Skip to content

Move Otl's three non-shaping tenants out: OtlData, OtlTags and the debugOTL trace (#160) #198

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
jakejackson1 merged 4 commits into from
Sep 17, 2026
Merged

Move Otl's three non-shaping tenants out: OtlData, OtlTags and the debugOTL trace (#160) #198

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
c93ff46
Choose the script and language system tags a run is laid out with in …
jakejackson1 Sep 16, 2026
2442de1
Build the debugOTL shaping trace in OtlDump, which the shaper echoes …
jakejackson1 Sep 16, 2026
c11fa3b
Cut, trim and edit laid-out runs through OtlData rather than the shap…
jakejackson1 Sep 16, 2026
a398165
Walk the original Indic tags in one pass, and share the Bidi tests' c…
jakejackson1 Sep 16, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
26 changes: 14 additions & 12 deletions src/Bidi.php
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,6 +2,7 @@

namespace Mpdf;

use Mpdf\Shaper\OtlData;
use Mpdf\Utils\UtfString;

/**
Expand All @@ -11,8 +12,9 @@
* resolved and in string form. Nothing here reads a font table - this is the Unicode algorithm over
* the classes Ucdn supplies, which is why it sits beside Ucdn rather than under Fonts.
*
* prepare() takes an Otl because resolving the explicit embedding controls means deleting them from
* the text afterwards, and removing a character from text and its OTLdata together is Otl's job.
* prepare() takes the document's text encoding because resolving the explicit embedding controls
* means deleting them from the text afterwards, which Shaper\OtlData does to the text and its OTLdata
* together, counting characters in that encoding.
*
* @see https://www.unicode.org/reports/tr9/
*
Expand Down Expand Up @@ -445,7 +447,7 @@ public static function sort($ta, $str, $dir, &$chunkOTLdata, $useGPOS)
* from Mpdf::printbuffer(); reorder() then rearranges, from WriteFlowingBlock() and
* finishFlowingBlock(), once line breaks have divided the paragraph into lines.
*/
public static function prepare(&$para, $dir, Otl $otl)
public static function prepare(&$para, $dir, $encoding)
{

// Set the initial paragraph embedding level
Expand Down Expand Up @@ -653,11 +655,11 @@ public static function prepare(&$para, $dir, Otl $otl)
// Paragraph separators are not included in the embedding.
// X9. Remove all RLE, LRE, RLO, LRO, and PDF codes.
if ($controlchars) {
$otl->removeChar($para[$nc][0], $para[$nc][18], "\xe2\x80\xaa");
$otl->removeChar($para[$nc][0], $para[$nc][18], "\xe2\x80\xab");
$otl->removeChar($para[$nc][0], $para[$nc][18], "\xe2\x80\xac");
$otl->removeChar($para[$nc][0], $para[$nc][18], "\xe2\x80\xad");
$otl->removeChar($para[$nc][0], $para[$nc][18], "\xe2\x80\xae");
OtlData::removeChar($para[$nc][0], $para[$nc][18], "\xe2\x80\xaa", $encoding);
OtlData::removeChar($para[$nc][0], $para[$nc][18], "\xe2\x80\xab", $encoding);
OtlData::removeChar($para[$nc][0], $para[$nc][18], "\xe2\x80\xac", $encoding);
OtlData::removeChar($para[$nc][0], $para[$nc][18], "\xe2\x80\xad", $encoding);
OtlData::removeChar($para[$nc][0], $para[$nc][18], "\xe2\x80\xae", $encoding);
preg_replace("/\x{202a}-\x{202e}/u", '', $para[$nc][0]);
}
}
Expand Down Expand Up @@ -1011,10 +1013,10 @@ public static function prepare(&$para, $dir, Otl $otl)
$numchunks = count($para);
if ($controlchars) {
for ($nc = 0; $nc < $numchunks; $nc++) {
$otl->removeChar($para[$nc][0], $para[$nc][18], "\xe2\x81\xa6");
$otl->removeChar($para[$nc][0], $para[$nc][18], "\xe2\x81\xa7");
$otl->removeChar($para[$nc][0], $para[$nc][18], "\xe2\x81\xa8");
$otl->removeChar($para[$nc][0], $para[$nc][18], "\xe2\x81\xa9");
OtlData::removeChar($para[$nc][0], $para[$nc][18], "\xe2\x81\xa6", $encoding);
OtlData::removeChar($para[$nc][0], $para[$nc][18], "\xe2\x81\xa7", $encoding);
OtlData::removeChar($para[$nc][0], $para[$nc][18], "\xe2\x81\xa8", $encoding);
OtlData::removeChar($para[$nc][0], $para[$nc][18], "\xe2\x81\xa9", $encoding);
preg_replace("/\x{2066}-\x{2069}/u", '', $para[$nc][0]);
}
// Remove any blank chunks made by removing directional codes
Expand Down
62 changes: 30 additions & 32 deletions src/Mpdf.php
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,6 +13,7 @@
use Mpdf\Output\Destination;
use Mpdf\PsrLogAwareTrait\MpdfPsrLogAwareTrait;
use Mpdf\QrCode;
use Mpdf\Shaper\OtlData;
use Mpdf\Utils\Arrays;
use Mpdf\Utils\NumericString;
use Mpdf\Utils\UtfString;
Expand Down Expand Up @@ -6141,8 +6142,8 @@ function MultiCell(

