Files
tippecanoe/jsontool.cpp
T
Erica FischerandCursor 65beb3f0d7 Migrate jsonpull to unique_ptr ownership
Replaces the shared_ptr-based json_object_ptr with a unique_ptr that
has a stateless custom deleter dispatching on json_object::type before
calling the right subclass destructor. Eliminates per-node atomic
reference-counting and the control-block allocation that shared_ptr
required for every node in the tree.

API now distinguishes owning and borrowing pointers explicitly:
- json_read / json_read_separators / json_hash_get return raw
  json_object * (borrowed from the parser-owned tree).
- json_read_tree / json_disconnect return json_object_ptr (caller
  takes ownership; back-pointers are cleared so the subtree can
  outlive the parser).
- json_free / json_context / json_stringify take raw pointers.
- The parser's container_stack holds raw pointers; jp->root keeps
  unique_ptr ownership of the most recent top-level value.

Internally, take_from_owner moves the unique_ptr out of whichever
parent vector / hash entry / parser root owned it, which both
json_free and json_disconnect rely on.

In the streaming parsers (parse_feature, parse_layers, the
geojson-loop callback), we are careful to free `j` only after we
have processed a complete Feature: json_read returns each token
as the tree is being built up, and freeing an intermediate node
would splice it out of the surrounding hash and corrupt the
in-progress feature.

Benchmark (tl_2022_us_county.json, -z0 --extend-zooms-if-still-dropping,
median of 5 runs on macOS arm64): 8.5s, vs 10.6s with shared_ptr
and 8.7s on the pre-refactor C baseline.

