Add '/' as explicit hyphen delimiter and relax the alphabetic-surround
requirement for '/' and '-' in buildExplicitBreakInfos so URL path
segments can be line-wrapped. Includes repeated-separator guard to
prevent breaks between consecutive identical separators (e.g. "http://").
Co-authored-by: Cursor <cursoragent@cursor.com>
Reduces ParsedText::layoutAndExtractLines CPU time 5–9% via two
independent optimizations from jpirnay's PR #1027:
- 128-entry direct-mapped word-width cache (4 KB BSS, FNV-1a hash)
absorbs redundant getTextAdvanceX calls across paragraphs
- Early exit in hyphenateWordAtIndex when prefix exceeds available
width (ascending byte-offset order guarantees monotonic widths)
- Reusable prefix string buffer eliminates per-candidate substr allocs
- Reserve hint for lineBreakIndices in computeLineBreaks
List-specific upstream changes (splice, iterator style) not applicable
as mod already uses std::vector (PR #1038). Benchmark infrastructure
excluded (removed by author in final commit).
Co-authored-by: Cursor <cursoragent@cursor.com>
- #1038 (partial): Add .erase() for consumed words in layoutAndExtractLines
to fix redundant early flush bug; fix wordContinues flag in hyphenateWordAtIndex
- #1037: Add combining mark handling for hyphenation (NFC-like precomposition)
and rendering (base glyph tracking in EpdFont, GfxRenderer including CCW)
- #1045: Shorten STR_FORGET_BUTTON labels across all 9 translation files
- #1019: Display file extensions in File Browser via getFileExtension helper
- Pull romanian.yaml from upstream/master (merged PR #987)
Co-authored-by: Cursor <cursoragent@cursor.com>
Port three new upstream commits and align the existing #1002 port:
- PR #1014: Strip unused CSS rules by filtering unsupported selector
types (+, >, [, :, #, ~, *, descendants) in processRuleBlockWithStyle.
Fix normalized() trailing whitespace to also strip newlines.
- PR #1018: Add deleteCache() to CssParser, move CSS_CACHE_VERSION to
static class member, remove stale cache on version mismatch, invalidate
section caches (Storage.removeDir) when CSS is rebuilt. Refactor
parseCssFiles() to early-return when cache exists.
- PR #990: Adapt classic theme continue-reading card width to cover
aspect ratio (clamped to 90% screen width), increase homeTopPadding
20->40, fix centering with rect.x offset for boxX/continueBoxX.
- #1002 alignment: Add tryInterpretLength() to skip non-numeric CSS
values (auto, inherit), add "both width and height set" image sizing
branch in ChapterHtmlSlimParser.
Co-authored-by: Cursor <cursoragent@cursor.com>
Parse CSS height/width into CssStyle for images and use aspect-ratio-
preserving logic when CSS dimensions are set. Falls back to viewport-fit
scaling when no CSS dimensions are present. Includes divide-by-zero
guards and viewport clamping with aspect ratio rescaling.
- Add imageHeight field to CssStyle/CssPropertyFlags
- Parse CSS height declarations into imageHeight
- Add imageHeight + width to cache serialization (bump cache v2->v3)
- Replace viewport-fit-only image scaling with CSS-aware sizing
Co-authored-by: Cursor <cursoragent@cursor.com>
Cherry-pick two bug fixes from upstream PR #992:
- fix(GfxRenderer): Null-safety in getSpaceWidth/getTextAdvanceX to
prevent Load access fault when bold/italic font variants lack certain
glyphs (upstream 3e2c518)
- fix(PNGdec): Increase PNG_MAX_BUFFERED_PIXELS to 16416 for 2048px
wide images and add pre-decode buffer overflow guard (upstream b8e743e)
Co-authored-by: Cursor <cursoragent@cursor.com>
- Remove stale Lyra3CoversTheme.h (functionality merged into LyraTheme)
- Fix UITheme.cpp to use LyraTheme for LYRA_3_COVERS theme variant
- Update open-x4-sdk submodule to 91e7e2b (drawImageTransparent support)
- Run clang-format on all source files
Co-authored-by: Cursor <cursoragent@cursor.com>
## Summary
**What is the goal of this PR?**
This change fixes an issue I noticed while reading where occasionally,
especially in italics, some words would have too much space between
them. The problem was that word width calculations were including any
negative X overhang, and combined with a space before the word, that can
lead to an inconsistently large space.
