Mirrors/esp-idf
1
0
Fork 0
mirror of https://github.com/espressif/esp-idf.git synced 2026-10-02 03:00:34 +03:00
Code Issues Packages Projects Releases Wiki Activity

feat(esp_http_client): detect oversized headers in tx buffer while sending request

Browse Source 
When CONFIG_ESP_HTTP_CLIENT_STRICT_HEADER_BUFFER is enabled,
esp_http_client_request_send() fails fast when a single request header is
larger than buffer_size_tx, instead of silently emitting a truncated request.

The condition is reported with a dedicated ESP_ERR_HTTP_HEADER_TOO_LONG error
code rather than overloading ESP_ERR_HTTP_WRITE_DATA, so callers can tell a
permanent header-sizing failure apart from a transient write error. errno is
cleared on this path so the async caller (which maps errno == EAGAIN to "retry
later") does not spin retrying a request that can never succeed.

Closes https://github.com/espressif/esp-idf/issues/18639

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This commit is contained in:
Ashish Sharma
2026-06-25 18:24:42 +08:00
co-authored by Claude Opus 4.8
parent c2e931b1b1
commit 8c181842f1
9 changed files with 237 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files
components/esp_http_client/esp_http_client.c
+33
View File
@@ -1817,7 +1817,26 @@ esp_err_t esp_http_client_request_send(esp_http_client_handle_t client, int writ
}
int wlen = client->buffer_size_tx - first_line_len;
#ifdef CONFIG_ESP_HTTP_CLIENT_STRICT_HEADER_BUFFER
int prev_header_index = client->header_index;
#endif
while ((client->header_index = http_header_generate_string(client->request->headers, client->header_index, client->request->buffer->data + first_line_len, &wlen))) {
#ifdef CONFIG_ESP_HTTP_CLIENT_STRICT_HEADER_BUFFER
/* No-progress detection: a positive return equal to (or below)
* the input index means the offending header at prev_header_index
* is larger than the buffer and pagination cannot advance. */
if (client->header_index <= prev_header_index) {
ESP_LOGD(TAG, "Header at index %d does not fit in tx buffer (size: %d)",
prev_header_index, client->buffer_size_tx);
/* This is a permanent failure, not a transient one. Clear errno so
* the async caller (which keys "retry later" off errno == EAGAIN)
* cannot misread a stale EAGAIN and spin retrying a request that
* can never succeed. */
errno = 0;
return ESP_ERR_HTTP_HEADER_TOO_LONG;
}
prev_header_index = client->header_index;
#endif
if (wlen <= 0) {
break;
}
@@ -1843,6 +1862,20 @@ esp_err_t esp_http_client_request_send(esp_http_client_handle_t client, int writ
wlen = client->buffer_size_tx;
}
#ifdef CONFIG_ESP_HTTP_CLIENT_STRICT_HEADER_BUFFER
/* Case where the very first header (at index 0) is larger than the
* buffer: the helper returns 0 with *buffer_len zeroed, so the loop
* exits without entering the body. Distinguish from a legitimate
* empty/already-done state by the zeroed wlen. */
if (client->header_index == 0 && wlen == 0) {
ESP_LOGD(TAG, "Header at index 0 does not fit in tx buffer (size: %d)", client->buffer_size_tx);
/* Permanent failure: clear errno so the async caller does not treat a
* stale EAGAIN as "retry later" and spin on an unsendable request. */
errno = 0;
return ESP_ERR_HTTP_HEADER_TOO_LONG;
}
#endif
client->data_written_index = 0;
client->data_write_left = client->post_len;
client->state = HTTP_STATE_REQ_COMPLETE_HEADER;