Co-authored-by: Cursor <cursoragent@cursor.com>
2026-05-30 20:44:25 -07:00

472 lines
10 KiB
C++

#include <stdio.h>
#include <stdlib.h>
#include <ctype.h>
#include <string.h>
#include <stdarg.h>
#include <unistd.h>
#include <string>
#include <getopt.h>
#include <vector>
#include "jsonpull/jsonpull.h"
#include "csv.hpp"
#include "text.hpp"
#include "geojson-loop.hpp"
#include "milo/dtoa_milo.h"
#include "errors.hpp"
int fail = EXIT_SUCCESS;
bool wrap = false;
const char *extract = NULL;
FILE *csvfile = NULL;
std::vector<std::string> header;
std::vector<std::string> fields;
int pe = false;
std::string buffered;
int buffered_type = -1;
// 0: nothing yet
// 1: buffered a line
// 2: wrote the line and the wrapper
int buffer_state = 0;
std::vector<unsigned long> decode32(const char *s) {
std::vector<unsigned long> utf32;
while (*s != '\0') {
unsigned long b = *(s++) & 0xFF;
if (b < 0x80) {
utf32.push_back(b);
} else if ((b & 0xe0) == 0xc0) {
unsigned long c = (b & 0x1f) << 6;
unsigned long b1 = *(s++) & 0xFF;
if ((b1 & 0xc0) == 0x80) {
c |= b1 & 0x3f;
utf32.push_back(c);
} else {
s--;
utf32.push_back(0xfffd);
}
} else if ((b & 0xf0) == 0xe0) {
unsigned long c = (b & 0x0f) << 12;
unsigned long b1 = *(s++) & 0xFF;
if ((b1 & 0xc0) == 0x80) {
c |= (b1 & 0x3f) << 6;
unsigned long b2 = *(s++) & 0xFF;
if ((b2 & 0xc0) == 0x80) {
c |= b2 & 0x3f;
utf32.push_back(c);
} else {
s -= 2;
utf32.push_back(0xfffd);
}
} else {
s--;
utf32.push_back(0xfffd);
}
} else if ((b & 0xf8) == 0xf0) {
unsigned long c = (b & 0x07) << 18;
unsigned long b1 = *(s++) & 0xFF;
if ((b1 & 0xc0) == 0x80) {
c |= (b1 & 0x3f) << 12;
unsigned long b2 = *(s++) & 0xFF;
if ((b2 & 0xc0) == 0x80) {
c |= (b2 & 0x3f) << 6;
unsigned long b3 = *(s++) & 0xFF;
if ((b3 & 0xc0) == 0x80) {
c |= b3 & 0x3f;
utf32.push_back(c);
} else {
s -= 3;
utf32.push_back(0xfffd);
}
} else {
s -= 2;
utf32.push_back(0xfffd);
}
} else {
s -= 1;
utf32.push_back(0xfffd);
}
} else {
utf32.push_back(0xfffd);
}
}
return utf32;
}
// This uses a really weird encoding for strings
// so that they will sort in UTF-32 order in spite of quoting
std::string sort_quote(const char *s) {
std::vector<unsigned long> utf32 = decode32(s);
std::string ret;
for (size_t i = 0; i < utf32.size(); i++) {
if (utf32[i] < 0xD800) {
char buf[8];
snprintf(buf, sizeof(buf), "\\u%04lu", utf32[i]);
ret.append(std::string(buf));
} else {
unsigned long c = utf32[i];
if (c <= 0x7f) {
ret.push_back(c);
} else if (c <= 0x7ff) {
ret.push_back(0xc0 | (c >> 6));
ret.push_back(0x80 | (c & 0x3f));
} else if (c <= 0xffff) {
ret.push_back(0xe0 | (c >> 12));
ret.push_back(0x80 | ((c >> 6) & 0x3f));
ret.push_back(0x80 | (c & 0x3f));
} else {
ret.push_back(0xf0 | (c >> 18));
ret.push_back(0x80 | ((c >> 12) & 0x3f));
ret.push_back(0x80 | ((c >> 6) & 0x3f));
ret.push_back(0x80 | (c & 0x3f));
}
}
}
return ret;
}
void out(std::string const &s, int type, json_object *properties) {
if (extract != NULL) {
std::string extracted = sort_quote("null");
bool found = false;
json_object *o = json_hash_get(properties, extract);
if (o != nullptr) {
found = true;
if (o->type == JSON_STRING) {
extracted = sort_quote(o->string().c_str());
} else {
// Numbers, booleans, null, and any other non-string
// values are rendered via json_stringify(); calling
// o->string() here would assert because the type-tagged
// accessor requires JSON_STRING.
extracted = sort_quote(json_stringify(o).c_str());
}
}
if (!found) {
static bool warned = false;
if (!warned) {
fprintf(stderr, "Warning: extract key \"%s\" not found in JSON\n", extract);
warned = true;
}
}
printf("{\"%s\":%s}\n", extracted.c_str(), s.c_str());
return;
}
if (!wrap) {
printf("%s\n", s.c_str());
return;
}
if (buffer_state == 0) {
buffered = s;
buffered_type = type;
buffer_state = 1;
return;
}
if (buffer_state == 1) {
if (buffered_type == 1) {
printf("{\"type\":\"FeatureCollection\",\"features\":[\n");
} else {
printf("{\"type\":\"GeometryCollection\",\"geometries\":[\n");
}
printf("%s\n", buffered.c_str());
buffer_state = 2;
}
printf(",\n%s\n", s.c_str());
if (type != buffered_type) {
fprintf(stderr, "Error: mix of bare geometries and features\n");
exit(EXIT_IMPOSSIBLE);
}
}
std::string prev_joinkey;
void join_csv(json_object *j) {
if (header.size() == 0) {
std::string s = csv_getline(csvfile);
if (s.size() == 0) {
fprintf(stderr, "Couldn't get column header from CSV file\n");
exit(EXIT_CSV);
}
std::string err = check_utf8(s);
if (err != "") {
fprintf(stderr, "%s\n", err.c_str());
exit(EXIT_UTF8);
}
header = csv_split(s.c_str());
for (size_t i = 0; i < header.size(); i++) {
header[i] = csv_dequote(header[i]);
}
if (header.size() == 0) {
fprintf(stderr, "No columns in CSV header \"%s\"\n", s.c_str());