## Additional Context
Screenshots of some problematic text:
| In CrossPoint 1.0 | With this change |
| -- | -- |
| <img
src="https://github.com/user-attachments/assets/87bf0e4b-341f-4ba9-b3ea-38c13bd26363"
width="400" /> | <img
src="https://github.com/user-attachments/assets/bf11ba20-c297-4ce1-aa07-43477ef86fc2"
width="400" /> |
---
### 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?** (e.g., Implements the new feature for
file uploading.)
* Fixes:
https://github.com/crosspoint-reader/crosspoint-reader/issues/947
**What changes are included?**
* Check to see if there's free heap memory before processing CSS (should
we be doing this type of check or is it better to just crash if we
exhaust the memory?)
* Skip CSS files larger than 128kb
## Additional Context
* I found that a copy of `Release it` contained a 250kb+ CSS file, from
the homepage of the publisher. It has nothing to do with the epub, so we
should just skip it
* Major question: Are there better ways to detect CSS that doesn't
belong in a book, or is this size-based approach valid?
* Another question: Are there any epubs we know of that legitimately
include >128kb CSS files?
Code changes themselves created with an agent, all investigation and
write-up done by human. If you (the maintainers) would prefer a
different fix for this issue, let me know.
---
### 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 >**_
* **What is the goal of this PR?**
Add proper hyphenation support for the Ukrainian language.
* **What changes are included?**
- Added Ukrainian hyphenation rules/dictionary
---
Did you use AI tools to help write this code? _**NO**_
Port 6 upstream PRs (PR #939 was already ported):
- #852: Complete HalPowerManager with RAII Lock class, WiFi check in
setPowerSaving, skipLoopDelay overrides for ClearCache/OtaUpdate,
and power lock in Activity render task loops
- #965: Fix paragraph formatting inside list items by tracking
listItemUntilDepth to prevent unwanted line breaks
- #972: Micro-optimizations: std::move in insertFont, const ref for
getDataFromBook parameter
- #971: Remove redundant hasPrintableChars pre-rendering pass from
EpdFont, EpdFontFamily, and GfxRenderer
- #977: Skip unsupported image formats before extraction, add
PARSE_BUFFER_SIZE constant and chapter parse timing
- #975: Fix UITheme memory leak by replacing raw pointer with
std::unique_ptr for currentTheme
Co-authored-by: Cursor <cursoragent@cursor.com>
## Summary
**What is the goal of this PR?**
Skip constructing a `std::string` just to get the underlying `c_str()`
buffer, when a string literal gives the same end result.
---
### 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
* If an EPUB has:
```
<dc:creator>J.R.R. Tolkien</dc:creator>
<dc:creator>Christopher Tolkien</dc:creator>
```
the current result for epub.author would provide : "J.R.R.
TolkienChristopher Tolkien" (no separator!)
* The fix will seperate multiple authors: "J.R.R. Tolkien, Christopher
Tolkien"
## Additional Context
* Simple fix in ContentOpfParser - I am not seeing any dependence on the
wrong concatenated result.
---
### 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
This PR applies a micro optimization on `SerializedHyphenationPatterns`,
which allow reading `rootOffset` directly without having to parse then
cache it.
It should not affect storage space since no new bytes are added.
This also gets rid of the linear cache search whenever
`liangBreakIndexes` is called. In theory, the performance should be
improved a bit, although it may be too small to be noticeable in
practice.
master branch:
```
english: 99.1023%
french: 100%
german: 97.7289%
russian: 97.2167%
spanish: 99.0236%
```
This PR:
```
english: 99.1023%
french: 100%
german: 97.7289%
russian: 97.2167%
spanish: 99.0236%
```
---
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 - mostly IDE
tab-autocompletions
Remove empty sentinel BMP file from generateThumbBmp() that blocked
placeholder generation for books without covers. Add removeBook() to
RecentBooksStore and clear book from recents on cache delete. Ensure
home screen always generates placeholder when thumbnail generation fails.
