## Summary
**What is the goal of this PR?**
Add Wi-Fi network management to the Web UI settings page, similar to
existing OPDS server management, so users can view, add, edit, and
delete saved Wi-Fi credentials from the browser. Closes
https://github.com/crosspoint-reader/crosspoint-reader/issues/1544 and
https://github.com/crosspoint-reader/crosspoint-reader/discussions/607
**What changes are included?**
- Added Wi-Fi API endpoints:
- GET /api/wifi Listing saved networks (without exposing plaintext
passwords)
- POST /api/wifi Creating/updating networks
- POST /api/wifi/delete Deleting networks by index
- Added a new Wi-Fi Networks management section web UI settings page
<img width="921" height="712" alt="image"
src="https://github.com/user-attachments/assets/42c20a63-f335-4389-8246-d38065d6b65a"
/>
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? _**< YES >**_
---------
Co-authored-by: Copilot <copilot@github.com>
## Summary
* **What is the goal of this PR?**
Enhance file browser long-press behavior so the delete action now works
for both files and directories.
* **What changes are included?**
Updated `FileBrowserActivity.cpp` to support for directory deletion on
long-press
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? _**NO**_
## Summary
Deduplicated battery drawing code for Roundraff theme with other themes.
Now `BaseTheme` provides non-virtual `drawBatteryLeft` and
`drawBatteryRight` methods, which use a shared static
`drawBatteryOutline` implementation, and defer to a virtual
`fillBatteryIcon` implementation. In Classic and Roundraff
`fillBatteryIcon` uses a solid fill, while Lyra and Lyra Extended use a
segmented fill.
The charging indicator inside the battery was missing for Roundraff, but
is now consistent with other themes because of the shared drawing
implementation.
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? _**PARTIALLY**_
## Summary
With the Roundraff theme selected and no recent books, the home screen
menu shows a "Continue Reading" option and menu handling is offset by
one ("Continue Reading" actually does "Browse Files", "File Transfer"
actually does "Settings", etc.) Simple fix here is to omit the "Continue
Reading" menu item when there are no recent books.
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? _**NO**_
When Power+Down is released with a slight stagger (Down before Power),
the Power release event leaked through to the activity loop, triggering
short-press Power actions like clip selection. Add a
screenshotComboActive flag that suppresses all input processing until
Power is fully released after a screenshot combo.
## Summary
* **What is the goal of this PR?** (e.g., Implements the new feature for
file uploading.)
* **What changes are included?**
## Additional Context
* Add any other information that might be helpful for the reviewer
(e.g., performance implications, potential risks,
specific areas to focus on).
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? _**YES**_
## Summary
Before, any sub-activity of a reader activity needed to override
`isReaderActivity` to maintain correct bookkeeping through
`ActivityManager::isReaderActivity`. We could easily miss this in any
new sub-activities. Instead, simplify so each reader activity correctly
reports and then `ActivityManager` checks for any reader activity in its
stack.
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? _**NO**_
## Summary
Simplify duplicated code in `XtcReaderActivity` to use
`ReaderUtils::detectPageTurn`. This implementation is now shared with
`EpubReaderActivity` and `TxtReaderActivity`. Also deduplicated the
chapter skip time constant.
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? _**PARTIALLY**_
## Summary
Update the Open X4 SDK to point to the CrossPoint organization fork.
This allows us to take SDK changes without being blocked on the upstream
repository.
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? _**NO**_
## Summary
* **What is the goal of this PR?**
Fixes a bug in the jpeg nearest-neighbor downscaler where source image
rows containing visible content may be cropped.
* **What changes are included?**
Split the single `fineScaleFP`/`invScaleFP` scale pair in
`JpegToFramebufferConverter.cpp` into separate X
(`fineScaleFPX`/`invScaleFPX`) and Y (`fineScaleFPY`/`invScaleFPY`)
pairs
## Additional Context
The one case that I encountered in my epub:
<img width="480" height="37" alt="img_6_0"
src="https://github.com/user-attachments/assets/0de3778c-cf3d-42dc-a1ba-4ba729f6b1c4"
/>
The image generated from pxc cache with the latest commit:
<img width="464" height="36" alt="img_6_0_master pxc"
src="https://github.com/user-attachments/assets/92ba06e5-be15-4db6-840e-882bdddb1baf"
/>
With this fix, the bottom outline of each character is still there:
<img width="464" height="36" alt="img_6_0_fixed pxc"
src="https://github.com/user-attachments/assets/4246ce55-5a20-480f-862f-7988f804c1fb"
/>
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? _**PARTIALLY**_
## Summary
* **What is the goal of this PR?** Fix the vertical misalignment between
the "File Manager" title and the action buttons (Upload / New Folder /
Delete Selected).