exit(EXIT_CSV);
}
}
json_object *properties = json_hash_get(j, "properties");
json_object *key = nullptr;
if (properties != nullptr) {
key = json_hash_get(properties, header[0].c_str());
}
if (key == nullptr) {
static bool warned = false;
if (!warned) {
fprintf(stderr, "Warning: couldn't find CSV key \"%s\" in JSON\n", header[0].c_str());
warned = true;
}
return;
}
std::string joinkey;
if (key->type == JSON_STRING) {
joinkey = key->string();
} else if (key->type == JSON_NUMBER) {
joinkey = milo::dtoa_milo(key->number());
} else {
joinkey = json_stringify(key);
}
if (joinkey < prev_joinkey) {
fprintf(stderr, "GeoJSON file is out of sort: \"%s\" follows \"%s\"\n", joinkey.c_str(), prev_joinkey.c_str());
exit(EXIT_IMPOSSIBLE);
}
prev_joinkey = joinkey;
if (fields.size() == 0 || joinkey > fields[0]) {
std::string prevkey;
if (fields.size() > 0) {
prevkey = fields[0];
}
while (true) {
std::string s = csv_getline(csvfile);
if (s.size() == 0) {
fields.clear();
break;
}
std::string err = check_utf8(s);
if (err != "") {
fprintf(stderr, "%s\n", err.c_str());
exit(EXIT_UTF8);
}
fields = csv_split(s.c_str());
for (size_t i = 0; i < fields.size(); i++) {
fields[i] = csv_dequote(fields[i]);
}
if (fields.size() > 0 && fields[0] < prevkey) {
fprintf(stderr, "CSV file is out of sort: \"%s\" follows \"%s\"\n", fields[0].c_str(), prevkey.c_str());
exit(EXIT_CSV);
}
if (fields.size() > 0 && fields[0] >= joinkey) {
break;
}
if (fields.size() > 0) {
prevkey = fields[0];
}
}
}
if (fields.size() > 0 && joinkey == fields[0]) {
properties->entries().reserve(properties->entries().size() + fields.size());
for (size_t i = 1; i < fields.size(); i++) {
std::string k = header[i];
std::string v = fields[i];
json_type attr_type = JSON_STRING;
if (v.size() > 0) {
if (v[0] == '"') {
v = csv_dequote(v);
} else if (is_number(v)) {
attr_type = JSON_NUMBER;
}
} else if (pe) {
attr_type = JSON_NULL;
}
if (attr_type != JSON_NULL) {
json_object_ptr ko(new json_string(properties, properties->parser));
ko->string() = k;
json_object_ptr vo;
if (attr_type == JSON_STRING) {
vo = json_object_ptr(new json_string(properties, properties->parser));
vo->string() = v;
} else if (attr_type == JSON_NUMBER) {
vo = json_object_ptr(new json_number(properties, properties->parser));
vo->set_number(atof(v.c_str()));
} else {
abort();
}
properties->entries().push_back({std::move(ko), std::move(vo)});
}
}
}
}
struct json_join_action : json_feature_action {
int add_feature(json_object *geometry, bool, json_object *, json_object *, json_object *, json_object *feature) {
if (feature != geometry) { // a real feature, not a bare geometry
if (csvfile != NULL) {
join_csv(feature);
}
out(json_stringify(feature), 1, json_hash_get(feature, "properties"));
} else {
out(json_stringify(geometry), 2, nullptr);
}
return 1;
}
void check_crs(json_object *) {
}
};
void process(FILE *fp, const char *fname) {
json_pull_ptr jp = json_begin_file(fp);
json_join_action jja;
jja.fname = fname;
parse_json(&jja, jp);
}
int main(int argc, char **argv) {
const char *csv = NULL;
struct option long_options[] = {
{"wrap", no_argument, 0, 'w'},
{"extract", required_argument, 0, 'e'},
{"csv", required_argument, 0, 'c'},
{"empty-csv-columns-are-null", no_argument, &pe, 1},
{"prevent", required_argument, 0, 'p'},
{0, 0, 0, 0},
};
std::string getopt_str;
for (size_t lo = 0; long_options[lo].name != NULL; lo++) {
if (long_options[lo].val > ' ') {
getopt_str.push_back(long_options[lo].val);
if (long_options[lo].has_arg == required_argument) {
getopt_str.push_back(':');
}
}
}
extern int optind;
int i;
while ((i = getopt_long(argc, argv, getopt_str.c_str(), long_options, NULL)) != -1) {
switch (i) {
case 0:
break;
case 'w':
wrap = true;
break;
case 'e':
extract = optarg;
break;
case 'c':
csv = optarg;
break;
case 'p':
if (strcmp(optarg, "e") == 0) {
pe = true;
} else {
fprintf(stderr, "%s: Unknown option for -p%s\n", argv[0], optarg);
exit(EXIT_ARGS);
}
break;
default:
fprintf(stderr, "Unexpected option -%c\n", i);
exit(EXIT_ARGS);
}
}
if (extract != NULL && wrap) {
fprintf(stderr, "%s: --wrap and --extract not supported together\n", argv[0]);
exit(EXIT_ARGS);
}
if (csv != NULL) {
csvfile = fopen(csv, "r");
if (csvfile == NULL) {
perror(csv);
exit(EXIT_OPEN);
}
}
if (optind >= argc) {
process(stdin, "standard input");
} else {
for (i = optind; i < argc; i++) {
FILE *f = fopen(argv[i], "r");
if (f == NULL) {
perror(argv[i]);
exit(EXIT_OPEN);
}
process(f, argv[i]);
fclose(f);
}
}
if (buffer_state == 1) {
printf("%s\n", buffered.c_str());
} else if (buffer_state == 2) {
printf("]}\n");
}
if (csvfile != NULL) {
if (fclose(csvfile) != 0) {
perror("close");
exit(EXIT_CLOSE);
}
}
return fail;
}