lsm/PdfPig
1
0
Fork 0
mirror of https://github.com/UglyToad/PdfPig.git synced 2025-10-07 07:34:36 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

Fix #1148

Browse Source
This commit is contained in:
BobLd
2025-09-13 09:43:14 +01:00
parent 22eab422a3
commit d5b97065bd
4 changed files with 40 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
src/UglyToad.PdfPig.Tests/Integration/GithubIssuesTests.cs
View File

@@ -7,6 +7,28 @@
public class GithubIssuesTests
{
[Fact]
public void Issue1148()
{
var path = IntegrationHelpers.GetSpecificTestDocumentPath("P2P-33713919.pdf");
using (var document = PdfDocument.Open(path, new ParsingOptions() { UseLenientParsing = true }))
{
var page = document.GetPage(2);
var letters = page.Letters;
var words = NearestNeighbourWordExtractor.Instance.GetWords(letters).ToArray();
var firstTableLine = words[42];
Assert.EndsWith("C<--,:", firstTableLine.Text); // Just to make sure we are looking at the correct line. Text might change as this is not actually correct
Assert.Equal(firstTableLine.BoundingBox.BottomLeft, new PdfPoint(x: 31.890118, y: 693.035685));
Assert.Equal(firstTableLine.BoundingBox.BottomRight, new PdfPoint(x: 563.3851179999991, y: 693.035685));
}
}
[Fact]
public void Issue1122()
{

BIN
src/UglyToad.PdfPig.Tests/Integration/SpecificTestDocuments/P2P-33713919.pdf Normal file
View File

Binary file not shown.

5
src/UglyToad.PdfPig/PdfFonts/CidFonts/Type2CidFont.cs
View File

@@ -61,9 +61,12 @@
this.defaultWidth = defaultWidth;
this.cidToGid = cidToGid;
// TODO: This should maybe take units per em into account?
var scale = 1 / (double)(fontProgram?.GetFontMatrixMultiplier() ?? 1000);
FontMatrix = TransformationMatrix.FromValues(scale, 0, 0, scale, 0, 0);
// NB: For the font matrixPdfBox always return 1/1000 with the comment '1000 upem, this is not strictly true'
// see https://github.com/apache/pdfbox/blob/a5379f5588ee4c98222ee61366ad3d82e0f2264e/pdfbox/src/main/java/org/apache/pdfbox/pdmodel/font/PDCIDFontType2.java#L191
// Always using 1/1000 breaks the 'ReadWordsFromOldGutnishPage1' test
}
public double GetWidthFromFont(int characterIdentifier)

18
src/UglyToad.PdfPig/PdfFonts/Composite/Type0Font.cs
View File

@@ -126,13 +126,23 @@
// Get the bounding box in glyph space
var boundingBox = CidFont.GetBoundingBox(characterIdentifier);
boundingBox = CidFont.GetFontMatrix(characterIdentifier).Transform(boundingBox);
var fontMatrix = CidFont.GetFontMatrix(characterIdentifier);
boundingBox = fontMatrix.Transform(boundingBox);
var width = CidFont.GetWidthFromFont(characterIdentifier);
var advanceWidth = GetFontMatrix().TransformX(width);
// BobLD: Not sure why we don't need CidFont.GetFontMatrix(characterCode)
// Might be related to https://github.com/veraPDF/veraPDF-library/issues/1010
double scale = fontMatrix.A;
if (Math.Abs(scale - 0.001) < 0.0001)
{
// BobLD: The value of scale is close enough to 0.001 to be able to use 0.001.
// Still not sure what is the correct logic, but this hack fixes issue #1148 (while not breaking "Old Gutnish Internet Explorer.pdf")
//
// Based on https://martin.hoppenheit.info/blog/2018/pdfa-validation-and-inconsistent-glyph-width-information/
// which quotes: section 6.2.11.5 of ISO 19005-2:2011 (PDF/A-2) clarifies this issue: “For ISO 19005, consistent is defined to be a difference of no more than 1/1000 unit.”
scale = 0.001;
}
var advanceWidth = scale * width;
var result = new CharacterBoundingBox(boundingBox, advanceWidth);