utils: plugins: Provide a choice for encoding raw UTF-8 strings#10665
utils: plugins: Provide a choice for encoding raw UTF-8 strings#10665
Conversation
cosmo0920
requested review from
PettitWesley,
braydonk,
edsiper,
fujimotos,
koleini,
leonardo-albertovich,
niedbalski,
nokute78,
patrick-stephens and
pwhelan
as code owners
WalkthroughAdds a new config option json.escape_unicode and plumbs it through core APIs and many plugins. Public headers and functions gain an escape_unicode parameter; calls are updated accordingly. Default is enabled. Tests and HTTP endpoints are adjusted. CI updates GCC/Clang versions for aarch64 workflow. Changes
Sequence Diagram(s)sequenceDiagram
autonumber
participant User as Config file
participant Core as flb_config
participant API as Pack/Utils APIs
participant Plugin as Output/Filter Plugins
User->>Core: Set "json.escape_unicode" (bool)
Core-->>Plugin: Pass struct flb_config*
Plugin->>API: flb_msgpack_*_to_json(..., escape_unicode=config->json_escape_unicode)
API-->>Plugin: JSON string (escaped or raw Unicode)
Plugin-->>User: Output/Send data
note over API: Default initialized to FLB_TRUE
Estimated code review effort🎯 4 (Complex) | ⏱️ ~60 minutes Suggested reviewers
Poem
Pre-merge checks and finishing touches❌ Failed checks (2 warnings)
✅ Passed checks (3 passed)
✨ Finishing touches
🧪 Generate unit tests
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 2
🧹 Nitpick comments (2)
plugins/out_bigquery/bigquery.c (2)
850-852: Avoid passing the wholestruct flb_config *when only one flag is needed
bigquery_format()now depends on the full global config, although it uses onlyjson_escape_unicode. This tightens coupling and complicates unit-testing. Consider passing the plainint escape_unicodeflag instead (mirroringflb_msgpack_raw_to_json_sds) or wrapping it into the existingctxstructure so that only plugin-local state flows through the call chain.
963-966:(void) config;is now misleadingThe cast was previously used to silence an “unused parameter” warning, but
configis now actively passed tobigquery_format(). Keeping the cast may confuse future readers and automated linters.- (void) config;
| out_buf = flb_msgpack_raw_to_json_sds(mp_sbuf.data, mp_sbuf.size, | ||
| config->json_escape_unicode); | ||
|
|
There was a problem hiding this comment.
Missing NULL-check before dereferencing config
config is assumed non-NULL when accessing config->json_escape_unicode. The engine passes a valid pointer in normal paths, but defensive code would make the function robust to accidental NULLs or future refactors.
- out_buf = flb_msgpack_raw_to_json_sds(mp_sbuf.data, mp_sbuf.size,
- config->json_escape_unicode);
+ if (!config) {
+ flb_plg_error(ctx->ins, "NULL config passed to bigquery_format()");
+ return -1;
+ }
+ out_buf = flb_msgpack_raw_to_json_sds(mp_sbuf.data, mp_sbuf.size,
+ config->json_escape_unicode);Committable suggestion skipped: line range outside the PR's diff.
🤖 Prompt for AI Agents
In plugins/out_bigquery/bigquery.c around lines 941 to 943, the code
dereferences the config pointer without checking if it is NULL, which can cause
crashes if config is NULL. Add a NULL check for config before accessing
config->json_escape_unicode, and handle the NULL case appropriately, such as
using a default value or returning an error, to make the function more robust.
There was a problem hiding this comment.
Actionable comments posted: 0
🧹 Nitpick comments (1)
tests/runtime/in_systemd.c (1)
35-35: LGTM! Function signature update correctly applied.The addition of the
FLB_TRUEparameter maintains the existing Unicode escaping behavior in this test, which is appropriate since this test focuses on duplicated key handling rather than Unicode encoding behavior.Consider adding a separate test case to verify both Unicode escaping behaviors (
FLB_TRUEandFLB_FALSE) if the systemd input plugin is expected to handle Unicode characters, ensuring comprehensive coverage of the new functionality introduced by this PR.
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (1)
tests/runtime/in_systemd.c(1 hunks)
🧰 Additional context used
🧬 Code Graph Analysis (1)
tests/runtime/in_systemd.c (1)
src/flb_pack.c (1)
flb_msgpack_raw_to_json_sds(814-873)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (28)
- GitHub Check: pr-windows-build / call-build-windows-package (Windows 64bit (Arm64), amd64_arm64, -DCMAKE_SYSTEM_NAME=Windows -DCMA...
