From 4d0e637aaab0ae16e796878a6c5665662958c422 Mon Sep 17 00:00:00 2001 From: Johannes Schindelin Date: Mon, 26 Apr 2021 22:08:18 +0200 Subject: [PATCH] scalar: implement a minimal JSON parser No grown-up C project comes without their own JSON parser. Just kidding! We need to parse a JSON result when determining which cache server to use. It would appear that searching for needles `"CacheServers":[`, `"Url":"` and `"GlobalDefault":true` _happens_ to work right now, it is fragile as it depends on no whitespace padding and on the order of the fields remaining as-is. Let's implement a super simple JSON parser (at the cost of being slightly inefficient) for that purpose. To avoid allocating a ton of memory, we implement a callback-based one. And to save on complexity, let's not even bother validating the input properly (we will just go ahead and instead rely on Azure Repos to produce correct JSON). Note: An alternative would have been to use existing solutions such as JSON-C, CentiJSON or JSMN. However, they are all a lot larger than the current solution; The smallest, JSMN, which does not even provide parsed string values (something we actually need) weighs in with 471 lines, while we get away with 182 + 29 lines for the C and the header file, respectively. Signed-off-by: Johannes Schindelin --- Makefile | 3 +- contrib/buildsystems/CMakeLists.txt | 2 +- json-parser.c | 183 ++++++++++++++++++++++++++++ json-parser.h | 29 +++++ 4 files changed, 215 insertions(+), 2 deletions(-) create mode 100644 json-parser.c create mode 100644 json-parser.h diff --git a/Makefile b/Makefile index 2dbb049b69b606..8edabb8d566f1f 100644 --- a/Makefile +++ b/Makefile @@ -2726,6 +2726,7 @@ GIT_OBJS += git.o .PHONY: git-objs git-objs: $(GIT_OBJS) +SCALAR_OBJS := json-parser.o SCALAR_OBJS += scalar.o .PHONY: scalar-objs scalar-objs: $(SCALAR_OBJS) @@ -2874,7 +2875,7 @@ $(REMOTE_CURL_PRIMARY): remote-curl.o http.o http-walker.o $(LAZYLOAD_LIBCURL_OB $(QUIET_LINK)$(CC) $(ALL_CFLAGS) -o $@ $(ALL_LDFLAGS) $(filter %.o,$^) \ $(CURL_LIBCURL) $(EXPAT_LIBEXPAT) $(LIBS) -scalar$X: scalar.o GIT-LDFLAGS $(GITLIBS) +scalar$X: $(SCALAR_OBJS) GIT-LDFLAGS $(GITLIBS) $(QUIET_LINK)$(CC) $(ALL_CFLAGS) -o $@ $(ALL_LDFLAGS) \ $(filter %.o,$^) $(LIBS) diff --git a/contrib/buildsystems/CMakeLists.txt b/contrib/buildsystems/CMakeLists.txt index ec690a76a76304..d0d620ed740980 100644 --- a/contrib/buildsystems/CMakeLists.txt +++ b/contrib/buildsystems/CMakeLists.txt @@ -798,7 +798,7 @@ target_link_libraries(git-sh-i18n--envsubst common-main) add_executable(git-shell ${CMAKE_SOURCE_DIR}/shell.c) target_link_libraries(git-shell common-main) -add_executable(scalar ${CMAKE_SOURCE_DIR}/scalar.c) +add_executable(scalar ${CMAKE_SOURCE_DIR}/scalar.c ${CMAKE_SOURCE_DIR}/json-parser.c) target_link_libraries(scalar common-main) if(CURL_FOUND) diff --git a/json-parser.c b/json-parser.c new file mode 100644 index 00000000000000..539320b7b2f940 --- /dev/null +++ b/json-parser.c @@ -0,0 +1,183 @@ +#include "git-compat-util.h" +#include "hex.h" +#include "json-parser.h" + +static int reset_iterator(struct json_iterator *it) +{ + it->p = it->begin = it->json; + strbuf_release(&it->key); + strbuf_release(&it->string_value); + it->type = JSON_NULL; + return -1; +} + +static int parse_json_string(struct json_iterator *it, struct strbuf *out) +{ + const char *begin = it->p; + + if (*(it->p)++ != '"') + return error("expected double quote: '%.*s'", 5, begin), + reset_iterator(it); + + strbuf_reset(&it->string_value); +#define APPEND(c) strbuf_addch(out, c) + while (*it->p != '"') { + switch (*it->p) { + case '\0': + return error("incomplete string: '%s'", begin), + reset_iterator(it); + case '\\': + it->p++; + if (*it->p == '\\' || *it->p == '"') + APPEND(*it->p); + else if (*it->p == 'b') + APPEND(8); + else if (*it->p == 't') + APPEND(9); + else if (*it->p == 'n') + APPEND(10); + else if (*it->p == 'f') + APPEND(12); + else if (*it->p == 'r') + APPEND(13); + else if (*it->p == 'u') { + unsigned char binary[2]; + int i; + + if (hex_to_bytes(binary, it->p + 1, 2) < 0) + return error("invalid: '%.