Fix recent documentation error

doc/html/pcre2_compile.html

@@ -98,8 +98,12 @@ private data structure that contains the compiled pattern, or NULL if an error
 was detected. In the error case, a text error message can be obtained by
 passing the value returned via the <i>errorcode</i> argument to the the
 <b>pcre2_get_error_message()</b> function. The offset (in code units) where the
-error was encountered is returned via the <i>erroroffset</i> argument. Both
+error was encountered is returned via the <i>erroroffset</i> argument.
-values are set to zero for a successful return.
+</P>
+<P>
+If there is no error, the value passed via <i>errorcode</i> returns the message
+"no error" if passed to <b>pcre2_get_error_message()</b>, and the value passed
+via <i>erroroffset</i> is zero.
 </P>
 <P>
 There is a complete description of the PCRE2 native API, with more detail on

doc/html/pcre2api.html

@@ -1383,8 +1383,7 @@ If <i>errorcode</i> or <i>erroroffset</i> is NULL, <b>pcre2_compile()</b> return
 NULL immediately. Otherwise, the variables to which these point are set to an
 error code and an offset (number of code units) within the pattern,
 respectively, when <b>pcre2_compile()</b> returns NULL because a compilation
-error has occurred. The values are both zero when compilation is successful
+error has occurred. 
-and <b>pcre2_compile()</b> returns a non-NULL value.
 </P>
 <P>
 There are nearly 100 positive error codes that <b>pcre2_compile()</b> may return
@@ -1399,15 +1398,18 @@ because the textual error messages that are obtained by calling the
 message"
 <a href="#geterrormessage">below)</a>
 should be self-explanatory. Macro names starting with PCRE2_ERROR_ are defined
-for both positive and negative error codes in <b>pcre2.h</b>.
+for both positive and negative error codes in <b>pcre2.h</b>. When compilation
+is successful <i>errorcode</i> is set to a value that returns the message "no
+error" if passed to <b>pcre2_get_error_message()</b>.
 </P>
 <P>
 The value returned in <i>erroroffset</i> is an indication of where in the
-pattern the error occurred. It is not necessarily the furthest point in the
+pattern an error occurred. When there is no error, zero is returned. A non-zero
-pattern that was read. For example, after the error "lookbehind assertion is
+value is not necessarily the furthest point in the pattern that was read. For
-not fixed length", the error offset points to the start of the failing
+example, after the error "lookbehind assertion is not fixed length", the error
-assertion. For an invalid UTF-8 or UTF-16 string, the offset is that of the
+offset points to the start of the failing assertion. For an invalid UTF-8 or
-first code unit of the failing character.
+UTF-16 string, the offset is that of the first code unit of the failing
+character.
 </P>
 <P>
 Some errors are not detected until the whole pattern has been scanned; in these

doc/pcre2.txt

@@ -1372,27 +1372,29 @@ COMPILING A PATTERN
        diately. Otherwise, the variables to which these point are  set  to  an
        error code and an offset (number of code units) within the pattern, re-
        spectively, when pcre2_compile() returns NULL because a compilation er-
-       ror has occurred. The values are both zero when compilation is success-
+       ror has occurred.
-       ful and pcre2_compile() returns a non-NULL value.
 
-       There are nearly 100 positive error codes that pcre2_compile() may  re-
+       There  are nearly 100 positive error codes that pcre2_compile() may re-
-       turn  if it finds an error in the pattern. There are also some negative
+       turn if it finds an error in the pattern. There are also some  negative
-       error codes that are used for invalid UTF strings when validity  check-
+       error  codes that are used for invalid UTF strings when validity check-
-       ing  is  in  force.  These  are  the same as given by pcre2_match() and
+       ing is in force. These are the  same  as  given  by  pcre2_match()  and
        pcre2_dfa_match(), and are described in the pcre2unicode documentation.
-       There  is  no  separate documentation for the positive error codes, be-
+       There is no separate documentation for the positive  error  codes,  be-
-       cause the textual error messages  that  are  obtained  by  calling  the
+       cause  the  textual  error  messages  that  are obtained by calling the
        pcre2_get_error_message() function (see "Obtaining a textual error mes-
-       sage" below) should be  self-explanatory.  Macro  names  starting  with
+       sage"  below)  should  be  self-explanatory.  Macro names starting with
-       PCRE2_ERROR_  are defined for both positive and negative error codes in
+       PCRE2_ERROR_ are defined for both positive and negative error codes  in
-       pcre2.h.
+       pcre2.h.  When  compilation  is  successful errorcode is set to a value
+       that returns the message "no error" if passed  to  pcre2_get_error_mes-
+       sage().
 
