Repository navigation
Releases
MangoX2 + Pico W Runtime v0.2.9
MangoX2 Runtime — Pico W v0.2.9
中文
正式版(Stable)。適用於 MangoX2 Runtime — Pico W。
主要更新
新增 Pot 與感測器 Change Callback Student API,保留既有 Motor/Drive 功能。
下載與 SHA-256
MangoBox_Runtime_v0.2.9_MangoX2_PicoW.uf2— SHA-256ee21000a4b32cc1ef6657ec629cfaa6925813d9247adc0d1dedede0eb00e9949
安裝方式
將對應 UF2 以 BOOTSEL 模式複製至指定 Pico 板;不同 Target 不可混用。
驗證範圍
- 正式版原始碼與檔案版本、SHA-256 由本機建置流程核對。
- Target-specific full hardware regression not performed; family representative hardware tests completed
- 本次依專案負責人的決定採本機驗證作為發布依據,未以 GitHub Actions 作為封版門檻。
English
Stable release for MangoX2 Runtime — Pico W.
Changes
新增 Pot 與感測器 Change Callback Student API,保留既有 Motor/Drive 功能。
Downloads and SHA-256
MangoBox_Runtime_v0.2.9_MangoX2_PicoW.uf2— SHA-256ee21000a4b32cc1ef6657ec629cfaa6925813d9247adc0d1dedede0eb00e9949
Installation
Copy the correct target UF2 to the Pico BOOTSEL USB volume. Do not interchange target images.
Validation scope
- Exact stable output files and checksums were verified by local build tools.
- Target-specific full hardware regression not performed; family representative hardware tests completed
- CI was explicitly waived as a release gate. This release does not claim green GitHub Actions.
MangoX2 + Pico 2 W Runtime v0.2.9
MangoX2 Runtime — Pico 2 W v0.2.9
中文
正式版(Stable)。適用於 MangoX2 Runtime — Pico 2 W。
主要更新
新增 Pot 與感測器 Change Callback Student API,保留既有 Motor/Drive 功能。
下載與 SHA-256
MangoBox_Runtime_v0.2.9_MangoX2_Pico2W.uf2— SHA-25649ef1bf7d63806546b4e140273b5e6952397838f1a2f1c7c4e16d8f96a885f99
安裝方式
將對應 UF2 以 BOOTSEL 模式複製至指定 Pico 板;不同 Target 不可混用。
驗證範圍
- 正式版原始碼與檔案版本、SHA-256 由本機建置流程核對。
- Target-specific full hardware regression not performed; family representative hardware tests completed
- 本次依專案負責人的決定採本機驗證作為發布依據,未以 GitHub Actions 作為封版門檻。
English
Stable release for MangoX2 Runtime — Pico 2 W.
Changes
新增 Pot 與感測器 Change Callback Student API,保留既有 Motor/Drive 功能。
Downloads and SHA-256
MangoBox_Runtime_v0.2.9_MangoX2_Pico2W.uf2— SHA-25649ef1bf7d63806546b4e140273b5e6952397838f1a2f1c7c4e16d8f96a885f99
Installation
Copy the correct target UF2 to the Pico BOOTSEL USB volume. Do not interchange target images.
Validation scope
- Exact stable output files and checksums were verified by local build tools.
- Target-specific full hardware regression not performed; family representative hardware tests completed
- CI was explicitly waived as a release gate. This release does not claim green GitHub Actions.
MangoX2 + Pico Runtime v0.2.9
MangoX2 Runtime — Pico v0.2.9
中文
正式版(Stable)。適用於 MangoX2 Runtime — Pico。
主要更新
新增 Pot 與感測器 Change Callback Student API,保留既有 Motor/Drive 功能。
下載與 SHA-256
MangoBox_Runtime_v0.2.9_MangoX2_Pico.uf2— SHA-25624fbdcbd2d310fddf9b32eb2012e4eaf6f76e28c093bcfa191d6ed01370b1778
安裝方式
將對應 UF2 以 BOOTSEL 模式複製至指定 Pico 板;不同 Target 不可混用。
驗證範圍
- 正式版原始碼與檔案版本、SHA-256 由本機建置流程核對。
- MangoX2 + Pico: Pot/Light change + raw Logger PASS (RC4 representative hardware)
- 本次依專案負責人的決定採本機驗證作為發布依據,未以 GitHub Actions 作為封版門檻。
English
Stable release for MangoX2 Runtime — Pico.
