C# 将每个单独单词的坐标提取到pdf文件中的文本块中
接下来,我试图获取C# 将每个单独单词的坐标提取到pdf文件中的文本块中,c#,pdf,itext,C#,Pdf,Itext,接下来,我试图获取文本块中的所有单词及其每个坐标(实际页面,顶部,底部,左侧,右侧) 由于textcunk可以是一个短语、一个单词或任何东西,因此我尝试手动执行此操作,依靠最后一个单词的矩形并每次剪切它。我注意到这个手动方法可能有很多问题(我需要手动计算特殊字符等等),所以我问自己ITextSharp是否提供了更简单的方法来执行这个操作 MyChunk和LocationTextExtractionsTramy继承的类如下所示: public class Chunk { public Gu
文本块
中的所有单词及其每个坐标(实际页面
,顶部
,底部
,左侧
,右侧
)
由于textcunk
可以是一个短语、一个单词或任何东西,因此我尝试手动执行此操作,依靠最后一个单词的矩形并每次剪切它。我注意到这个手动方法可能有很多问题(我需要手动计算特殊字符等等),所以我问自己ITextSharp是否提供了更简单的方法来执行这个操作
MyChunk
和LocationTextExtractionsTramy
继承的类如下所示:
public class Chunk
{
public Guid Id { get; set; }
public Rectangle Rect { get; set; }
public TextRenderInfo Render { get; set; }
public BaseFont BF { get; set; }
public string Text { get; set; }
public int FontSize { get; set; }
public Chunk(Rectangle rect, TextRenderInfo renderInfo)
{
this.Rect = rect;
this.Render = renderInfo;
this.Text = Render.GetText();
Initialize();
}
public Chunk(Rectangle rect, TextRenderInfo renderInfo, string text)
{
this.Rect = rect;
this.Render = renderInfo;
this.Text = text;
Initialize();
}
private void Initialize()
{
this.Id = Guid.NewGuid();
this.BF = Render.GetFont();
this.FontSize = ObtainFontSize();
}
private int ObtainFontSize()
{
return Convert.ToInt32(this.Render.GetSingleSpaceWidth() * 12 / this.BF.GetWidthPoint(" ", 12));
}
}
public class LocationTextExtractionPersonalizada : LocationTextExtractionStrategy
{
//Save each coordinate
public List<Chunk> ChunksInPage = new List<Chunk>();
//Automatically called on each chunk on PDF
public override void RenderText(TextRenderInfo renderInfo)
{
base.RenderText(renderInfo);
if (string.IsNullOrWhiteSpace(renderInfo.GetText())
|| renderInfo == null)
return;
//Get chunk Vectors
var bottomLeft = renderInfo.GetDescentLine().GetStartPoint();
var topRight = renderInfo.GetAscentLine().GetEndPoint();
//Create Rectangle based on previous Vectors
var rect = new Rectangle(
bottomLeft[Vector.I1],
bottomLeft[Vector.I2],
topRight[Vector.I1],
topRight[Vector.I2]);
if (rect == null)
return;
//Add each chunk with its coordinates
ChunksInPage.Add(new Chunk(rect, renderInfo));
}
}
公共类块
{
公共Guid Id{get;set;}
公共矩形Rect{get;set;}
公共文本RenderInfo呈现{get;set;}
公共BaseFont BF{get;set;}
公共字符串文本{get;set;}
公共int FontSize{get;set;}
公共块(矩形rect,TextRenderInfo renderInfo)
{
this.Rect=Rect;
this.Render=renderInfo;
this.Text=Render.GetText();
初始化();
}
公共块(矩形rect、TextRenderInfo renderInfo、字符串text)
{
this.Rect=Rect;
this.Render=renderInfo;
this.Text=文本;
初始化();
}
私有void初始化()
{
this.Id=Guid.NewGuid();
this.BF=Render.GetFont();
this.FontSize=获取FontSize();
}
私有int获取方大小()
{
返回Convert.ToInt32(this.Render.GetSingleSpaceWidth()*12/this.BF.GetWidthPoint(“,12));
}
}
公共类LocationTextExtractionPersonalizada:LocationTextExtractionStrategy
{
//保存每个坐标
public List ChunksInPage=new List();
//自动调用PDF上的每个块
公共覆盖无效RenderText(TextRenderInfo renderInfo)
{
base.RenderText(renderInfo);
if(string.IsNullOrWhiteSpace(renderInfo.GetText())