* **What changes are included?**
- `h2`: `margin-top: 0` → `margin: 0` — removes the default browser
`margin-bottom` that was inflating the height of the header-left flex
container, pushing the calculated flex center below the visual midpoint
of the title text
- `.page-header-left`: `align-items: baseline` → `align-items: center` —
ensures the title and breadcrumbs are center-aligned with each other and
with the buttons
### Before
<img width="1198" height="560" alt="Screenshot From 2026-04-30 22-29-28"
src="https://github.com/user-attachments/assets/b9a43069-cd93-4028-ab21-75be253ed1f0"
/>
### After
<img width="1198" height="560" alt="Screenshot From 2026-04-30 22-48-49"
src="https://github.com/user-attachments/assets/b9977c65-e479-4490-8ec3-15ae99cfccd7"
/>
## Additional Context
Small CSS-only change, no JavaScript or HTML structure touched.
---
### AI Usage
Did you use AI tools to help write this code? _**YES**_
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
## Summary
Added RAM delta output to firmware_size_history.py output, e.g.:
```
% scripts/firmware_size_history.py --commits adcd796 8e18472
[info] Will restore to 'sd-card-fonts' when finished.
[info] Stashing uncommitted changes...
[info] Building 2 commits...
[1/2] adcd7961c9 feat: self-heal from transient WiFi loss, add dBm indicator during WebServerActivity (#1780)
Building (env: default)...
Flash: 5,751,209, RAM: 97,996 bytes
[2/2] 8e184724d1 Merge branch 'master' into sd-card-fonts
Building (env: default)...
Flash: 5,762,395, RAM: 97,988 bytes
[info] Restoring 'sd-card-fonts'...
[info] Restoring stashed changes...
Commit Flash Delta RAM Delta Title
────────── ─────────── ─────── ─────────── ─────── ────────────────────────────────────────
adcd7961c9 5,751,209 97,996 feat: self-heal from transient WiFi loss, add dBm indicator during WebServerActivity (#1780)
8e184724d1 5,762,395 +11,186 97,988 -8 Merge branch 'master' into sd-card-fonts
```
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? _**YES**_
## Summary
* **What is the goal of this PR?**
I was seeing hangs during File Transfer. Troubleshooting via serial
showed it had to do with the SHUTTING_DOWN state. Strengthened the WiFi
State Machine a bit and added self-healing. Additionally, I found it
obnoxious to need to keep referring to serial for my wifi strength, so a
dBm meter is added opposing the SSID when on the File Transfer page.
* **What changes are included?**
The dBm meter was at risk of causing rapid screen updates (if, say, we
hovered right around a threshold), so I've implemented a basic
hysteresis around this. RISING/FALLING would be more canonical variable
names, but are reserved in the framework.
## Additional Context
My X4 has a terrible antenna, apparently, and I was running into this
failure condition pretty regularly. Wifi "bars" were chosen as a
relatively pan-cultural glyph rather than relying on localization.
Tested on hardware at -83 dBm under sustained EPUB upload (60 books, ~30
MB). Transient losses up to ~14s now ride through; pre-fix the same
losses required a power-cycle after as little as a 2s blip.
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? Nope. Commits are hand
written. Claude was used to build a local test harness for validation
only during iteration.
## Summary
Simplify code calculating compile-time array sizes with
`sizeof(array)/sizeof(element)` to use `std::size`.