Changes
新增 Pot 與感測器 Change Callback Student API,保留既有 Motor/Drive 功能。
Downloads and SHA-256
MangoBox_Runtime_v0.2.9_MangoX2_Pico.uf2— SHA-25624fbdcbd2d310fddf9b32eb2012e4eaf6f76e28c093bcfa191d6ed01370b1778
Installation
Copy the correct target UF2 to the Pico BOOTSEL USB volume. Do not interchange target images.
Validation scope
- Exact stable output files and checksums were verified by local build tools.
- MangoX2 + Pico: Pot/Light change + raw Logger PASS (RC4 representative hardware)
- CI was explicitly waived as a release gate. This release does not claim green GitHub Actions.
MangoLite + Pico W Runtime v0.6.4
MangoLite Runtime — Pico W v0.6.4
中文
正式版(Stable)。適用於 MangoLite Runtime — Pico W。
主要更新
新增 Pot 與感測器 Change Callback Student API,保留 BLE Control v1 與 Motor/Drive 校正。
下載與 SHA-256
MangoLite_Runtime_v0.6.4_PicoW.uf2— SHA-256480faa6a98a515ea863cdeb51a8dd111ad7313fcc3faa631333eb75cdb189779
安裝方式
將對應 UF2 以 BOOTSEL 模式複製至指定 Pico 板;不同 Target 不可混用。
驗證範圍
- 正式版原始碼與檔案版本、SHA-256 由本機建置流程核對。
- MangoLite + Pico W: Pot/Light change + raw Logger PASS (RC4 representative hardware)
- 本次依專案負責人的決定採本機驗證作為發布依據,未以 GitHub Actions 作為封版門檻。
English
Stable release for MangoLite Runtime — Pico W.
Changes
新增 Pot 與感測器 Change Callback Student API,保留 BLE Control v1 與 Motor/Drive 校正。
Downloads and SHA-256
MangoLite_Runtime_v0.6.4_PicoW.uf2— SHA-256480faa6a98a515ea863cdeb51a8dd111ad7313fcc3faa631333eb75cdb189779
Installation
Copy the correct target UF2 to the Pico BOOTSEL USB volume. Do not interchange target images.
Validation scope
- Exact stable output files and checksums were verified by local build tools.
- MangoLite + Pico W: Pot/Light change + raw Logger PASS (RC4 representative hardware)
- CI was explicitly waived as a release gate. This release does not claim green GitHub Actions.
MangoLite + Pico 2 W Runtime v0.6.4
MangoLite Runtime — Pico 2 W v0.6.4
中文
正式版(Stable)。適用於 MangoLite Runtime — Pico 2 W。
主要更新
新增 Pot 與感測器 Change Callback Student API,保留 BLE Control v1 與 Motor/Drive 校正。
下載與 SHA-256
MangoLite_Runtime_v0.6.4_Pico2W.uf2— SHA-25692cfe6bf077adb7e181275446d0e12771d71c3f6159e67e3afc1e0a8d0ae4ede
安裝方式
將對應 UF2 以 BOOTSEL 模式複製至指定 Pico 板;不同 Target 不可混用。
驗證範圍
- 正式版原始碼與檔案版本、SHA-256 由本機建置流程核對。
- Target-specific full hardware regression not performed; family representative hardware tests completed
- 本次依專案負責人的決定採本機驗證作為發布依據,未以 GitHub Actions 作為封版門檻。
English
Stable release for MangoLite Runtime — Pico 2 W.
Changes
新增 Pot 與感測器 Change Callback Student API,保留 BLE Control v1 與 Motor/Drive 校正。
Downloads and SHA-256
MangoLite_Runtime_v0.6.4_Pico2W.uf2— SHA-25692cfe6bf077adb7e181275446d0e12771d71c3f6159e67e3afc1e0a8d0ae4ede
Installation
Copy the correct target UF2 to the Pico BOOTSEL USB volume. Do not interchange target images.
Validation scope
- Exact stable output files and checksums were verified by local build tools.