||renderInfo==null)
返回;
//获取块向量
var bottomLeft=renderInfo.GetDescentLine().GetStartPoint();
var topRight=renderInfo.GetAscentLine().GetEndPoint();
//基于前面的向量创建矩形
var rect=新矩形(
左下[Vector.I1],
左下[Vector.I2],
右上[Vector.I1],
右上[Vector.I2]);
if(rect==null)
返回;
//添加每个块及其坐标
添加(新块(rect,renderInfo));
}
}
因此,一旦我得到了文件等等,我会这样做:
private void ProcessContent()
{
for (int page= 1; page <= pdfReader.NumberOfPages; page++)
{
var strategy = new LocationTextExtractionPersonalizada();
var currentPageText = PdfTextExtractor.GetTextFromPage(
pdfReader,
page,
strategy);
//Here is where I want to get each word with its coordinates
var chunksWords= ChunkRawToWord(strategy.ChunksInPage);
}
}
private List<Chunk> ChunkRawToWord(IList<Chunk> chunks)
{
if (chunks == null || chunks[0] == null)
return null;
var words = new List<Chunk>();
//Poor RegEx pattern to get the word and its wathever
string pattern = @"[@&\w+]*(-*\/*\s*\:*\;*\,*\.*\(*\)*\%*\>*\<*)?";
var something = chunks[0].Render.GetCharacterRenderInfos();
for (int i = 0; i < chunks.Count; i++)
{
var wordsInChunk = Regex.Matches(
chunks[i].Text,
pattern,
RegexOptions.IgnoreCase);
var rectangleChunk = new Rectangle(chunks[i].Rect);
for (int j = 0; j < wordsInChunk.Count; j++)
{
if (string.IsNullOrWhiteSpace(wordsInChunk[j].Value))
continue;
var word = new Chunk(
rectangleChunk,
chunks[i].Render,
wordsInChunk[j].ToString());
if (j == 0)
{
word.Rect.Right = word.BF.GetWidthPoint(word.Text, word.FontSize);
words.Add(word);
continue;
}
if (words.Count <= 0)
continue;
word.Rect.Left = words[j - 1].Rect.Right;
word.Rect.Right = words[j - 1].Rect.Right + word.BF.GetWidthPoint(word.Text, word.FontSize);
words.Add(word);
}
}
return words;
}
private void ProcessContent()
{
对于(int page=1;page您可以使用方法textrendrinfo.GetCharacterRenderInfos()
获取区块中每个字符的TextRenderInfo
集合。然后可以将单个字符重新组合为单词,并使用该单词中第一个和最后一个TextRenderInfo
的坐标计算包含单词的矩形
在自定义文本提取策略中:
var _separators = new[] { "-", "(", ")", "/", " ", ":", ";", ",", "."};
protected virtual void ParseRenderInfo(TextRenderInfo currentInfo)
{
var resultInfo = new List<TextRenderInfo>();
var chars = currentInfo.GetCharacterRenderInfos();
foreach (var charRenderInfo in chars)
{
resultInfo.Add(charRenderInfo);
var currentChar = charRenderInfo.GetText();
if (_separators.Contains(currentChar))
{
ProcessWord(currentInfo, resultInfo);
resultInfo.Clear();
}
}
ProcessWord(currentInfo, resultInfo);
}
private void ProcessWord(TextRenderInfo charChunk, List<TextRenderInfo> wordChunks)
{
var firstRender = wordChunks.FirstOrDefault();
var lastRender = wordChunks.LastOrDefault();
if (firstRender == null || lastRender == null)
{
return;
}
var startCoords = firstRender.GetDescentLine().GetStartPoint();
var endCoords = lastRender.GetAscentLine().GetEndPoint();
var wordText = string.Join("", wordChunks.Select(x => x.GetText()));
var wordLocation = new LocationTextExtractionStrategy.TextChunkLocationDefaultImp(startCoords, endCoords, charChunk.GetSingleSpaceWidth());