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? _**PARTIALLY**_
## Summary
* Add new block style stack to track accumulated styles
* When dropping into nested block, apply new styles on last BlockStyle
in the stack and push to the back
* When leaving block, pop the stack
* Fix issue where the image % width always went off screen width, use
current container width based on accumulated styles
* Bump section cache version to regenerate pages
## Additional Context
Fixes https://github.com/crosspoint-reader/crosspoint-reader/pull/1581
more correctly, and includes fix from
https://github.com/crosspoint-reader/crosspoint-reader/pull/1580
Consider:
```html
<div class="c1"> <!-- marginLeft: 10px -->
<div class="c2"> <!-- marginLeft: 20px -->
<p class="c3">text</p> <!-- marginLeft: 5px -->
</div>
<p class="c4">text2</p> <!-- marginLeft: 5px -->
</div>
```
| Element | Expected | Before #1581 (leaked) | After #1581 (reset) |
This PR (style stack) |
| --- | --- | --- | --- | --- |
| `c3` | 10+20+5 = 35px | 35px | 25px | 35px |
| `c4` | 10+5 = 15px | 35px (wrong, leaked c2) | 5px (wrong, lost c1) |
15px |
This PR replaces the reset-on-close approach with a proper block style
stack. When a block element opens, its resolved style is pushed onto the
stack. When it closes, the stack pops back to the parent's style. This
correctly accumulates nested margins/padding while preventing style
leakage to siblings.
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? Yes
---------
Co-authored-by: Zach Nelson <zach@zdnelson.com>
## Summary
* Add new block style stack to track accumulated styles
* When dropping into nested block, apply new styles on last BlockStyle
in the stack and push to the back
* When leaving block, pop the stack
* Fix issue where the image % width always went off screen width, use
current container width based on accumulated styles
* Bump section cache version to regenerate pages
## Additional Context
Fixes https://github.com/crosspoint-reader/crosspoint-reader/pull/1581
more correctly, and includes fix from
https://github.com/crosspoint-reader/crosspoint-reader/pull/1580
Consider:
```html
<div class="c1"> <!-- marginLeft: 10px -->
<div class="c2"> <!-- marginLeft: 20px -->
<p class="c3">text</p> <!-- marginLeft: 5px -->
</div>
<p class="c4">text2</p> <!-- marginLeft: 5px -->
</div>
```
| Element | Expected | Before #1581 (leaked) | After #1581 (reset) |
This PR (style stack) |
| --- | --- | --- | --- | --- |
| `c3` | 10+20+5 = 35px | 35px | 25px | 35px |
| `c4` | 10+5 = 15px | 35px (wrong, leaked c2) | 5px (wrong, lost c1) |
15px |
This PR replaces the reset-on-close approach with a proper block style
stack. When a block element opens, its resolved style is pushed onto the
stack. When it closes, the stack pops back to the parent's style. This
correctly accumulates nested margins/padding while preventing style
leakage to siblings.
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? Yes
---------
Co-authored-by: Zach Nelson <zach@zdnelson.com>
## Summary
Small cleanup to avoid a dynamically allocated static structure for auto
page-turn rate values.
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? _**NO**_
## Summary
Small simplification to node sorting algorithm in
GfxRenderer::fillPolygon.
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? _**NO**_
## Summary
**What is the goal of this PR?**
Restore compatibility with older XTC files that use the legacy 48-byte
header layout, so they open correctly instead of failing
header/page-table validation.
**What changes are included?**
- Updates the XTC parser to accept legacy files where pageTableOffset
starts at 0x30 instead of the newer full header size.
- Avoids treating legacy header bytes as a valid chapterOffset, so older
files are not misclassified as having chapters.
- Switches XTC file size and seek operations to 64-bit-safe wrapper
calls in HalStorage, which matches the XTC format’s 64-bit offset fields
and keeps page/chapter lookups from narrowing offsets.
- Keeps the existing bounds checks and improves related logging when XTC
page loads fail.
## Additional Context
- This is a narrow compatibility fix for XTC handling. The functional
change is in the XTC parser; the HalStorage changes are supporting
wrapper methods needed by that parser update.
- Risk should be low outside XTC reading, since the HAL changes only
expose existing underlying file APIs rather than changing storage
behavior.
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? **YES**
## Summary
The GitHub recent release API returns lots of data, including the full
release notes. For 1.2.0, that data is 30,530 bytes. The existing
OtaUpdater HTTP handler and single-buffer JSON parsing can't reliably
handle that much data in the constrained ESP32 environment.
This change does a few things:
1. Adds a very simple lib/JsonParser/StreamingJsonParser.cpp with
SAX-style callbacks to read JSON data incrementally.
2. Adds a very simple lib/JsonParser/ReleaseJsonParser.cpp building on
StreamingJsonParser, which specifically parses the GitHub release JSON
for the release version, URL, and size.