*s'", + 6, it->p - 1), + reset_iterator(it); + it->p += 4; + + i = (binary[0] << 8) | binary[1]; + if (i < 0x80) + APPEND(i); + else if (i < 0x0800) { + APPEND(0xc0 | ((i >> 6) & 0x1f)); + APPEND(0x80 | (i & 0x3f)); + } else if (i < 0x10000) { + APPEND(0xe0 | ((i >> 12) & 0x0f)); + APPEND(0x80 | ((i >> 6) & 0x3f)); + APPEND(0x80 | (i & 0x3f)); + } else { + APPEND(0xf0 | ((i >> 18) & 0x07)); + APPEND(0x80 | ((i >> 12) & 0x3f)); + APPEND(0x80 | ((i >> 6) & 0x3f)); + APPEND(0x80 | (i & 0x3f)); + } + } + break; + default: + APPEND(*it->p); + } + it->p++; + } + + it->end = it->p++; + return 0; +} + +static void skip_whitespace(struct json_iterator *it) +{ + while (isspace(*it->p)) + it->p++; +} + +int iterate_json(struct json_iterator *it) +{ + skip_whitespace(it); + it->begin = it->p; + + switch (*it->p) { + case '\0': + return reset_iterator(it), 0; + case 'n': + if (!starts_with(it->p, "null")) + return error("unexpected value: %.*s", 4, it->p), + reset_iterator(it); + it->type = JSON_NULL; + it->end = it->p = it->begin + 4; + break; + case 't': + if (!starts_with(it->p, "true")) + return error("unexpected value: %.*s", 4, it->p), + reset_iterator(it); + it->type = JSON_TRUE; + it->end = it->p = it->begin + 4; + break; + case 'f': + if (!starts_with(it->p, "false")) + return error("unexpected value: %.*s", 5, it->p), + reset_iterator(it); + it->type = JSON_FALSE; + it->end = it->p = it->begin + 5; + break; + case '-': case '.': + case '0': case '1': case '2': case '3': case '4': + case '5': case '6': case '7': case '8': case '9': + it->type = JSON_NUMBER; + it->end = it->p = it->begin + strspn(it->p, "-.0123456789"); + break; + case '"': + it->type = JSON_STRING; + if (parse_json_string(it, &it->string_value) < 0) + return -1; + break; + case '[': { + const char *save = it->begin; + size_t key_offset = it->key.len; + int i = 0, res; + + for (it->p++, skip_whitespace(it); *it->p != ']'; i++) { + strbuf_addf(&it->key, "[%d]", i); + + if ((res = iterate_json(it))) + return reset_iterator(it), res; + strbuf_setlen(&it->key, key_offset); + + skip_whitespace(it); + if (*it->p == ',') + it->p++; + } + + it->type = JSON_ARRAY; + it->begin = save; + it->end = it->p; + it->p++; + break; + } + case '{': { + const char *save = it->begin; + size_t key_offset = it->key.len; + int res; + + strbuf_addch(&it->key, '.'); + for (it->p++, skip_whitespace(it); *it->p != '}'; ) { + strbuf_setlen(&it->key, key_offset + 1); + if (parse_json_string(it, &it->key) < 0) + return -1; + skip_whitespace(it); + if (*(it->p)++ != ':') + return error("expected colon: %.*s", 5, it->p), + reset_iterator(it); + + if ((res = iterate_json(it))) + return res; + + skip_whitespace(it); + if (*it->p == ',') + it->p++; + } + strbuf_setlen(&it->key, key_offset); + + it->type = JSON_OBJECT; + it->begin = save; + it->end = it->p; + it->p++; + break; + } + } + + return it->fn(it); +} diff --git a/json-parser.h b/json-parser.h new file mode 100644 index 00000000000000..ce1fdc5ee23928 --- /dev/null +++ b/json-parser.h @@ -0,0 +1,29 @@ +#ifndef JSON_PARSER_H +#define JSON_PARSER_H + +#include "strbuf.h" + +struct json_iterator { + const char *json, *p, *begin, *end; + struct strbuf key, string_value; + enum { + JSON_NULL = 0, + JSON_FALSE, + JSON_TRUE, + JSON_NUMBER, + JSON_STRING, + JSON_ARRAY, + JSON_OBJECT + } type; + int (*fn)(struct json_iterator *it); + void *fn_data; +}; +#define JSON_ITERATOR_INIT(json_, fn_, fn_data_) { \ + .json = json_, .p = json_, \ + .key = STRBUF_INIT, .string_value = STRBUF_INIT, \ + .fn = fn_, .fn_data = fn_data_ \ +} + +int iterate_json(struct json_iterator *it); + +#endif