- GitHub Check: pr-windows-build / call-build-windows-package (Windows 32bit, x86, x86-windows-static, 3.31.6)
- GitHub Check: pr-windows-build / call-build-windows-package (Windows 64bit, x64, x64-windows-static, 3.31.6)
- GitHub Check: PR - fuzzing test
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SIMD=Off, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SANITIZE_THREAD=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SANITIZE_THREAD=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SIMD=Off, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SIMD=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SANITIZE_MEMORY=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_COVERAGE=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DSANITIZE_UNDEFINED=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SANITIZE_MEMORY=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SIMD=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_JEMALLOC=Off, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SMALL=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SMALL=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_JEMALLOC=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_JEMALLOC=Off, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DSANITIZE_UNDEFINED=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_JEMALLOC=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DSANITIZE_ADDRESS=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DSANITIZE_ADDRESS=On, 3.31.6, gcc, g++)
- GitHub Check: pr-compile-system-libs (-DFLB_PREFER_SYSTEM_LIBS=On, 3.31.6, gcc, g++, ubuntu-24.04, clang-14)
- GitHub Check: pr-compile-system-libs (-DFLB_PREFER_SYSTEM_LIBS=On, 3.31.6, gcc, g++, ubuntu-22.04, clang-12)
- GitHub Check: pr-compile-system-libs (-DFLB_PREFER_SYSTEM_LIBS=On, 3.31.6, clang, clang++, ubuntu-22.04, clang-12)
- GitHub Check: pr-compile-system-libs (-DFLB_PREFER_SYSTEM_LIBS=On, 3.31.6, clang, clang++, ubuntu-24.04, clang-14)
- GitHub Check: pr-compile-centos-7
Signed-off-by: Hiroshi Hatake <hiroshi@chronosphere.io>
Signed-off-by: Hiroshi Hatake <hiroshi@chronosphere.io>
Signed-off-by: Hiroshi Hatake <hiroshi@chronosphere.io>
Signed-off-by: Hiroshi Hatake <hiroshi@chronosphere.io>
Signed-off-by: Hiroshi Hatake <hiroshi@chronosphere.io>
Signed-off-by: Hiroshi Hatake <hiroshi@chronosphere.io>
Signed-off-by: Hiroshi Hatake <hiroshi@chronosphere.io>
Signed-off-by: Hiroshi Hatake <hiroshi@chronosphere.io>
Signed-off-by: Hiroshi Hatake <hiroshi@chronosphere.io>
Signed-off-by: Hiroshi Hatake <hiroshi@chronosphere.io>
Signed-off-by: Hiroshi Hatake <hiroshi@chronosphere.io>
Signed-off-by: Hiroshi Hatake <hiroshi@chronosphere.io>
Signed-off-by: Hiroshi Hatake <hiroshi@chronosphere.io>
Signed-off-by: Hiroshi Hatake <hiroshi@chronosphere.io>
Signed-off-by: Hiroshi Hatake <hiroshi@chronosphere.io>
Signed-off-by: Hiroshi Hatake <hiroshi@chronosphere.io>
cosmo0920
force-pushed
the
cosmo0920-provide-raw-utf8-choice
branch
from
3cf2ad4 to
4510a2a
Compare
There was a problem hiding this comment.
Actionable comments posted: 2
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
src/flb_utils.c (1)
791-1097: Fix two bugs in escaped path: surrogate pairs and index advance.
- Codepoints > 0xFFFF are emitted as a single “\uXXXX” (wrong) and there’s an unreachable branch checking c > 0xFFFF (c is a byte). Emit surrogate pairs in the accepted decode path.
- After processing a multibyte sequence, the loop does i += processed_bytes followed by i++ at the end, skipping one extra byte.
Apply this diff:
- else if (c >= 0x80 && c <= 0xFFFF) { + else if ((c & 0x80) != 0) { hex_bytes = flb_utf8_len(&str[i]); @@ - if (state == FLB_UTF8_ACCEPT) { - len = snprintf(tmp, sizeof(tmp), "\\u%.4x", codepoint); + if (state == FLB_UTF8_ACCEPT) { + if (codepoint <= 0xFFFF) { + len = snprintf(tmp, sizeof(tmp), "\\u%.4x", codepoint); + } + else { + high = 0xD800 + ((codepoint - 0x10000) >> 10); + low = 0xDC00 + ((codepoint - 0x10000) & 0x3FF); + len = snprintf(tmp, sizeof(tmp), "\\u%.4x\\u%.4x", high, low); + } if (available < len) { return FLB_FALSE; } memcpy(p, tmp, len); p += len; offset += len; available -= len; - } - else { - flb_warn("[pack] Invalid UTF-8 bytes found, skipping."); - } - - i += processed_bytes; + } + else { + /* replacement char for invalid sequence */ + if (available < 3) { + return FLB_FALSE; + } + memcpy(p, "\xEF\xBF\xBD", 3); + p += 3; + offset += 3; + available -= 3; + } + + /* we consumed hex_bytes; adjust for the loop's trailing i++ */ + i += processed_bytes - 1; } - } - /* Handle sequences beyond 0xFFFF */ - else if (c > 0xFFFF) { - utf_sequence_length = flb_utf8_len(str + i); - /* ... unreachable block removed ... */ - } + }
🧹 Nitpick comments (20)
plugins/out_nrlogs/newrelic.c (2)
145-147: Make config parameter const to document immutability.This helper only reads from config; marking it const tightens the contract and prevents accidental writes. Static scope keeps impact local.
-static flb_sds_t newrelic_compose_payload(struct flb_newrelic *ctx, - const void *data, size_t bytes, - struct flb_config *config) +static flb_sds_t newrelic_compose_payload(struct flb_newrelic *ctx, + const void *data, size_t bytes, + const struct flb_config *config)
236-237: Defensive: avoid potential NULL dereference on config.Core should pass a valid config, but a defensive default prevents crashes in tests/mocks.
- json = flb_msgpack_raw_to_json_sds(mp_sbuf.data, mp_sbuf.size, - config->json_escape_unicode); + int escape_unicode = FLB_TRUE; + if (config != NULL) { + escape_unicode = config->json_escape_unicode; + } + json = flb_msgpack_raw_to_json_sds(mp_sbuf.data, mp_sbuf.size, + escape_unicode);plugins/out_kinesis_firehose/firehose_api.h (1)
40-41: API addition (config param) is correct — declarations/callers updated; optional: make config const
- Verified: signature updated in plugins/out_kinesis_firehose/firehose_api.h and definition in firehose_api.c; caller updated in plugins/out_kinesis_firehose/firehose.c (no remaining 4‑arg calls found).
- Optional refactor: if the config pointer is only read, change the parameter to const across affected signatures: process_and_send_records (firehose_api.h/.c), process_event (firehose_api.c), add_event (firehose_api.c) and any internal helpers that accept config.
plugins/out_loki/loki.c (3)
480-482: Guard config and skip non-string map keys in pack_maps
- Dereferencing config without a null-check can crash if a caller ever passes NULL.
- pack_label_key assumes string keys; non-string keys should be skipped to avoid UB.
