feat: add methods in libary to accept custom stream. either as parser or custom stream

Author: galkahana

TextExtraction/TableExtraction.cpp

@@ -95,18 +95,21 @@ EStatusCode TableExtraction::ExtractTablePlacements(PDFParser* inParser, long in
 
 static const string scEmpty = "";
 
-EStatusCode TableExtraction::ExtractTables(const std::string& inFilePath, long inStartPage, long inEndPage) {
-    EStatusCode status = eSuccess;
-    InputFile sourceFile;
-
-    LatestWarnings.clear();
-    LatestError.code = eErrorNone;
-    LatestError.description = scEmpty;
-
+void TableExtraction::ClearState() {
     textsForPages.clear();
     tableLinesForPages.clear();
     tablesForPages.clear();
     mediaBoxesForPages.clear();
+    LatestWarnings.clear();
+    LatestError.code = eErrorNone;
+    LatestError.description = scEmpty;
+}
+
+EStatusCode TableExtraction::ExtractTables(const std::string& inFilePath, long inStartPage, long inEndPage) {
+    EStatusCode status = eSuccess;
+    InputFile sourceFile;
+
+    ClearState();
 
     do {
         status = sourceFile.OpenFile(inFilePath);
@@ -136,6 +139,43 @@ EStatusCode TableExtraction::ExtractTables(const std::string& inFilePath, long i
     return status;
 }
 
+PDFHummus::EStatusCode TableExtraction::ExtractTables(PDFParser* inParser, long inStartPage, long inEndPage) {
+    ClearState();
+
+    PDFHummus::EStatusCode status = ExtractTablePlacements(inParser, inStartPage, inEndPage);
+    if(status == eSuccess) {
+        ComposeTables();
+    }
+
+    return status;
+}
+
+PDFHummus::EStatusCode TableExtraction::ExtractTables(IByteReaderWithPosition* inStream, long inStartPage, long inEndPage)  {
+    EStatusCode status = eSuccess;
+
+    ClearState();
+
+    do {
+        PDFParser parser;
+        status = parser.StartPDFParsing(inStream);
+        if(status != eSuccess)
+        {
+            LatestError.code = eErrorInternalPDFWriter;
+            LatestError.description = string("Failed to parse file");
+            break;
+        }
+
+        status = ExtractTablePlacements(&parser, inStartPage, inEndPage);
+        if(status != eSuccess)
+            break;
+
+        ComposeTables();
+    } while(false);
+
+    return status;    
+}
+
+
 
 void TableExtraction::ComposeTables() {
     TableComposer tableComposer;

TextExtraction/TableExtraction.h

@@ -16,6 +16,7 @@
 #include "ErrorsAndWarnings.h"
 
 class PDFParser;
+class IByteReaderWithPosition;
 
 #include <sstream>
 #include <string>
@@ -33,6 +34,8 @@ class TableExtraction : public ITextInterpreterHandler, IGraphicContentInterpret
         virtual ~TableExtraction();
 
         PDFHummus::EStatusCode ExtractTables(const std::string& inFilePath, long inStartPage=0, long inEndPage=-1);
+        PDFHummus::EStatusCode ExtractTables(PDFParser* inParser, long inStartPage=0, long inEndPage=-1);
+        PDFHummus::EStatusCode ExtractTables(IByteReaderWithPosition* inStream, long inStartPage=0, long inEndPage=-1);
 
         ExtractionError LatestError;
         ExtractionWarningList LatestWarnings;  
@@ -65,5 +68,6 @@ class TableExtraction : public ITextInterpreterHandler, IGraphicContentInterpret
 
         PDFHummus::EStatusCode ExtractTablePlacements(PDFParser* inParser, long inStartPage, long inEndPage);
         void ComposeTables();
+        void ClearState();
 
 };

TextExtraction/TextExtraction.cpp

@@ -80,15 +80,18 @@ EStatusCode TextExtraction::ExtractTextPlacements(PDFParser* inParser, long inSt
 
 static const string scEmpty = "";
 
-EStatusCode TextExtraction::ExtractText(const std::string& inFilePath, long inStartPage, long inEndPage) {
-    EStatusCode status = eSuccess;
-    InputFile sourceFile;
-
+void TextExtraction::ClearState() {
+    textsForPages.clear();
     LatestWarnings.clear();
     LatestError.code = eErrorNone;
     LatestError.description = scEmpty;
+}
 
-    textsForPages.clear();
+EStatusCode TextExtraction::ExtractText(const std::string& inFilePath, long inStartPage, long inEndPage) {
+    EStatusCode status = eSuccess;
+    InputFile sourceFile;
+
+    ClearState();
 
     do {
         status = sourceFile.OpenFile(inFilePath);
@@ -117,6 +120,37 @@ EStatusCode TextExtraction::ExtractText(const std::string& inFilePath, long inSt
     return status;
 }
 
+PDFHummus::EStatusCode TextExtraction::ExtractText(PDFParser* inParser, long inStartPage, long inEndPage) {
+    ClearState();
+
+    return ExtractTextPlacements(inParser, inStartPage, inEndPage);
+}
+
+PDFHummus::EStatusCode TextExtraction::ExtractText(IByteReaderWithPosition* inStream, long inStartPage, long inEndPage) {
+    EStatusCode status = eSuccess;
+    InputFile sourceFile;
+
+    ClearState();
+
+    do {
+        PDFParser parser;
+        status = parser.StartPDFParsing(inStream);
+        if(status != eSuccess)
+        {
+            LatestError.code = eErrorInternalPDFWriter;
+            LatestError.description = string("Failed to parse file");
+            break;
+        }
+
+        status = ExtractTextPlacements(&parser, inStartPage, inEndPage);
+        if(status != eSuccess)
+            break;
+
+    } while(false);
+
+    return status;
+}
+
 static const string scCRLN = "\r\n";
 
 std::string TextExtraction::GetResultsAsText(int bidiFlag, TextComposer::ESpacing spacingFlag) {

TextExtraction/TextExtraction.h

@@ -11,6 +11,7 @@
 #include "ErrorsAndWarnings.h"
 
 class PDFParser;
+class IByteReaderWithPosition;
 
 #include <sstream>
 #include <string>
@@ -27,6 +28,8 @@ class TextExtraction : public ITextInterpreterHandler, IGraphicContentInterprete
         virtual ~TextExtraction();
 
         PDFHummus::EStatusCode ExtractText(const std::string& inFilePath, long inStartPage=0, long inEndPage=-1);
+        PDFHummus::EStatusCode ExtractText(PDFParser* inParser, long inStartPage=0, long inEndPage=-1);
+        PDFHummus::EStatusCode ExtractText(IByteReaderWithPosition* inStream, long inStartPage=0, long inEndPage=-1);
 
         ExtractionError LatestError;
         ExtractionWarningList LatestWarnings;  
@@ -55,4 +58,5 @@ class TextExtraction : public ITextInterpreterHandler, IGraphicContentInterprete
         double currentPageScopeBox[4];
 
         PDFHummus::EStatusCode ExtractTextPlacements(PDFParser* inParser, long inStartPage, long inEndPage);
+        void ClearState();
 };