/* -- OTL -- */
if (isset($OTLdata)) {
$tmpOTLdata = $this->otl->sliceOTLdata($OTLdata, $j, $i - $j);
$this->otl->trimOTLdata($tmpOTLdata, false, true);
$tmpOTLdata = OtlData::slice($OTLdata, $j, $i - $j);
OtlData::trim($tmpOTLdata, false, true);
$this->magic_reverse_dir($tmp, $directionality, $tmpOTLdata);
}
/* -- END OTL -- */
Expand Down Expand Up @@ -6191,8 +6192,8 @@ function MultiCell(

/* -- OTL -- */
if (isset($OTLdata)) {
$tmpOTLdata = $this->otl->sliceOTLdata($OTLdata, $j, $i - $j);
$this->otl->trimOTLdata($tmpOTLdata, false, true);
$tmpOTLdata = OtlData::slice($OTLdata, $j, $i - $j);
OtlData::trim($tmpOTLdata, false, true);
$this->magic_reverse_dir($tmp, $directionality, $tmpOTLdata);
}
/* -- END OTL -- */
Expand All @@ -6206,8 +6207,8 @@ function MultiCell(

/* -- OTL -- */
if (isset($OTLdata)) {
$tmpOTLdata = $this->otl->sliceOTLdata($OTLdata, $j, $sep - $j);
$this->otl->trimOTLdata($tmpOTLdata, false, true);
$tmpOTLdata = OtlData::slice($OTLdata, $j, $sep - $j);
OtlData::trim($tmpOTLdata, false, true);
}
/* -- END OTL -- */

Expand Down Expand Up @@ -6382,8 +6383,8 @@ function MultiCell(

/* -- OTL -- */
if (isset($OTLdata)) {
$tmpOTLdata = $this->otl->sliceOTLdata($OTLdata, $j, $i - $j);
$this->otl->trimOTLdata($tmpOTLdata, false, true);
$tmpOTLdata = OtlData::slice($OTLdata, $j, $i - $j);
OtlData::trim($tmpOTLdata, false, true);
$this->magic_reverse_dir($tmp, $directionality, $tmpOTLdata);
}
/* -- END OTL -- */
Expand Down Expand Up @@ -6646,7 +6647,7 @@ function finishFlowingBlock($endofblock = false, $next = '')
$content[count($content) - 1] = substr($content[count($content) - 1], 0, (strlen($content[count($content) - 1]) - $strip));
/* -- OTL -- */
if (isset($this->CurrentFont['useOTL']) && $this->CurrentFont['useOTL']) {
$this->otl->trimOTLdata($cOTLdata[count($cOTLdata) - 1], false, true);
OtlData::trim($cOTLdata[count($cOTLdata) - 1], false, true);
}
/* -- END OTL -- */
}
Expand Down Expand Up @@ -6735,9 +6736,9 @@ function finishFlowingBlock($endofblock = false, $next = '')
/* -- OTL -- */
// mPDF 5.7.1
if (isset($this->CurrentFont['useOTL']) && $this->CurrentFont['useOTL']) {
$this->otl->removeChar($chunk, $cOTLdata[$k], "\xc2\xad");
$this->otl->removeChar($chunk, $cOTLdata[$k], "\xe2\x80\x8b");
$this->otl->replaceSpace($chunk, $cOTLdata[$k]);
OtlData::removeChar($chunk, $cOTLdata[$k], "\xc2\xad", $this->mb_enc);
OtlData::removeChar($chunk, $cOTLdata[$k], "\xe2\x80\x8b", $this->mb_enc);
OtlData::nbspToSpace($chunk, $cOTLdata[$k], $this->mb_enc);
$content[$k] = $chunk;
} /* -- END OTL -- */ else { // *OTL*
$content[$k] = $chunk = str_replace("\xc2\xad", '', $chunk);
Expand Down Expand Up @@ -8381,7 +8382,7 @@ function WriteFlowingBlock($s, $sOTLdata)
$savedPreFont[] = $font[$cutcontentctr];
/* -- OTL -- */
if (!empty($sOTLdata)) {
$savedPreOTLdata[] = $this->otl->splitOTLdata($cOTLdata[$cutcontentctr], $cutcharctr, $cutcharctr);
$savedPreOTLdata[] = OtlData::split($cOTLdata[$cutcontentctr], $cutcharctr, $cutcharctr);
}
/* -- END OTL -- */

