more C like C++

[tedu]

Here is a more C like C++ implementation. :)

It's a fair bit faster because std::string is slow. Still using hash_map because it's convenient.

include <stdio.h>

include

using namespace std;

include <ext/hash_map>

using namespace __gnu_cxx;

int
readline(FILE _fp, char *_item, char *_feat, char *_val)
{
char *p;
static char buf[1024];

    if (!fgets(buf, sizeof(buf), fp))
            return 0;
    *item = buf;
    p = strchr(buf, ' ');
    *p = 0;

    *feat = p + 1;
    p = strchr(p + 1, ' ');
    *p = 0;

    *val = p + 1;
    p = strchr(p + 1, '\n');
    *p = 0;
    return 1;

}
struct eqstr
{
bool operator()(const char* s1, const char* s2) const
{
return strcmp(s1, s2) == 0;
}
};

int main(int argc, const char argv)
{
hash_map<const char, int, hash<const char>, eqstr> imap;
hash_map<const char*, int, hash<const char*>, eqstr> jmap;
hash_map<const char*, int, hash<const char*>, eqstr> I;
int J=0;
FILE *vocab = fopen("vocab", "w");

typeof(imap.end()) imapend = imap.end();
typeof(jmap.end()) jmapend = jmap.end();

for (int file_i=1; file_i<argc; file_i++) {
printf("%s\n",argv[file_i]);

const char *filename = argv[file_i];
FILE *in = fopen(filename, "r");
char outname[1024];
snprintf(outname, sizeof(outname), "%sn", filename);
FILE *out = fopen(outname, "w");

I[strdup(filename)] = 0;

char *item, *feat, *val;

while (readline(in, &item, &feat, &val)) {
  char key[1024];
  snprintf(key, sizeof(key), "%s %s", filename, item);
  if (imap.find(key) == imapend) {
    imap[strdup(key)] = ++I[filename];
  }
  if (jmap.find(feat) == jmapend) {
    jmap[strdup(feat)] = ++J;
    fprintf(vocab, "%s\n", feat);
  }
  fprintf(out, "%d %d %s\n", imap[key], jmap[feat], val);
}

}
return 0;
}

[brendano]

cool! if you can submit this as a pull request i'll merge. (it would be nice to confirm it gets the same answer as the other implementations...)

[brendano]

cool! if you can submit this as a pull request i'll merge. (it would be nice to confirm it gets the same answer as the other implementations...)