_chunks.Add(new CustomTextChunk(wordText, wordLocation));
}
var_分隔符=new[]{-”,“(“,”,“/”,“,”:”,“;”,“,”,“,”};
受保护的虚拟void ParseRenderInfo(TextRenderInfo currentInfo)
{
var resultInfo=新列表();
var chars=currentInfo.getCharacterRenderInfo();
foreach(变量charRenderInfo,单位为chars)
{
结果信息添加(charRenderInfo);
var currentChar=charRenderInfo.GetText();
if(_separators.Contains(currentChar))
{
进程字(currentInfo、resultInfo);
resultInfo.Clear();
}
}
进程字(currentInfo、resultInfo);
}
私有void ProcessWord(TextRenderInfo charChunk,List wordchunk)
{
var firstRender=wordChunks.FirstOrDefault();
var lastdrender=wordChunks.LastOrDefault();
如果(firstRender==null | | lastRender==null)
{
返回;
}
var startCords=firstRender.GetDescentLine().GetStartPoint();
var endCoords=lastRender.GetAscentLine().GetEndPoint();
var wordText=string.Join(“,wordChunks.Select(x=>x.GetText());
var wordLocation=new LocationTextExtractionStrategy.TextChunkLocationDefaultImp(startWord、endCoords、charChunk.GetSingleSpaceWidth());
_添加(新的CustomTextChunk(wordText,wordLocation));
}
您可以使用方法TextRenderInfo.GetCharacterRenderInfos()
获取区块中每个字符的TextRenderInfo
集合。然后可以将单个字符重新组合为单词,并使用该单词中第一个和最后一个TextRenderInfo
的坐标计算包含单词的矩形
在自定义文本提取策略中:
var _separators = new[] { "-", "(", ")", "/", " ", ":", ";", ",", "."};
protected virtual void ParseRenderInfo(TextRenderInfo currentInfo)
{
var resultInfo = new List<TextRenderInfo>();
var chars = currentInfo.GetCharacterRenderInfos();
foreach (var charRenderInfo in chars)
{
resultInfo.Add(charRenderInfo);
var currentChar = charRenderInfo.GetText();
if (_separators.Contains(currentChar))
{
ProcessWord(currentInfo, resultInfo);
resultInfo.Clear();
}
}
ProcessWord(currentInfo, resultInfo);
}
private void ProcessWord(TextRenderInfo charChunk, List<TextRenderInfo> wordChunks)
{
var firstRender = wordChunks.FirstOrDefault();
var lastRender = wordChunks.LastOrDefault();
if (firstRender == null || lastRender == null)
{
return;
}
var startCoords = firstRender.GetDescentLine().GetStartPoint();
var endCoords = lastRender.GetAscentLine().GetEndPoint();
var wordText = string.Join("", wordChunks.Select(x => x.GetText()));
var wordLocation = new LocationTextExtractionStrategy.TextChunkLocationDefaultImp(startCoords, endCoords, charChunk.GetSingleSpaceWidth());
_chunks.Add(new CustomTextChunk(wordText, wordLocation));
}
var_分隔符=new[]{-”,“(“,”,“/”,“,”:”,“;”,“,”,“,”};
受保护的虚拟void ParseRenderInfo(TextRenderInfo currentInfo)
{
var resultInfo=新列表();
var chars=currentInfo.getCharacterRenderInfo();
foreach(变量charRenderInfo,单位为chars)
{
结果信息添加(charRenderInfo);
var currentChar=charRenderInfo.GetText();
if(_separators.Contains(currentChar))
{
进程字(currentInfo、resultInfo);
resultInfo.Clear();
}
}
进程字(currentInfo、resultInfo);
}
私有void ProcessWord(TextRenderInfo charChunk,List wordchunk)
{
var firstRender=wordChunks.FirstOrDefault();
var lastdrender=wordChunks.LastOrDefault();
如果(firstRender==null | | lastRender==null)
{