3. Updates OtaUpdater.cpp to use an instance of ReleaseJsonParser to
incrementally parse the large response it may receive from GitHub.
Building from this commit while overriding my local version to 1.1.9, I
was able to run the OTA update process ~5 times in a row successfully.
Fixes#1561 (second part, after #1805).
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? _**YES**_
## Summary
Omit any duplicate strings in non-English localization lookup tables.
For non-English lookup offsets, tag bit 15 to indicate the offset
applies to the English localization table.
Saves 18,766 bytes of flash.
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? _**PARTIALLY**_
## Summary
The file type was shown twice in the file browser table: as a bold badge
in the **Name** column and as plain text in the **Type** column.
**Before:** badge in Name column + redundant text in Type column
**After:** badge lives exclusively in the Type column
| | Before | After |
|---|---|---|
| Folder | Name: `📁 folder` **FOLDER** · Type: `Folder` | Name: `📁
folder` · Type: **FOLDER** |
| EPUB | Name: `📗 book.epub` **EPUB** · Type: `EPUB` | Name: `📗
book.epub` · Type: **EPUB** |
| Other | Name: `📄 file.txt` · Type: `TXT` | unchanged |
### Before
<img width="1125" height="571" alt="Screenshot From 2026-04-30 22-07-59"
src="https://github.com/user-attachments/assets/794c66d3-5ad5-4536-a142-708989b9dee8"
/>
### After
<img width="1125" height="571" alt="Screenshot From 2026-04-30 22-08-45"
src="https://github.com/user-attachments/assets/f9fb2898-b165-478a-813b-f2290ad3e42c"
/>
## Changes
- Removed `epub-badge` and `folder-badge` from the Name column
- Moved badges into the Type column, replacing the redundant plain text
- Non-EPUB files are unaffected (extension text remains in Type column)
🤖 Generated with [Claude Code](https://claude.com/claude-code)
---------
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
## Summary
Another follow up to #1408. That PR revised the language.bin settings
file to version 2. After merging, it occurred to me that it would make
more sense to have the language inside the overall settings.json file,
easy for users to see and change directly.
This change adds migration from language.bin to settings.json. It only
migrates from the pre-#1408 v1 language.bin format, because v2 has only
been in for a couple of commits and I don't think we need long-term
maintenance code to handle it.
#1408 had a potential long-term maintenance issue for migration. It
assumed that the enum order of languages never changes, only grows. So,
e.g. "Portuguese (Portugal)" could never be added next to "Portuguese
(Brazil)", only appended. This change includes a hardcoded frozen
ordering of the language indices as they existed at 2f969a9, and
migrates to storing a language code string in the settings.json instead
for future reordering flexibility.
Similar to the settings.bin to settings.json migration, which renames
settings.bin to settings.bin.bak, this change renames language.bin to
language.bin.bak after adding the language setting to settings.json.
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? _**PARTIALLY**_
## Summary
Quick follow up to #1408. We overlooked the fact that the change wasn't
using the LOG_ macros. Also, removed redundant file.close() calls, as in
23aad21.
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? _**PARTIALLY**_
## Summary
* **What is the goal of this PR?** Reduce the flash footprint of the
i18n string data and tooling improvements to `gen_i18n.py`.
* **What changes are included?**
### 1. `lib/I18n/I18n.cpp` — use offset-based lookup
`I18n::get()` previously dereferenced a `const char* const*` pointer
array. It now uses a two-field `LangStrings` struct (a flat char blob +
a `uint16_t` offset table) generated for each language:
```cpp
// before
const char* const* strings = getStringArray(_language);
return strings[index];
// after
const LangStrings lang = getLanguageStrings(_language);
return lang.data + lang.offsets[index];
```
Lookup cost is unchanged — still O(1), one array load and one addition.
### 2. `scripts/gen_i18n.py` — new generated layout
Each language's string data is now emitted as:
- **`STRINGS_XX_DATA[]`** — a single `const char[]` blob of all strings
concatenated with `\0` separators.
- **`OFFSETS_XX[]`** — a `uint16_t` array of one byte-offset per `StrId`
into the blob.
Previously each language had a `const char* const STRINGS_XX[]` pointer
array (4 bytes/entry on ESP32-C3).
