AaronPo/the-biergarten-app
1
0
Fork 0
mirror of https://github.com/aaronpo97/the-biergarten-app.git synced 2026-10-15 01:54:43 +00:00
Code Issues Packages Projects Releases Wiki Activity

C++ code formatting

Browse Source
This commit is contained in:
Aaron Po
2026-08-24 01:13:41 -04:00
parent b64c42ebde
commit 03ca966708
12 changed files with 136 additions and 143 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
tooling/pipeline/includes/web_client/http_web_client.h
View File

@@ -51,10 +51,9 @@ class HttpWebClient final : public WebClient {
* @param headers Additional request headers. * @param headers Additional request headers.
* @return Response body on HTTP 2xx; throws std::runtime_error otherwise. * @return Response body on HTTP 2xx; throws std::runtime_error otherwise.
*/ */
std::string Post( std::string Post(const std::string& url, const std::string& body,
const std::string& url, const std::string& body, const std::vector<std::pair<std::string, std::string>>&
const std::vector<std::pair<std::string, std::string>>& headers) headers) override;
override;
/** /**
* @brief Percent-encodes a single URI component (query parameter value or * @brief Percent-encodes a single URI component (query parameter value or

7
tooling/pipeline/src/application_options/parse_arguments.cc
View File

@@ -121,8 +121,8 @@ std::optional<ApplicationOptions> ParseArguments(
// Enforce mutual exclusivity before any further configuration is // Enforce mutual exclusivity before any further configuration is
// applied. // applied.
const int selected_generator_count = const int selected_generator_count = (use_mocked ? 1 : 0) +
(use_mocked ? 1 : 0) + (!model_path.empty() ? 1 : 0) + (!model_path.empty() ? 1 : 0) +
(use_openai ? 1 : 0); (use_openai ? 1 : 0);
if (selected_generator_count > 1) { if (selected_generator_count > 1) {
@@ -190,7 +190,8 @@ std::optional<ApplicationOptions> ParseArguments(
: use_openai ? GeneratorMode::kOpenAI : use_openai ? GeneratorMode::kOpenAI
: GeneratorMode::kLlama; : GeneratorMode::kLlama;
options.generator.model_path = model_path; options.generator.model_path = model_path;
options.generator.openai_model = var_map["openai-model"].as<std::string>(); options.generator.openai_model =
var_map["openai-model"].as<std::string>();
// Only populate sampling config when the user explicitly overrides at // Only populate sampling config when the user explicitly overrides at
// least one value. Leaving it as std::nullopt lets LlamaGenerator fall // least one value. Leaving it as std::nullopt lets LlamaGenerator fall

1
tooling/pipeline/src/biergarten_pipeline_orchestrator/query_cities_with_countries.cc
View File

@@ -43,6 +43,5 @@ std::vector<City> BiergartenPipelineOrchestrator::QueryLocations() {
" Locations available: {}\n Sampled Locations: {}", " Locations available: {}\n Sampled Locations: {}",
all_locations.size(), sampled_locations.size())}); all_locations.size(), sampled_locations.size())});
return sampled_locations; return sampled_locations;
} }

9
tooling/pipeline/src/data_generation/generation_json_validation.cc
View File

@@ -8,14 +8,13 @@
#include <algorithm> #include <algorithm>
#include <array> #include <array>
#include <boost/json.hpp>
#include <cctype> #include <cctype>
#include <format> #include <format>
#include <span> #include <span>
#include <string> #include <string>
#include <string_view> #include <string_view>
#include <boost/json.hpp>
namespace { namespace {
std::string Trim(std::string_view value) { std::string Trim(std::string_view value) {
constexpr std::string_view whitespace = " \t\n\r\f\v"; constexpr std::string_view whitespace = " \t\n\r\f\v";
@@ -55,7 +54,8 @@ std::string CondenseWhitespace(std::string_view text) {
return out; return out;
} }
// Keeps truncation past the string's midpoint, preserving the critical opening content. // Keeps truncation past the string's midpoint, preserving the critical opening
// content.
constexpr size_t kTruncationGuardDivisor = 2; constexpr size_t kTruncationGuardDivisor = 2;
bool ReadRequiredTrimmedStringField(const boost::json::object& obj, bool ReadRequiredTrimmedStringField(const boost::json::object& obj,
@@ -64,8 +64,7 @@ bool ReadRequiredTrimmedStringField(const boost::json::object& obj,
const boost::json::value* field = obj.if_contains(key); const boost::json::value* field = obj.if_contains(key);
if (field == nullptr || !field->is_string()) { if (field == nullptr || !field->is_string()) {
if (error_out != nullptr) { if (error_out != nullptr) {
*error_out = *error_out = std::format("Missing or invalid string field: {}", key);
std::format("Missing or invalid string field: {}", key);
} }
return false; return false;
} }

