Fork of the espurna firmware for `mhsw` switches
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

514 lines
14 KiB

Rework settings (#2282) * wip based on early draft. todo benchmarking * fixup eraser, assume keys are unique * fix cursor copy, test removal at random * small benchmark via permutations. todo lambdas and novirtual * fix empty condition / reset * overwrite optimizations, fix move offsets overflows * ...erase using 0xff instead of 0 * test the theory with code, different length kv were bugged * try to check for out-of-bounds writes / reads * style * trying to fix mover again * clarify length, defend against reading len on edge * fix uncommited rewind change * prove space assumptions * more concise traces, fix move condition (agrh!!!) * slightly more internal knowledge (estimates API?) * make sure cursor is only valid within the range * ensure 0 does not blow things * go back up * cursor comments * comments * rewrite writes through cursor * in del too * estimate kv storage requirements, return available size * move raw erase / move into a method, allow ::set to avoid scanning storage twice * refactor naming, use in code * amend storage slicing test * fix crash handler offsets, cleanup configuration * start -> begin * eeprom readiness * dependencies * unused * SPI_FLASH constants for older Core * vtables -> templates * less include dependencies * gcov help, move estimate outside of the class * writer position can never match, use begin + offset * tweak save_crash to trigger only once in a serious crash * doh, header function should be inline * foreach api, tweak structs for public api * use test helper class * when not using foreach, move cursor reset closer to the loop using read_kv * coverage comments, fix typo in tests decltype * ensure set() does not break with offset * make codacy happy again
4 years ago
Rework settings (#2282) * wip based on early draft. todo benchmarking * fixup eraser, assume keys are unique * fix cursor copy, test removal at random * small benchmark via permutations. todo lambdas and novirtual * fix empty condition / reset * overwrite optimizations, fix move offsets overflows * ...erase using 0xff instead of 0 * test the theory with code, different length kv were bugged * try to check for out-of-bounds writes / reads * style * trying to fix mover again * clarify length, defend against reading len on edge * fix uncommited rewind change * prove space assumptions * more concise traces, fix move condition (agrh!!!) * slightly more internal knowledge (estimates API?) * make sure cursor is only valid within the range * ensure 0 does not blow things * go back up * cursor comments * comments * rewrite writes through cursor * in del too * estimate kv storage requirements, return available size * move raw erase / move into a method, allow ::set to avoid scanning storage twice * refactor naming, use in code * amend storage slicing test * fix crash handler offsets, cleanup configuration * start -> begin * eeprom readiness * dependencies * unused * SPI_FLASH constants for older Core * vtables -> templates * less include dependencies * gcov help, move estimate outside of the class * writer position can never match, use begin + offset * tweak save_crash to trigger only once in a serious crash * doh, header function should be inline * foreach api, tweak structs for public api * use test helper class * when not using foreach, move cursor reset closer to the loop using read_kv * coverage comments, fix typo in tests decltype * ensure set() does not break with offset * make codacy happy again
4 years ago
Terminal: change command-line parser (#2247) Change the underlying command line handling: - switch to a custom parser, inspired by redis / sds - update terminalRegisterCommand signature, pass only bare minimum - clean-up `help` & `commands`. update settings `set`, `get` and `del` - allow our custom test suite to run command-line tests - clean-up Stream IO to allow us to print large things into debug stream (for example, `eeprom.dump`) - send parsing errors to the debug log As a proof of concept, introduce `TERMINAL_MQTT_SUPPORT` and `TERMINAL_WEB_API_SUPPORT` - MQTT subscribes to the `<root>/cmd/set` and sends response to the `<root>/cmd`. We can't output too much, as we don't have any large-send API. - Web API listens to the `/api/cmd?apikey=...&line=...` (or PUT, params inside the body). This one is intended as a possible replacement of the `API_SUPPORT`. Internals introduce a 'task' around the AsyncWebServerRequest object that will simulate what WiFiClient does and push data into it continuously, switching between CONT and SYS. Both are experimental. We only accept a single command and not every command is updated to use Print `ctx.output` object. We are also somewhat limited by the Print / Stream overall, perhaps I am overestimating the usefulness of Arduino compatibility to such an extent :) Web API handler can also sometimes show only part of the result, whenever the command tries to yield() by itself waiting for something. Perhaps we would need to create a custom request handler for that specific use-case.
4 years ago
Rework settings (#2282) * wip based on early draft. todo benchmarking * fixup eraser, assume keys are unique * fix cursor copy, test removal at random * small benchmark via permutations. todo lambdas and novirtual * fix empty condition / reset * overwrite optimizations, fix move offsets overflows * ...erase using 0xff instead of 0 * test the theory with code, different length kv were bugged * try to check for out-of-bounds writes / reads * style * trying to fix mover again * clarify length, defend against reading len on edge * fix uncommited rewind change * prove space assumptions * more concise traces, fix move condition (agrh!!!) * slightly more internal knowledge (estimates API?) * make sure cursor is only valid within the range * ensure 0 does not blow things * go back up * cursor comments * comments * rewrite writes through cursor * in del too * estimate kv storage requirements, return available size * move raw erase / move into a method, allow ::set to avoid scanning storage twice * refactor naming, use in code * amend storage slicing test * fix crash handler offsets, cleanup configuration * start -> begin * eeprom readiness * dependencies * unused * SPI_FLASH constants for older Core * vtables -> templates * less include dependencies * gcov help, move estimate outside of the class * writer position can never match, use begin + offset * tweak save_crash to trigger only once in a serious crash * doh, header function should be inline * foreach api, tweak structs for public api * use test helper class * when not using foreach, move cursor reset closer to the loop using read_kv * coverage comments, fix typo in tests decltype * ensure set() does not break with offset * make codacy happy again
4 years ago
Terminal: change command-line parser (#2247) Change the underlying command line handling: - switch to a custom parser, inspired by redis / sds - update terminalRegisterCommand signature, pass only bare minimum - clean-up `help` & `commands`. update settings `set`, `get` and `del` - allow our custom test suite to run command-line tests - clean-up Stream IO to allow us to print large things into debug stream (for example, `eeprom.dump`) - send parsing errors to the debug log As a proof of concept, introduce `TERMINAL_MQTT_SUPPORT` and `TERMINAL_WEB_API_SUPPORT` - MQTT subscribes to the `<root>/cmd/set` and sends response to the `<root>/cmd`. We can't output too much, as we don't have any large-send API. - Web API listens to the `/api/cmd?apikey=...&line=...` (or PUT, params inside the body). This one is intended as a possible replacement of the `API_SUPPORT`. Internals introduce a 'task' around the AsyncWebServerRequest object that will simulate what WiFiClient does and push data into it continuously, switching between CONT and SYS. Both are experimental. We only accept a single command and not every command is updated to use Print `ctx.output` object. We are also somewhat limited by the Print / Stream overall, perhaps I am overestimating the usefulness of Arduino compatibility to such an extent :) Web API handler can also sometimes show only part of the result, whenever the command tries to yield() by itself waiting for something. Perhaps we would need to create a custom request handler for that specific use-case.