Co-authored-by: Cursor <cursoragent@cursor.com>
Cherry-pick merge from pablohc/crosspoint-reader@2d8cbcf, based on
upstream PR #556 by martinbrook with pablohc's refresh optimization.
- Add JPEG decoder (picojpeg) and PNG decoder (PNGdec) with 4-level
grayscale Bayer dithering for e-ink display
- Add pixel caching system (.pxc files) for fast image re-rendering
- Integrate image extraction from EPUB HTML parser (<img> tag support)
- Add ImageBlock/PageImage types with serialization support
- Add image-aware refresh optimization (double FAST_REFRESH technique)
- Add experimental displayWindow() partial refresh support
- Bump section cache version 12->13 to invalidate stale caches
- Resolve TAG_PageImage=3 to avoid conflict with mod's TAG_PageTableRow=2
Co-authored-by: Cursor <cursoragent@cursor.com>
Replace the "[Table omitted]" placeholder with full table rendering:
- Two-pass layout: buffer table content during SAX parsing, then
calculate column widths and lay out cells after </table> closes
- Colspan support for cells spanning multiple columns
- Forced line breaks within cells (<br>, <p>, <div> etc.)
- Center-align full-width spanning rows (section headers/titles)
- Width hints from HTML attributes and CSS (col, td, th width)
- Two-pass fair-share column width distribution that prevents
narrow columns from being excessively squeezed
- Double-encoded entity handling
- PageTableRow with grid-line rendering and serialization support
- Asymmetric vertical cell padding to balance font leading
Co-authored-by: Cursor <cursoragent@cursor.com>
## Summary
Closes#766. Thank you for the help @bramschulting!
**What is the goal of this PR?**
- First and foremost, fix issue #766.
- Through working on that, I realized the current CSS parsing/loading
code can be improved dramatically for large files and still had
additional performance improvements to be made, even with EPUBs with
small CSS.
**What changes are included?**
- Stream CSS parsing and reuse normalization buffers to cut allocations
- Add rule limits and selector validation to release rules and free up
memory when needed
- Skip CSS parsing/loading entirely when "Book's Embedded Style" is off
## Additional Context
- My test EPUB has been updated
[here](https://github.com/jdk2pq/css-test-epub) to include a very large
CSS file to test this out
---
### 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**_, Codex
Add OMIT_BOOKERLY, OMIT_NOTOSANS, OMIT_OPENDYSLEXIC flags to
selectively exclude font families, and OMIT_HYPH_DE/EN/ES/FR/IT/RU
flags to exclude individual hyphenation language tries.
The mod build environment excludes OpenDyslexic (~1.03 MB) and all
hyphenation tries (~282 KB), reducing flash usage by ~1.3 MB.
Font Family setting switched from Enum to DynamicEnum with
index-to-value mapping to handle arbitrary font exclusion without
breaking the settings UI or persisted values.
Co-authored-by: Cursor <cursoragent@cursor.com>
## Summary
Closes#743.
**What is the goal of this PR?**
- Add back handling for HTML entities in expat. This was originally part
of the code that got removed
[here](https://github.com/crosspoint-reader/crosspoint-reader/pull/274)
- Handle ` ` characters to resolve issue #743
**What changes are included?**
- Brought back HTML entity table from previous commit and refactored it
to use a static const char * table with linear lookup to reduce heap
allocations.
- Used `XML_SetDefaultHandlerExpand` in expat to parse out the entities
correctly, without needing them defined in DOCTYPE
- Added handling for ` ` so that the text stays together and
doesn't break onto a new line with text separated by an ` `
## Additional Context
- This supersedes [this
PR](https://github.com/crosspoint-reader/crosspoint-reader/pull/751)
that simply handled `nbsp;` as whitespace. Instead, we want that
character to serve its true purpose and affect the line-breaking
algorithm.
- Updated my test EPUB [here](https://github.com/jdk2pq/css-test-epub)
with ` ` characters examples at the end of the book
---
### 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**_, Claude Code
## Summary
* Definition and use of a central LOG function, that can later be
extended or completely be removed (for public use where debugging
information may not be required) to save flash by suppressing the
-DENABLE_SERIAL_LOG like in the slim branch
* **What changes are included?**
## Additional Context
* By using the central logger the usual:
```
#include <HardwareSerial.h>
...
