or whatever) with garbage. (And then this > maybe could make 1.5.0.) > > Seem reasonable? I don't think so as you would copy back all the graphics files (maybe converted to postscript from some bitmap format and thus huge) and all sort of things that gets copied to the tempdir. Why not implementing what I really proposed, i.e., 1) make a list of all file names which are present in the tempdir just before calling htlatex 2) call htlatex 3) get a new list of files and copy back only those not present in the first list In this way the cruft copied back would be greatly reduced as .log, .dvi, and .aux file would not be copied. Using your patch about usetempdir and the wrapper I posted in the other thread, I was able to successfully get an html export. So, if you think that the above is too much work to be done in C++, another option would be applying that patch and converting to python that shell script, such that it can be used in all platforms as the html converter. -- Enrico
