eric/Crosspoint
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge branch 'master' of https://github.com/jpirnay/crosspoint-reader into feat-rendering-perf

Browse Source
This commit is contained in:
jpirnay
2026-04-24 15:47:47 +02:00
parent 23a77c6aa5 4cc8bcbddd
commit c4498cd535
70 changed files with 3665 additions and 487 deletions
Download Patch File Download Diff File Expand all files Collapse all files
lib/Epub/Epub/FootnoteEntry.h
+5 -2
View File
@@ -2,9 +2,12 @@
#include <cstring>
#define FOOTNOTE_NUMBER_LEN 32
#define FOOTNOTE_HREF_LEN 96
struct FootnoteEntry {
char number[24];
char href[64];
char number[FOOTNOTE_NUMBER_LEN];
char href[FOOTNOTE_HREF_LEN];
FootnoteEntry() {
number[0] = '\0';
lib/Epub/Epub/Section.cpp
+1 -1
View File
@@ -12,7 +12,7 @@
#include "parsers/ChapterHtmlSlimParser.h"
namespace {
constexpr uint8_t SECTION_FILE_VERSION = 21;
constexpr uint8_t SECTION_FILE_VERSION = 22;
constexpr uint32_t HEADER_SIZE = sizeof(uint8_t) + // SECTION_FILE_VERSION
sizeof(int) + // fontId
sizeof(float) + // lineCompression
lib/Epub/Epub/parsers/ChapterHtmlSlimParser.cpp
+25 -30
View File
@@ -170,6 +170,17 @@ void ChapterHtmlSlimParser::flushPartWordBuffer() {
nextWordContinues = false;
}
// Emit the current page, keeping paragraphLutPerPage and completedPageCount in lockstep.
// Callers must ensure currentPage is non-null and carries content; the helper resets
// currentPage to a fresh Page and zeroes currentPageNextY so the caller can keep building.
void ChapterHtmlSlimParser::emitPage(uint32_t xhtmlByteOffset) {
paragraphLutPerPage.push_back({xhtmlByteOffset, xpathParagraphIndex});
completePageFn(std::move(currentPage));
completedPageCount++;
currentPage.reset(new Page());
currentPageNextY = 0;
}
// start a new text block if needed
void ChapterHtmlSlimParser::startNewTextBlock(const BlockStyle& blockStyle) {
nextWordContinues = false; // New block = new paragraph, no continuation
@@ -198,10 +209,7 @@ void ChapterHtmlSlimParser::startNewTextBlock(const BlockStyle& blockStyle) {
if (!pendingAnchorId.empty()) {
if (std::find(tocAnchors.begin(), tocAnchors.end(), pendingAnchorId) != tocAnchors.end()) {
if (currentPage && !currentPage->elements.empty()) {
completePageFn(std::move(currentPage));
completedPageCount++;
currentPage.reset(new Page());
currentPageNextY = 0;
emitPage(lastBodyChildByteOffset);
}
}
anchorData.push_back({std::move(pendingAnchorId), static_cast<uint16_t>(completedPageCount)});
@@ -218,10 +226,7 @@ void ChapterHtmlSlimParser::startNewTextBlock(const BlockStyle& blockStyle) {
if (!pendingAnchorId.empty() &&
std::find(tocAnchors.begin(), tocAnchors.end(), pendingAnchorId) != tocAnchors.end()) {
if (currentPage && !currentPage->elements.empty()) {
completePageFn(std::move(currentPage));
completedPageCount++;
currentPage.reset(new Page());
currentPageNextY = 0;
emitPage(lastBodyChildByteOffset);
}
}
// Record deferred anchor after previous block is flushed (and any TOC page break)
@@ -593,15 +598,11 @@ void XMLCALL ChapterHtmlSlimParser::startElement(void* userData, const XML_Char*
(self->currentPageNextY + totalImageHeightWithSpacing > self->viewportHeight)) {
LOG_DBG("EHP", "Image page break: currentY=%d needed=%d viewportH=%d", self->currentPageNextY,
totalImageHeightWithSpacing, self->viewportHeight);
self->paragraphLutPerPage.push_back({self->lastBodyChildByteOffset, self->xpathParagraphIndex});
self->completePageFn(std::move(self->currentPage));
self->completedPageCount++;
self->currentPage.reset(new Page());
self->emitPage(self->lastBodyChildByteOffset);
if (!self->currentPage) {
LOG_ERR("EHP", "Failed to create new page");
return;
}
self->currentPageNextY = 0;
} else if (!self->currentPage) {
self->currentPage.reset(new Page());
if (!self->currentPage) {
@@ -742,9 +743,9 @@ void XMLCALL ChapterHtmlSlimParser::startElement(void* userData, const XML_Char*
}
self->insideFootnoteLink = true;
self->footnoteLinkDepth = self->depth;
strncpy(self->currentFootnoteLinkHref, href, sizeof(self->currentFootnoteLinkHref) - 1);