- Target-specific full hardware regression not performed; family representative hardware tests completed
- CI was explicitly waived as a release gate. This release does not claim green GitHub Actions.
MangoBox Device Manager v0.5.3
MangoBox Device Manager v0.5.3
中文
正式版(Stable)。適用於 MangoBox Device Manager。
主要更新
新增 Pot 感測設定、監看與 Student API Python 範例,支援 Light / Sound / Distance / Joystick Change Callback 範例。保留原有 Motor Direction Calibration;Installer 與 Portable 由同一份建置產出。
下載與 SHA-256
MangoBox_Device_Manager_v0.5.3_Setup.exe— SHA-2567a17177e724828be4c1c94dce99855e70ec501954e91765f487de775b4ca2438MangoBox_Device_Manager_v0.5.3_Portable.zip— SHA-25666fb81a81bebb24017c831c63efafc388010d309a755eff6fda311959c7853eb
安裝方式
Windows:一般使用者建議下載 Installer,免安裝/教室部署使用 Portable。
驗證範圍
- 正式版原始碼與檔案版本、SHA-256 由本機建置流程核對。
- 裝置管理器本機建置與封裝 PASS;正式安裝/啟動/解除安裝視覺 smoke 尚待確認。
- 本次依專案負責人的決定採本機驗證作為發布依據,未以 GitHub Actions 作為封版門檻。
English
Stable release for MangoBox Device Manager.
Changes
新增 Pot 感測設定、監看與 Student API Python 範例,支援 Light / Sound / Distance / Joystick Change Callback 範例。保留原有 Motor Direction Calibration;Installer 與 Portable 由同一份建置產出。
Downloads and SHA-256
MangoBox_Device_Manager_v0.5.3_Setup.exe— SHA-2567a17177e724828be4c1c94dce99855e70ec501954e91765f487de775b4ca2438MangoBox_Device_Manager_v0.5.3_Portable.zip— SHA-25666fb81a81bebb24017c831c63efafc388010d309a755eff6fda311959c7853eb
Installation
Use the Windows Installer for ordinary deployment, or Portable ZIP for no-install/classroom use.
Validation scope
- Exact stable output files and checksums were verified by local build tools.
- Windows installed-app smoke is recommended and is not yet recorded as PASS.
- CI was explicitly waived as a release gate. This release does not claim green GitHub Actions.
MangoThonny v0.4.3.1
MangoThonny v0.4.3.1 發行說明
日期:2026-09-28
狀態:Stable
作業模式與學習支援更新
MangoThonny v0.4.3.1 是建立在 v0.4.3 之上的教學工作流程更新。本版強化 Assignment Mode,降低學生因操作順序造成作業或 Learning Log 遺漏的機會,加入可依學生目前程式碼提供繁體中文提示的離線 Error Advisor,並記錄學生如何使用這些協助,方便教師於繳交後進行教學診斷與學習歷程分析。
作業模式
- 作業功能由「工具」選單移至主工具列,讓學生能直接看到作業狀態與操作入口。
- 開始新作業時,自動建立作業工作區與
main.py。 - 作業開始或匯入後,學號與作業代碼會鎖定,避免中途誤改。
- 作業代碼會自動轉為大寫。
- Assignment Mode 啟用時,左側檔案瀏覽器限制在目前作業 Project Root。
- 開啟、儲存或編輯作業範圍外檔案時會受到防呆限制,避免外部檔案破壞作業歷程。
- 離開 Assignment Mode 時,會關閉作業專案相關程式頁並回到一般工作區狀態。
- 「帶走/匯入」流程會保留同一個
assignment_id與既有學習歷程,可跨電腦延續同一份作業。 - 正式繳交 ZIP 成功產生後,作業才會標記為 completed。
- 匯出的作業會記錄當時使用的 MangoThonny 版本,便於後續追蹤。
作業資源與容量政策
作業專案會在繳交前套用可設定的檔案與專案限制:
- 程式/資料檔:單檔 2 MB;
- 圖片:單檔 2 MB;
- 音訊:單檔 5 MB;
- 影片:單檔 10 MB;
- 可匯出的作業內容:總計 15 MB。
MangoThonny 在匯入素材時會立即檢查,也會偵測學生直接透過 Windows 檔案總管複製進作業資料夾的檔案。若檔案類型或大小不符合規則,會顯示繁體中文提示。系統產生的 Mango App Export 輸出不納入作業繳交與容量計算。
MangoBox 繁體中文 Error Advisor
發生 Python 執行錯誤時,右側 Assistant 會顯示 MangoBox 的繁體中文除錯協助;Shell 仍保留原始 Python traceback,讓學生同時保有真實錯誤訊息與教學引導。
Error Advisor:
- 支援常見 Python 錯誤類型,包括 Syntax、Indentation、Name、Type、Index、Value、Import、File、Attribute、OSError/I/O 等;