#### Flash savings
| Table type | Size per language | 19 languages |
|---|---|---|
| `const char*` pointer array (before) | `339 × 4 = 1,356 B` | **25,764
B** |
| `uint16_t` offset table (after) | `339 × 2 = 678 B` | **12,882 B** |
| **Saved** | | **12,882 B (~12.6 KB)** |
String data size is unchanged — 133,092 B across 19 languages.
**Total: 158,856 B → 145,974 B** (pointer tables → offset tables).
### 3. Build-time stripping of unused strings
`gen_i18n.py` now scans the `src/` and `lib/` trees for `STR_*`
references and, during a PlatformIO build, automatically omits the 52
strings that are defined in YAML but never referenced in code. This
further reduces the compiled output from 339 → 287 string keys per
language.
---
## `gen_i18n.py` CLI reference
```
python gen_i18n.py [translations_dir [output_dir]] [options]
```
| Argument / Flag | Default | Description |
|---|---|---|
| `translations_dir` | `lib/I18n/translations` | Path to the
per-language YAML files |
| `output_dir` | `lib/I18n/` | Where to write the generated `.h` /
`.cpp` files |
| `--src-dirs DIR [DIR …]` | `src lib` | Directories scanned for `STR_*`
usage |
| `--strip-unused` | off | Remove unreferenced `STR_*` keys from
generated output |
| `--verbose` / `-v` | off | Print per-key INFO/WARNING messages and
`Generated:` lines |
The PlatformIO build (SCons `pre:` hook) calls `main(strip_unused=True)`
automatically, so unused strings are always stripped from firmware
builds without any manual flag.
**Default run output** (no flags) shows a per-language summary table:
```
Language Code Own Fallback Unused Data (B)
------------------ ---- --- -------- ------ --------
English EN 339 0 52 5,266
Belarusian BE 310 29 52 9,673
…
Total: 339 | Used in code: 287 | Never used: 52
Flash (now): 133,092 B strings + 12,882 B offset tables (uint16_t) = 145,974 B
Flash (before): 133,092 B strings + 25,764 B pointer tables (ptr32) = 158,856 B
Saved by offset tables: 12,882 B
```
---
### AI Usage
Did you use AI tools to help write this code? **YES** (GitHub Copilot)
---------
Co-authored-by: Zach Nelson <zach@zdnelson.com>
## Summary
I love the new screenshot feature but I can tell that soon I'm gonna
have a folder full of screenshots and not know what's what. It would be
great if the folder could self organize.
When a screenshot is taken while reading, the filename would now include
the book title, chapter (EPUB), page number, and progress percentage.
Example: My-Great-Book_ch3_p5_42pct_12345.bmp
* **What is the goal of this PR?** (e.g., Implements the new feature for
file uploading.)
* **What changes are included?**
## Additional Context
* Non-reader screens keep the existing screenshot-<millis>.bmp naming.
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? _**< YES >**_
I reviewed all the changes but I don't have any experience with this
project so this is my best guess
## Summary
* **What is the goal of this PR?** Fix KOReader sync failing on HTTPS
servers due to TLS out-of-memory on ESP32-C3
* **What changes are included?**
- Replace `WiFiClientSecure`/`HTTPClient` with `esp_http_client`
(ESP-IDF native API) for all KOSync HTTP requests
- Use 2KB TLS buffers instead of the default 16KB — KOSync payloads are
tiny JSON (<1KB), so this is more than sufficient
- Use `esp_crt_bundle_attach` for proper TLS certificate verification
(replaces `setInsecure()`)
- Strip trailing slashes from server URL to prevent double-slash in API
paths (e.g. `https://server.com//users/auth`)
- Add `lastHttpCode` static field for diagnostics
- Add free heap logging to help debug memory issues
### Problem
The ESP32-C3 has ~46KB free heap after WiFi is initialized.
`WiFiClientSecure` allocates 16KB for TLS RX + 16KB for TLS TX = 32KB
just for the TLS buffers, leaving almost no room for the actual TLS
handshake (which needs additional dynamic allocations). This causes
KOReader sync to:
1. Fail silently with network errors on HTTPS servers (including the
default `sync.koreader.rocks`)
2. Occasionally crash with heap exhaustion
### Solution
`esp_http_client` (the ESP-IDF native HTTP client) allows configuring
`buffer_size` and `buffer_size_tx` independently. Setting both to 2KB
(total 4KB) leaves plenty of heap for the TLS handshake while still
being sufficient for KOSync's small JSON payloads.