4 years ago
Rework settings (#2282) * wip based on early draft. todo benchmarking * fixup eraser, assume keys are unique * fix cursor copy, test removal at random * small benchmark via permutations. todo lambdas and novirtual * fix empty condition / reset * overwrite optimizations, fix move offsets overflows * ...erase using 0xff instead of 0 * test the theory with code, different length kv were bugged * try to check for out-of-bounds writes / reads * style * trying to fix mover again * clarify length, defend against reading len on edge * fix uncommited rewind change * prove space assumptions * more concise traces, fix move condition (agrh!!!) * slightly more internal knowledge (estimates API?) * make sure cursor is only valid within the range * ensure 0 does not blow things * go back up * cursor comments * comments * rewrite writes through cursor * in del too * estimate kv storage requirements, return available size * move raw erase / move into a method, allow ::set to avoid scanning storage twice * refactor naming, use in code * amend storage slicing test * fix crash handler offsets, cleanup configuration * start -> begin * eeprom readiness * dependencies * unused * SPI_FLASH constants for older Core * vtables -> templates * less include dependencies * gcov help, move estimate outside of the class * writer position can never match, use begin + offset * tweak save_crash to trigger only once in a serious crash * doh, header function should be inline * foreach api, tweak structs for public api * use test helper class * when not using foreach, move cursor reset closer to the loop using read_kv * coverage comments, fix typo in tests decltype * ensure set() does not break with offset * make codacy happy again
4 years ago
Rework settings (#2282) * wip based on early draft. todo benchmarking * fixup eraser, assume keys are unique * fix cursor copy, test removal at random * small benchmark via permutations. todo lambdas and novirtual * fix empty condition / reset * overwrite optimizations, fix move offsets overflows * ...erase using 0xff instead of 0 * test the theory with code, different length kv were bugged * try to check for out-of-bounds writes / reads * style * trying to fix mover again * clarify length, defend against reading len on edge * fix uncommited rewind change * prove space assumptions * more concise traces, fix move condition (agrh!!!) * slightly more internal knowledge (estimates API?) * make sure cursor is only valid within the range * ensure 0 does not blow things * go back up * cursor comments * comments * rewrite writes through cursor * in del too * estimate kv storage requirements, return available size * move raw erase / move into a method, allow ::set to avoid scanning storage twice * refactor naming, use in code * amend storage slicing test * fix crash handler offsets, cleanup configuration * start -> begin * eeprom readiness * dependencies * unused * SPI_FLASH constants for older Core * vtables -> templates * less include dependencies * gcov help, move estimate outside of the class * writer position can never match, use begin + offset * tweak save_crash to trigger only once in a serious crash * doh, header function should be inline * foreach api, tweak structs for public api * use test helper class * when not using foreach, move cursor reset closer to the loop using read_kv * coverage comments, fix typo in tests decltype * ensure set() does not break with offset * make codacy happy again
4 years ago
Rework settings (#2282) * wip based on early draft. todo benchmarking * fixup eraser, assume keys are unique * fix cursor copy, test removal at random * small benchmark via permutations. todo lambdas and novirtual * fix empty condition / reset * overwrite optimizations, fix move offsets overflows * ...erase using 0xff instead of 0 * test the theory with code, different length kv were bugged * try to check for out-of-bounds writes / reads * style * trying to fix mover again * clarify length, defend against reading len on edge * fix uncommited rewind change * prove space assumptions * more concise traces, fix move condition (agrh!!!) * slightly more internal knowledge (estimates API?) * make sure cursor is only valid within the range * ensure 0 does not blow things * go back up * cursor comments * comments * rewrite writes through cursor * in del too * estimate kv storage requirements, return available size * move raw erase / move into a method, allow ::set to avoid scanning storage twice * refactor naming, use in code * amend storage slicing test * fix crash handler offsets, cleanup configuration * start -> begin * eeprom readiness * dependencies * unused * SPI_FLASH constants for older Core * vtables -> templates * less include dependencies * gcov help, move estimate outside of the class * writer position can never match, use begin + offset * tweak save_crash to trigger only once in a serious crash * doh, header function should be inline * foreach api, tweak structs for public api * use test helper class * when not using foreach, move cursor reset closer to the loop using read_kv * coverage comments, fix typo in tests decltype * ensure set() does not break with offset * make codacy happy again
4 years ago
Rework settings (#2282) * wip based on early draft. todo benchmarking * fixup eraser, assume keys are unique * fix cursor copy, test removal at random * small benchmark via permutations. todo lambdas and novirtual * fix empty condition / reset * overwrite optimizations, fix move offsets overflows * ...erase using 0xff instead of 0 * test the theory with code, different length kv were bugged * try to check for out-of-bounds writes / reads * style * trying to fix mover again * clarify length, defend against reading len on edge * fix uncommited rewind change * prove space assumptions * more concise traces, fix move condition (agrh!!!) * slightly more internal knowledge (estimates API?) * make sure cursor is only valid within the range * ensure 0 does not blow things * go back up * cursor comments * comments * rewrite writes through cursor * in del too * estimate kv storage requirements, return available size * move raw erase / move into a method, allow ::set to avoid scanning storage twice * refactor naming, use in code * amend storage slicing test * fix crash handler offsets, cleanup configuration * start -> begin * eeprom readiness * dependencies * unused * SPI_FLASH constants for older Core * vtables -> templates * less include dependencies * gcov help, move estimate outside of the class * writer position can never match, use begin + offset * tweak save_crash to trigger only once in a serious crash * doh, header function should be inline * foreach api, tweak structs for public api * use test helper class * when not using foreach, move cursor reset closer to the loop using read_kv * coverage comments, fix typo in tests decltype * ensure set() does not break with offset * make codacy happy again
4 years ago
7 years ago
Rework settings (#2282) * wip based on early draft. todo benchmarking * fixup eraser, assume keys are unique * fix cursor copy, test removal at random * small benchmark via permutations. todo lambdas and novirtual * fix empty condition / reset * overwrite optimizations, fix move offsets overflows * ...erase using 0xff instead of 0 * test the theory with code, different length kv were bugged * try to check for out-of-bounds writes / reads * style * trying to fix mover again * clarify length, defend against reading len on edge * fix uncommited rewind change * prove space assumptions * more concise traces, fix move condition (agrh!!!) * slightly more internal knowledge (estimates API?) * make sure cursor is only valid within the range * ensure 0 does not blow things * go back up * cursor comments * comments * rewrite writes through cursor * in del too * estimate kv storage requirements, return available size * move raw erase / move into a method, allow ::set to avoid scanning storage twice * refactor naming, use in code * amend storage slicing test * fix crash handler offsets, cleanup configuration * start -> begin * eeprom readiness * dependencies * unused * SPI_FLASH constants for older Core * vtables -> templates * less include dependencies * gcov help, move estimate outside of the class * writer position can never match, use begin + offset * tweak save_crash to trigger only once in a serious crash * doh, header function should be inline * foreach api, tweak structs for public api * use test helper class * when not using foreach, move cursor reset closer to the loop using read_kv * coverage comments, fix typo in tests decltype * ensure set() does not break with offset * make codacy happy again
4 years ago