- 採離線規則與 Python AST/靜態分析,不會執行學生程式碼來猜測修法;
- 在可取得資訊時,可分析學生目前的錯誤行與前面簡單的變數指定內容;
- 以三階段方式引導學生:
- 先找出問題在哪裡;
- 選擇適合的修改方向;
- 修改後重新執行並確認結果;
- 可提供 source-aware 提示,例如判斷
int與str型態差異、發現可能拼錯的變數名稱、說明串列長度與索引超界; - 發生錯誤時會隱藏重複的 Thonny 英文 suggestions 與 Warnings,讓右側只保留 MangoBox 教學內容;
- 原始 Python traceback 仍保留在 Shell,方便學生逐步學習真正的 Python 錯誤訊息。
Error Advisor 學習歷程
每一次除錯協助都可以和同一個 Assignment/Session/Run 關聯,並記錄可觀察的學生互動,包括:
- Advisor 是否產生、是否真的顯示;
- 實際可見停留時間;
- 是否展開「找問題/怎麼修改/怎麼確認」區塊;
- 可選的「有幫助/還是不懂」回饋;
- 下一次 Run 的結果:
- 下一次 Run 已修正;
- 轉為其他錯誤;
- 同錯誤持續;
- 尚待下一次 Run。
進階 Learning Log 會將這些資料以描述性指標呈現,不會把它直接宣稱為 Error Advisor 所造成的學習成效。
正式作業繳交除了保留原始 Learning Log,也會包含 advisor_summary.json 與 advisor_usage_summary,方便日後重新進行教學或研究分析。
UI 與使用性
- 主工具列直接顯示作業狀態、歷程記錄狀態、存檔狀態、專案狀態、Run/Event 數量與容量。
- 可操作的作業按鈕使用 Mango 橙色;Disabled 按鈕使用深灰底與高對比淺色文字。
- MangoBox 作業管理頁採用一致的狀態配色。
- 專案容量使用進度條顯示。
- 教師/研究用途的進階 Learning Log 與學生繳交流程保持分離。
- Error Advisor 的「有幫助/還是不懂」固定在 Assistant 底部,不會再被 Warnings 或非同步畫面更新蓋掉。
相容性
- Thonny base:
4.1.7 - bundled Python:
3.10.11 - Host Student API:
0.4.6 - canonical Host import:
mangobox - legacy Host import:
mangox2 - Event Schema:
1.1 - Standard Teaching Packages v1
- Mango App Exporter
- 保留
v0.4.3的 Host Python/High-Level MicroPython lifecycle 與 compact sensor projection 行為
Runtime firmware 仍採獨立版本管理。目前 repository 的 Runtime baseline 為 MangoX2 v0.2.8 與 MangoLite v0.6.3。
驗證
v0.4.3.1-rc14 與正式版已完成:
- MangoThonny Learning Log regression:PASS
- Assignment/projection/transfer/resource-policy regression:PASS
- Host frontend 與 High-Level MicroPython lifecycle regression:PASS
- source-aware 繁體中文 Error Advisor regression:PASS
- Advisor engagement/feedback/follow-up analytics regression:PASS
- Portable Build + Package Verification:PASS
- Assignment Mode GUI smoke:PASS
- 中文 Error Advisor GUI smoke:PASS
- 固定式 feedback footer:PASS
- 進階 Learning Log session filter:PASS
- Advisor 使用/後續結果摘要:PASS
- 正式 Windows Installer build:PASS
- 正式 Installer 安裝/啟動/捷徑/版本/升級/解除安裝/保留資料/重新安裝 smoke:PASS
正式 Artifact source commit:
d1838524e08416b6428ec9098399c9444b60fa12
正式檔案
Portable:
MangoThonny-Portable-v0.4.3.1.zip
SHA-256: b34e43209b7aa8dfbd7d9ae343e15617de5b0882e56f60eec08d3c34db6629a6
Windows Installer:
MangoThonny-Setup-v0.4.3.1.exe
SHA-256: fa1740dde1f314e18993391a757e48e145ba74785e53dd86464be687955924fc
Size: 92.2 MB
Language: Traditional Chinese (bundled zh-TW overlay)
MangoThonny v0.4.3.1 Release Notes
Date: 2026-09-28
Status: Stable
Assignment Mode and learning-support update
MangoThonny v0.4.3.1 is a focused teaching-workflow update on top of v0.4.3. It strengthens Assignment Mode, reduces common submission/logging mistakes, adds an offline Traditional-Chinese Error Advisor that can use the learner's current source code as context, and records how learners interact with the guidance so teachers can review the learning process after submission.