Apply:
static void pack_maps(struct flb_loki *ctx, msgpack_packer *mp_pck, char *tag, int tag_len, msgpack_object *map, struct flb_mp_map_header *mh, - struct mk_list *list, - struct flb_config *config) + struct mk_list *list, + struct flb_config *config) { + /* Defensive: config is expected non-NULL, but tolerate just in case */ + int escape = config ? config->json_escape_unicode : FLB_CONF_UNICODE_STR_JSON_ESCAPE; @@ - pack_label_key(mp_pck, (char*) accessed_map_kv.key.via.str.ptr, + /* Ensure key is string */ + if (accessed_map_kv.key.type != MSGPACK_OBJECT_STR) { + continue; + } + pack_label_key(mp_pck, (char*) accessed_map_kv.key.via.str.ptr, accessed_map_kv.key.via.str.size); @@ - accessed_map_val_json = flb_msgpack_to_json_str(1024, - &accessed_map_kv.val, - config->json_escape_unicode); + accessed_map_val_json = flb_msgpack_to_json_str(1024, + &accessed_map_kv.val, + escape);Also applies to: 542-545
1422-1424: Null-safe use of config in pack_recordAvoid unconditional deref of config; use a local escape flag.
static int pack_record(struct flb_loki *ctx, msgpack_packer *mp_pck, msgpack_object *rec, flb_sds_t *dynamic_tenant_id, struct flb_mp_accessor *remove_mpa, struct flb_config *config) { + int escape = config ? config->json_escape_unicode : FLB_CONF_UNICODE_STR_JSON_ESCAPE; @@ - line = flb_msgpack_to_json_str(size_hint, rec, config->json_escape_unicode); + line = flb_msgpack_to_json_str(size_hint, rec, escape);Also applies to: 1514-1514
2252-2254: Avoid leaking dynamic_tenant_id in format testOn success, dynamic_tenant_id isn’t freed.
*out_data = payload; *out_size = flb_sds_len(payload); + if (dynamic_tenant_id != NULL) { + flb_sds_destroy(dynamic_tenant_id); + } return 0;tests/internal/mp.c (2)
221-223: Assert JSON conversion success in tests.Capture and assert the return value from flb_msgpack_to_json to avoid silent failures.
- flb_msgpack_to_json(&final_json[0], sizeof(final_json), &result_final.data, FLB_TRUE); + ret = flb_msgpack_to_json(&final_json[0], sizeof(final_json), &result_final.data, FLB_TRUE); + TEST_CHECK(ret > 0);- flb_msgpack_to_json(&final_json[0], sizeof(final_json), &result_final.data, FLB_TRUE); + ret = flb_msgpack_to_json(&final_json[0], sizeof(final_json), &result_final.data, FLB_TRUE); + TEST_CHECK(ret > 0);- flb_msgpack_to_json(&final_json[0], sizeof(final_json), &result_final.data, FLB_TRUE); + ret = flb_msgpack_to_json(&final_json[0], sizeof(final_json), &result_final.data, FLB_TRUE); + TEST_CHECK(ret > 0);Also applies to: 313-314, 415-416
522-524: Consider adding a test variant with escape disabled.Add a subtest using FLB_FALSE to validate raw UTF-8 emission path alongside the escaped path.
tests/internal/opentelemetry.c (1)
191-192: Broaden coverage: test with unicode escaping off.These expectations exercise FLB_TRUE only. Add a parameterized run (FLB_FALSE) for inputs containing non-ASCII to ensure parity and surface differences intentionally introduced by the option.
Also applies to: 212-214, 311-314, 572-572, 606-606, 611-611, 616-616, 621-621
plugins/out_influxdb/influxdb.c (1)
245-248: Minor hygiene: null out freed temp.Set str to NULL after flb_free to avoid accidental reuse in future edits.
- flb_free(str); - str_size = 0; + flb_free(str); + str = NULL; + str_size = 0;src/flb_help.c (1)
828-829: Use config->json_escape_unicode instead of hardcoded FLB_TRUE.Make help schema JSON respect the global setting for consistency with other endpoints.
- json = flb_msgpack_raw_to_json_sds(mp_sbuf.data, mp_sbuf.size, FLB_TRUE); + json = flb_msgpack_raw_to_json_sds(mp_sbuf.data, mp_sbuf.size, config->json_escape_unicode);src/http_server/api/v2/reload.c (1)
116-118: Honor global json_escape_unicode in HTTP responses.Avoid hardcoded FLB_TRUE to keep behavior consistent with the configured setting.
- out_buf = flb_msgpack_raw_to_json_sds(mp_sbuf.data, mp_sbuf.size, FLB_TRUE); + out_buf = flb_msgpack_raw_to_json_sds(mp_sbuf.data, mp_sbuf.size, config->json_escape_unicode);- out_buf = flb_msgpack_raw_to_json_sds(mp_sbuf.data, mp_sbuf.size, FLB_TRUE); + out_buf = flb_msgpack_raw_to_json_sds(mp_sbuf.data, mp_sbuf.size, config->json_escape_unicode);Also applies to: 151-152
src/http_server/api/v1/storage.c (1)
107-110: Consider propagating json_escape_unicode to storage metrics JSON.This path always escapes. If feasible, thread config->json_escape_unicode into this callback (e.g., via hs context or by delaying JSON conversion to response time) for consistency.
src/http_server/api/v1/plugins.c (1)
93-93: Use config->json_escape_unicode for consistency with global setting.This endpoint currently forces escaping. Since you already have config, consider honoring the global flag.
Apply this diff:
- out_buf = flb_msgpack_raw_to_json_sds(mp_sbuf.data, mp_sbuf.size, FLB_TRUE); + out_buf = flb_msgpack_raw_to_json_sds(mp_sbuf.data, mp_sbuf.size, config->json_escape_unicode);src/http_server/api/v1/uptime.c (1)
91-91: Consider honoring json_escape_unicode instead of hard-coding.Uptime payload is ASCII today, but for consistency with other endpoints, consider using the global flag.
Apply this diff:
- out_buf = flb_msgpack_raw_to_json_sds(mp_sbuf.data, mp_sbuf.size, FLB_TRUE); + out_buf = flb_msgpack_raw_to_json_sds(mp_sbuf.data, mp_sbuf.size, config->json_escape_unicode);tests/internal/utils.c (4)
217-231: Pass-through of escape_unicode is correct; add minor robustness checks.The new parameter is plumbed correctly into flb_utils_write_str. To harden the test helper:
- Guard against allocation failure.