Serial.printf("[%lu] [WCS] Obfuscating/deobfuscating %zu bytes\n", millis(), data.size());
```
would then become
```
#include <Logging.h>
...
LOG_DBG("WCS", "Obfuscating/deobfuscating %zu bytes", data.size());
```
You do have ``LOG_DBG`` for debug messages, ``LOG_ERR`` for error
messages and ``LOG_INF`` for informational messages. Depending on the
verbosity level defined (see below) soe of these message types will be
suppressed/not-compiled.
* The normal compilation (default) will create a firmware.elf file of
42.194.356 bytes, the same code via slim will create 42.024.048 bytes -
170.308 bytes less
* Firmware.bin : 6.469.984 bytes for default, 6.418.672 bytes for slim -
51.312 bytes less
### 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_
---------
Co-authored-by: Xuan Son Nguyen <son@huggingface.co>
Implements StarDict-based dictionary lookup from the reader menu,
adapted from upstream PR #857 with /.dictionary/ folder path,
std::vector compatibility (PR #802), HTML definition rendering,
orientation-aware button hints, side button hints with CCW text
rotation, sparse index caching to SD card, pronunciation line
filtering, and reorganized reader menu with bookmark stubs.
Co-authored-by: Cursor <cursoragent@cursor.com>
Noticed that the Indexing... popup went missing despite 3-5 seconds
delay. Reducing to 10KB, so we get a popup for delays > ~2s.
### 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
Replace std::list with std::vector for the words, wordStyles,
wordXpos, and wordContinues containers in TextBlock and ParsedText.
Vectors provide contiguous memory layout for better cache locality
and O(1) random access, eliminating per-node heap allocation and
the 16-byte prev/next pointer overhead of doubly-linked list nodes.
The indexed access also removes the need for a separate continuesVec
copy that was previously built from the list for O(1) layout access.
## Summary
* **What is the goal of this PR?** Add Italian language hyphenation
support to improve text rendering for Italian books.
* **What changes are included?**
* Added Italian hyphenation trie (hyph-it.trie.h) generated from Typst's
hypher patterns
* Registered italianHyphenator in LanguageRegistry.cpp for language tag
it
* Added Italian to the hyphenation evaluation test suite
* Added Italian test data file with 5000 test cases
## 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? _**NO**_
---------
Co-authored-by: drbourbon <fabio@MacBook-Air-di-Fabio.local>
## Summary
- Closes#730
**What is the goal of this PR?**
- Adds percentage-based value support to CSS properties that accept
percentages (padding, margin, text-indent)
**What changes are included?**
- Adds `Percent` as another CSS unit
- Passes the viewport width to `fromCssStyle` so that we can resolve
percentage-based values
- Adds a fallback of using an emspace for text-indent if we have an
unresolvable value for whatever reason
## Additional Context
- This was missed in my CSS support feature, and the fallback when we
encounter a percentage value is to use px instead. This means 5% (which
would be ~30px on the screen) turns into 5px. When percentages are used
in `text-indent`, this fallback behavior makes the indent look like a
single space character. Whoops! 😬
My test EPUB has been updated
[here](https://github.com/jdk2pq/css-test-epub) with percentage based
CSS values at the end of the book.
---
### 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**_, Claude Code
I had an epub that just showed a blank cover and wouldnt work for the
sleep screen either, turns out it was an epub3 and I guess we didn't
support that. Super simple fix here
## Summary
Continue my changes to introduce the HAL infrastructure from
https://github.com/crosspoint-reader/crosspoint-reader/pull/522
This PR touches quite a lot of files, but most of them are just name
changing. It should not have any impacts to the end behavior.
## Additional Context
My plan is to firstly add this small shim layer, which sounds useless at
first, but then I'll implement an emulated driver which can be helpful
for testing and for development.
Currently, on my fork, I'm using a FS driver that allow "mounting" a
local directory from my computer to the device, much like the `-v` mount
option on docker. This allows me to quickly reset `.crosspoint`
directory if anything goes wrong. I plan to upstream this feature when
this PR get merged.