Assignment Mode
- Assignment controls are moved to the main toolbar instead of being hidden under Tools.
- Starting a new assignment creates the assignment workspace and
main.pyautomatically. - Student ID and assignment code are locked after the assignment starts or is imported.
- Assignment codes are normalized to uppercase.
- The file browser is constrained to the current Assignment Project while Assignment Mode is active.
- Opening or saving files outside the Project is guarded so out-of-project edits do not silently break the assignment learning history.
- Leaving Assignment Mode closes assignment-owned editor tabs and returns MangoThonny to the normal workspace state.
- Transfer/import workflows preserve the stable
assignment_idand learning history across computers. - Final submission marks the assignment completed only after the submission package is created successfully.
- The exported assignment records the MangoThonny version used to create it.
Resource and project policy
Assignment projects apply configurable file and project limits before submission:
- program / data file: 2 MB per file;
- image: 2 MB per file;
- audio: 5 MB per file;
- video: 10 MB per file;
- exportable assignment content: 15 MB total.
MangoThonny checks imported files immediately and also detects files copied directly into the workspace by Windows Explorer. Unsupported or oversized files are reported with learner-facing Traditional-Chinese guidance. System-generated Mango App Export output is excluded from assignment submission/quota accounting.
Traditional-Chinese Error Advisor
The Assistant panel now presents MangoBox learning guidance for execution errors while the Shell continues to preserve the original Python traceback.
The Error Advisor:
- covers common Python exception families including Syntax, Indentation, Name, Type, Index, Value, Import, File, Attribute, OSError / I/O and related runtime errors;
- uses offline rules and Python AST/static analysis; it does not execute learner code to infer a fix;
- can inspect the learner's current source line and simple prior assignments when available;
- provides a teaching sequence of:
- identify where the problem is;
- choose an appropriate correction direction;
- verify the result after editing;
- gives source-aware hints such as distinguishing
intvsstr, detecting likely misspelled variable names, and reporting list/index context; - suppresses duplicate English Thonny error suggestions and Warnings during an error episode so the right panel remains focused on MangoBox guidance;
- keeps the original Python traceback visible in the Shell for authentic debugging practice.
Error Advisor learning-history instrumentation
Each advisor episode can be linked to the same Assignment / Session / Run identity and records observable learner interaction, including:
- advisor available / actually shown;
- visible dwell time;
- section open/close interaction;
- optional feedback: 「有幫助」 / 「還是不懂」;
- next-run outcome:
- resolved on the next Run;
- changed to a different error;
- same error persisted;
- still pending the next Run.
The advanced local Learning Log view summarizes advisor usage as descriptive evidence only. These metrics do not by themselves claim that the Advisor caused a learning outcome.
Final assignment submissions include advisor_summary.json and advisor_usage_summary in addition to the underlying canonical learning events, so later research/teaching analysis can be recomputed from source evidence.
UI and usability
- Assignment status, recording state, save state, project state, Run/Event counts and capacity are visible in the main toolbar.
- Enabled Assignment actions use Mango orange; disabled actions use dark gray with high-contrast light text.
- Assignment Manager follows the same state-color language.
- Project capacity is shown with a progress indicator.
- Teacher/research-only advanced Learning Log remains separate from the student submission action.
