Unbreak non-trigram matches after we changed to asynchronous access().

[plocate] / plocate-build.cpp

diff --git a/plocate-build.cpp b/plocate-build.cpp
index 2f8bf318caedbaa2048769923659ba22637825fb..e1e2f644ef2ca891f93d48fdaab15e4f681bf418 100644 (file)
--- a/plocate-build.cpp
+++ b/plocate-build.cpp
@@ -1,25 +1,24 @@
 #include "db.h"
+#include "turbopfor-encode.h"
 
 #include <algorithm>
-#include <arpa/inet.h>
 #include <assert.h>
 #include <chrono>
-#include <endian.h>
-#include <fcntl.h>
+#include <getopt.h>
+#include <iosfwd>
 #include <math.h>
 #include <memory>
+#include <stdint.h>
 #include <stdio.h>
+#include <stdlib.h>
 #include <string.h>
 #include <string>
+#include <string_view>
 #include <sys/stat.h>
-#include <sys/types.h>
-#include <unistd.h>
-#include <unordered_map>
+#include <utility>
 #include <vector>
 #include <zstd.h>
 
-#include "turbopfor-encode.h"
-
 #define P4NENC_BOUND(n) ((n + 127) / 128 + (n + 32) * sizeof(uint32_t))
 #define dprintf(...)
 //#define dprintf(...) fprintf(stderr, __VA_ARGS__);
@@ -145,7 +144,17 @@ void PostingListBuilder::write_header(uint32_t docid)
 class Corpus {
 public:
        Corpus(FILE *outfp, size_t block_size)
-               : invindex(new PostingListBuilder *[NUM_TRIGRAMS]), outfp(outfp), block_size(block_size) {}
+               : invindex(new PostingListBuilder *[NUM_TRIGRAMS]), outfp(outfp), block_size(block_size)
+       {
+               fill(invindex.get(), invindex.get() + NUM_TRIGRAMS, nullptr);
+       }
+       ~Corpus()
+       {
+               for (unsigned i = 0; i < NUM_TRIGRAMS; ++i) {
+                       delete invindex[i];
+               }
+       }
+
        void add_file(string filename);
        void flush_block();
 
@@ -365,6 +374,7 @@ void do_build(const char *infile, const char *outfile, int block_size)
        hdr.version = -1;  // Mark as broken.
        hdr.hashtable_size = 0;  // Not known yet.
        hdr.extra_ht_slots = num_overflow_slots;
+       hdr.num_docids = 0;
        hdr.hash_table_offset_bytes = -1;  // We don't know these offsets yet.
        hdr.filename_index_offset_bytes = -1;
        fwrite(&hdr, sizeof(hdr), 1, outfp);
@@ -485,8 +495,66 @@ void do_build(const char *infile, const char *outfile, int block_size)
        dprintf("\n");
 }
 
+void usage()
+{
+       printf(
+               "Usage: plocate-build MLOCATE_DB PLOCATE_DB\n"
+               "\n"
+               "Generate plocate index from mlocate.db, typically /var/lib/mlocate/mlocate.db.\n"
+               "Normally, the destination should be /var/lib/mlocate/plocate.db.\n"
+               "\n"
+               "  -b, --block-size SIZE  number of filenames to store in each block (default 32)\n"
+               "      --help             print this help\n"
+               "      --version          print version information\n");
+}
+
+void version()
+{
+       printf("plocate-build %s\n", PLOCATE_VERSION);
+       printf("Copyright 2020 Steinar H. Gunderson\n");
+       printf("License GPLv2+: GNU GPL version 2 or later <https://gnu.org/licenses/gpl.html>.\n");
+       printf("This is free software: you are free to change and redistribute it.\n");
+       printf("There is NO WARRANTY, to the extent permitted by law.\n");
+}
+
 int main(int argc, char **argv)
 {
-       do_build(argv[1], argv[2], 32);
+       static const struct option long_options[] = {
+               { "block-size", required_argument, 0, 'b' },
+               { "help", no_argument, 0, 'h' },
+               { "version", no_argument, 0, 'V' },
+               { 0, 0, 0, 0 }
+       };
+
+       int block_size = 32;
+
+       setlocale(LC_ALL, "");
+       for (;;) {
+               int option_index = 0;
+               int c = getopt_long(argc, argv, "b:hV", long_options, &option_index);
+               if (c == -1) {
+                       break;
+               }
+               switch (c) {
+               case 'b':
+                       block_size = atoi(optarg);
+                       break;
+               case 'h':
+                       usage();
+                       exit(0);
+               case 'v':
+                       version();
+                       exit(0);
+               default:
+                       exit(1);
+               }
+       }
+
+       if (argc - optind != 2) {
+               usage();
+               exit(1);
+       }
+
+       do_build(argv[optind], argv[optind + 1], block_size);
        exit(EXIT_SUCCESS);
 }