doc: document limitations with PCRE2_MATCH_INVALID_UTF

Closes: #187

Author: carenas

doc/pcre2api.3

@@ -1615,6 +1615,7 @@ PCRE2_EXTRA_MATCH_WORD are also supported. Any other options cause an error.
 .sp
 This option forces PCRE2_UTF (see below) and also enables support for matching
 by \fBpcre2_match()\fP in subject strings that contain invalid UTF sequences.
+(Assuming the matching text is suitably aligned).
 This facility is not supported for DFA matching. For details, see the
 .\" HREF
 \fBpcre2unicode\fP

doc/pcre2unicode.3

@@ -442,6 +442,12 @@ would match an instance of WORD that is surrounded by invalid UTF code units.
 Using PCRE2_MATCH_INVALID_UTF, an application can run matches on arbitrary
 data, knowing that any matched strings that are returned are valid UTF. This
 can be useful when searching for UTF text in executable or other binary files.
+.P
+The test for validity of Unicode characters is done for whole code units, so
+if the valid Unicode string is not aligned to a multiple of the code point size
+it won't be identified as valid. If that is a possibility on your data and
+unaligned access is not a problem in your system, you might want to try the
+matches with different starting byte offsets as a work around.
 .
 .
 .SH AUTHOR