From ed300116bd55e66e9a28b49d2263a02b8b55c785 Mon Sep 17 00:00:00 2001
From: Dirk Zimoch <dirk.zimoch@psi.ch>
Date: Tue, 21 May 2019 09:24:54 +0200
Subject: [PATCH] add tolower and toupper to regsubst

---
 docs/formats.html      | 12 +++++++++++-
 src/RegexpConverter.cc | 40 ++++++++++++++++++++++++++++++++++++----
 2 files changed, 47 insertions(+), 5 deletions(-)

diff --git a/docs/formats.html b/docs/formats.html
index c08b923..cac1788 100644
--- a/docs/formats.html
+++ b/docs/formats.html
@@ -601,7 +601,7 @@ It is only available if a PCRE library is installed.
 <div class="box">
 <p>
 If PCRE is not available for your host or cross architecture, download
-the sourcecode from <a target="ex" href="http://www.pcre.org/">www.pcre.org</a>
+the sourcecode from <a target="ex" href="https://www.pcre.org/">www.pcre.org</a>
 and try my EPICS compatible <a target="ex"
 href="http://epics.web.psi.ch/software/streamdevice/pcre/Makefile">Makefile</a>
 to compile it like a normal EPICS support module.
@@ -656,6 +656,13 @@ Matches of the <em>regex</em> are replaced by the string <em>subst</em> with all
 <code>\1</code> through <code>\9</code> replaced with the match of the corresponding
 sub-expression <span class="new"> if such a sub-expression exists.
 
+Occurrences of <code>\U<i>n</i></code>, <code>\L<i>n</i></code>, <code>\u<i>n</i></code>,
+or <code>\l<i>n</i></code> with <code><i>n</i></code> being a number <code>0</code>
+through <code>9</code> or <code>&</code> are replaced with the corresponding sub-expression
+converted to all upper case, all lower case, first letter upper case, or first letter lower
+case, respectively.</span>
+</p>
+<p><span class="new">
 Due to limitations of the parser, <code>\1</code> and <code>\x01</code> are the same 
 which makes it difficult to use literal bytes with values lower than 10 in <em>subst</em>.
 Therefore <code>\0</code> aways means a literal byte (incompatible change from earlier version!)
@@ -722,6 +729,9 @@ which is not at the end of a word.
 <code>%#/([^+-])*([+-])/\2\1/</code> moves a postfix sign to the front.
 (<code>1.23-</code> becomes <code>-1.23</code>)<br>
 </div>
+<div class="indent">
+<code>%#-2/.*/\U0/</code> converts the previous 2 characters to upper case.
+</div>
 <a name="mantexp"></a>
 <h2>15. MantissaExponent DOUBLE converter (<code>%m</code>)</h2>
 <p>
diff --git a/src/RegexpConverter.cc b/src/RegexpConverter.cc
index d6f5bf9..d3e9f2d 100644
--- a/src/RegexpConverter.cc
+++ b/src/RegexpConverter.cc
@@ -20,9 +20,10 @@
 
 #include "StreamFormatConverter.h"
 #include "StreamError.h"
-#include "string.h"
 #include "pcre.h"
+#include <string.h>
 #include <limits.h>
+#include <ctype.h>
 
 #define Z PRINTF_SIZE_T_PREFIX
 
@@ -219,9 +220,40 @@ static void regsubst(const StreamFormat& fmt, StreamBuffer& buffer, size_t start
                 if (s[r] == esc)
                 {
                     unsigned char ch = s[r+1];
-                    debug("found escaped \\%u, in range 1-%d?\n", ch, rc-1);
-                    if (ch != 0 && ch < rc) // escaped 1 - 9 : replace with subexpr
+                    if (strchr("ulUL", ch))
                     {
+                        unsigned char br = s[r+2] - '0';
+                        if (br == (unsigned char)('&'-'0')) br = 0;
+                        debug("found case conversion \\%c%u\n", ch, br);
+                        if (br >= rc)
+                        {
+                            s.remove(r, 1);
+                            continue;
+                        }
+                        br *= 2;
+                        rl = ovector[br+1] - ovector[br];
+                        s.replace(r, 3, buffer(start+c+ovector[br]), rl);
+                        switch (ch)
+                        {
+                            case 'u':
+                                if (islower(s[r])) s[r] = toupper(s[r]);
+                                break;
+                            case 'l':
+                                if (isupper(s[r])) s[r] = tolower(s[r]);
+                                break;
+                            case 'U':
+                                for (int i = 0; i < rl; i++)
+                                    if (islower(s[r+i])) s[r+i] = toupper(s[r+i]);
+                                break;
+                            case 'L':
+                                for (int i = 0; i < rl; i++)
+                                    if (isupper(s[r+i])) s[r+i] = tolower(s[r+i]);
+                                break;
+                        }
+                    }
+                    else if (ch != 0 && ch < rc) // escaped 1 - 9 : replace with subexpr
+                    {
+                        debug("found escaped \\%u\n", ch);
                         ch *= 2;
                         rl = ovector[ch+1] - ovector[ch];
                         debug("yes, replace \\%d: \"%s\"\n", ch/2, buffer.expand(start+c+ovector[ch], rl)());
@@ -230,7 +262,7 @@ static void regsubst(const StreamFormat& fmt, StreamBuffer& buffer, size_t start
                     }
                     else
                     {
-                        debug("no, use literal \\%u\n", ch);
+                        debug("use literal \\%u\n", ch);
                         s.remove(r, 1); // just remove escape
                     }
                 }