Rework settings (#2282) * wip based on early draft. todo benchmarking * fixup eraser, assume keys are unique * fix cursor copy, test removal at random * small benchmark via permutations. todo lambdas and novirtual * fix empty condition / reset * overwrite optimizations, fix move offsets overflows * ...erase using 0xff instead of 0 * test the theory with code, different length kv were bugged * try to check for out-of-bounds writes / reads * style * trying to fix mover again * clarify length, defend against reading len on edge * fix uncommited rewind change * prove space assumptions * more concise traces, fix move condition (agrh!!!) * slightly more internal knowledge (estimates API?) * make sure cursor is only valid within the range * ensure 0 does not blow things * go back up * cursor comments * comments * rewrite writes through cursor * in del too * estimate kv storage requirements, return available size * move raw erase / move into a method, allow ::set to avoid scanning storage twice * refactor naming, use in code * amend storage slicing test * fix crash handler offsets, cleanup configuration * start -> begin * eeprom readiness * dependencies * unused * SPI_FLASH constants for older Core * vtables -> templates * less include dependencies * gcov help, move estimate outside of the class * writer position can never match, use begin + offset * tweak save_crash to trigger only once in a serious crash * doh, header function should be inline * foreach api, tweak structs for public api * use test helper class * when not using foreach, move cursor reset closer to the loop using read_kv * coverage comments, fix typo in tests decltype * ensure set() does not break with offset * make codacy happy again
4 years ago
Rework settings (#2282) * wip based on early draft. todo benchmarking * fixup eraser, assume keys are unique * fix cursor copy, test removal at random * small benchmark via permutations. todo lambdas and novirtual * fix empty condition / reset * overwrite optimizations, fix move offsets overflows * ...erase using 0xff instead of 0 * test the theory with code, different length kv were bugged * try to check for out-of-bounds writes / reads * style * trying to fix mover again * clarify length, defend against reading len on edge * fix uncommited rewind change * prove space assumptions * more concise traces, fix move condition (agrh!!!) * slightly more internal knowledge (estimates API?) * make sure cursor is only valid within the range * ensure 0 does not blow things * go back up * cursor comments * comments * rewrite writes through cursor * in del too * estimate kv storage requirements, return available size * move raw erase / move into a method, allow ::set to avoid scanning storage twice * refactor naming, use in code * amend storage slicing test * fix crash handler offsets, cleanup configuration * start -> begin * eeprom readiness * dependencies * unused * SPI_FLASH constants for older Core * vtables -> templates * less include dependencies * gcov help, move estimate outside of the class * writer position can never match, use begin + offset * tweak save_crash to trigger only once in a serious crash * doh, header function should be inline * foreach api, tweak structs for public api * use test helper class * when not using foreach, move cursor reset closer to the loop using read_kv * coverage comments, fix typo in tests decltype * ensure set() does not break with offset * make codacy happy again
4 years ago
Rework settings (#2282) * wip based on early draft. todo benchmarking * fixup eraser, assume keys are unique * fix cursor copy, test removal at random * small benchmark via permutations. todo lambdas and novirtual * fix empty condition / reset * overwrite optimizations, fix move offsets overflows * ...erase using 0xff instead of 0 * test the theory with code, different length kv were bugged * try to check for out-of-bounds writes / reads * style * trying to fix mover again * clarify length, defend against reading len on edge * fix uncommited rewind change * prove space assumptions * more concise traces, fix move condition (agrh!!!) * slightly more internal knowledge (estimates API?) * make sure cursor is only valid within the range * ensure 0 does not blow things * go back up * cursor comments * comments * rewrite writes through cursor * in del too * estimate kv storage requirements, return available size * move raw erase / move into a method, allow ::set to avoid scanning storage twice * refactor naming, use in code * amend storage slicing test * fix crash handler offsets, cleanup configuration * start -> begin * eeprom readiness * dependencies * unused * SPI_FLASH constants for older Core * vtables -> templates * less include dependencies * gcov help, move estimate outside of the class * writer position can never match, use begin + offset * tweak save_crash to trigger only once in a serious crash * doh, header function should be inline * foreach api, tweak structs for public api * use test helper class * when not using foreach, move cursor reset closer to the loop using read_kv * coverage comments, fix typo in tests decltype * ensure set() does not break with offset * make codacy happy again
4 years ago
Rework settings (#2282) * wip based on early draft. todo benchmarking * fixup eraser, assume keys are unique * fix cursor copy, test removal at random * small benchmark via permutations. todo lambdas and novirtual * fix empty condition / reset * overwrite optimizations, fix move offsets overflows * ...erase using 0xff instead of 0 * test the theory with code, different length kv were bugged * try to check for out-of-bounds writes / reads * style * trying to fix mover again * clarify length, defend against reading len on edge * fix uncommited rewind change * prove space assumptions * more concise traces, fix move condition (agrh!!!) * slightly more internal knowledge (estimates API?) * make sure cursor is only valid within the range * ensure 0 does not blow things * go back up * cursor comments * comments * rewrite writes through cursor * in del too * estimate kv storage requirements, return available size * move raw erase / move into a method, allow ::set to avoid scanning storage twice * refactor naming, use in code * amend storage slicing test * fix crash handler offsets, cleanup configuration * start -> begin * eeprom readiness * dependencies * unused * SPI_FLASH constants for older Core * vtables -> templates * less include dependencies * gcov help, move estimate outside of the class * writer position can never match, use begin + offset * tweak save_crash to trigger only once in a serious crash * doh, header function should be inline * foreach api, tweak structs for public api * use test helper class * when not using foreach, move cursor reset closer to the loop using read_kv * coverage comments, fix typo in tests decltype * ensure set() does not break with offset * make codacy happy again
4 years ago
Rework settings (#2282) * wip based on early draft. todo benchmarking * fixup eraser, assume keys are unique * fix cursor copy, test removal at random * small benchmark via permutations. todo lambdas and novirtual * fix empty condition / reset * overwrite optimizations, fix move offsets overflows * ...erase using 0xff instead of 0 * test the theory with code, different length kv were bugged * try to check for out-of-bounds writes / reads * style * trying to fix mover again * clarify length, defend against reading len on edge * fix uncommited rewind change * prove space assumptions * more concise traces, fix move condition (agrh!!!) * slightly more internal knowledge (estimates API?) * make sure cursor is only valid within the range * ensure 0 does not blow things * go back up * cursor comments * comments * rewrite writes through cursor * in del too * estimate kv storage requirements, return available size * move raw erase / move into a method, allow ::set to avoid scanning storage twice * refactor naming, use in code * amend storage slicing test * fix crash handler offsets, cleanup configuration * start -> begin * eeprom readiness * dependencies * unused * SPI_FLASH constants for older Core * vtables -> templates * less include dependencies * gcov help, move estimate outside of the class * writer position can never match, use begin + offset * tweak save_crash to trigger only once in a serious crash * doh, header function should be inline * foreach api, tweak structs for public api * use test helper class * when not using foreach, move cursor reset closer to the loop using read_kv * coverage comments, fix typo in tests decltype * ensure set() does not break with offset * make codacy happy again
4 years ago
Rework settings (#2282) * wip based on early draft. todo benchmarking * fixup eraser, assume keys are unique * fix cursor copy, test removal at random * small benchmark via permutations. todo lambdas and novirtual * fix empty condition / reset * overwrite optimizations, fix move offsets overflows * ...erase using 0xff instead of 0 * test the theory with code, different length kv were bugged * try to check for out-of-bounds writes / reads * style * trying to fix mover again * clarify length, defend against reading len on edge * fix uncommited rewind change * prove space assumptions * more concise traces, fix move condition (agrh!!!) * slightly more internal knowledge (estimates API?) * make sure cursor is only valid within the range * ensure 0 does not blow things * go back up * cursor comments * comments * rewrite writes through cursor * in del too * estimate kv storage requirements, return available size * move raw erase / move into a method, allow ::set to avoid scanning storage twice * refactor naming, use in code * amend storage slicing test * fix crash handler offsets, cleanup configuration * start -> begin * eeprom readiness * dependencies * unused * SPI_FLASH constants for older Core * vtables -> templates * less include dependencies * gcov help, move estimate outside of the class * writer position can never match, use begin + offset * tweak save_crash to trigger only once in a serious crash * doh, header function should be inline * foreach api, tweak structs for public api * use test helper class * when not using foreach, move cursor reset closer to the loop using read_kv * coverage comments, fix typo in tests decltype * ensure set() does not break with offset * make codacy happy again
