1. searching...
O.K., I am in the process of making an update to my ZipMaster program.
The thing that bugs me most about it is that it is so darn slow! I search
the file string by string comparing every word to each other. Does anyone
else know a faster way to look for certain words withen text files?
Oh yeah, and I added support fo finding files loaded by Babor's font.e and
by Micheal Bolin's awesome pic_load.e, but does anyone know any other
file-loading functions that should be added?
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
The Reaper (J. Lays) http://www.geocities.com/TimesSquare/Alley/4444/
reaper at auracom.com Check out my Euphoria Games page at:
-= http://www.geocities.com/TimesSquare/Alley/4444/eugames.html
........................
. .. -||..........__...... "There are those who spend their life
. / ||......../-- \\.:::: Dreaming of their desire;
. ..| ||...... / | |.::: You can either live to make it happen,
.| _-||.......|| / /.:::: Or burn in misery's fire."
..| |..||...... -\_- \ |\-.:::
.| |.[< \ .../ \.::
.||.|||\|\ | - - . \.::::
...|.\|| | \ | | |.:::.
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
2. Re: searching...
reaper at LOKI.ATCON.COM wrote:
>
> O.K., I am in the process of making an update to my ZipMaster program.
>
> The thing that bugs me most about it is that it is so darn slow! I search
> the file string by string comparing every word to each other. Does anyone
> else know a faster way to look for certain words withen text files?
>
Yes, there are several way's.
1. searching by the Knuth-Morris-Pratt method.
2. searching by the Boyer-Moore method, BM-search.
Use 2.
BM-search uses the method to start the comparing of the string at the
end and not from the begin of the string.
You can find these methods and modula-2 code in the book: Algorithms
and data structures, from Niklaus Wirth. And there are several
homepages from univerities that show these information.
R.S. Boyer en J.S.Moore. A fast string searching algorithm. Comm.
ACM, 20, 10 (okt 1977), pag 762-772.
Good luck,
Marcel Kollenaar
3. Re: searching...
At 08:17 97-05-09 -0400, you wrote:
>O.K., I am in the process of making an update to my ZipMaster program.
>
>The thing that bugs me most about it is that it is so darn slow! I search
>the file string by string comparing every word to each other. Does anyone
>else know a faster way to look for certain words withen text files?
Instead of checking each line character by character your program could
eliminate most of them with a simple test like:
sequence line -- contain line read from file
if not match("include",line) and not find('"',line) then
return -- skip this line
end if
Your zip master is only concern by lines that begin with the word
"include" and others that have double quotes. Resources files names have to
be between
double quotes.
And zip master doesn't need to know wich procedure or function load the
requested resource. Zip master only have to check if the double quotes enclose
a valid file name. Your program could have a function that check the contain
of any double quotes found in the file to see if it is a file name, a directory
name or else. Directories names are added to a sequence containing path_list
for future search of files. File names are searched on hard disk in any
directory member of path_list and current directory.
the preceding method is based on the following assumptions:
1) any line beginning with the word "include" is followed by file name and
nothing else.
2) Any double quote containing a path name is certainly assigned to a constant
or variable that point to directory where the program search files.
3) Any double quote containing a string that is a valid dos file name is
possibly a resource name. It can be confirmed by searching on disk for such
a file.
I have tested this method with success.
Jacques Deschenes
Baie-Comeau, Quebec
Canada
desja at quebectel.com
4. Re: searching...
At 10:21 AM 5/11/97 -0400, you wrote:
> Instead of checking each line character by character your program could
>eliminate most of them with a simple test like:
>
> sequence line -- contain line read from file
>
> if not match("include",line) and not find('"',line) then
> return -- skip this line
> end if
This is good stuff, do that's no surprise considering it's coming from you.
By the way, your SFX2.E is absolutly fantastic!!!
Anyways, the only problem here is that it will also check comments. Some
people put comments after their "include whatever.e" (like me) and also put
filenames that they are not using yet behind comments, though the file is
still there.
I solved most of ZipMaster's problems by getting rid of the code that writes
the line number to the screen. I put that there because the program was so
slow, but it turns out that was what was slowing the program down in the
first place. Ugh!
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
The Reaper (J. Lays) http://www.geocities.com/TimesSquare/Alley/4444/
reaper at auracom.com Check out my Euphoria Games page at:
-= http://www.geocities.com/TimesSquare/Alley/4444/eugames.html
........................
. .. -||..........__...... "There is a shadow that looms over life,
. / ||......../-- \\.:::: Of knowledge that will never be found;
. ..| ||...... / | |.::: You can guess or waste your life on it,
.| _-||.......|| / /.:::: Or look beyond into what's next."
..| |..||...... -\_- \ |\-.:::
.| |.[< \ .../ \.::
.||.|||\|\ | - - . \.::::
...|.\|| | \ | | |.:::.
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