---
### 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
Closes#182. Closes#710. Closes#711.
## Summary
**What is the goal of this PR?**
- A longer-term, more robust fix for the issue with spurious spaces
appearing after style changes. Replaces solution from #694.
**What changes are included?**
- Add continuation flags to determine if to add a space after a word or
if the word connects to the previous word. Replaces simple solution that
only considered ending punctuation.
- Fixed an issue with greedy line-breaking algorithm where punctuation
could appear on the next line, separated from the word, if there was a
style change between the word and punctuation
---
### 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**_, Claude Code
Closes#712
## Summary
**What is the goal of this PR?**
- To add new settings for toggling on/off embedded CSS styles in the
reader. This gives more control and customization to the user over how
the ereader experience looks.
**What changes are included?**
- Added new "Embedded Style" option to the Reader settings
- Added new "Book's Style" option for "Paragraph Alignment"
- User's selected "Paragraph Alignment" will take precedence and
override the embedded CSS `text-align` property, _unless_ the user has
"Book's Style" set as their "Paragraph Alignment"
## Additional Context
https://github.com/user-attachments/assets/9e404b13-c7e0-41c7-9406-4715f389166a
Addresses feedback from the community about the new CSS feature:
https://github.com/crosspoint-reader/crosspoint-reader/pull/700
---
### 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**_, Claude Code
Fixes issue #182
## Summary
**What is the goal of this PR?**
When inline styles change mid-paragraph, words like periods, commas, and
quotes could end up as separate tokens. The justified text algorithm was
treating these as regular words, adding space before them.
**What changes are included?**
Now tracks which words are "attaching punctuation" (., , ! ? ; : " ' and
smart quotes) and excludes them from gap counting. These punctuation
marks attach directly to the preceding word without spacing.
## Additional Context
This is split out from code in #411 to address this comment
https://github.com/crosspoint-reader/crosspoint-reader/pull/411#discussion_r2751166631
---
### 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**_, Claude Code
## Summary
### What is the goal of this PR?
- Visual UI overhaul
- UI theme selection
### What changes are included?
- Added a setting "UI Theme": Classic, Lyra
- The classic theme is the current Crosspoint theme
- The Lyra theme implements these mockups:
https://www.figma.com/design/UhxoV4DgUnfrDQgMPPTXog/Lyra-Theme?node-id=2003-7596&t=4CSOZqf0n9uQMxDt-0
by Discord users yagofarias, ruby and gan_shu
- New functions in GFXRenderer to render rounded rectangles, greyscale
fills (using dithering) and thick lines
- Basic UI components are factored into BaseTheme methods which can be
overridden by each additional theme. Methods that are not overridden
will fallback to BaseTheme behavior. This means any new
features/components in CrossPoint only need to be developed for the
"Classic" BaseTheme.
- Additional themes can easily be developed by the community using this
foundation
## Additional Context
- Only the Home, Library and main Settings screens have been implemented
so far, this will be extended to the transfer screens and chapter
selection screen later on, but we need to get the ball rolling somehow
:)
- Loading extra covers on the home screen in the Lyra theme takes a
little more time (about 2 seconds), I added a loading bar popup (reusing
the Indexing progress bar from the reader view, factored into a neat UI
component) but the popup adds ~400ms to the loading time.
- ~~Home screen thumbnails will need to be generated separately for each
theme, because they are displayed in different sizes. Because we're
using dithering, displaying a thumb with the wrong size causes the
picture to look janky or dark as it does on the screenshots above. No
worries this will be fixed in a future PR.~~ Thumbs are now generated
with a size parameter
- UI Icons will need to be implemented in a future PR.
---
### 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**_
This is not a vibe coded PR. Copilot was used for autocompletion to save
time but I reviewed, understood and edited all generated code.
---------
Co-authored-by: Dave Allie <dave@daveallie.com>
## Summary
* **What is the goal of this PR?**
- Adds basic CSS parsing to EPUBs and determine the CSS rules when
rendering to the screen so that text is styled correctly. Currently
supports bold, underline, italics, margin, padding, and text alignment
## Additional Context
- My main reason for wanting this is that the book I'm currently
reading, Carl's Doomsday Scenario (2nd in the Dungeon Crawler Carl
series), relies _a lot_ on styled text for telling parts of the story.