4 years ago
Rework settings (#2282) * wip based on early draft. todo benchmarking * fixup eraser, assume keys are unique * fix cursor copy, test removal at random * small benchmark via permutations. todo lambdas and novirtual * fix empty condition / reset * overwrite optimizations, fix move offsets overflows * ...erase using 0xff instead of 0 * test the theory with code, different length kv were bugged * try to check for out-of-bounds writes / reads * style * trying to fix mover again * clarify length, defend against reading len on edge * fix uncommited rewind change * prove space assumptions * more concise traces, fix move condition (agrh!!!) * slightly more internal knowledge (estimates API?) * make sure cursor is only valid within the range * ensure 0 does not blow things * go back up * cursor comments * comments * rewrite writes through cursor * in del too * estimate kv storage requirements, return available size * move raw erase / move into a method, allow ::set to avoid scanning storage twice * refactor naming, use in code * amend storage slicing test * fix crash handler offsets, cleanup configuration * start -> begin * eeprom readiness * dependencies * unused * SPI_FLASH constants for older Core * vtables -> templates * less include dependencies * gcov help, move estimate outside of the class * writer position can never match, use begin + offset * tweak save_crash to trigger only once in a serious crash * doh, header function should be inline * foreach api, tweak structs for public api * use test helper class * when not using foreach, move cursor reset closer to the loop using read_kv * coverage comments, fix typo in tests decltype * ensure set() does not break with offset * make codacy happy again
4 years ago
Rework settings (#2282) * wip based on early draft. todo benchmarking * fixup eraser, assume keys are unique * fix cursor copy, test removal at random * small benchmark via permutations. todo lambdas and novirtual * fix empty condition / reset * overwrite optimizations, fix move offsets overflows * ...erase using 0xff instead of 0 * test the theory with code, different length kv were bugged * try to check for out-of-bounds writes / reads * style * trying to fix mover again * clarify length, defend against reading len on edge * fix uncommited rewind change * prove space assumptions * more concise traces, fix move condition (agrh!!!) * slightly more internal knowledge (estimates API?) * make sure cursor is only valid within the range * ensure 0 does not blow things * go back up * cursor comments * comments * rewrite writes through cursor * in del too * estimate kv storage requirements, return available size * move raw erase / move into a method, allow ::set to avoid scanning storage twice * refactor naming, use in code * amend storage slicing test * fix crash handler offsets, cleanup configuration * start -> begin * eeprom readiness * dependencies * unused * SPI_FLASH constants for older Core * vtables -> templates * less include dependencies * gcov help, move estimate outside of the class * writer position can never match, use begin + offset * tweak save_crash to trigger only once in a serious crash * doh, header function should be inline * foreach api, tweak structs for public api * use test helper class * when not using foreach, move cursor reset closer to the loop using read_kv * coverage comments, fix typo in tests decltype * ensure set() does not break with offset * make codacy happy again
4 years ago
6 years ago
Terminal: change command-line parser (#2247) Change the underlying command line handling: - switch to a custom parser, inspired by redis / sds - update terminalRegisterCommand signature, pass only bare minimum - clean-up `help` & `commands`. update settings `set`, `get` and `del` - allow our custom test suite to run command-line tests - clean-up Stream IO to allow us to print large things into debug stream (for example, `eeprom.dump`) - send parsing errors to the debug log As a proof of concept, introduce `TERMINAL_MQTT_SUPPORT` and `TERMINAL_WEB_API_SUPPORT` - MQTT subscribes to the `<root>/cmd/set` and sends response to the `<root>/cmd`. We can't output too much, as we don't have any large-send API. - Web API listens to the `/api/cmd?apikey=...&line=...` (or PUT, params inside the body). This one is intended as a possible replacement of the `API_SUPPORT`. Internals introduce a 'task' around the AsyncWebServerRequest object that will simulate what WiFiClient does and push data into it continuously, switching between CONT and SYS. Both are experimental. We only accept a single command and not every command is updated to use Print `ctx.output` object. We are also somewhat limited by the Print / Stream overall, perhaps I am overestimating the usefulness of Arduino compatibility to such an extent :) Web API handler can also sometimes show only part of the result, whenever the command tries to yield() by itself waiting for something. Perhaps we would need to create a custom request handler for that specific use-case.
4 years ago
Rework settings (#2282) * wip based on early draft. todo benchmarking * fixup eraser, assume keys are unique * fix cursor copy, test removal at random * small benchmark via permutations. todo lambdas and novirtual * fix empty condition / reset * overwrite optimizations, fix move offsets overflows * ...erase using 0xff instead of 0 * test the theory with code, different length kv were bugged * try to check for out-of-bounds writes / reads * style * trying to fix mover again * clarify length, defend against reading len on edge * fix uncommited rewind change * prove space assumptions * more concise traces, fix move condition (agrh!!!) * slightly more internal knowledge (estimates API?) * make sure cursor is only valid within the range * ensure 0 does not blow things * go back up * cursor comments * comments * rewrite writes through cursor * in del too * estimate kv storage requirements, return available size * move raw erase / move into a method, allow ::set to avoid scanning storage twice * refactor naming, use in code * amend storage slicing test * fix crash handler offsets, cleanup configuration * start -> begin * eeprom readiness * dependencies * unused * SPI_FLASH constants for older Core * vtables -> templates * less include dependencies * gcov help, move estimate outside of the class * writer position can never match, use begin + offset * tweak save_crash to trigger only once in a serious crash * doh, header function should be inline * foreach api, tweak structs for public api * use test helper class * when not using foreach, move cursor reset closer to the loop using read_kv * coverage comments, fix typo in tests decltype * ensure set() does not break with offset * make codacy happy again
4 years ago
Terminal: change command-line parser (#2247) Change the underlying command line handling: - switch to a custom parser, inspired by redis / sds - update terminalRegisterCommand signature, pass only bare minimum - clean-up `help` & `commands`. update settings `set`, `get` and `del` - allow our custom test suite to run command-line tests - clean-up Stream IO to allow us to print large things into debug stream (for example, `eeprom.dump`) - send parsing errors to the debug log As a proof of concept, introduce `TERMINAL_MQTT_SUPPORT` and `TERMINAL_WEB_API_SUPPORT` - MQTT subscribes to the `<root>/cmd/set` and sends response to the `<root>/cmd`. We can't output too much, as we don't have any large-send API. - Web API listens to the `/api/cmd?apikey=...&line=...` (or PUT, params inside the body). This one is intended as a possible replacement of the `API_SUPPORT`. Internals introduce a 'task' around the AsyncWebServerRequest object that will simulate what WiFiClient does and push data into it continuously, switching between CONT and SYS. Both are experimental. We only accept a single command and not every command is updated to use Print `ctx.output` object. We are also somewhat limited by the Print / Stream overall, perhaps I am overestimating the usefulness of Arduino compatibility to such an extent :) Web API handler can also sometimes show only part of the result, whenever the command tries to yield() by itself waiting for something. Perhaps we would need to create a custom request handler for that specific use-case.
