Thanks, I didn't know where that constant was coming from.<br><br>I am attaching the updated patch. The same description applies to it:<br><br>Checks were added for hexidecimal values that did not have valid<br>
characters, and for EOF received before the final closing brace of the<br>
rich text stream. The error values were tested on richedit versions 1,<br>
2, 3 & 4.1, and they were all the same for these cases.<br>
---<br>
dlls/riched20/editor.c | 7 +++++--<br>
dlls/riched20/reader.c | 8 +++++---<br>
2 files changed, 10 insertions(+), 5 deletions(-)<br><br><div class="gmail_quote">On Wed, Jun 18, 2008 at 2:41 AM, Rob Shearman <<a href="mailto:robertshearman@gmail.com">robertshearman@gmail.com</a>> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Dylan Smith wrote:<br>
> + if (!inStream.editstream->dwError && parser.stackTop > 0)<br>
> + inStream.editstream->dwError = -2147024858;<br>
<br>
This should be HRESULT_FROM_WIN32(ERROR_HANDLE_EOF).<br>
<br>
--<br>
<font color="#888888">Rob Shearman<br>
</font></blockquote></div><br>