When text is bolded, it's supposed to be a message that's rendered
"on-screen" in the story. When characters are "chatting" with each
other, the text is bolded and their names are underlined. Plus, normal
emphasis is provided with italicizing words here and there. So, this
greatly improves my experience reading this book on the Xteink, and I
figured it was useful enough for others too.
- For transparency: I'm a software engineer, but I'm mostly frontend and
TypeScript/JavaScript. It's been _years_ since I did any C/C++, so I
would not be surprised if I'm doing something dumb along the way in this
code. Please don't hesitate to ask for changes if something looks off. I
heavily relied on Claude Code for help, and I had a lot of inspiration
from how [microreader](https://github.com/CidVonHighwind/microreader)
achieves their CSS parsing and styling. I did give this as good of a
code review as I could and went through everything, and _it works on my
machine_ 😄
### Before
### After
---
### AI Usage
Did you use AI tools to help write this code? **YES**, Claude Code
## Summary
* refactors Indexing popups into ScreenComponents (they had different
implementations in different files)
* removes Indexing popup for small chapters
* only show Indexing popup (without progress bar) for large chapters
(using same minimum file size condition as for progress bar before)
## Additional Context
* Having to show even single popup message and redraw the screen slows
down the flow significantly
* Testing results:
* Opening large chapter with progress bar - 11 seconds
* Same chapter without progress bar, only single Indexing popup - 5
seconds
---
### AI Usage
Did you use AI tools to help write this code? _**< PARTIALLY>**_
## Summary
* Correctly render italics on image alt placeholders
* Parser incorrectly handled depth of self-closing tags
* Self-closing tags immediately call start and end tag
## Additional Context
* Previously, it would incorrectly make the whole chapter bold/italics,
or not italicised the image alt
---
### 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
Optimizes EPUB metadata indexing for large books (2000+ chapters) from
~30 minutes to ~50 seconds by replacing O(n²) algorithms with O(n log n)
hash-indexed lookups.
Fixes#134
## Problem
Three phases had O(n²) complexity due to nested loops:
| Phase | Operation | Before (2768 chapters) |
|-------|-----------|------------------------|
| OPF Pass | For each spine ref, scan all manifest items | ~25 min |
| TOC Pass | For each TOC entry, scan all spine items | ~5 min |
| buildBookBin | For each spine item, scan ZIP central directory | ~8.4
min |
Total: **~30+ minutes** for first-time indexing of large EPUBs.
## Solution
Replace linear scans with sorted hash indexes + binary search:
- **OPF Pass**: Build `{hash(id), len, offset}` index from manifest,
binary search for each spine ref
- **TOC Pass**: Build `{hash(href), len, spineIndex}` index from spine,
binary search for each TOC entry
- **buildBookBin**: New `ZipFile::fillUncompressedSizes()` API - single
ZIP central directory scan with batch hash matching
All indexes use FNV-1a hashing with length as secondary key to minimize
collisions. Indexes are freed immediately after each phase.
## Results
**Shadow Slave EPUB (2768 chapters):**
| Phase | Before | After | Speedup |
|-------|--------|-------|---------|
| OPF pass | ~25 min | 10.8 sec | ~140x |
| TOC pass | ~5 min | 4.7 sec | ~60x |
| buildBookBin | 506 sec | 34.6 sec | ~15x |
| **Total** | **~30+ min** | **~50 sec** | **~36x** |
**Normal EPUB (87 chapters):** 1.7 sec - no regression.
## Memory
Peak temporary memory during indexing:
- OPF index: ~33KB (2770 items × 12 bytes)
- TOC index: ~33KB (2768 items × 12 bytes)
- ZIP batch: ~44KB (targets + sizes arrays)
All indexes cleared immediately after each phase. No OOM risk on
ESP32-C3.