4 years ago
Rework settings (#2282) * wip based on early draft. todo benchmarking * fixup eraser, assume keys are unique * fix cursor copy, test removal at random * small benchmark via permutations. todo lambdas and novirtual * fix empty condition / reset * overwrite optimizations, fix move offsets overflows * ...erase using 0xff instead of 0 * test the theory with code, different length kv were bugged * try to check for out-of-bounds writes / reads * style * trying to fix mover again * clarify length, defend against reading len on edge * fix uncommited rewind change * prove space assumptions * more concise traces, fix move condition (agrh!!!) * slightly more internal knowledge (estimates API?) * make sure cursor is only valid within the range * ensure 0 does not blow things * go back up * cursor comments * comments * rewrite writes through cursor * in del too * estimate kv storage requirements, return available size * move raw erase / move into a method, allow ::set to avoid scanning storage twice * refactor naming, use in code * amend storage slicing test * fix crash handler offsets, cleanup configuration * start -> begin * eeprom readiness * dependencies * unused * SPI_FLASH constants for older Core * vtables -> templates * less include dependencies * gcov help, move estimate outside of the class * writer position can never match, use begin + offset * tweak save_crash to trigger only once in a serious crash * doh, header function should be inline * foreach api, tweak structs for public api * use test helper class * when not using foreach, move cursor reset closer to the loop using read_kv * coverage comments, fix typo in tests decltype * ensure set() does not break with offset * make codacy happy again
4 years ago
Terminal: change command-line parser (#2247) Change the underlying command line handling: - switch to a custom parser, inspired by redis / sds - update terminalRegisterCommand signature, pass only bare minimum - clean-up `help` & `commands`. update settings `set`, `get` and `del` - allow our custom test suite to run command-line tests - clean-up Stream IO to allow us to print large things into debug stream (for example, `eeprom.dump`) - send parsing errors to the debug log As a proof of concept, introduce `TERMINAL_MQTT_SUPPORT` and `TERMINAL_WEB_API_SUPPORT` - MQTT subscribes to the `<root>/cmd/set` and sends response to the `<root>/cmd`. We can't output too much, as we don't have any large-send API. - Web API listens to the `/api/cmd?apikey=...&line=...` (or PUT, params inside the body). This one is intended as a possible replacement of the `API_SUPPORT`. Internals introduce a 'task' around the AsyncWebServerRequest object that will simulate what WiFiClient does and push data into it continuously, switching between CONT and SYS. Both are experimental. We only accept a single command and not every command is updated to use Print `ctx.output` object. We are also somewhat limited by the Print / Stream overall, perhaps I am overestimating the usefulness of Arduino compatibility to such an extent :) Web API handler can also sometimes show only part of the result, whenever the command tries to yield() by itself waiting for something. Perhaps we would need to create a custom request handler for that specific use-case.
4 years ago
Rework settings (#2282) * wip based on early draft. todo benchmarking * fixup eraser, assume keys are unique * fix cursor copy, test removal at random * small benchmark via permutations. todo lambdas and novirtual * fix empty condition / reset * overwrite optimizations, fix move offsets overflows * ...erase using 0xff instead of 0 * test the theory with code, different length kv were bugged * try to check for out-of-bounds writes / reads * style * trying to fix mover again * clarify length, defend against reading len on edge * fix uncommited rewind change * prove space assumptions * more concise traces, fix move condition (agrh!!!) * slightly more internal knowledge (estimates API?) * make sure cursor is only valid within the range * ensure 0 does not blow things * go back up * cursor comments * comments * rewrite writes through cursor * in del too * estimate kv storage requirements, return available size * move raw erase / move into a method, allow ::set to avoid scanning storage twice * refactor naming, use in code * amend storage slicing test * fix crash handler offsets, cleanup configuration * start -> begin * eeprom readiness * dependencies * unused * SPI_FLASH constants for older Core * vtables -> templates * less include dependencies * gcov help, move estimate outside of the class * writer position can never match, use begin + offset * tweak save_crash to trigger only once in a serious crash * doh, header function should be inline * foreach api, tweak structs for public api * use test helper class * when not using foreach, move cursor reset closer to the loop using read_kv * coverage comments, fix typo in tests decltype * ensure set() does not break with offset * make codacy happy again
4 years ago
Terminal: change command-line parser (#2247) Change the underlying command line handling: - switch to a custom parser, inspired by redis / sds - update terminalRegisterCommand signature, pass only bare minimum - clean-up `help` & `commands`. update settings `set`, `get` and `del` - allow our custom test suite to run command-line tests - clean-up Stream IO to allow us to print large things into debug stream (for example, `eeprom.dump`) - send parsing errors to the debug log As a proof of concept, introduce `TERMINAL_MQTT_SUPPORT` and `TERMINAL_WEB_API_SUPPORT` - MQTT subscribes to the `<root>/cmd/set` and sends response to the `<root>/cmd`. We can't output too much, as we don't have any large-send API. - Web API listens to the `/api/cmd?apikey=...&line=...` (or PUT, params inside the body). This one is intended as a possible replacement of the `API_SUPPORT`. Internals introduce a 'task' around the AsyncWebServerRequest object that will simulate what WiFiClient does and push data into it continuously, switching between CONT and SYS. Both are experimental. We only accept a single command and not every command is updated to use Print `ctx.output` object. We are also somewhat limited by the Print / Stream overall, perhaps I am overestimating the usefulness of Arduino compatibility to such an extent :) Web API handler can also sometimes show only part of the result, whenever the command tries to yield() by itself waiting for something. Perhaps we would need to create a custom request handler for that specific use-case.
4 years ago
Rework settings (#2282) * wip based on early draft. todo benchmarking * fixup eraser, assume keys are unique * fix cursor copy, test removal at random * small benchmark via permutations. todo lambdas and novirtual * fix empty condition / reset * overwrite optimizations, fix move offsets overflows * ...erase using 0xff instead of 0 * test the theory with code, different length kv were bugged * try to check for out-of-bounds writes / reads * style * trying to fix mover again * clarify length, defend against reading len on edge * fix uncommited rewind change * prove space assumptions * more concise traces, fix move condition (agrh!!!) * slightly more internal knowledge (estimates API?) * make sure cursor is only valid within the range * ensure 0 does not blow things * go back up * cursor comments * comments * rewrite writes through cursor * in del too * estimate kv storage requirements, return available size * move raw erase / move into a method, allow ::set to avoid scanning storage twice * refactor naming, use in code * amend storage slicing test * fix crash handler offsets, cleanup configuration * start -> begin * eeprom readiness * dependencies * unused * SPI_FLASH constants for older Core * vtables -> templates * less include dependencies * gcov help, move estimate outside of the class * writer position can never match, use begin + offset * tweak save_crash to trigger only once in a serious crash * doh, header function should be inline * foreach api, tweak structs for public api * use test helper class * when not using foreach, move cursor reset closer to the loop using read_kv * coverage comments, fix typo in tests decltype * ensure set() does not break with offset * make codacy happy again
4 years ago
Terminal: change command-line parser (#2247) Change the underlying command line handling: - switch to a custom parser, inspired by redis / sds - update terminalRegisterCommand signature, pass only bare minimum - clean-up `help` & `commands`. update settings `set`, `get` and `del` - allow our custom test suite to run command-line tests - clean-up Stream IO to allow us to print large things into debug stream (for example, `eeprom.dump`) - send parsing errors to the debug log As a proof of concept, introduce `TERMINAL_MQTT_SUPPORT` and `TERMINAL_WEB_API_SUPPORT` - MQTT subscribes to the `<root>/cmd/set` and sends response to the `<root>/cmd`. We can't output too much, as we don't have any large-send API. - Web API listens to the `/api/cmd?apikey=...&line=...` (or PUT, params inside the body). This one is intended as a possible replacement of the `API_SUPPORT`. Internals introduce a 'task' around the AsyncWebServerRequest object that will simulate what WiFiClient does and push data into it continuously, switching between CONT and SYS. Both are experimental. We only accept a single command and not every command is updated to use Print `ctx.output` object. We are also somewhat limited by the Print / Stream overall, perhaps I am overestimating the usefulness of Arduino compatibility to such an extent :) Web API handler can also sometimes show only part of the result, whenever the command tries to yield() by itself waiting for something. Perhaps we would need to create a custom request handler for that specific use-case.
