Skip to content

Curl: fix pagination problems #206

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
carlopi merged 5 commits into from
Jan 15, 2026
Merged

Curl: fix pagination problems #206

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
fe93f3c
curl: fix missing url
DinosL Jan 9, 2026
b91491b
add auth information to header
DinosL Jan 14, 2026
ee22299
don't encode custom headers as query params
DinosL Jan 14, 2026
a9b9990
remove unecessary test
DinosL Jan 14, 2026
c089000
Use the response handler to get the pagination info
DinosL Jan 14, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
67 changes: 25 additions & 42 deletions src/httpfs_curl_client.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -191,16 +191,13 @@ class HTTPFSCurlClient : public HTTPClient {
}

unique_ptr<HTTPResponse> Get(GetRequestInfo &info) override {
ResetRequestInfo();
if (state) {
state->get_count++;
}

auto curl_headers = TransformHeadersCurl(info.headers);
auto curl_headers = TransformHeadersCurl(info.headers, info.params);
request_info->url = info.url;
if (!info.params.extra_headers.empty()) {
auto curl_params = TransformParamsCurl(info.params);
request_info->url += "?" + curl_params;
}

CURLcode res;
{
Expand All @@ -226,6 +223,13 @@ class HTTPFSCurlClient : public HTTPClient {
state->total_bytes_received += bytes_received;
}

if (info.response_handler) {
auto response = TransformResponseCurl(res);
if (!info.response_handler(*response)) {
return response;
}
}

Comment on lines +226 to +232
Copy link
Collaborator

@carlopi carlopi Jan 14, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Could you possibly check the similar code in httpfs_httplib_client.cpp:

        unique_ptr<HTTPResponse> Get(GetRequestInfo &info) override {
                if (state) {
                        state->get_count++;
                }
                auto headers = TransformHeaders(info.headers, info.params);
                if (!info.response_handler && !info.content_handler) {
                        return TransformResult(client->Get(info.path, headers));
                } else {
                        return TransformResult(client->Get(
                            info.path.c_str(), headers,
                            [&](const duckdb_httplib_openssl::Response &response) {
                                    auto http_response = TransformResponse(response);
                                    return info.response_handler(*http_response);
                            },
                            [&](const char *data, size_t data_length) {
                                    if (state) {
                                            state->total_bytes_received += data_length;
                                    }
                                    return info.content_handler(const_data_ptr_cast(data), data_length);
                            }));
                }
        }

In particular, I am not sure if they can be made more similar to each other the GET handling.

const char *data = request_info->body.c_str();
if (info.content_handler) {
info.content_handler(const_data_ptr_cast(data), bytes_received);
Expand All @@ -235,20 +239,17 @@ class HTTPFSCurlClient : public HTTPClient {
}

unique_ptr<HTTPResponse> Put(PutRequestInfo &info) override {
ResetRequestInfo();
if (state) {
state->put_count++;
state->total_bytes_sent += info.buffer_in_len;
}

auto curl_headers = TransformHeadersCurl(info.headers);
auto curl_headers = TransformHeadersCurl(info.headers, info.params);
// Add content type header from info
curl_headers.Add("Content-Type: " + info.content_type);
// transform parameters
request_info->url = info.url;
if (!info.params.extra_headers.empty()) {
auto curl_params = TransformParamsCurl(info.params);
request_info->url += "?" + curl_params;
}

CURLcode res;
{
Expand All @@ -272,17 +273,14 @@ class HTTPFSCurlClient : public HTTPClient {
}

unique_ptr<HTTPResponse> Head(HeadRequestInfo &info) override {
ResetRequestInfo();
if (state) {
state->head_count++;
}

auto curl_headers = TransformHeadersCurl(info.headers);
auto curl_headers = TransformHeadersCurl(info.headers, info.params);
request_info->url = info.url;
// transform parameters
if (!info.params.extra_headers.empty()) {
auto curl_params = TransformParamsCurl(info.params);
request_info->url += "?" + curl_params;
}

CURLcode res;
{
Expand All @@ -306,17 +304,14 @@ class HTTPFSCurlClient : public HTTPClient {
}

unique_ptr<HTTPResponse> Delete(DeleteRequestInfo &info) override {
ResetRequestInfo();
if (state) {
state->delete_count++;
}

auto curl_headers = TransformHeadersCurl(info.headers);
auto curl_headers = TransformHeadersCurl(info.headers, info.params);
// transform parameters
request_info->url = info.url;
if (!info.params.extra_headers.empty()) {
auto curl_params = TransformParamsCurl(info.params);
request_info->url += "?" + curl_params;
}

CURLcode res;
{
Expand All @@ -343,20 +338,17 @@ class HTTPFSCurlClient : public HTTPClient {
}

unique_ptr<HTTPResponse> Post(PostRequestInfo &info) override {
ResetRequestInfo();
if (state) {
state->post_count++;
state->total_bytes_sent += info.buffer_in_len;
}

auto curl_headers = TransformHeadersCurl(info.headers);
auto curl_headers = TransformHeadersCurl(info.headers, info.params);
const string content_type = "Content-Type: application/octet-stream";
curl_headers.Add(content_type.c_str());
// transform parameters
request_info->url = info.url;
if (!info.params.extra_headers.empty()) {
auto curl_params = TransformParamsCurl(info.params);
request_info->url += "?" + curl_params;
}

CURLcode res;
{
Expand All @@ -382,7 +374,9 @@ class HTTPFSCurlClient : public HTTPClient {
}

private:
CURLRequestHeaders TransformHeadersCurl(const HTTPHeaders &header_map) {
CURLRequestHeaders TransformHeadersCurl(const HTTPHeaders &header_map, const HTTPParams &params) {
auto &httpfs_params = params.Cast<HTTPFSParams>();

std::vector<std::string> headers;
for (auto &entry : header_map) {
const std::string new_header = entry.first + ": " + entry.second;
Expand All @@ -392,23 +386,12 @@ class HTTPFSCurlClient : public HTTPClient {
for (auto &header : headers) {
curl_headers.Add(header);
}
return curl_headers;
}

string TransformParamsCurl(const HTTPParams &params) {
string result = "";
unordered_map<string, string> escaped_params;
bool first_param = true;
for (auto &entry : params.extra_headers) {
const string key = entry.first;
const string value = curl_easy_escape(*curl, entry.second.c_str(), 0);
if (!first_param) {
result += "&";
if (!httpfs_params.pre_merged_headers) {
for (auto &entry : params.extra_headers) {
curl_headers.Add(entry.first + ": " + entry.second);
}
result += key + "=" + value;
first_param = false;
}
return result;
return curl_headers;
}

void ResetRequestInfo() {
Expand Down Expand Up @@ -441,7 +424,7 @@ class HTTPFSCurlClient : public HTTPClient {
response->headers.Insert(header.first, header.second);
}
}
ResetRequestInfo();
// ResetRequestInfo();
return response;
}

Expand Down
3 changes: 1 addition & 2 deletions src/s3fs.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1236,13 +1236,12 @@ string AWSListObjectV2::Request(string &path, HTTPParams &http_params, S3AuthPar
req_params += "&delimiter=%2F";
}

string listobjectv2_url = req_path + "?" + req_params;

auto header_map =
CreateS3Header(req_path, req_params, parsed_url.host, "s3", "GET", s3_auth_params, "", "", "", "");

// Get requests use fresh connection
string full_host = parsed_url.http_proto + parsed_url.host;
string listobjectv2_url = full_host + req_path + "?" + req_params;
std::stringstream response;
GetRequestInfo get_request(
full_host, listobjectv2_url, header_map, http_params,
Expand Down
Loading