- Also assert off equals strlen(buf) to catch accidental NULs or length mismatches.
Apply:
static void write_str_test_cases_w_buf_size(struct write_str_case *cases, int buf_size, int escape_unicode) { - char *buf = flb_calloc(buf_size + 1, sizeof(char)); + char *buf = flb_calloc(buf_size + 1, sizeof(char)); int size = buf_size + 1; int off; int ret; @@ - ret = flb_utils_write_str(buf, &off, buf_size, tcase->input, tcase->input_len, - escape_unicode); + ret = flb_utils_write_str(buf, &off, buf_size, tcase->input, tcase->input_len, + escape_unicode); + + if (buf == NULL) { + TEST_CHECK(buf != NULL); + return; + } @@ if (!TEST_CHECK(strlen(buf) == strlen(tcase->output))) { @@ } + if (!TEST_CHECK((size_t) off == strlen(buf))) { + TEST_MSG("off (%d) != strlen(buf) (%zu)", off, strlen(buf)); + }
398-436: Raw UTF‑8 happy-path coverage is solid; add a couple of JSON-escape cases.Since raw mode must still escape JSON control/meta chars, consider adding newline and backslash cases.
Apply:
/* Test string with a quote */ { "\"hello\"", 7, "\\\"hello\\\"", FLB_TRUE }, + /* Newline must be escaped even in raw mode */ + { + "line 1\nline 2", 13, + "line 1\\nline 2", + FLB_TRUE + }, + /* Backslash must be escaped */ + { + "a\\b", 3, + "a\\\\b", + FLB_TRUE + },
438-503: Minor: duplicated test case; optional extra NUL-char case.
- Case 2 and Case 6 are identical inputs/expectations; drop one to avoid redundant coverage.
- Optional: add a raw-mode case containing U+0000 to assert it’s escaped as \u0000 per JSON.
Apply:
@@ - /* - * Case 6: Sequence with invalid continuation byte - */ - { - /* Input: "a" + 日(E6 97 A5) + ASCII "b" */ - "a\xE6\x97""b", 4, - "a""\xEF\xBF\xBD""\xEF\xBF\xBD""b", - FLB_TRUE - }, + /* (removed duplicate of Case 2) */ @@ /* End of cases */ { 0 } };And (optional) append before the terminator:
+ /* + * Case X: NUL must be escaped in JSON + */ + { + "a\0b", 3, + "a\\u0000b", + FLB_TRUE + },Please confirm that flb_utils_write_str_raw collapses incomplete sequences into two U+FFFD bytes (your expectations) and not a single replacement for the whole invalid span; adjust expected outputs if the implementation differs.
927-929: Test registration OK; consider aligning names.Display name "write_raw_str_invalid_bytes" maps to test_write_raw_str_invalid_sequences. For consistency, either rename the test function or the display name.
plugins/processor_content_modifier/cm_utils.c (1)
153-155: Honor config->json_escape_unicode: add opt-in helper and keep BC wrapper.Hard-coding FLB_TRUE here diverges from config->json_escape_unicode and can produce double-escaped sequences when JSON is re-serialized. Repository search shows no other callers of cm_utils_variant_convert_to_json, so adding an opt-in helper is safe.
Apply this refactor (keep BC wrapper):
+cfl_sds_t cm_utils_variant_convert_to_json_with_opts(struct cfl_variant *value, int escape_unicode) +{ + cfl_sds_t json_result; + mpack_writer_t writer; + char *data = NULL; + size_t size = 0; + + mpack_writer_init_growable(&writer, &data, &size); + pack_cfl_variant(&writer, value); + mpack_writer_destroy(&writer); + + json_result = flb_msgpack_raw_to_json_sds(data, size, escape_unicode); + MPACK_FREE(data); + return json_result; +}cfl_sds_t cm_utils_variant_convert_to_json(struct cfl_variant *value) { - cfl_sds_t json_result; - mpack_writer_t writer; - char *data; - size_t size; - - data = NULL; - size = 0; - - mpack_writer_init_growable(&writer, &data, &size); - pack_cfl_variant(&writer, value); - mpack_writer_destroy(&writer); - - /* Using JSON escape here to keep backward compatibility */ - json_result = flb_msgpack_raw_to_json_sds(data, size, FLB_TRUE); - MPACK_FREE(data); - return json_result; + /* Keep BC: default to escaping unless callers opt in to config-driven behavior */ + return cm_utils_variant_convert_to_json_with_opts(value, FLB_TRUE); }Example call-site when config is available:
out = cm_utils_variant_convert_to_json_with_opts(val, config->json_escape_unicode);
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (77)
.github/workflows/unit-tests.yaml(3 hunks)include/fluent-bit/flb_config.h(2 hunks)include/fluent-bit/flb_pack.h(1 hunks)include/fluent-bit/flb_utils.h(1 hunks)plugins/filter_expect/expect.c(10 hunks)plugins/filter_nightfall/nightfall_api.c(1 hunks)plugins/filter_wasm/filter_wasm.c(1 hunks)plugins/out_azure/azure.c(3 hunks)plugins/out_azure_blob/azure_blob.c(1 hunks)plugins/out_azure_kusto/azure_kusto.c(4 hunks)plugins/out_azure_logs_ingestion/azure_logs_ingestion.c(3 hunks)plugins/out_bigquery/bigquery.c(3 hunks)plugins/out_calyptia/calyptia.c(3 hunks)plugins/out_chronicle/chronicle.c(7 hunks)plugins/out_cloudwatch_logs/cloudwatch_api.c(11 hunks)plugins/out_cloudwatch_logs/cloudwatch_api.h(1 hunks)plugins/out_cloudwatch_logs/cloudwatch_logs.c(1 hunks)plugins/out_datadog/datadog.c(1 hunks)plugins/out_es/es.c(1 hunks)plugins/out_file/file.c(3 hunks)plugins/out_http/http.c(4 hunks)plugins/out_influxdb/influxdb.c(1 hunks)plugins/out_kafka/kafka.c(1 hunks)plugins/out_kafka_rest/kafka.c(3 