30
tooling/pipeline/src/data_generation/openai/call_chat_completions_api.cc
View File

@@ -6,9 +6,11 @@
* *
* API docs last accessed 2026-08-23. * API docs last accessed 2026-08-23.
* Docs: https://platform.openai.com/docs/api-reference/chat/create * Docs: https://platform.openai.com/docs/api-reference/chat/create
* Structured Outputs: https://platform.openai.com/docs/guides/structured-outputs * Structured Outputs:
* https://platform.openai.com/docs/guides/structured-outputs
*/ */
#include <boost/json.hpp>
#include <format> #include <format>
#include <stdexcept> #include <stdexcept>
#include <string> #include <string>
@@ -16,8 +18,6 @@
#include <utility> #include <utility>
#include <vector> #include <vector>
#include <boost/json.hpp>
#include "data_generation/openai_generator.h" #include "data_generation/openai_generator.h"
namespace { namespace {
@@ -48,7 +48,6 @@ std::string OpenAIGenerator::CallChatCompletionsApi(
const std::string& system_prompt, const std::string& user_prompt, const std::string& system_prompt, const std::string& user_prompt,
std::string_view json_schema, std::string_view schema_name, std::string_view json_schema, std::string_view schema_name,
int max_tokens) { int max_tokens) {
boost::system::error_code schema_error; boost::system::error_code schema_error;
boost::json::value schema_value = boost::json::value schema_value =
boost::json::parse(json_schema, schema_error); boost::json::parse(json_schema, schema_error);
@@ -60,7 +59,8 @@ std::string OpenAIGenerator::CallChatCompletionsApi(
schema_error.message()); schema_error.message());
} }
// response_format: https://platform.openai.com/docs/guides/structured-outputs // response_format:
// https://platform.openai.com/docs/guides/structured-outputs
boost::json::object json_schema_wrapper; boost::json::object json_schema_wrapper;
json_schema_wrapper["name"] = std::string(schema_name); json_schema_wrapper["name"] = std::string(schema_name);
json_schema_wrapper["strict"] = true; json_schema_wrapper["strict"] = true;
@@ -127,14 +127,14 @@ std::string OpenAIGenerator::CallChatCompletionsApi(
message_field != nullptr && message_field->is_string()) { message_field != nullptr && message_field->is_string()) {
error_message = std::string(message_field->as_string()); error_message = std::string(message_field->as_string());
} }
throw std::runtime_error( throw std::runtime_error(std::format(
std::format("OpenAIGenerator: API request failed: {}", error_message)); "OpenAIGenerator: API request failed: {}", error_message));
} }
const boost::json::value* choices_field = response_obj. const boost::json::value* choices_field =
if_contains("choices"); response_obj.if_contains("choices");
bool choices_field_missing_or_invalid = bool choices_field_missing_or_invalid = choices_field == nullptr ||
choices_field == nullptr || !choices_field->is_array() || !choices_field->is_array() ||
choices_field->get_array().empty(); choices_field->get_array().empty();
if (choices_field_missing_or_invalid) { if (choices_field_missing_or_invalid) {
@@ -161,12 +161,12 @@ std::string OpenAIGenerator::CallChatCompletionsApi(
// Refusal field: https://platform.openai.com/docs/guides/structured-outputs // Refusal field: https://platform.openai.com/docs/guides/structured-outputs
const boost::json::value* refusal_field = message_obj.if_contains("refusal"); const boost::json::value* refusal_field = message_obj.if_contains("refusal");
bool model_refused_request = bool model_refused_request = refusal_field != nullptr &&
refusal_field != nullptr && refusal_field->is_string() && refusal_field->is_string() &&
!refusal_field->as_string().empty(); !refusal_field->as_string().empty();
if (model_refused_request) { if (model_refused_request) {
throw std::runtime_error(std::format( throw std::runtime_error(
"OpenAIGenerator: request refused by safety policies: {}", std::format("OpenAIGenerator: request refused by safety policies: {}",
std::string(refusal_field->as_string()))); std::string(refusal_field->as_string())));
} }