Expand All @@ -8402,7 +8403,7 @@ function WriteFlowingBlock($s, $sOTLdata)
}

if (!empty($sOTLdata)) {
$savedPreOTLdata[] = $this->otl->splitOTLdata($cOTLdata[(count($cOTLdata) - 1)], mb_strlen($currContent, $this->mb_enc));
$savedPreOTLdata[] = OtlData::split($cOTLdata[(count($cOTLdata) - 1)], mb_strlen($currContent, $this->mb_enc));
}

if (strpos($contentB[(count($contentB) - 1)], 'R') !== false) { // ???
Expand All @@ -8427,7 +8428,7 @@ function WriteFlowingBlock($s, $sOTLdata)
$next = count($savedPreContent) - 1;
$savedPreContent[$next] = '-' . $savedPreContent[$next];
if (!empty($savedPreOTLdata[$next])) {
$this->otl->prependOTLchar($savedPreOTLdata[$next], $charData, 'C');
OtlData::prependChar($savedPreOTLdata[$next], $charData, 'C');
}
}
}
Expand All @@ -8445,7 +8446,7 @@ function WriteFlowingBlock($s, $sOTLdata)
if (!is_array($breakfound)) {
$savedFont = $this->saveFont();
if (!empty($sOTLdata)) {
$savedOTLdata = $this->otl->splitOTLdata($cOTLdata[(count($cOTLdata) - 1)], mb_strlen($currContent, $this->mb_enc));
$savedOTLdata = OtlData::split($cOTLdata[(count($cOTLdata) - 1)], mb_strlen($currContent, $this->mb_enc));
}
}

Expand All @@ -8465,7 +8466,7 @@ function WriteFlowingBlock($s, $sOTLdata)
} // *CJK-FONTS*
/* -- OTL -- */
if (isset($this->CurrentFont['useOTL']) && $this->CurrentFont['useOTL']) {
$this->otl->trimOTLdata($cOTLdata[count($cOTLdata) - 1], false, true); // NB also does U+3000
OtlData::trim($cOTLdata[count($cOTLdata) - 1], false, true); // NB also does U+3000
}
/* -- END OTL -- */