        The value returned in erroroffset is an indication of where in the pat-
-       tern  the  error  occurred. It is not necessarily the furthest point in
+       tern an error occurred. When there is no error,  zero  is  returned.  A
-       the pattern that was read. For example, after the error "lookbehind as-
+       non-zero  value  is  not  necessarily the furthest point in the pattern
-       sertion  is  not fixed length", the error offset points to the start of
+       that was read. For example, after the error  "lookbehind  assertion  is
-       the failing assertion. For an invalid UTF-8 or UTF-16 string, the  off-
+       not  fixed length", the error offset points to the start of the failing
-       set is that of the first code unit of the failing character.
+       assertion. For an invalid UTF-8 or UTF-16 string, the offset is that of
+       the first code unit of the failing character.
 
        Some  errors are not detected until the whole pattern has been scanned;
        in these cases, the offset passed back is the length  of  the  pattern.

doc/pcre2_compile.3

@@ -86,8 +86,11 @@ private data structure that contains the compiled pattern, or NULL if an error
 was detected. In the error case, a text error message can be obtained by
 passing the value returned via the \fIerrorcode\fP argument to the the
 \fBpcre2_get_error_message()\fP function. The offset (in code units) where the
-error was encountered is returned via the \fIerroroffset\fP argument. Both
+error was encountered is returned via the \fIerroroffset\fP argument.
-values are set to zero for a successful return.
+.P
+If there is no error, the value passed via \fIerrorcode\fP returns the message
+"no error" if passed to \fBpcre2_get_error_message()\fP, and the value passed
+via \fIerroroffset\fP is zero.
 .P
 There is a complete description of the PCRE2 native API, with more detail on
 each option, in the

doc/pcre2api.3

@@ -1323,8 +1323,7 @@ If \fIerrorcode\fP or \fIerroroffset\fP is NULL, \fBpcre2_compile()\fP returns
 NULL immediately. Otherwise, the variables to which these point are set to an
 error code and an offset (number of code units) within the pattern,
 respectively, when \fBpcre2_compile()\fP returns NULL because a compilation
-error has occurred. The values are both zero when compilation is successful
+error has occurred. 
-and \fBpcre2_compile()\fP returns a non-NULL value.
 .P
 There are nearly 100 positive error codes that \fBpcre2_compile()\fP may return
 if it finds an error in the pattern. There are also some negative error codes
@@ -1343,14 +1342,17 @@ message"
 below)
 .\"
 should be self-explanatory. Macro names starting with PCRE2_ERROR_ are defined
-for both positive and negative error codes in \fBpcre2.h\fP.
+for both positive and negative error codes in \fBpcre2.h\fP. When compilation
+is successful \fIerrorcode\fP is set to a value that returns the message "no
+error" if passed to \fBpcre2_get_error_message()\fP.
 .P
 The value returned in \fIerroroffset\fP is an indication of where in the
-pattern the error occurred. It is not necessarily the furthest point in the
+pattern an error occurred. When there is no error, zero is returned. A non-zero
-pattern that was read. For example, after the error "lookbehind assertion is
+value is not necessarily the furthest point in the pattern that was read. For
-not fixed length", the error offset points to the start of the failing
+example, after the error "lookbehind assertion is not fixed length", the error
-assertion. For an invalid UTF-8 or UTF-16 string, the offset is that of the
+offset points to the start of the failing assertion. For an invalid UTF-8 or
-first code unit of the failing character.
+UTF-16 string, the offset is that of the first code unit of the failing
+character.
 .P
 Some errors are not detected until the whole pattern has been scanned; in these
 cases, the offset passed back is the length of the pattern. Note that the