This also fixes the `setInsecure()` anti-pattern —
`esp_crt_bundle_attach` provides proper certificate verification using
the ESP-IDF's built-in CA bundle, so credentials and reading history are
no longer sent over unverified TLS connections.
### Files changed
| File | Change |
|------|--------|
| `lib/KOReaderSync/KOReaderSyncClient.cpp` | Replace
WiFiClientSecure/HTTPClient with esp_http_client; add ResponseBuffer,
base64 encoder, createClient helper |
| `lib/KOReaderSync/KOReaderSyncClient.h` | Add `lastHttpCode` static
field |
| `lib/KOReaderSync/KOReaderCredentialStore.cpp` | Strip trailing
slashes from base URL |
## Additional Context
Tested on a CrossPoint X4 (ESP32-C3 with 4MB flash). Before this change,
KOSync auth to `sync.koreader.rocks` (HTTPS) would fail ~80% of the
time. After: works reliably.
The `base64Encode` helper is needed because `esp_http_client` doesn't
have a built-in `setAuthorization()` method like `HTTPClient` does. This
is used for the HTTP Basic Auth header required by Calibre-Web-Automated
KOSync servers.
Fixes#581
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it helps set the right
context for reviewers.
Did you use AI tools to help write this code? _**PARTIALLY**_ AI
assisted with the esp_http_client migration pattern and base64 encoder.
The root cause analysis (TLS buffer OOM) and solution design were
manual.
---------
Co-authored-by: trilwu <trilwu@users.noreply.github.com>
Co-authored-by: Justin Mitchell <justin@jmitch.com>
This allows KOSync to work for files downloaded from OPDS when set to
use filename matching
## Summary
* **What is the goal of this PR?** Brings the names of files downloaded
via OPDS into line with the KOReader OPDS plugin so the files sync when
KOSync is set to filename matching
* **What changes are included?** Change the filename generated by the
OPDS browser from "\<title\> - \<author\>" to "\<author\> - \<title\>"
so it matches KOReader.
## Additional Context
* Add any other information that might be helpful for the reviewer
(e.g., performance implications, potential risks,
specific areas to focus on).
Both KOReader and crosspoint sanitise the filename by removing special
characters. This step seems to already use the same rules.
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? NO
## Summary
- **What is the goal of this PR?** Fix chapter-opener text collapsing to
1-2 words per line in EPUBs that apply large em-based horizontal CSS
insets.
- **What changes are included?** Caps `marginLeft`, `marginRight`,
`paddingLeft`, and `paddingRight` at 2em in `BlockStyle::fromCssStyle`
(`lib/Epub/Epub/blocks/BlockStyle.h`). Vertical margins/padding are
unchanged - the bug is horizontal-only.
## Additional Context
**Repro:** *Mother Night* by Kurt Vonnegut, Chapter 21 ("My best
friend..."). The chapter-opening element's embedded CSS sets a large
horizontal inset.
- Settings: Embedded Style = On, Justify alignment (default).
- Before: 1-2 words per line with a visible river between them.
- After: body text fills the usable page width like every other
paragraph.
**Why the clamp lives in `fromCssStyle`:** This is the single point
where CSS lengths resolve to pixels, so clamping here keeps both
`effectiveWidth` call sites in `ChapterHtmlSlimParser` (`:1131-1133`
makePages, `:841-844` long-block split) consistent with the
`leftInset()` xOffset. A width-only clamp at either call site would
leave text pushed right by a large xOffset.
**Test plan:**
- [x] Flashed to Xteink X4. Chapter 21 of *Mother Night* renders
normally with Embedded Style on.
- Users with cached layouts of affected books need to delete
`.crosspoint/epub_<hash>/sections/` (or the whole `.crosspoint/`) on the
SD card to pick up the fix.
**Before / After:**
<p float="left">
<img height="400" alt="IMG_0320"
src="https://github.com/user-attachments/assets/de8815ae-f2f4-4845-be1f-449c5a25d191"
/>
<img height="400" alt="IMG_0324"
src="https://github.com/user-attachments/assets/95f5f4cb-8aa1-418e-b611-18e0c342cbef"
/>
</p>
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it helps set the right
context for reviewers.