4 years ago
Rework settings (#2282) * wip based on early draft. todo benchmarking * fixup eraser, assume keys are unique * fix cursor copy, test removal at random * small benchmark via permutations. todo lambdas and novirtual * fix empty condition / reset * overwrite optimizations, fix move offsets overflows * ...erase using 0xff instead of 0 * test the theory with code, different length kv were bugged * try to check for out-of-bounds writes / reads * style * trying to fix mover again * clarify length, defend against reading len on edge * fix uncommited rewind change * prove space assumptions * more concise traces, fix move condition (agrh!!!) * slightly more internal knowledge (estimates API?) * make sure cursor is only valid within the range * ensure 0 does not blow things * go back up * cursor comments * comments * rewrite writes through cursor * in del too * estimate kv storage requirements, return available size * move raw erase / move into a method, allow ::set to avoid scanning storage twice * refactor naming, use in code * amend storage slicing test * fix crash handler offsets, cleanup configuration * start -> begin * eeprom readiness * dependencies * unused * SPI_FLASH constants for older Core * vtables -> templates * less include dependencies * gcov help, move estimate outside of the class * writer position can never match, use begin + offset * tweak save_crash to trigger only once in a serious crash * doh, header function should be inline * foreach api, tweak structs for public api * use test helper class * when not using foreach, move cursor reset closer to the loop using read_kv * coverage comments, fix typo in tests decltype * ensure set() does not break with offset * make codacy happy again
4 years ago
Terminal: change command-line parser (#2247) Change the underlying command line handling: - switch to a custom parser, inspired by redis / sds - update terminalRegisterCommand signature, pass only bare minimum - clean-up `help` & `commands`. update settings `set`, `get` and `del` - allow our custom test suite to run command-line tests - clean-up Stream IO to allow us to print large things into debug stream (for example, `eeprom.dump`) - send parsing errors to the debug log As a proof of concept, introduce `TERMINAL_MQTT_SUPPORT` and `TERMINAL_WEB_API_SUPPORT` - MQTT subscribes to the `<root>/cmd/set` and sends response to the `<root>/cmd`. We can't output too much, as we don't have any large-send API. - Web API listens to the `/api/cmd?apikey=...&line=...` (or PUT, params inside the body). This one is intended as a possible replacement of the `API_SUPPORT`. Internals introduce a 'task' around the AsyncWebServerRequest object that will simulate what WiFiClient does and push data into it continuously, switching between CONT and SYS. Both are experimental. We only accept a single command and not every command is updated to use Print `ctx.output` object. We are also somewhat limited by the Print / Stream overall, perhaps I am overestimating the usefulness of Arduino compatibility to such an extent :) Web API handler can also sometimes show only part of the result, whenever the command tries to yield() by itself waiting for something. Perhaps we would need to create a custom request handler for that specific use-case.
4 years ago
Rework settings (#2282) * wip based on early draft. todo benchmarking * fixup eraser, assume keys are unique * fix cursor copy, test removal at random * small benchmark via permutations. todo lambdas and novirtual * fix empty condition / reset * overwrite optimizations, fix move offsets overflows * ...erase using 0xff instead of 0 * test the theory with code, different length kv were bugged * try to check for out-of-bounds writes / reads * style * trying to fix mover again * clarify length, defend against reading len on edge * fix uncommited rewind change * prove space assumptions * more concise traces, fix move condition (agrh!!!) * slightly more internal knowledge (estimates API?) * make sure cursor is only valid within the range * ensure 0 does not blow things * go back up * cursor comments * comments * rewrite writes through cursor * in del too * estimate kv storage requirements, return available size * move raw erase / move into a method, allow ::set to avoid scanning storage twice * refactor naming, use in code * amend storage slicing test * fix crash handler offsets, cleanup configuration * start -> begin * eeprom readiness * dependencies * unused * SPI_FLASH constants for older Core * vtables -> templates * less include dependencies * gcov help, move estimate outside of the class * writer position can never match, use begin + offset * tweak save_crash to trigger only once in a serious crash * doh, header function should be inline * foreach api, tweak structs for public api * use test helper class * when not using foreach, move cursor reset closer to the loop using read_kv * coverage comments, fix typo in tests decltype * ensure set() does not break with offset * make codacy happy again
4 years ago
Terminal: change command-line parser (#2247) Change the underlying command line handling: - switch to a custom parser, inspired by redis / sds - update terminalRegisterCommand signature, pass only bare minimum - clean-up `help` & `commands`. update settings `set`, `get` and `del` - allow our custom test suite to run command-line tests - clean-up Stream IO to allow us to print large things into debug stream (for example, `eeprom.dump`) - send parsing errors to the debug log As a proof of concept, introduce `TERMINAL_MQTT_SUPPORT` and `TERMINAL_WEB_API_SUPPORT` - MQTT subscribes to the `<root>/cmd/set` and sends response to the `<root>/cmd`. We can't output too much, as we don't have any large-send API. - Web API listens to the `/api/cmd?apikey=...&line=...` (or PUT, params inside the body). This one is intended as a possible replacement of the `API_SUPPORT`. Internals introduce a 'task' around the AsyncWebServerRequest object that will simulate what WiFiClient does and push data into it continuously, switching between CONT and SYS. Both are experimental. We only accept a single command and not every command is updated to use Print `ctx.output` object. We are also somewhat limited by the Print / Stream overall, perhaps I am overestimating the usefulness of Arduino compatibility to such an extent :) Web API handler can also sometimes show only part of the result, whenever the command tries to yield() by itself waiting for something. Perhaps we would need to create a custom request handler for that specific use-case.