- Feedback controls are placed in a persistent Assistant footer so delayed Warnings/rendering cannot erase them.
Compatibility retained
- Thonny base:
4.1.7; - bundled Python:
3.10.11; - Host Student API:
0.4.6; - canonical Host import:
mangobox; - legacy Host import:
mangox2; - Event Schema:
1.1; - Standard Teaching Packages v1;
- Mango App Exporter;
- existing Host Python / High-Level MicroPython lifecycle and compact sensor-projection behavior from
v0.4.3.
Runtime firmware remains versioned independently. The current repository Runtime baseline is MangoX2 v0.2.8 and MangoLite v0.6.3.
Validation and stable artifact build
The v0.4.3.1-rc14 line completed:
- complete MangoThonny Learning Log regression: PASS;
- Assignment / projection / transfer / resource-policy regression: PASS;
- Host frontend and High-Level MicroPython lifecycle regression: PASS;
- source-aware Chinese Error Advisor regression: PASS;
- advisor engagement / feedback / follow-up analytics regression: PASS;
- Portable Build + Package Verification: PASS;
- GUI smoke for Assignment Mode state/locking/resource guards: PASS;
- GUI smoke for Chinese Error Advisor: PASS;
- persistent feedback footer: PASS;
- session-filter fix for advanced Learning Log: PASS;
- advisor usage/follow-up summary display: PASS.
The accepted pre-release source baseline is:
cb075914ee6c7eef3216bc789a0a7338d17a1b5f
Final artifacts
The stable v0.4.3.1 artifacts were rebuilt from the final stable source merge commit:
source_commit = d1838524e08416b6428ec9098399c9444b60fa12
Exac...
MangoX2 + Pico W Runtime v0.2.8
MangoX2 Runtime v0.2.8 — MangoX2 + Pico W
中文
MangoX2 Runtime v0.2.8 將車體左右輪映射與單輪反轉正式納入 Drive Runtime,讓不同車體安裝方向可透過 Device Manager 校正,不需要修改學生程式。
主要更新
- 新增
drive.left_motor/drive.right_motor。 - 新增
drive.left_inverted/drive.right_inverted。 - 預設值保持 v0.2.7 既有車體行為相容。
- Raw
M1/M2保留實體 channel 語意;forward/backward/ tank / arc / spin 使用校正後邏輯車輪語意。 - Drive profile 變更後,下一個 Drive command 即套用新設定,不需 reboot。
驗證
- Motor / Drive automated regression:PASS。
- Pico / Pico W / Pico 2 W 三 target clean local build:PASS。
- 代表性 MangoX2 Motor Direction Calibration 實機 smoke:PASS。
下載與 SHA-256
MangoBox_Runtime_v0.2.8_MangoX2_PicoW.uf2- SHA-256:
cbbfe8e6ca8b996a40b118656ec55afbec8916fe1dd21421b729b24ebe557df8
- SHA-256:
English
MangoX2 Runtime v0.2.8 adds stable logical wheel mapping and independent per-wheel inversion to the Drive Runtime, allowing chassis wiring and installation differences to be calibrated without changing learner programs.
Highlights
- Adds
drive.left_motor/drive.right_motor. - Adds
drive.left_inverted/drive.right_inverted. - Preserves the legacy v0.2.7 default vehicle behavior.
- Raw
M1/M2keep physical-channel semantics, while vehicle movement commands use calibrated logical wheel semantics. - Updated Drive configuration is applied on the next Drive command without reboot.
Validation
- Motor / Drive automated regression: PASS.
- Clean local builds for Pico / Pico W / Pico 2 W: PASS.
- Representative MangoX2 Motor Direction Calibration hardware smoke: PASS.
