lsm/PdfPig
1
0
Fork 0
mirror of https://github.com/UglyToad/PdfPig.git synced 2025-06-28 15:30:17 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

merge pull request #67 from BobLd/master

Browse Source 
Fix error in DocstrumBB
This commit is contained in:
Eliot Jones 2019-09-11 12:30:22 +01:00 committed by GitHub
commit f822ad48ea
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 7 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
src/UglyToad.PdfPig/DocumentLayoutAnalysis/DocstrumBB.cs
View File

@ -5,6 +5,7 @@ using System.Linq;
using System.Threading.Tasks; using System.Threading.Tasks;
using UglyToad.PdfPig.Content; using UglyToad.PdfPig.Content;
using UglyToad.PdfPig.Geometry; using UglyToad.PdfPig.Geometry;
using UglyToad.PdfPig.Util;
namespace UglyToad.PdfPig.DocumentLayoutAnalysis namespace UglyToad.PdfPig.DocumentLayoutAnalysis
{ {
@ -47,6 +48,8 @@ namespace UglyToad.PdfPig.DocumentLayoutAnalysis
public IReadOnlyList<TextBlock> GetBlocks(IEnumerable<Word> pageWords, double wlAngleLB, double wlAngleUB, public IReadOnlyList<TextBlock> GetBlocks(IEnumerable<Word> pageWords, double wlAngleLB, double wlAngleUB,
double blAngleLB, double blAngleUB, double blMultiplier) double blAngleLB, double blAngleUB, double blMultiplier)
{ {
if (pageWords.Count() == 0) return EmptyArray<TextBlock>.Instance;
var pageWordsArr = pageWords.Where(w => !string.IsNullOrWhiteSpace(w.Text)).ToArray(); // remove white spaces var pageWordsArr = pageWords.Where(w => !string.IsNullOrWhiteSpace(w.Text)).ToArray(); // remove white spaces
var withinLineDistList = new ConcurrentBag<double[]>(); var withinLineDistList = new ConcurrentBag<double[]>();
@ -102,7 +105,7 @@ namespace UglyToad.PdfPig.DocumentLayoutAnalysis
// 2. Rebuild lines, using max distance = +Inf as we know all words will be in the // 2. Rebuild lines, using max distance = +Inf as we know all words will be in the
// same block. Filtering will still be done based on angle. // same block. Filtering will still be done based on angle.
var mergedLines = GetLines(mergedWords.ToArray(), wlAngleLB, wlAngleUB, double.MaxValue); var mergedLines = GetLines(mergedWords.ToArray(), double.MaxValue, wlAngleLB, wlAngleUB);
blocks[b] = new TextBlock(mergedLines.ToList()); blocks[b] = new TextBlock(mergedLines.ToList());
// Remove // Remove

3
src/UglyToad.PdfPig/DocumentLayoutAnalysis/RecursiveXYCut.cs
View File

@ -2,6 +2,7 @@
using System.Collections.Generic; using System.Collections.Generic;
using System.Linq; using System.Linq;
using UglyToad.PdfPig.Content; using UglyToad.PdfPig.Content;
using UglyToad.PdfPig.Util;
namespace UglyToad.PdfPig.DocumentLayoutAnalysis namespace UglyToad.PdfPig.DocumentLayoutAnalysis
{ {
@ -64,6 +65,8 @@ namespace UglyToad.PdfPig.DocumentLayoutAnalysis
Func<IEnumerable<decimal>, decimal> dominantFontWidthFunc, Func<IEnumerable<decimal>, decimal> dominantFontWidthFunc,
Func<IEnumerable<decimal>, decimal> dominantFontHeightFunc) Func<IEnumerable<decimal>, decimal> dominantFontHeightFunc)
{ {
if (pageWords.Count() == 0) return EmptyArray<TextBlock>.Instance;
XYLeaf root = new XYLeaf(pageWords); // Create a root node. XYLeaf root = new XYLeaf(pageWords); // Create a root node.
XYNode node = VerticalCut(root, minimumWidth, dominantFontWidthFunc, dominantFontHeightFunc); XYNode node = VerticalCut(root, minimumWidth, dominantFontWidthFunc, dominantFontHeightFunc);