## Note on Threshold
All optimizations are gated by `LARGE_SPINE_THRESHOLD = 400` to preserve
existing behavior for small books. However, the algorithms work
correctly for any book size and are faster even for small books:
| Book Size | Old O(n²) | New O(n log n) | Improvement |
|-----------|-----------|----------------|-------------|
| 10 ch | 100 ops | 50 ops | 2x |
| 100 ch | 10K ops | 800 ops | 12x |
| 400 ch | 160K ops | 4K ops | 40x |
If preferred, the threshold could be removed to use the optimized path
universally.
## Testing
- [x] Shadow Slave (2768 chapters): 50s first-time indexing, loads and
navigates correctly
- [x] Normal book (87 chapters): 1.7s indexing, no regression
- [x] Build passes
- [x] clang-format passes
## Files Changed
- `lib/Epub/Epub/parsers/ContentOpfParser.h/.cpp` - OPF manifest index
- `lib/Epub/Epub/BookMetadataCache.h/.cpp` - TOC index + batch size
lookup
- `lib/ZipFile/ZipFile.h/.cpp` - New `fillUncompressedSizes()` API
- `lib/Epub/Epub.cpp` - Timing logs
<details>
<summary><b>Algorithm Details</b> (click to expand)</summary>
### Phase 1: OPF Pass - Manifest to Spine Lookup
**Problem**: Each `<itemref idref="ch001">` in spine must find matching
`<item id="ch001" href="...">` in manifest.
```
OLD: For each of 2768 spine refs, scan all 2770 manifest items
= 7.6M string comparisons
NEW: While parsing manifest, build index:
{ hash("ch001"), len=5, file_offset=120 }
Sort index, then binary search for each spine ref:
2768 × log₂(2770) ≈ 2768 × 11 = 30K comparisons
```
### Phase 2: TOC Pass - TOC Entry to Spine Index Lookup
**Problem**: Each TOC entry with `href="chapter0001.xhtml"` must find
its spine index.
```
OLD: For each of 2768 TOC entries, scan all 2768 spine entries
= 7.6M string comparisons
NEW: At beginTocPass(), read spine once and build index:
{ hash("OEBPS/chapter0001.xhtml"), len=25, spineIndex=0 }
Sort index, binary search for each TOC entry:
2768 × log₂(2768) ≈ 30K comparisons
Clear index at endTocPass() to free memory.
```
### Phase 3: buildBookBin - ZIP Size Lookup
**Problem**: Need uncompressed file size for each spine item (for
reading progress). Sizes are in ZIP central directory.
```
OLD: For each of 2768 spine items, scan ZIP central directory (2773 entries)
= 7.6M filename reads + string comparisons
Time: 506 seconds
NEW:
Step 1: Build targets from spine
{ hash("OEBPS/chapter0001.xhtml"), len=25, index=0 }
Sort by (hash, len)
Step 2: Single pass through ZIP central directory
For each entry:
- Compute hash ON THE FLY (no string allocation)
- Binary search targets
- If match: sizes[target.index] = uncompressedSize
Step 3: Use sizes array directly (O(1) per spine item)
Total: 2773 entries × log₂(2768) ≈ 33K comparisons
Time: 35 seconds
```
### Why Hash + Length?
Using 64-bit FNV-1a hash + string length as a composite key:
- Collision probability: ~1 in 2⁶⁴ × typical_path_lengths
- No string storage needed in index (just 12-16 bytes per entry)
- Integer comparisons are faster than string comparisons
- Verification on match handles the rare collision case
</details>
---
_AI-assisted development. All changes tested on hardware._
## Summary
* **What is the goal of this PR?** Add Spanish language hyphenation
support to improve text rendering for Spanish books.
* **What changes are included?**
- Added Spanish hyphenation trie (`hyph-es.trie.h`) generated from
Typst's hypher patterns
- Registered `spanishHyphenator` in `LanguageRegistry.cpp` for language
tag `es`
- Added Spanish to the hyphenation evaluation test suite
- Added Spanish test data file with 5000 test cases
## Additional Context
* **Test Results:** Spanish hyphenation achieves 99.02% F1 Score (97.72%
perfect matches out of 5000 test cases)
* **Compatibility:** Works automatically for EPUBs with
`<dc:language>es</dc:language>` (or es-ES, es-MX, etc.)
