inteligr8/alfresco-transform-core
1
0
Fork 0
mirror of https://github.com/Alfresco/alfresco-transform-core.git synced 2025-08-14 17:58:27 +00:00
Code Issues Packages Projects Releases Wiki Activity

[MNT-22398] Switched methods. Logging

Browse Source
This commit is contained in:
Tiago Salvado
2022-08-29 17:50:48 +01:00
parent b3cbf91102
commit 8732b6f38c
1 changed files with 53 additions and 53 deletions
Download Patch File Download Diff File Expand all files Collapse all files

106
alfresco-transform-misc/alfresco-transform-misc/src/main/java/org/alfresco/transformer/transformers/TextToPdfContentTransformer.java
View File

@@ -181,6 +181,59 @@ public class TextToPdfContentTransformer implements SelectableTransformer
return new InputStreamReader(is);
}
/**
* Skips the BOM for UTF-8 encoding
*/
private InputStream handleUTF8BOM(InputStream is)
{
return new PushbackInputStream(is, UTF8_READ_AHEAD_BYTES)
{
boolean bomRead;
@Override
public int read(byte[] bytes, int off, int len) throws IOException
{
int i = 0;
int b = 0;
for (; i < len; i++)
{
b = read();
if (b == -1)
{
break;
}
bytes[off + i] = (byte) b;
}
return i == 0 && b == -1 ? -1 : i;
}
@Override
public int read() throws IOException
{
if (!bomRead)
{
bomRead = true;
byte[] bytes = new byte[UTF8_READ_AHEAD_BYTES];
int end = in.read(bytes, 0, UTF8_READ_AHEAD_BYTES);
if (bytes[0] == EF && bytes[1] == BB && bytes[2] == BF)
{
logger.warn("UTF-8 BOM detected, it will be skipped");
}
else
{
for (int i = end - 1; i >= 0; i--)
{
unread(bytes[i]);
}
}
}
return super.read();
}
};
}
/**
* Handles the situation where there is a BOM even though the encoding indicates that normally there should not be
* one for UTF-16BE and UTF-16LE. For extra flexibility includes UTF-16 too which optionally has the BOM. Rather
@@ -302,59 +355,6 @@ public class TextToPdfContentTransformer implements SelectableTransformer
};
}
/**
* Skips the BOM for UTF-8 encoding
*/
private InputStream handleUTF8BOM(InputStream is)
{
return new PushbackInputStream(is, UTF8_READ_AHEAD_BYTES)
{
boolean bomRead;
@Override
public int read(byte[] bytes, int off, int len) throws IOException
{
int i = 0;
int b = 0;
for (; i < len; i++)
{
b = read();
if (b == -1)
{
break;
}
bytes[off + i] = (byte) b;
}
return i == 0 && b == -1 ? -1 : i;
}
@Override
public int read() throws IOException
{
if (!bomRead)
{
bomRead = true;
byte[] bytes = new byte[UTF8_READ_AHEAD_BYTES];
int end = in.read(bytes, 0, UTF8_READ_AHEAD_BYTES);
if (bytes[0] == EF && bytes[1] == BB && bytes[2] == BF)
{
logger.warn("UTF-8 BOM detected");
}
else
{
for (int i = end - 1; i >= 0; i--)
{
unread(bytes[i]);
}
}
}
return super.read();
}
};
}
private static class PagedTextToPDF extends TextToPDF
{
// REPO-1066: duplicating the following lines from org.apache.pdfbox.tools.TextToPDF because they made them private