4 years ago
  1. /*
  2. SETTINGS MODULE
  3. Copyright (C) 2016-2019 by Xose Pérez <xose dot perez at gmail dot com>
  4. */
  5. #include "espurna.h"
  6. #include "crash.h"
  7. #include "terminal.h"
  8. #include "storage_eeprom.h"
  9. #include <algorithm>
  10. #include <vector>
  11. #include <cstdlib>
  12. #include <ArduinoJson.h>
  13. BrokerBind(ConfigBroker);
  14. // -----------------------------------------------------------------------------
  15. namespace settings {
  16. // Depending on features enabled, we may end up with different left boundary
  17. // Settings are written right-to-left, so we only have issues when there are a lot of key-values
  18. // XXX: slightly hacky, because we EEPROMr.length() is 0 before we enter setup() code
  19. kvs_type kv_store(
  20. EepromStorage{},
  21. #if DEBUG_SUPPORT
  22. EepromReservedSize + CrashReservedSize,
  23. #else
  24. EepromReservedSize,
  25. #endif
  26. EepromSize
  27. );
  28. } // namespace settings
  29. size_t settingsSize() {
  30. return settings::kv_store.size() - settings::kv_store.available();
  31. }
  32. // --------------------------------------------------------------------------
  33. namespace settings {
  34. namespace internal {
  35. uint32_t u32fromString(const String& string, int base) {
  36. const char *ptr = string.c_str();
  37. char *value_endptr = nullptr;
  38. // invalidate the whole string when invalid chars are detected
  39. const auto value = strtoul(ptr, &value_endptr, base);
  40. if (value_endptr == ptr || value_endptr[0] != '\0') {
  41. return 0;
  42. }
  43. return value;
  44. }
  45. // --------------------------------------------------------------------------
  46. template <>
  47. float convert(const String& value) {
  48. return atof(value.c_str());
  49. }
  50. template <>
  51. double convert(const String& value) {
  52. return atof(value.c_str());
  53. }
  54. template <>
  55. int convert(const String& value) {
  56. return value.toInt();
  57. }
  58. template <>
  59. long convert(const String& value) {
  60. return value.toInt();
  61. }
  62. template <>
  63. bool convert(const String& value) {
  64. return convert<int>(value) == 1;
  65. }
  66. template <>
  67. unsigned long convert(const String& value) {
  68. if (!value.length()) {
  69. return 0;
  70. }
  71. int base = 10;
  72. if (value.length() > 2) {
  73. if (value.startsWith("0b")) {
  74. base = 2;
  75. } else if (value.startsWith("0o")) {
  76. base = 8;
  77. } else if (value.startsWith("0x")) {
  78. base = 16;
  79. }
  80. }
  81. return u32fromString((base == 10) ? value : value.substring(2), base);
  82. }
  83. template <>
  84. unsigned int convert(const String& value) {
  85. return convert<unsigned long>(value);
  86. }
  87. template <>
  88. unsigned short convert(const String& value) {
  89. return convert<unsigned long>(value);
  90. }
  91. template <>
  92. unsigned char convert(const String& value) {
  93. return convert<unsigned long>(value);
  94. }
  95. } // namespace settings::internal
  96. } // namespace settings
  97. // -----------------------------------------------------------------------------
  98. /*
  99. struct SettingsKeys {
  100. struct iterator {
  101. iterator(size_t total) :
  102. total(total)
  103. {}
  104. iterator& operator++() {
  105. if (total && (current_index < (total - 1))) {
  106. ++current_index
  107. current_value = settingsKeyName(current_index);
  108. return *this;
  109. }
  110. return end();
  111. }
  112. iterator operator++(int) {
  113. iterator val = *this;
  114. ++(*this);
  115. return val;
  116. }
  117. operator String() {
  118. return (current_index < total) ? current_value : empty_value;
  119. }
  120. bool operator ==(iterator& const other) const {
  121. return (total == other.total) && (current_index == other.current_index);
  122. }
  123. bool operator !=(iterator& const other) const {
  124. return !(*this == other);
  125. }
  126. using difference_type = size_t;
  127. using value_type = size_t;
  128. using pointer = const size_t*;
  129. using reference = const size_t&;
  130. using iterator_category = std::forward_iterator_tag;
  131. const size_t total;
  132. String empty_value;
  133. String current_value;
  134. size_t current_index = 0;
  135. };
  136. iterator begin() {
  137. return iterator {total};
  138. }
  139. iterator end() {
  140. return iterator {0};
  141. }
  142. };
  143. */
  144. // Note: we prefer things sorted via this function, not kv_store.keys() directly
  145. std::vector<String> settingsKeys() {
  146. auto keys = settings::kv_store.keys();
  147. std::sort(keys.begin(), keys.end(), [](const String& rhs, const String& lhs) -> bool {
  148. return lhs.compareTo(rhs) > 0;
  149. });
  150. return keys;
  151. }
  152. static std::vector<settings_key_match_t> _settings_matchers;
  153. void settingsRegisterDefaults(const settings_key_match_t& matcher) {
  154. _settings_matchers.push_back(matcher);
  155. }
  156. String settingsQueryDefaults(const String& key) {
  157. for (auto& matcher : _settings_matchers) {
  158. if (matcher.match(key.c_str())) {
  159. return matcher.key(key);
  160. }
  161. }
  162. return String();
  163. }
  164. // -----------------------------------------------------------------------------
  165. // Key-value API
  166. // -----------------------------------------------------------------------------
  167. String settings_key_t::toString() const {
  168. if (_index < 0) {
  169. return _value;
  170. } else {
  171. return _value + _index;
  172. }
  173. }
  174. settings_move_key_t _moveKeys(const String& from, const String& to, unsigned char index) {
  175. return settings_move_key_t {{from, index}, {to, index}};
  176. }
  177. void moveSetting(const String& from, const String& to) {
  178. const auto value = getSetting(from);
  179. if (value.length() > 0) setSetting(to, value);
  180. delSetting(from);
  181. }
  182. void moveSetting(const String& from, const String& to, unsigned char index) {
  183. const auto keys = _moveKeys(from, to, index);
  184. const auto value = getSetting(keys.first);
  185. if (value.length() > 0) setSetting(keys.second, value);
  186. delSetting(keys.first);
  187. }
  188. void moveSettings(const String& from, const String& to) {
  189. unsigned char index = 0;
  190. while (index < 100) {
  191. const auto keys = _moveKeys(from, to, index);
  192. const auto value = getSetting(keys.first);
  193. if (value.length() == 0) break;
  194. setSetting(keys.second, value);
  195. delSetting(keys.first);
  196. ++index;
  197. }
  198. }
  199. template<>
  200. String getSetting(const settings_key_t& key, String defaultValue) {
  201. auto result = settings::kv_store.get(key.toString());
  202. if (!result) {
  203. return defaultValue;
  204. }
  205. return result.value;
  206. }
  207. template
  208. bool getSetting(const settings_key_t& key, bool defaultValue);
  209. template
  210. int getSetting(const settings_key_t& key, int defaultValue);
  211. template
  212. long getSetting(const settings_key_t& key, long defaultValue);
  213. template
  214. unsigned char getSetting(const settings_key_t& key, unsigned char defaultValue);
  215. template
  216. unsigned short getSetting(const settings_key_t& key, unsigned short defaultValue);
  217. template