Download and SHA-256
MangoBox_Runtime_v0.2.8_MangoX2_PicoW.uf2- SHA-256:
cbbfe8e6ca8b996a40b118656ec55afbec8916fe1dd21421b729b24ebe557df8
- SHA-256:
MangoX2 + Pico 2 W Runtime v0.2.8
MangoX2 Runtime v0.2.8 — MangoX2 + Pico 2 W
中文
MangoX2 Runtime v0.2.8 將車體左右輪映射與單輪反轉正式納入 Drive Runtime,讓不同車體安裝方向可透過 Device Manager 校正,不需要修改學生程式。
主要更新
- 新增
drive.left_motor/drive.right_motor。 - 新增
drive.left_inverted/drive.right_inverted。 - 預設值保持 v0.2.7 既有車體行為相容。
- Raw
M1/M2保留實體 channel 語意;forward/backward/ tank / arc / spin 使用校正後邏輯車輪語意。 - Drive profile 變更後,下一個 Drive command 即套用新設定,不需 reboot。
驗證
- Motor / Drive automated regression:PASS。
- Pico / Pico W / Pico 2 W 三 target clean local build:PASS。
- 代表性 MangoX2 Motor Direction Calibration 實機 smoke:PASS。
下載與 SHA-256
MangoBox_Runtime_v0.2.8_MangoX2_Pico2W.uf2- SHA-256:
58e61072961ded6d6f4a76c6f88c6556956b1be215e431ac622331dc4cc7dc2c
- SHA-256:
English
MangoX2 Runtime v0.2.8 adds stable logical wheel mapping and independent per-wheel inversion to the Drive Runtime, allowing chassis wiring and installation differences to be calibrated without changing learner programs.
Highlights
- Adds
drive.left_motor/drive.right_motor. - Adds
drive.left_inverted/drive.right_inverted. - Preserves the legacy v0.2.7 default vehicle behavior.
- Raw
M1/M2keep physical-channel semantics, while vehicle movement commands use calibrated logical wheel semantics. - Updated Drive configuration is applied on the next Drive command without reboot.
Validation
- Motor / Drive automated regression: PASS.
- Clean local builds for Pico / Pico W / Pico 2 W: PASS.
- Representative MangoX2 Motor Direction Calibration hardware smoke: PASS.
Download and SHA-256
MangoBox_Runtime_v0.2.8_MangoX2_Pico2W.uf2- SHA-256:
58e61072961ded6d6f4a76c6f88c6556956b1be215e431ac622331dc4cc7dc2c
- SHA-256:
MangoX2 + Pico Runtime v0.2.8
MangoX2 Runtime v0.2.8 — MangoX2 + Pico
中文
MangoX2 Runtime v0.2.8 將車體左右輪映射與單輪反轉正式納入 Drive Runtime,讓不同車體安裝方向可透過 Device Manager 校正,不需要修改學生程式。
主要更新
- 新增
drive.left_motor/drive.right_motor。 - 新增
drive.left_inverted/drive.right_inverted。 - 預設值保持 v0.2.7 既有車體行為相容。
- Raw
M1/M2保留實體 channel 語意;forward/backward/ tank / arc / spin 使用校正後邏輯車輪語意。 - Drive profile 變更後,下一個 Drive command 即套用新設定,不需 reboot。
驗證
- Motor / Drive automated regression:PASS。
- Pico / Pico W / Pico 2 W 三 target clean local build:PASS。
- 代表性 MangoX2 Motor Direction Calibration 實機 smoke:PASS。
下載與 SHA-256
MangoBox_Runtime_v0.2.8_MangoX2_Pico.uf2- SHA-256:
f94b582fcf3cb253c43c62747b170781e061498fe68f996a6d787f7e4b3fbcbd
- SHA-256:
English
MangoX2 Runtime v0.2.8 adds stable logical wheel mapping and independent per-wheel inversion to the Drive Runtime, allowing chassis wiring and installation differences to be calibrated without changing learner programs.
Highlights
- Adds
drive.left_motor/drive.right_motor. - Adds
drive.left_inverted/drive.right_inverted. - Preserves the legacy v0.2.7 default vehicle behavior.
- Raw
M1/M2keep physical-channel semantics, while vehicle movement commands use calibrated logical wheel semantics. - Updated Drive configuration is applied on the next Drive command without reboot.
Validation
- Motor / Drive automated regression: PASS.
- Clean local builds for Pico / Pico W / Pico 2 W: PASS.
- Representative MangoX2 Motor Direction Calibration hardware smoke: PASS.
Download and SHA-256
MangoBox_Runtime_v0.2.8_MangoX2_Pico.uf2- SHA-256:
f94b582fcf3cb253c43c62747b170781e061498fe68f996a6d787f7e4b3fbcbd
- SHA-256: