javier/Stirling-PDF
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Fixes and others (#83)

Browse Source 
Features
-------------
Custom application name via APP_NAME docker env
(These next 3 are done with OCRMyPDF)
Extra features to OCR for scanned page cleanup (tilt/noise fixing)
Adding OCR ability to read and output to text file
Added Dedicated PDF/A conversion page 

Bug fixes
--------------
Fix concurrent calls on Libre and OCRMyPDF
jbig fix for compressions
Fix for compression metadata issues due to forced conversions to PDF/A


Other
--------
Removal of UK US language and just using "English" due to extra development time
Still issue with concurrent files for PDF to image... will fix later sorry
This commit is contained in:
Anthony Stirling
2023-04-01 21:02:54 +01:00
committed by GitHub
parent 0b4e3de455
commit 6d5dbd9729
23 changed files with 531 additions and 537 deletions
Download Patch File Download Diff File Expand all files Collapse all files

131
src/main/java/stirling/software/SPDF/utils/ProcessExecutor.java
View File

@@ -1,65 +1,100 @@
package stirling.software.SPDF.utils;
import java.io.BufferedReader;
import java.util.concurrent.ConcurrentHashMap;
import java.io.IOException;
import java.io.InputStreamReader;
import java.nio.charset.StandardCharsets;
import java.util.ArrayList;
import java.util.List;
import java.util.Map;
import java.util.concurrent.Semaphore;
public class ProcessExecutor {
public static int runCommandWithOutputHandling(List<String> command) throws IOException, InterruptedException {
ProcessBuilder processBuilder = new ProcessBuilder(command);
Process process = processBuilder.start();
public enum Processes {
LIBRE_OFFICE,
OCR_MY_PDF
}
// Read the error stream and standard output stream concurrently
List<String> errorLines = new ArrayList<>();
List<String> outputLines = new ArrayList<>();
private static final Map<Processes, ProcessExecutor> instances = new ConcurrentHashMap<>();
Thread errorReaderThread = new Thread(() -> {
try (BufferedReader errorReader = new BufferedReader(new InputStreamReader(process.getErrorStream(), StandardCharsets.UTF_8))) {
String line;
while ((line = errorReader.readLine()) != null) {
errorLines.add(line);
}
} catch (IOException e) {
e.printStackTrace();
}
});
Thread outputReaderThread = new Thread(() -> {
try (BufferedReader outputReader = new BufferedReader(new InputStreamReader(process.getInputStream(), StandardCharsets.UTF_8))) {
String line;
while ((line = outputReader.readLine()) != null) {
outputLines.add(line);
}
} catch (IOException e) {
e.printStackTrace();
}
});
private final Semaphore semaphore;
errorReaderThread.start();
outputReaderThread.start();
// Wait for the conversion process to complete
int exitCode = process.waitFor();
// Wait for the reader threads to finish
errorReaderThread.join();
outputReaderThread.join();
if (outputLines.size() > 0) {
String outputMessage = String.join("\n", outputLines);
System.out.println("Command output:\n" + outputMessage);
}
if (errorLines.size() > 0) {
String errorMessage = String.join("\n", errorLines);
System.out.println("Command error output:\n" + errorMessage);
if (exitCode != 0) {
throw new IOException("Command process failed with exit code " + exitCode + ". Error message: " + errorMessage);
}
}
private ProcessExecutor(int semaphoreLimit) {
this.semaphore = new Semaphore(semaphoreLimit);
}
public static ProcessExecutor getInstance(Processes processType) {
return instances.computeIfAbsent(processType, key -> {
int semaphoreLimit = switch (key) {
case LIBRE_OFFICE -> 1;
case OCR_MY_PDF -> 2;
};
return new ProcessExecutor(semaphoreLimit);
});
}
public int runCommandWithOutputHandling(List<String> command) throws IOException, InterruptedException {
int exitCode = 1;
semaphore.acquire();
try {
System.out.print("Running command: " + String.join(" ", command));
ProcessBuilder processBuilder = new ProcessBuilder(command);
Process process = processBuilder.start();
// Read the error stream and standard output stream concurrently
List<String> errorLines = new ArrayList<>();
List<String> outputLines = new ArrayList<>();
Thread errorReaderThread = new Thread(() -> {
try (BufferedReader errorReader = new BufferedReader(new InputStreamReader(process.getErrorStream(), StandardCharsets.UTF_8))) {
String line;
while ((line = errorReader.readLine()) != null) {
errorLines.add(line);
}
} catch (IOException e) {
e.printStackTrace();
}
});
Thread outputReaderThread = new Thread(() -> {
try (BufferedReader outputReader = new BufferedReader(new InputStreamReader(process.getInputStream(), StandardCharsets.UTF_8))) {
String line;
while ((line = outputReader.readLine()) != null) {
outputLines.add(line);
}
} catch (IOException e) {
e.printStackTrace();
}
});
errorReaderThread.start();
outputReaderThread.start();
// Wait for the conversion process to complete
exitCode = process.waitFor();
// Wait for the reader threads to finish
errorReaderThread.join();
outputReaderThread.join();
if (outputLines.size() > 0) {
String outputMessage = String.join("\n", outputLines);
System.out.println("Command output:\n" + outputMessage);
}
if (errorLines.size() > 0) {
String errorMessage = String.join("\n", errorLines);
System.out.println("Command error output:\n" + errorMessage);
if (exitCode != 0) {
throw new IOException("Command process failed with exit code " + exitCode + ". Error message: " + errorMessage);
}
}
} finally {
semaphore.release();
}
return exitCode;
}