hunks)plugins/out_kinesis_firehose/firehose.c(1 hunks)plugins/out_kinesis_firehose/firehose_api.c(7 hunks)plugins/out_kinesis_firehose/firehose_api.h(1 hunks)plugins/out_kinesis_streams/kinesis.c(1 hunks)plugins/out_kinesis_streams/kinesis_api.c(7 hunks)plugins/out_kinesis_streams/kinesis_api.h(1 hunks)plugins/out_lib/out_lib.c(2 hunks)plugins/out_logdna/logdna.c(4 hunks)plugins/out_loki/loki.c(11 hunks)plugins/out_nats/nats.c(3 hunks)plugins/out_nrlogs/newrelic.c(3 hunks)plugins/out_null/null.c(1 hunks)plugins/out_opensearch/opensearch.c(2 hunks)plugins/out_oracle_log_analytics/oci_logan.c(1 hunks)plugins/out_pgsql/pgsql.c(1 hunks)plugins/out_s3/s3.c(3 hunks)plugins/out_skywalking/skywalking.c(5 hunks)plugins/out_slack/slack.c(1 hunks)plugins/out_splunk/splunk.c(5 hunks)plugins/out_stackdriver/stackdriver.c(4 hunks)plugins/out_stdout/stdout.c(1 hunks)plugins/out_tcp/tcp.c(4 hunks)plugins/out_udp/udp.c(3 hunks)plugins/out_vivo_exporter/vivo.c(5 hunks)plugins/out_websocket/websocket.c(1 hunks)plugins/processor_content_modifier/cm_utils.c(1 hunks)src/flb_config.c(3 hunks)src/flb_help.c(1 hunks)src/flb_pack.c(13 hunks)src/flb_record_accessor.c(1 hunks)src/flb_sds.c(1 hunks)src/flb_utils.c(3 hunks)src/fluent-bit.c(1 hunks)src/http_server/api/v1/metrics.c(1 hunks)src/http_server/api/v1/plugins.c(1 hunks)src/http_server/api/v1/storage.c(1 hunks)src/http_server/api/v1/trace.c(2 hunks)src/http_server/api/v1/uptime.c(1 hunks)src/http_server/api/v2/reload.c(2 hunks)src/http_server/flb_hs_endpoints.c(1 hunks)tests/internal/fuzzers/flb_json_fuzzer.c(3 hunks)tests/internal/fuzzers/msgpack_parse_fuzzer.c(1 hunks)tests/internal/fuzzers/utils_fuzzer.c(1 hunks)tests/internal/log_event_decoder.c(2 hunks)tests/internal/mp.c(4 hunks)tests/internal/msgpack_append_message.c(1 hunks)tests/internal/opentelemetry.c(8 hunks)tests/internal/pack.c(6 hunks)tests/internal/stream_processor.c(1 hunks)tests/internal/utils.c(6 hunks)tests/runtime/filter_lua.c(5 hunks)tests/runtime/in_opentelemetry.c(3 hunks)tests/runtime/in_systemd.c(1 hunks)
🚧 Files skipped from review as they are similar to previous changes (57)
- src/http_server/api/v1/trace.c
- tests/internal/fuzzers/flb_json_fuzzer.c
- plugins/out_kafka/kafka.c
- src/flb_sds.c
- plugins/out_slack/slack.c
- plugins/out_null/null.c
- plugins/filter_wasm/filter_wasm.c
- plugins/out_cloudwatch_logs/cloudwatch_logs.c
- tests/internal/stream_processor.c
- plugins/out_cloudwatch_logs/cloudwatch_api.h
- plugins/out_opensearch/opensearch.c
- plugins/out_azure/azure.c
- include/fluent-bit/flb_config.h
- .github/workflows/unit-tests.yaml
- plugins/out_file/file.c
- plugins/out_pgsql/pgsql.c
- tests/internal/fuzzers/utils_fuzzer.c
- plugins/out_kafka_rest/kafka.c
- plugins/out_kinesis_streams/kinesis.c
- plugins/out_skywalking/skywalking.c
- include/fluent-bit/flb_pack.h
- plugins/out_stackdriver/stackdriver.c
- plugins/out_udp/udp.c
- plugins/out_kinesis_streams/kinesis_api.h
- plugins/out_calyptia/calyptia.c
- plugins/out_kinesis_firehose/firehose.c
- plugins/out_oracle_log_analytics/oci_logan.c
- src/flb_record_accessor.c
- tests/runtime/in_opentelemetry.c
- plugins/out_nats/nats.c
- plugins/out_http/http.c
- plugins/out_kinesis_firehose/firehose_api.c
- tests/runtime/filter_lua.c
- tests/internal/pack.c
- plugins/out_azure_kusto/azure_kusto.c
- plugins/filter_expect/expect.c
- plugins/out_logdna/logdna.c
- plugins/out_es/es.c
- plugins/out_azure_logs_ingestion/azure_logs_ingestion.c
- plugins/out_vivo_exporter/vivo.c
- plugins/filter_nightfall/nightfall_api.c
- plugins/out_datadog/datadog.c
- plugins/out_splunk/splunk.c
- plugins/out_azure_blob/azure_blob.c
- tests/internal/log_event_decoder.c
- plugins/out_cloudwatch_logs/cloudwatch_api.c
- tests/internal/msgpack_append_message.c
- plugins/out_tcp/tcp.c
- plugins/out_lib/out_lib.c
- src/http_server/flb_hs_endpoints.c
- tests/internal/fuzzers/msgpack_parse_fuzzer.c
- plugins/out_kinesis_streams/kinesis_api.c
- plugins/out_bigquery/bigquery.c
- src/http_server/api/v1/metrics.c
- plugins/out_chronicle/chronicle.c
- tests/runtime/in_systemd.c
- src/flb_config.c
🧰 Additional context used
🧬 Code graph analysis (15)
src/flb_help.c (1)
src/flb_pack.c (1)
flb_msgpack_raw_to_json_sds(817-876)
src/http_server/api/v2/reload.c (1)
src/flb_pack.c (1)
flb_msgpack_raw_to_json_sds(817-876)
plugins/processor_content_modifier/cm_utils.c (1)
src/flb_pack.c (1)
flb_msgpack_raw_to_json_sds(817-876)
plugins/out_nrlogs/newrelic.c (1)
src/flb_pack.c (1)
flb_msgpack_raw_to_json_sds(817-876)
src/http_server/api/v1/storage.c (1)
src/flb_pack.c (1)
flb_msgpack_raw_to_json_sds(817-876)
src/fluent-bit.c (1)
src/flb_pack.c (1)
flb_msgpack_raw_to_json_sds(817-876)
tests/internal/mp.c (1)
src/flb_pack.c (2)
flb_msgpack_to_json(802-815)flb_msgpack_raw_to_json_sds(817-876)
src/http_server/api/v1/plugins.c (1)
src/flb_pack.c (1)
flb_msgpack_raw_to_json_sds(817-876)
src/flb_utils.c (1)
include/fluent-bit/flb_simd.h (4)
flb_vector8_load(166-177)flb_vector8_has_le(301-338)flb_vector8_has(288-299)flb_vector8_is_highbit_set(270-283)
