* Move filesystem mounting code into a separate file to simplify the
main source file.
* Clean up SDMMC and SDSPI related code. Move pin configuration into
Kconfig.
* Use same base_path ('/data') for either SPIFFS or SD. Remove the
check in file_server.c about the base path.
* Update README according to the changes above.
Feature note: Sets up a simple DNS server that answers all questions with the ESP softAP's IP as well as a HTTP server which redirects all requests to a captive portal welcoming page.
Triggers captive portal detection (log-in screen popup) for iOS, Android and Windows.
Includes example test for testing DNS redirect and HTTP redirect.
Closes https://github.com/espressif/esp-idf/issues/7099
Closes https://github.com/espressif/esp-idf/issues/4075
Now the uri handler gets called immediately after the handshake. In the handler we can identify that this was the handshake by checking req->method as it is still a GET from the first part of the handshake the client has sent. On a normal websocket-frame-receive (when normally a websocket uriHandler gets called) this field is set to 0
Closes https://github.com/espressif/esp-idf/issues/6597
Because this is the 'simple' server example, it is probably where users will start out to understand the HTTP Server API; and also where they might come if they are having problems.
So I think it is useful to explain to users what to do if they see "httpd_parse: parse_block: request URI/header too long", which can easily happen if they submit a POST from an HTML form (for example). That certainly would have helped me!
Merges https://github.com/espressif/esp-idf/pull/6547
Signed-off-by: Shubham Kulkarni <shubham.kulkarni@espressif.com>
the binary size check in example test was removed long time ago. Now we
have updated ttfw_idf to raise exception when performance standard is
not found. These fake performance check will be regarded as error.
Remove them now.
Add logic to set `FIN` flag automatically for transmitted WS frames, but
if `fragmented` option set indicating an expert/manual mode, then the
`FIN` flag is set according to the `final` option.
Since the recent refactoring of spi bus_lock comonent introduced heavy logging under debug/verbose level of verbosity, the http-server test executed in the CI became less stable due to complicated regex. Fixed by removing the first optional group
esp_netif_init() returns standard esp_err_t error code (unlike tcpip_adapter init), so shall be checked for the return value
Also to make the initialization code more consistent.
This MR is intended to fix incorrect parsing of HTTP requests when empty header values are present.
The issue is was due to asymmetric behavior of `http_parser` library, which in case of:
non-empty header values : invokes callbacks with the pointer to the start of a value
empty header values : invokes callbacks with pointer to the start of next header or section
Since HTTP server relies on this pointer (along with length of the value) to locate the end of a value, and replace the line terminators (CRLFs) with null characters, the second case needed to be handled correctly.
Closes IDFGH-1539
Closes https://github.com/espressif/esp-idf/issues/3803
The build system automatically determines offsets of partitions from
the partition table, so no manual changes are needed. Instead, add a
note that partition offsets may need to be updated when increasing
the bootloader size.
This change prevents buffer overflows in case of really long file paths.
Other changes:
* Remove query (?) and fragment (#) component from URI when converting to file path
* /index.html and favicon.ico can be overridden by files with same name and path in SPIFFS
* README.md updated
* Fixed regex constraints for parsing IP and other parameters from monitor log.
* httplib connection timeouts set
* Redundent tests (commented out earlier) have been removed from advanced_tests. These tests are already run during unit testing.
This introduces the following changes in the example test scripts :
* Dependency on python requests library removed in favor of httplib
* Bug fixed in the logic responsible for receiving and processing http chunked responses
* Default timeouts increased Note : Due to connectivity issues (between runner host and DUT) in the runner environment, some of the advanced_tests are being ignored. These tests are intended for verifying the expected limits of the http_server capabilities, and implement sending and receiving of large HTTP packets and malformed requests, running multiple parallel sessions, etc. It is advised that all these tests be run locally, when making changes or adding new features to this component.