<img width="115" height="189" alt="imagen"
src="https://github.com/user-attachments/assets/9b92e7fc-b98d-48af-8d53-dfdc2e68abee"
/>
| Metric | Value |
|--------|-------|
| Perfect matches | 97.72% |
| Overall Precision | 99.33% |
| Overall Recall | 99.42% |
| Overall F1 Score | 99.38% |
---
### AI Usage
Did you use AI tools to help write this code? _**PARTIALLY**_
AI assisted with:
- Guiding and compile
- Preparing the PR description
## Summary
Should address issues discussed in #168 and potentially fix#478.
---
### 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
* Fixes#519
* Refactors repeated code into new function:
`ChapterHtmlSlimParser::flushPartWordBuffer()`
## Additional Context
* The `<br/>` tag is self closing and _in-line_, so the existing logic
for closing block tags does not get applied to `<br/>` tags.
* This PR adds the _in-line_ logic to:
* Flush the word preceding the `<br/>` tag from `partWordBuffer` to
`currentTextBlock` before calling `startNewTextBlock`
* **New function**: `ChapterHtmlSlimParser::flushPartWordBuffer()`
* **Purpose**: Consolidates the logic for flushing `partWordBuffer` to
`currentTextBlock`
* **Impact**: Simplifies `ChapterHtmlSlimParser::characterData(…)`,
`ChapterHtmlSlimParser::startElement(…)`, and
`ChapterHtmlSlimParser::endElement(…)` by integrating reused code into
single function
---
### 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 additional punctuation marks to the list of characters that can be
immediately followed by a line break even where there is no explicit
space
## Additional Context
* Huge appreciation to @osteotek for his amazing work on hyphenation.
Reading on the device is so much better now.
* I am getting bad line breaks when ellipses (…) are between words and
book file does not explicitly include some kind of breaking space.
* Per
[discussion](https://github.com/crosspoint-reader/crosspoint-reader/pull/305#issuecomment-3765411406),
several new characters are added in this PR to the `isExplicitHyphen`
list to allow line breaks immediately after them:
Character | Unicode | Usage | Why include it?
-- | -- | -- | --
Solidus (Slash) | U+002F | / | Essential for breaking URLs and "and/or"
constructs.
Backslash | U+005C | \ | Critical for technical text, file paths, and
coding documentation.
Underscore | U+005F | _ | Prevents "runaway" line lengths in usernames
or code snippets.
Middle Dot | U+00B7 | · | Acts as a semantic separator in dictionaries
or stylistic lists.
Ellipsis | U+2026 | … | Prevents justification failure when dialogue
lacks following spaces.
Midline Horizontal Ellipsis | U+22EF | ⋯ | Useful for mathematical
sequences and technical notation.
### Example:
This shows an example of what line breaking looks like *with* this PR.
Note the line break after "matter…" (which would not previously have
been allowed). It's particularly important here because the book
includes non-breaking spaces in "Mr. Aldrich" and "Mr. Rockefeller."
---
### 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
* Adds (optional) Hyphenation for English, French, German, Russian
languages
## Additional Context
* Included hyphenation dictionaries add approximately 280kb to the flash
usage (German alone takes 200kb)
* Trie encoded dictionaries are adopted from hypher project
(https://github.com/typst/hypher)
* Soft hyphens (and other explicit hyphens) take precedence over
dict-based hyphenation. Overall, the hyphenation rules are quite
aggressive, as I believe it makes more sense on our smaller screen.
---------
Co-authored-by: Dave Allie <dave@daveallie.com>
## Summary
- Adds KOReader progress sync integration, allowing CrossPoint to sync
reading positions with other
KOReader-compatible devices
- Stores credentials securely with XOR obfuscation
- Uses KOReader's partial MD5 document hashing for cross-device book
matching
- Syncs position via percentage with estimated XPath for compatibility
# Features
- Settings: KOReader Username, Password, and Authenticate options
- Sync from chapters menu: "Sync Progress" option appears when
credentials are configured
- Bidirectional sync: Can apply remote progress or upload local progress
---------
Co-authored-by: Dave Allie <dave@daveallie.com>