15
tooling/pipeline/src/data_generation/openai/generate_brewery.cc
View File

@@ -17,8 +17,7 @@
namespace { namespace {
std::string FormatLocalLanguageCodes( std::string FormatLocalLanguageCodes(const std::vector<std::string>& codes) {
const std::vector<std::string>& codes) {
if (codes.empty()) { if (codes.empty()) {
return "Not provided"; return "Not provided";
} }
@@ -36,7 +35,8 @@ std::string FormatLocalLanguageCodes(
constexpr int kBreweryMaxTokens = 2800; constexpr int kBreweryMaxTokens = 2800;
// Structured outputs already guarantee schema-valid JSON, so retries here only cover transient failures or the model returning placeholder text. // Structured outputs already guarantee schema-valid JSON, so retries here only
// cover transient failures or the model returning placeholder text.
constexpr int kMaxAttempts = 2; constexpr int kMaxAttempts = 2;
} // namespace } // namespace
@@ -51,7 +51,8 @@ BreweryResult OpenAIGenerator::GenerateBrewery(
FormatLocalLanguageCodes(location.local_languages); FormatLocalLanguageCodes(location.local_languages);
/** /**
* Loads the backend-agnostic brewery system prompt via the injected prompt directory. * Loads the backend-agnostic brewery system prompt via the injected prompt
* directory.
*/ */
const std::string system_prompt = const std::string system_prompt =
prompt_directory_->Load("BREWERY_GENERATION"); prompt_directory_->Load("BREWERY_GENERATION");
@@ -70,9 +71,9 @@ BreweryResult OpenAIGenerator::GenerateBrewery(
std::string last_error; std::string last_error;
for (int attempt = 0; attempt < kMaxAttempts; ++attempt) { for (int attempt = 0; attempt < kMaxAttempts; ++attempt) {
raw = CallChatCompletionsApi(system_prompt, user_prompt, raw =
kBreweryJsonSchema, "brewery_result", CallChatCompletionsApi(system_prompt, user_prompt, kBreweryJsonSchema,
kBreweryMaxTokens); "brewery_result", kBreweryMaxTokens);
if (logger_) { if (logger_) {
logger_->Log({.level = LogLevel::Debug, logger_->Log({.level = LogLevel::Debug,
.phase = PipelinePhase::BreweryAndBeerGeneration, .phase = PipelinePhase::BreweryAndBeerGeneration,

6
tooling/pipeline/src/data_generation/openai/generate_user.cc
View File

@@ -16,7 +16,8 @@
namespace { namespace {
constexpr int kUserMaxTokens = 1200; constexpr int kUserMaxTokens = 1200;
// Structured outputs already guarantee schema-valid JSON, so retries here only cover transient failures or placeholder text. // Structured outputs already guarantee schema-valid JSON, so retries here only
// cover transient failures or placeholder text.
constexpr int kMaxAttempts = 2; constexpr int kMaxAttempts = 2;
} // namespace } // namespace
@@ -47,8 +48,7 @@ UserResult OpenAIGenerator::GenerateUser(const EnrichedCity& city,
const std::string retry_context = std::format( const std::string retry_context = std::format(
"Name: {} {}\nCity: {}, {}\nPersona: {}", name.first_name, "Name: {} {}\nCity: {}, {}\nPersona: {}", name.first_name,
name.last_name, city.location.city, city.location.country, name.last_name, city.location.city, city.location.country, persona.name);
persona.name);
std::string raw; std::string raw;
std::string last_error; std::string last_error;

4
tooling/pipeline/src/data_generation/openai/openai_generator.cc
View File

@@ -3,12 +3,12 @@
* @brief OpenAIGenerator constructor. * @brief OpenAIGenerator constructor.
*/ */
#include "data_generation/openai_generator.h"
#include <memory> #include <memory>
#include <string> #include <string>
#include <utility> #include <utility>
#include "data_generation/openai_generator.h"
OpenAIGenerator::OpenAIGenerator( OpenAIGenerator::OpenAIGenerator(
std::string api_key, std::string model, std::shared_ptr<ILogger> logger, std::string api_key, std::string model, std::shared_ptr<ILogger> logger,
std::unique_ptr<IPromptDirectory> prompt_directory, std::unique_ptr<IPromptDirectory> prompt_directory,