Did you use AI tools to help write this code? _**< YES >**_
I used Claude Code (Opus 4.7) to evaluate the codebase and find the
relevant code related to this rendering. From there, it was human
designed, reviewed and tested.
Co-authored-by: rhoopr <>
## Summary
The Python scripts in the current repo have many requirements that are
not mentioned in any requirements.txt files, so I have therefore added
them to the directories that need them.
Question: Should these changes maybe moved into a "root"
requirements.txt file?
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? _**NO**_
## Summary
- Add new `roundedraff` theme.
- Fix sleep screen artifact where book cover showed grid lines in
`Cover` mode with `Crop`.
## Additional Context
## Key changes
- Added `src/components/themes/roundedraff/` theme implementation.
- Updated sleep cover rendering logic in:
- `src/activities/boot_sleep/SleepActivity.cpp`
- `src/activities/boot_sleep/SleepActivity.h`
- Improved cover generation/regeneration paths in:
- `lib/Epub/Epub.cpp`
- `lib/Epub/Epub.h`
- `lib/Txt/Txt.cpp`
- `lib/Txt/Txt.h`
## Verification
- Sleep screen set to Book cover
- Sleep screen cover mode set to Crop
- Confirmed grid artifacts no longer appear on cover sleep screen.
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? _**< YES >**_
---------
Co-authored-by: CaptainFrito <yzq6x5zypy@privaterelay.appleid.com>
Co-authored-by: Zach Nelson <zach@zdnelson.com>
Co-authored-by: Uri Tauber <142022451+Uri-Tauber@users.noreply.github.com>
## Summary
* **What is the goal of this PR?**
Enables the PlatformIO build cache to speed up compilation.
## Additional Context
Significant improvement when switching branches, as unchanged files
won't need re-compilation.
See
[Docs](https://docs.platformio.org/en/latest/projectconf/sections/platformio/options/directory/build_cache_dir.html).
Note: May need to manually delete .cache folder if updating toolchain or
framework.
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? _**< NO >**_
## Summary
* Includes short SHA in version string for better version tracking, so
it looks like `1.1.0-dev-feat-kosync-xpath-05c6cf8`
* Closes
https://github.com/crosspoint-reader/crosspoint-reader/issues/1247
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? _**< YES >**_
## Summary
* Add swedish translation of strings added by "feat: Support for
multiple OPDS servers (#1209)"
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? _**NO**_
## Summary
* Adds pre-commit git hook that automatically runs
./bin/clang-format-fix
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? _**< YES >**_
---------
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
## Summary
* **What is the goal of this PR?**
Fix the visual keyboard cursor not advancing when typing a space. Typing
one space leaves the cursor at position 0; typing two spaces advances it
only by one space width. This affects all input types except URL.
* **What changes are included?**
Replace `getTextWidth()` with `getTextAdvanceX()` in four locations
within `KeyboardEntryActivity::render()` for cursor positioning and
line-wrapping calculations.
## Additional Context
* **Root cause**: `getTextWidth()` returns the bounding-box width of the
drawn glyphs. The space glyph has `width=0` and `height=0` (it's
invisible), so `getTextWidth(" ") == 0`. The trailing `advanceX` of the
last character is only flushed when the *next* character is processed,
so a string ending in space reports zero width. `getTextAdvanceX()`
correctly includes the final glyph's advance, matching how `drawText()`
actually positions characters.
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? _**YES**_
Adds royalty.dev as a funding platform for contributors by adding custom
funding link and badge to README
A maintainer would need to click the "Are you the maintainer?" button on
https://app.royalty.dev/crosspoint-reader/crosspoint-reader to install
the github app which pulls the PRs in and ranks them.
(Disclosure, Royalty is a product I've recently built and am using for
my own OSS and currently gating custom font builds and nightlys on
crosspointreader.com)
This page currently can collect donations but without the integration it
wont notify contributors and it wont rescore fresh PRs.
## Summary
* Add the swedish translations of the text strings added by #1697
---
### AI Usage
While CrossPoint doesn't have restrictions on AI tools in contributing,
please be transparent about their usage as it
helps set the right context for reviewers.
Did you use AI tools to help write this code? _**NO**_