tests/internal/utils.c (1)
src/flb_utils.c (1)
flb_utils_write_str(1278-1287)
plugins/out_loki/loki.c (1)
src/flb_pack.c (2)
flb_msgpack_to_json_str(1250-1291)flb_msgpack_raw_to_json_sds(817-876)
plugins/out_s3/s3.c (1)
src/flb_pack.c (1)
flb_pack_msgpack_to_json_format(960-1241)
tests/internal/opentelemetry.c (2)
src/flb_pack.c (1)
flb_msgpack_to_json_str(1250-1291)src/opentelemetry/flb_opentelemetry_utils.c (1)
flb_otel_utils_find_map_entry_by_key(28-73)
include/fluent-bit/flb_utils.h (1)
src/flb_utils.c (1)
flb_utils_write_str_buf(1289-1327)
src/http_server/api/v1/uptime.c (1)
src/flb_pack.c (1)
flb_msgpack_raw_to_json_sds(817-876)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (79)
- GitHub Check: PR - container builds / Windows container images (2025)
- GitHub Check: PR - container builds / Windows container images (2022)
- GitHub Check: PR - packages build Linux / debian/trixie package build and stage to S3
- GitHub Check: PR - packages build Linux / debian/bookworm.arm64v8 package build and stage to S3
- GitHub Check: PR - packages build Linux / almalinux/8 package build and stage to S3
- GitHub Check: PR - packages build Linux / raspbian/bookworm package build and stage to S3
- GitHub Check: PR - packages build Linux / debian/trixie.arm64v8 package build and stage to S3
- GitHub Check: PR - packages build Linux / debian/buster package build and stage to S3
- GitHub Check: PR - packages build Linux / debian/bullseye package build and stage to S3
- GitHub Check: PR - packages build Linux / debian/bullseye.arm64v8 package build and stage to S3
- GitHub Check: PR - packages build Linux / ubuntu/22.04 package build and stage to S3
- GitHub Check: PR - packages build Linux / ubuntu/24.04 package build and stage to S3
- GitHub Check: PR - packages build Linux / ubuntu/24.04.arm64v8 package build and stage to S3
- GitHub Check: PR - packages build Linux / debian/buster.arm64v8 package build and stage to S3
- GitHub Check: PR - packages build Linux / ubuntu/22.04.arm64v8 package build and stage to S3
- GitHub Check: PR - packages build Linux / almalinux/9 package build and stage to S3
- GitHub Check: PR - packages build Linux / debian/bookworm package build and stage to S3
- GitHub Check: PR - packages build Linux / almalinux/9.arm64v8 package build and stage to S3
- GitHub Check: PR - packages build Linux / rockylinux/10 package build and stage to S3
- GitHub Check: PR - packages build Linux / almalinux/10.arm64v8 package build and stage to S3
- GitHub Check: PR - packages build Linux / almalinux/8.arm64v8 package build and stage to S3
- GitHub Check: PR - packages build Linux / almalinux/10 package build and stage to S3
- GitHub Check: PR - packages build Linux / rockylinux/10.arm64v8 package build and stage to S3
- GitHub Check: PR - packages build Linux / rockylinux/8.arm64v8 package build and stage to S3
- GitHub Check: PR - packages build Linux / centos/10.arm64v8 package build and stage to S3
- GitHub Check: PR - packages build Linux / rockylinux/9 package build and stage to S3
- GitHub Check: PR - packages build Linux / centos/9.arm64v8 package build and stage to S3
- GitHub Check: PR - packages build Linux / rockylinux/8 package build and stage to S3
- GitHub Check: PR - packages build Linux / centos/9 package build and stage to S3
- GitHub Check: PR - packages build Linux / rockylinux/9.arm64v8 package build and stage to S3
- GitHub Check: PR - packages build Linux / centos/8.arm64v8 package build and stage to S3
- GitHub Check: PR - packages build Linux / centos/7 package build and stage to S3
- GitHub Check: PR - packages build Linux / centos/8 package build and stage to S3
- GitHub Check: PR - packages build Linux / centos/10 package build and stage to S3
- GitHub Check: PR - packages build Linux / amazonlinux/2023 package build and stage to S3
- GitHub Check: PR - packages build Linux / amazonlinux/2023.arm64v8 package build and stage to S3
- GitHub Check: PR - packages build Linux / amazonlinux/2.arm64v8 package build and stage to S3
- GitHub Check: PR - packages build Linux / centos/7.arm64v8 package build and stage to S3
- GitHub Check: PR - packages build Linux / amazonlinux/2 package build and stage to S3
- GitHub Check: PR - container builds / arm/v7/production container image build
- GitHub Check: PR - container builds / arm64/production container image build
- GitHub Check: PR - container builds / arm/v7/debug container image build
- GitHub Check: PR - container builds / arm64/debug container image build
- GitHub Check: PR - container builds / amd64/production container image build
- GitHub Check: PR - container builds / amd64/debug container image build
- GitHub Check: PR - packages build Windows / call-build-windows-package (Windows 64bit, x64, x64-windows-static, 3.31.6)
- GitHub Check: PR - packages build Windows / call-build-windows-package (Windows 32bit, x86, x86-windows-static, 3.31.6)
- GitHub Check: PR - packages build Windows / call-build-windows-package (Windows 64bit (Arm64), amd64_arm64, -DCMAKE_SYSTEM_NAME=Windows -DCMA...