14
tooling/pipeline/src/main.cc
View File

@@ -166,10 +166,10 @@ int main(const int argc, char** argv) {
inj.template create<std::unique_ptr<WebClient>>(), inj.template create<std::unique_ptr<WebClient>>(),
log_producer); log_producer);
}), }),
di::bind<DataGenerator>().to( di::bind<DataGenerator>().to([&options, &model_path, &sampling,
[&options, &model_path, &sampling, &prompt_directory, &prompt_directory,
&log_producer]( &log_producer](const auto& inj)
const auto& inj) -> std::unique_ptr<DataGenerator> { -> std::unique_ptr<DataGenerator> {
if (options.generator.mode == GeneratorMode::kMock) { if (options.generator.mode == GeneratorMode::kMock) {
log_producer->Log({.level = LogLevel::Info, log_producer->Log({.level = LogLevel::Info,
.phase = PipelinePhase::Startup, .phase = PipelinePhase::Startup,
@@ -179,8 +179,7 @@ int main(const int argc, char** argv) {
} }
if (options.generator.mode == GeneratorMode::kOpenAI) { if (options.generator.mode == GeneratorMode::kOpenAI) {
log_producer->Log( log_producer->Log({.level = LogLevel::Info,
{.level = LogLevel::Info,
.phase = PipelinePhase::Startup, .phase = PipelinePhase::Startup,
.message = std::format( .message = std::format(
"Generator: OpenAIGenerator | model={}", "Generator: OpenAIGenerator | model={}",
@@ -194,7 +193,8 @@ int main(const int argc, char** argv) {
} }
#ifdef BIERGARTEN_MOCK_ONLY #ifdef BIERGARTEN_MOCK_ONLY
// Only reachable when mode == Llama, which requires llama.cpp and is excluded from a BIERGARTEN_MOCK_ONLY build. // Only reachable when mode == Llama, which requires llama.cpp and
// is excluded from a BIERGARTEN_MOCK_ONLY build.
throw std::runtime_error( throw std::runtime_error(
"LlamaGenerator is unavailable in a BIERGARTEN_MOCK_ONLY " "LlamaGenerator is unavailable in a BIERGARTEN_MOCK_ONLY "
"build"); "build");

18
tooling/pipeline/src/services/curated_data/curated_json_data_service.cc
View File

@@ -125,8 +125,7 @@ const LocationsList& CuratedJsonDataService::LoadLocations() {
if (!cache_.locations.empty()) { if (!cache_.locations.empty()) {
return cache_.locations; return cache_.locations;
} }
boost::json::value root_val = boost::json::value root_val = ParseJsonFile(filepaths_.locations_path);
ParseJsonFile(filepaths_.locations_path);
if (!root_val.is_object()) { if (!root_val.is_object()) {
throw std::runtime_error( throw std::runtime_error(
@@ -135,14 +134,12 @@ const LocationsList& CuratedJsonDataService::LoadLocations() {
boost::json::object root_obj = root_val.as_object(); boost::json::object root_obj = root_val.as_object();
if (!root_obj.contains("cities")) if (!root_obj.contains("cities")) {
{
throw std::runtime_error( throw std::runtime_error(
"Invalid locations JSON: root object must contain a 'cities' key"); "Invalid locations JSON: root object must contain a 'cities' key");
} }
if (!root_obj["cities"].is_array()) if (!root_obj["cities"].is_array()) {
{
throw std::runtime_error( throw std::runtime_error(
"Invalid locations JSON: 'cities' key must be an array"); "Invalid locations JSON: 'cities' key must be an array");
} }
@@ -179,8 +176,7 @@ const PersonasList& CuratedJsonDataService::LoadPersonas() {
return cache_.personas; return cache_.personas;
} }
const boost::json::value root = const boost::json::value root = ParseJsonFile(filepaths_.personas_path);
ParseJsonFile(filepaths_.personas_path);
if (!root.is_array()) { if (!root.is_array()) {
throw std::runtime_error( throw std::runtime_error(
@@ -215,8 +211,7 @@ const ForenamesByCountryMap& CuratedJsonDataService::LoadForenamesByCountry() {
return cache_.forenames_by_country; return cache_.forenames_by_country;
} }
const boost::json::value root = const boost::json::value root = ParseJsonFile(filepaths_.forenames_path);
ParseJsonFile(filepaths_.forenames_path);
if (!root.is_object()) { if (!root.is_object()) {
throw std::runtime_error( throw std::runtime_error(
@@ -263,8 +258,7 @@ const SurnamesByCountryMap& CuratedJsonDataService::LoadSurnamesByCountry() {
return cache_.surnames_by_country; return cache_.surnames_by_country;
} }
const boost::json::value root = const boost::json::value root = ParseJsonFile(filepaths_.surnames_path);
ParseJsonFile(filepaths_.surnames_path);
if (!root.is_object()) { if (!root.is_object()) {
throw std::runtime_error( throw std::runtime_error(

4
tooling/pipeline/src/services/sqlite/process_record.cc
View File

@@ -180,8 +180,8 @@ uint64_t SqliteExportService::ProcessRecord(const BreweryRecord& brewery) {
sqlite_export_service_internal::Bind( sqlite_export_service_internal::Bind(
insert_brewery_address_stmt_, insert_brewery_address_stmt_,
sqlite_export_service_internal::BoundParam<double>{ sqlite_export_service_internal::BoundParam<double>{
.index = sqlite_export_service_internal:: .index =
kBreweryAddressLatitudeBindIndex, sqlite_export_service_internal::kBreweryAddressLatitudeBindIndex,
.value = brewery.address.latitude, .value = brewery.address.latitude,
.action = "Failed to bind SQLite brewery address latitude"}); .action = "Failed to bind SQLite brewery address latitude"});