  218. unsigned int getSetting(const settings_key_t& key, unsigned int defaultValue);
  219. template
  220. unsigned long getSetting(const settings_key_t& key, unsigned long defaultValue);
  221. template
  222. float getSetting(const settings_key_t& key, float defaultValue);
  223. template
  224. double getSetting(const settings_key_t& key, double defaultValue);
  225. String getSetting(const settings_key_t& key) {
  226. static const String defaultValue("");
  227. return getSetting(key, defaultValue);
  228. }
  229. String getSetting(const settings_key_t& key, const char* defaultValue) {
  230. return getSetting(key, String(defaultValue));
  231. }
  232. String getSetting(const settings_key_t& key, const __FlashStringHelper* defaultValue) {
  233. return getSetting(key, String(defaultValue));
  234. }
  235. template<>
  236. bool setSetting(const settings_key_t& key, const String& value) {
  237. return settings::kv_store.set(key.toString(), value);
  238. }
  239. bool delSetting(const settings_key_t& key) {
  240. return settings::kv_store.del(key.toString());
  241. }
  242. bool hasSetting(const settings_key_t& key) {
  243. return settings::kv_store.has(key.toString());
  244. }
  245. void saveSettings() {
  246. #if not SETTINGS_AUTOSAVE
  247. eepromCommit();
  248. #endif
  249. }
  250. void resetSettings() {
  251. auto* ptr = EEPROMr.getDataPtr();
  252. std::fill(ptr + EepromReservedSize, ptr + EepromSize, 0xFF);
  253. EEPROMr.commit();
  254. }
  255. // -----------------------------------------------------------------------------
  256. // API
  257. // -----------------------------------------------------------------------------
  258. bool settingsRestoreJson(JsonObject& data) {
  259. // Note: we try to match what /config generates, expect {"app":"ESPURNA",...}
  260. const char* app = data["app"];
  261. if (!app || strcmp(app, APP_NAME) != 0) {
  262. DEBUG_MSG_P(PSTR("[SETTING] Wrong or missing 'app' key\n"));
  263. return false;
  264. }
  265. // .../config will add this key, but it is optional
  266. if (data["backup"].as<bool>()) {
  267. resetSettings();
  268. }
  269. // These three are just metadata, no need to actually store them
  270. for (auto element : data) {
  271. if (strcmp(element.key, "app") == 0) continue;
  272. if (strcmp(element.key, "version") == 0) continue;
  273. if (strcmp(element.key, "backup") == 0) continue;
  274. setSetting(element.key, element.value.as<char*>());
  275. }
  276. saveSettings();
  277. DEBUG_MSG_P(PSTR("[SETTINGS] Settings restored successfully\n"));
  278. return true;
  279. }
  280. bool settingsRestoreJson(char* json_string, size_t json_buffer_size) {
  281. // XXX: as of right now, arduinojson cannot trigger callbacks for each key individually
  282. // Manually separating kv pairs can allow to parse only a small chunk, since we know that there is only string type used (even with bools / ints). Can be problematic when parsing data that was not generated by us.
  283. // Current parsing method is limited only by keys (~sizeof(uintptr_t) bytes per key, data is not copied when string is non-const)
  284. DynamicJsonBuffer jsonBuffer(json_buffer_size);
  285. JsonObject& root = jsonBuffer.parseObject((char *) json_string);
  286. if (!root.success()) {
  287. DEBUG_MSG_P(PSTR("[SETTINGS] JSON parsing error\n"));
  288. return false;
  289. }
  290. return settingsRestoreJson(root);
  291. }
  292. void settingsGetJson(JsonObject& root) {
  293. // Get sorted list of keys
  294. auto keys = settingsKeys();
  295. // Add the key-values to the json object
  296. for (unsigned int i=0; i<keys.size(); i++) {
  297. String value = getSetting(keys[i]);
  298. root[keys[i]] = value;
  299. }
  300. }
  301. void settingsProcessConfig(const settings_cfg_list_t& config, settings_filter_t filter) {
  302. for (auto& entry : config) {
  303. String value = getSetting(entry.key, entry.default_value);
  304. if (filter) {
  305. value = filter(value);
  306. }
  307. if (value.equals(entry.setting)) continue;
  308. entry.setting = std::move(value);
  309. }
  310. }
  311. // -----------------------------------------------------------------------------
  312. // Initialization
  313. // -----------------------------------------------------------------------------
  314. void settingsSetup() {
  315. terminalRegisterCommand(F("CONFIG"), [](const terminal::CommandContext& ctx) {
  316. // TODO: enough of a buffer?
  317. DynamicJsonBuffer jsonBuffer(1024);
  318. JsonObject& root = jsonBuffer.createObject();
  319. settingsGetJson(root);
  320. root.prettyPrintTo(ctx.output);
  321. terminalOK(ctx);
  322. });
  323. terminalRegisterCommand(F("KEYS"), [](const terminal::CommandContext& ctx) {
  324. auto keys = settingsKeys();
  325. ctx.output.println(F("Current settings:"));
  326. for (unsigned int i=0; i<keys.size(); i++) {
  327. const auto value = getSetting(keys[i]);
  328. ctx.output.printf("> %s => \"%s\"\n", (keys[i]).c_str(), value.c_str());
  329. }
  330. auto available [[gnu::unused]] = settings::kv_store.available();
  331. ctx.output.printf("Number of keys: %u\n", keys.size());
  332. ctx.output.printf("Available: %u bytes (%u%%)\n", available, (100 * available) / settings::kv_store.size());
  333. terminalOK(ctx);
  334. });
  335. terminalRegisterCommand(F("DEL"), [](const terminal::CommandContext& ctx) {
  336. if (ctx.argc != 2) {
  337. terminalError(ctx, F("del <key> [<key>...]"));
  338. return;
  339. }
  340. int result = 0;
  341. for (auto it = (ctx.argv.begin() + 1); it != ctx.argv.end(); ++it) {
  342. result += settings::kv_store.del(*it);
  343. }
  344. if (result) {
  345. terminalOK(ctx);
  346. } else {
  347. terminalError(ctx, F("no keys were removed"));
  348. }
  349. });
  350. terminalRegisterCommand(F("SET"), [](const terminal::CommandContext& ctx) {
  351. if (ctx.argc != 3) {
  352. terminalError(ctx, F("set <key> <value>"));
  353. return;
  354. }
  355. if (settings::kv_store.set(ctx.argv[1], ctx.argv[2])) {
  356. terminalOK(ctx);
  357. return;
  358. }
  359. terminalError(ctx, F("could not set the key"));
  360. });
  361. terminalRegisterCommand(F("GET"), [](const terminal::CommandContext& ctx) {
  362. if (ctx.argc < 2) {
  363. terminalError(ctx, F("Wrong arguments"));
  364. return;
  365. }
  366. for (auto it = (ctx.argv.begin() + 1); it != ctx.argv.end(); ++it) {
  367. const String& key = *it;
  368. auto result = settings::kv_store.get(key);
  369. if (!result) {
  370. const auto maybeDefault = settingsQueryDefaults(key);
  371. if (maybeDefault.length()) {
  372. ctx.output.printf("> %s => %s (default)\n", key.c_str(), maybeDefault.c_str());
  373. } else {
  374. ctx.output.printf("> %s =>\n", key.c_str());
  375. }
  376. continue;
  377. }
  378. ctx.output.printf("> %s => \"%s\"\n", key.c_str(), result.value.c_str());
  379. }
  380. terminalOK(ctx);
  381. });
  382. terminalRegisterCommand(F("RELOAD"), [](const terminal::CommandContext&) {
  383. espurnaReload();
  384. terminalOK();
  385. });
  386. terminalRegisterCommand(F("FACTORY.RESET"), [](const terminal::CommandContext&) {
  387. resetSettings();
  388. terminalOK();
  389. });
  390. #if not SETTINGS_AUTOSAVE
  391. terminalRegisterCommand(F("SAVE"), [](const terminal::CommandContext&) {
  392. eepromCommit();
  393. terminalOK();
  394. });
  395. #endif
  396. }