- GitHub Check: PR - packages build MacOS / call-build-macos-package (Apple Silicon macOS runner, macos-14, 3.31.6)
- GitHub Check: PR - packages build MacOS / call-build-macos-package (Intel macOS runner, macos-14-large, 3.31.6)
- GitHub Check: pr-windows-build / call-build-windows-package (Windows 64bit (Arm64), amd64_arm64, -DCMAKE_SYSTEM_NAME=Windows -DCMA...
- GitHub Check: pr-windows-build / call-build-windows-package (Windows 32bit, x86, x86-windows-static, 3.31.6)
- GitHub Check: pr-windows-build / call-build-windows-package (Windows 64bit, x64, x64-windows-static, 3.31.6)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SIMD=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_ARROW=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SIMD=Off, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DSANITIZE_UNDEFINED=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SANITIZE_MEMORY=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DSANITIZE_ADDRESS=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SIMD=Off, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SIMD=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SANITIZE_THREAD=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SANITIZE_THREAD=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SMALL=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SMALL=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DSANITIZE_ADDRESS=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DSANITIZE_UNDEFINED=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SANITIZE_MEMORY=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_COVERAGE=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_JEMALLOC=Off, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_JEMALLOC=Off, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_JEMALLOC=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_JEMALLOC=On, 3.31.6, gcc, g++)
- GitHub Check: PR - fuzzing test
- GitHub Check: pr-compile-system-libs (-DFLB_PREFER_SYSTEM_LIBS=On, 3.31.6, gcc, g++, ubuntu-24.04, clang-14)
- GitHub Check: pr-compile-system-libs (-DFLB_PREFER_SYSTEM_LIBS=On, 3.31.6, clang, clang++, ubuntu-22.04, clang-12)
- GitHub Check: pr-compile-system-libs (-DFLB_PREFER_SYSTEM_LIBS=On, 3.31.6, clang, clang++, ubuntu-24.04, clang-14)
- GitHub Check: pr-compile-system-libs (-DFLB_PREFER_SYSTEM_LIBS=On, 3.31.6, gcc, g++, ubuntu-22.04, clang-12)
- GitHub Check: pr-compile-centos-7
🔇 Additional comments (24)
plugins/out_nrlogs/newrelic.c (1)
373-375: LGTM: config is correctly plumbed to payload builder.Pass-through aligns with the global toggle for escaping. Please confirm end-to-end that New Relic accepts raw UTF‑8 when json_escape_unicode is Off.
plugins/out_loki/loki.c (7)
568-571: Config plumb-through looks correctForwarding config into pack_maps is consistent.
1643-1645: Config propagation through loki_compose_payload: LGTMSignature change and usage align with the new API.
1735-1738: Nested calls now receive config: LGTM
1773-1775: Else-path nested calls receive config: LGTM
1781-1783: Final JSON generation honors escape flag: LGTM
1866-1868: cb_loki_flush passes config correctly: LGTM
2141-2145: structured_metadata_map_keys entry: LGTMNo functional concerns.
src/flb_pack.c (2)
630-691: Propagating escape_unicode through STR/BIN is correct; verify BIN with non‑UTF‑8 stays valid JSON.If MSGPACK_OBJECT_BIN contains non‑UTF‑8 bytes and escape_unicode == FLB_FALSE, ensure flb_utils_write_str still produces valid JSON (e.g., escapes invalid sequences). Add a focused test for this case.
Also applies to: 692-700
817-876: LGTM — API threaded consistently end-to-end; ast-grep found no legacy arity calls.
Scanned for old arities (flb_msgpack_to_json, flb_msgpack_raw_to_json_sds, flb_pack_msgpack_to_json_format, flb_msgpack_to_json_str); no matches. Approve changes; manually verify external/generated code and plugins not present in the repo.plugins/out_influxdb/influxdb.c (1)
216-222: Correct: honor json_escape_unicode for string fields.Passing config->json_escape_unicode to flb_utils_write_str_buf aligns the plugin with the global option.
plugins/out_websocket/websocket.c (1)
253-259: LGTM: websocket output respects global unicode escaping.Passing config->json_escape_unicode to flb_pack_msgpack_to_json_format is correct and consistent with other outputs.
src/fluent-bit.c (1)
379-381: Explicitly forcing JSON escaping here is fine; just confirming intent.The inline comment says “Keep backward compatibility,” and passing FLB_TRUE hard-codes escaping regardless of config. If that’s the intended UX for CLI help output, LGTM; otherwise consider threading config->json_escape_unicode for consistency with the rest of the PR.
plugins/out_stdout/stdout.c (1)
260-262: LGTM: stdout JSON respects json_escape_unicode.Plumbing the config flag into flb_pack_msgpack_to_json_format looks correct.
plugins/out_s3/s3.c (3)
3349-3351: Signature change LGTM.Adding struct flb_config *config to propagate json_escape_unicode is correct.
3756-3760: LGTM: caller updated to pass config to extract_log_key.Correctly threads the new parameter.