Expand Down Expand Up @@ -8515,9 +8516,9 @@ function WriteFlowingBlock($s, $sOTLdata)
if (!$this->usingCoreFont) {
/* -- OTL -- */
if ((isset($this->CurrentFont['useOTL']) && $this->CurrentFont['useOTL']) || !empty($sOTLdata)) {
$this->otl->removeChar($chunk, $cOTLdata[$k], "\xc2\xad");
$this->otl->removeChar($chunk, $cOTLdata[$k], "\xe2\x80\x8b");
$this->otl->replaceSpace($chunk, $cOTLdata[$k]); // NBSP -> space
OtlData::removeChar($chunk, $cOTLdata[$k], "\xc2\xad", $this->mb_enc);
OtlData::removeChar($chunk, $cOTLdata[$k], "\xe2\x80\x8b", $this->mb_enc);
OtlData::nbspToSpace($chunk, $cOTLdata[$k], $this->mb_enc);
if (preg_match("/([" . $this->pregCURSchars . "])/u", $chunk)) {
$inclCursive = true;
}
Expand Down Expand Up @@ -8938,7 +8939,7 @@ function WriteFlowingBlock($s, $sOTLdata)
if ($this->checkCJK && $currContent == "\xe3\x80\x80") {
$currContent = '';
if (isset($this->CurrentFont['useOTL']) && $this->CurrentFont['useOTL']) {
$this->otl->trimOTLdata($cOTLdata[count($cOTLdata) - 1], true, false); // left trim U+3000
OtlData::trim($cOTLdata[count($cOTLdata) - 1], true, false); // left trim U+3000
}
}
/* -- END CJK-FONTS -- */
Expand Down Expand Up @@ -13734,7 +13735,7 @@ function WriteHTML($html, $mode = HTMLParserMode::DEFAULT_MODE, $init = true, $c
}
$e = $this->otl->applyOTL($e, $this->CurrentFont['useOTL']);
$this->OTLdata = $this->otl->OTLdata;
$this->otl->removeChar($e, $this->OTLdata, "\xef\xbb\xbf"); // Remove ZWNBSP (also Byte order mark FEFF)
OtlData::removeChar($e, $this->OTLdata, "\xef\xbb\xbf", $this->mb_enc); // Remove ZWNBSP (also Byte order mark FEFF)
} /* -- END OTL -- */
else {
// removes U+200E/U+200F LTR and RTL mark and U+200C/U+200D Zero-width Joiner and Non-joiner
Expand All @@ -13758,7 +13759,7 @@ function WriteHTML($html, $mode = HTMLParserMode::DEFAULT_MODE, $init = true, $c
if ($this->specialcontent == "type=select") {
$e = ltrim($e);
if (!empty($this->OTLdata)) {
$this->otl->trimOTLdata($this->OTLdata, true, false);
OtlData::trim($this->OTLdata, true, false);
} // *OTL*
$stringwidth = $this->GetStringWidth($e);
if (!isset($this->selectoption['MAXWIDTH']) || $stringwidth > $this->selectoption['MAXWIDTH']) {
Expand Down Expand Up @@ -13796,7 +13797,7 @@ function WriteHTML($html, $mode = HTMLParserMode::DEFAULT_MODE, $init = true, $c
if (($this->ignorefollowingspaces) && !$this->ispre) {
$e = ltrim($e);
if (!empty($this->OTLdata)) {
$this->otl->trimOTLdata($this->OTLdata, true, false);
OtlData::trim($this->OTLdata, true, false);
} // *OTL*
}
if ($e || $e === '0') {
Expand Down Expand Up @@ -13862,7 +13863,7 @@ function WriteHTML($html, $mode = HTMLParserMode::DEFAULT_MODE, $init = true, $c
if ($this->ignorefollowingspaces && !$this->ispre) {
$e = ltrim($e);
if (!empty($this->OTLdata)) {
$this->otl->trimOTLdata($this->OTLdata, true, false);
OtlData::trim($this->OTLdata, true, false);
} // *OTL*
}
if ($e || $e === '0') {
Expand Down Expand Up @@ -16259,10 +16260,7 @@ function printbuffer($arrayaux, $blockstate = 0, $is_table = false, $table_draft
// mPDF 6
// Process bidirectional text ready for bidi-re-ordering (which is done after line-breaks are established in WriteFlowingBlock etc.)
if (($blockdir == 'rtl' || $this->biDirectional) && !$table_draft) {
if (empty($this->otl)) {
$this->otl = new Otl($this, $this->fontCache);
}
Bidi::prepare($arrayaux, $blockdir, $this->otl);
Bidi::prepare($arrayaux, $blockdir, $this->mb_enc);
$array_size = count($arrayaux);
}

Expand Down Expand Up @@ -16310,7 +16308,7 @@ function printbuffer($arrayaux, $blockstate = 0, $is_table = false, $table_draft
if ($i == 0 && $vetor[0] != "\n" && ! $this->ispre) {
$vetor[0] = ltrim($vetor[0]);
if (!empty($vetor[18])) {
$this->otl->trimOTLdata($vetor[18], true, false);
OtlData::trim($vetor[18], true, false);
} // *OTL*
}

Expand Down Expand Up @@ -17055,7 +17053,7 @@ function PaintDivBB($divider = '', $blockstate = 0, $blvl = 0)
$legend = $this->blk[$blvl]['border_legend']; // Same structure array as textbuffer
$txt = $legend[0] = ltrim($legend[0]);
if (!empty($legend[18])) {
$this->otl->trimOTLdata($legend[18], true, false);
OtlData::trim($legend[18], true, false);
} // *OTL*
// Set font, size, style, color
$this->SetFont($legend[4], $legend[2], $legend[11]);
Expand Down Expand Up @@ -19207,7 +19205,7 @@ function TableCheckMinWidth($maxwidth, $forcewrap = 0, $textbuffer = [], $checkl
}
$line = trim($line);
if (!empty($OTLdata)) {
$this->otl->trimOTLdata($OTLdata, true, true);
OtlData::trim($OTLdata, true, true);
} // *OTL*
// SET FONT SIZE/STYLE from $chunk[n]
// FONTSIZE
Expand Down
Loading
Loading