self->currentFootnoteLinkHref[sizeof(self->currentFootnoteLinkHref) - 1] = '\0';
self->currentFootnoteLinkText[0] = '\0';
strncpy(self->currentFootnote.href, href, sizeof(self->currentFootnote.href) - 1);
self->currentFootnote.href[sizeof(self->currentFootnote.href) - 1] = '\0';
self->currentFootnote.number[0] = '\0';
self->currentFootnoteLinkTextLen = 0;
// Apply underline style to visually indicate the link
@@ -985,11 +986,11 @@ void XMLCALL ChapterHtmlSlimParser::characterData(void* userData, const XML_Char
}
// Extract footnote link text
for (int i = start; (self->currentFootnoteLinkTextLen < sizeof(self->currentFootnoteLinkText) - 1) && (i <= end);
for (int i = start; (self->currentFootnoteLinkTextLen < sizeof(self->currentFootnote.number) - 1) && (i <= end);
++i) {
self->currentFootnoteLinkText[self->currentFootnoteLinkTextLen++] = s[i];
self->currentFootnote.number[self->currentFootnoteLinkTextLen++] = s[i];
}
self->currentFootnoteLinkText[self->currentFootnoteLinkTextLen] = '\0';
self->currentFootnote.number[self->currentFootnoteLinkTextLen] = '\0';
}
for (int i = 0; i < len; i++) {
@@ -1220,11 +1221,11 @@ void XMLCALL ChapterHtmlSlimParser::endElement(void* userData, const XML_Char* n
// Closing a footnote link — create entry from collected text and href
if (self->insideFootnoteLink && self->depth == self->footnoteLinkDepth) {
if (self->currentFootnoteLinkText[0] != '\0' && self->currentFootnoteLinkHref[0] != '\0') {
if (self->currentFootnote.number[0] != '\0' && self->currentFootnote.href[0] != '\0') {
FootnoteEntry entry;
strncpy(entry.number, self->currentFootnoteLinkText, sizeof(entry.number) - 1);
strncpy(entry.number, self->currentFootnote.number, sizeof(entry.number) - 1);
entry.number[sizeof(entry.number) - 1] = '\0';
strncpy(entry.href, self->currentFootnoteLinkHref, sizeof(entry.href) - 1);
strncpy(entry.href, self->currentFootnote.href, sizeof(entry.href) - 1);
entry.href[sizeof(entry.href) - 1] = '\0';
int wordIndex =
self->wordsExtractedInBlock + (self->currentTextBlock ? static_cast<int>(self->currentTextBlock->size()) : 0);
@@ -1423,9 +1424,7 @@ bool ChapterHtmlSlimParser::parseAndBuildPages() {
anchorData.push_back({std::move(pendingAnchorId), static_cast<uint16_t>(completedPageCount)});
pendingAnchorId.clear();
}
paragraphLutPerPage.push_back({0u, xpathParagraphIndex}); // post-parse: no byte offset available
completePageFn(std::move(currentPage));
completedPageCount++;
emitPage(0u); // post-parse: no byte offset available
currentPage.reset();
currentTextBlock.reset();
}
@@ -1444,11 +1443,7 @@ ParsedText::LineProcessResult ChapterHtmlSlimParser::addLineToPage(std::shared_p
}
if (currentPageNextY + lineHeight > viewportHeight) {
paragraphLutPerPage.push_back({lastBodyChildByteOffset, xpathParagraphIndex});
completePageFn(std::move(currentPage));
completedPageCount++;
currentPage.reset(new Page());
currentPageNextY = 0;
emitPage(lastBodyChildByteOffset);
}
const bool noRoomForAnotherLine =
lib/Epub/Epub/parsers/ChapterHtmlSlimParser.h
+5 -2
View File
@@ -113,9 +113,8 @@ class ChapterHtmlSlimParser {
// Footnote link tracking
bool insideFootnoteLink = false;
int footnoteLinkDepth = -1;
char currentFootnoteLinkText[24] = {};
FootnoteEntry currentFootnote = {};
int currentFootnoteLinkTextLen = 0;
char currentFootnoteLinkHref[64] = {};
std::vector<std::pair<int, FootnoteEntry>> pendingFootnotes; // <wordIndex, entry>
int wordsExtractedInBlock = 0;
@@ -128,6 +127,10 @@ class ChapterHtmlSlimParser {
void startNewTextBlock(const BlockStyle& blockStyle);
void flushPartWordBuffer();
void makePages();
// Emit currentPage to the consumer while keeping paragraphLutPerPage and completedPageCount
// in lockstep. Every page break MUST go through this helper; open-coded completePageFn
// calls risk desynchronising paragraphLutPerPage and failing the size check in Section.cpp.
void emitPage(uint32_t xhtmlByteOffset);
// XML callbacks
static void XMLCALL startElement(void* userData, const XML_Char* name, const XML_Char** atts);
static void XMLCALL characterData(void* userData, const XML_Char* s, int len);