3766-3768: LGTM: S3 JSON formatting respects json_escape_unicode.Propagating the global flag here matches the PR objective.
src/flb_utils.c (3)
1161-1273: LGTM: raw UTF‑8 writer path.Valid UTF‑8 is copied verbatim; essential ASCII escapes preserved; invalid/truncated uses U+FFFD. Matches the feature goal.
1278-1287: Wrapper selection by flag looks good.Clean split between escaped/raw paths via escape_unicode.
1289-1310: Buffer growth loop remains correct with new flag.Delegates to the 6‑arg writer and retries on FLB_FALSE.
include/fluent-bit/flb_utils.h (1)
60-63: Public API change — flb_utils_write_str now has an extra parameter (breaking ABI).In-repo callers are already updated (e.g. include/fluent-bit/flb_utils.h, src/flb_utils.c, src/flb_sds.c:315, plugins/out_cloudwatch_logs/cloudwatch_api.c:709, tests/internal/utils.c). This is still a breaking change for out-of-tree plugins linking against the shared library.
Either bump the SONAME and call out the change in release notes, or add a compatibility wrapper with the old 5-arg signature that forwards to the new 6-arg function (choosing a sensible default for escape_unicode) and keep it exported.
tests/internal/utils.c (2)
205-210: Good harness split: escaped vs raw paths are explicit.Adding escape_unicode to the helper and introducing write_raw_str_test_cases keeps existing tests intact and makes raw-mode coverage straightforward.
Also applies to: 212-214
271-273: Call sites updated with the new flag look correct.All direct flb_utils_write_str calls in this test use FLB_TRUE (escaped mode), preserving prior behavior.
Also applies to: 277-279, 282-284, 288-296
plugins/processor_content_modifier/cm_utils.c (1)
153-155: API update is correct; preserves prior behavior.Passing FLB_TRUE to the new flb_msgpack_raw_to_json_sds(..., escape_unicode) matches the former default and keeps backward compatibility.
| msgpack_object *map, | ||
| struct flb_config *config) | ||
| { |
There was a problem hiding this comment.
🛠️ Refactor suggestion
Fix wrong return type of pack_structured_metadata
Function returns 0 but declares flb_sds_t. Make it int for clarity and to prevent misuse.
-static flb_sds_t pack_structured_metadata(struct flb_loki *ctx,
+static int pack_structured_metadata(struct flb_loki *ctx,
msgpack_packer *mp_pck,
char *tag, int tag_len,
msgpack_object *map,
struct flb_config *config)Note: pack_labels has the same pattern; consider making it int as well.
🤖 Prompt for AI Agents
In plugins/out_loki/loki.c around lines 561-563, the function
pack_structured_metadata is declared to return flb_sds_t but currently returns
integer constants (0), which is incorrect; change its return type to int, update
any return statements to return appropriate status codes (e.g., 0 on success, -1
on error), and update the function's callers if they expect a string to instead
check the integer status; also apply the same change to pack_labels (which
follows the same pattern) so both functions consistently return int status
codes.
| ret = flb_msgpack_to_json(val_buf + val_offset, | ||
| msgpack_size - val_offset, &val); | ||
| msgpack_size - val_offset, &val, | ||
| config->json_escape_unicode); | ||
| if (ret < 0) { |
There was a problem hiding this comment.
Handle buffer growth when JSON serialization needs more space.
flb_msgpack_to_json returns <= 0 when the buffer is too small. Current code treats only ret < 0 as error and does not retry with a larger buffer, which can truncate output.
Apply this diff:
- ret = flb_msgpack_to_json(val_buf + val_offset,
- msgpack_size - val_offset, &val,
- config->json_escape_unicode);
- if (ret < 0) {
- break;
- }
+ while (1) {
+ ret = flb_msgpack_to_json(val_buf + val_offset,
+ msgpack_size - val_offset, &val,
+ config->json_escape_unicode);
+ if (ret > 0) {
+ break;
+ }
+ /* grow buffer and retry */
+ size_t new_size = msgpack_size * 2;
+ char *tmp_buf = flb_realloc(val_buf, new_size);
+ if (!tmp_buf) {
+ flb_errno();
+ ret = -1;
+ break;
+ }
+ val_buf = tmp_buf;
+ msgpack_size = new_size;
+ }
+ if (ret < 0) {
+ break;
+ }📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| ret = flb_msgpack_to_json(val_buf + val_offset, | |
| msgpack_size - val_offset, &val); | |
| msgpack_size - val_offset, &val, | |
| config->json_escape_unicode); | |
| if (ret < 0) { | |
| while (1) { | |
| ret = flb_msgpack_to_json(val_buf + val_offset, | |
| msgpack_size - val_offset, &val, | |
| config->json_escape_unicode); | |
| if (ret > 0) { | |
| break; | |
| } | |
| /* grow buffer and retry */ | |
| size_t new_size = msgpack_size * 2; | |
| char *tmp_buf = flb_realloc(val_buf, new_size); | |
| if (!tmp_buf) { | |
| flb_errno(); | |
| ret = -1; | |
| break; | |
| } | |
| val_buf = tmp_buf; | |
| msgpack_size = new_size; | |
| } | |
| if (ret < 0) { | |
| break; | |
| } |
4 participants
This is because RFC 8259 requests the following both of cases:
ref: https://datatracker.ietf.org/doc/html/rfc8259#section-8
Currently, we implemented pattern 1 of RFC 8259. But That RFC also permits to handle raw UTF-8 strings.
So, we need to support not escaping choice especially for the multibyte characters.
Closes #10631.
Enter
[N/A]in the box, if an item is not applicable to your change.Testing
Before we can approve your change; please submit the following in a comment:
with the GBK encoded file:
Then, the generated file contains:
With leaks which is a detector of memory leaks on macOS:
From valgrind:
If this is a change to packaging of containers or native binaries then please confirm it works for all targets.
ok-package-testlabel to test for all targets (requires maintainer to do).Documentation
Backporting
Fluent Bit is licensed under Apache 2.0, by submitting this pull request I understand that this code will be released under the terms of that license.
Summary by CodeRabbit