typo fixed

Split long_format into signed_format and unsigned_format to hangle sign extension from EPICS LONG or ULONG to long or unsigned long correctly on systems with 64 bit long.
fix tests
2015-09-15 07:21:41 +00:00 · 2015-08-31 15:25:23 +00:00 · 2015-08-18 14:34:11 +00:00 · 2015-08-18 14:33:48 +00:00 · 2015-08-18 14:32:24 +00:00 · 2015-08-18 14:30:54 +00:00
65 changed files with 591 additions and 306 deletions
--- a/doc/PSI.gif
+++ b/doc/PSI.gif
--- a/doc/PSI.png
+++ b/doc/PSI.png
--- a/doc/aai.html
+++ b/doc/aai.html
@ -10,7 +10,7 @@
 </head>
 <body>
 <iframe src="nav.html" id="navleft"></iframe>
-<h1>StreamDevice: aai Records</h1>
+<h1>aai Records</h1>
 <p>
 <b>Note:</b> aai record support is disabled per default.
 Enable it in <code>src/CONFIG_STREAM</code>.
--- a/doc/aao.html
+++ b/doc/aao.html
@ -10,7 +10,7 @@
 </head>
 <body>
 <iframe src="nav.html" id="navleft"></iframe>
-<h1>StreamDevice: aao Records</h1>
+<h1>aao Records</h1>
 <p>
 <b>Note:</b> aao record support is disabled per default.
 Enable it in <code>src/CONFIG_STREAM</code>.
--- a/doc/ai.html
+++ b/doc/ai.html
@ -10,7 +10,7 @@
 </head>
 <body>
 <iframe src="nav.html" id="navleft"></iframe>
-<h1>StreamDevice: ai Records</h1>
+<h1>ai Records</h1>

 <h2>Normal Operation</h2>
 <p>
@ -41,6 +41,12 @@ written or read value.
   (=20.0/0xFFFF) maps 0x0000 to -10.0, 0x7FFF to 0.0 and 0xFFFF to 10.0.
   Using unsigned formats with values &ge; 0x800000 gives different results
   on 64 bit machines.
+   <p class="new">
+   If <code>LINR=="NO CONVERSION"</code> (the default), <code>VAL</code>
+   is directly converted from and to <code>long</code> without going through
+   <code>RVAL</code>. This allows for more bits on 64 bit machines.
+   To get the old behavior, use <code>LINR=="LINEAR"</code>.
+   </p>
  </dd>
 <dt>ENUM format (e.g. <code>%{</code>):</dt>
  <dd>
--- a/doc/ao.html
+++ b/doc/ao.html
@ -10,7 +10,7 @@
 </head>
 <body>
 <iframe src="nav.html" id="navleft"></iframe>
-<h1>StreamDevice: ao Records</h1>
+<h1>ao Records</h1>

 <h2>Normal Operation</h2>
 <p>
@ -40,6 +40,12 @@ written or read value.
   (=20.0/0xFFFF) maps -10.0 to 0x0000, 0.0 to 0x7FFF and 10.0 to 0xFFFF.
   Using unsigned formats with values &ge; 0x800000 gives different results
   on 64 bit machines.
+   <p class="new">
+   If <code>LINR=="NO CONVERSION"</code> (the default), <code>OVAL</code>
+   is directly converted to <code>long</code> without going through
+   <code>RVAL</code>. This allows for more bits on 64 bit machines.
+   To get the old behavior, use <code>LINR=="LINEAR"</code>.
+   </p>
  </dd>
 <dt>ENUM format (e.g. <code>%{</code>):</dt>
  <dd>
--- a/doc/bi.html
+++ b/doc/bi.html
@ -10,7 +10,7 @@
 </head>
 <body>
 <iframe src="nav.html" id="navleft"></iframe>
-<h1>StreamDevice: bi Records</h1>
+<h1>bi Records</h1>

 <h2>Normal Operation</h2>
 <p>
--- a/doc/bo.html
+++ b/doc/bo.html
@ -10,7 +10,7 @@
 </head>
 <body>
 <iframe src="nav.html" id="navleft"></iframe>
-<h1>StreamDevice: bo Records</h1>
+<h1>bo Records</h1>

 <h2>Normal Operation</h2>
 <p>
--- a/doc/calcout.html
+++ b/doc/calcout.html
@ -10,7 +10,7 @@
 </head>
 <body>
 <iframe src="nav.html" id="navleft"></iframe>
-<h1>StreamDevice: calcout Records</h1>
+<h1>calcout Records</h1>

 <p>
 <b>Note:</b> Device support for calcout records is only available for
--- a/doc/epics3_13.html
+++ b/doc/epics3_13.html
@ -10,7 +10,7 @@
 </head>
 <body>
 <iframe src="nav.html" id="navleft"></iframe>
-<h1>StreamDevice: Using EPICS 3.13</h1>
+<h1>Using EPICS 3.13</h1>

 <a name="pre"></a>
 <h2>1. Prerequisites</h2>
--- a/doc/formatconverter.html
+++ b/doc/formatconverter.html
@ -10,7 +10,7 @@
 </head>
 <body>
 <iframe src="nav.html" id="navleft"></iframe>
-<h1>StreamDevice: Format Converter API</h1>
+<h1>Format Converter API</h1>

 <a name="class"></a>
 <h2>Converter Class</h2>
--- a/doc/formats.html
+++ b/doc/formats.html
@ -10,7 +10,7 @@
 </head>
 <body>
 <iframe src="nav.html" id="navleft"></iframe>
-<h1>StreamDevice: Format Converters</h1>
+<h1>Format Converters</h1>

 <a name="syntax"></a>
 <h2>1. Format Syntax</h2>
@ -158,6 +158,9 @@ attribute (see
 target="ex">Record Reference Manual</a>), the record will be processed.
 It is your responsibility that the data type of the record field is
 compatible to the the data type of the converter.
+<span class="new">STRING formats are compatible with arrays of CHAR or UCHAR.<span>
+</p>
+<p>
 Note that using this syntax is by far not as efficient as using the
 default field.
 At the moment it is not possible to set <code>otherrecord</code> to an alarm
@ -195,7 +198,7 @@ is accepted.
 </p>
 <p class="new">
 When a maximum field width is given, leading whitespace only counts to the
-field witdth when the space flag is given.
+field witdth when the space flag is used.
 </p>

 <a name="stdl"></a>
@ -210,6 +213,10 @@ field witdth when the space flag is given.
 <p>
 With the <code>#</code> flag, octal values are prefixed with <code>0</code>
 and hexadecimal values with <code>0x</code> or <code>0X</code>.
+<p class="new">
+Unlike printf, <code>%x</code> and <code>%X</code> truncate the
+output to the the given width (number of least significant half bytes).
+</p>
 </p>
 <p>
 <b>Input:</b> <code>%d</code> matches signed decimal, <code>%u</code> matches
@ -230,7 +237,7 @@ is accepted.
 </p>
 <p class="new">
 When a maximum field width is given, leading whitespace only counts to the
-field witdth when the space flag is given.
+field witdth when the space flag is used.
 </p>

 <a name="stds"></a>
@ -248,7 +255,7 @@ The default <em>width</em> is infinite for <code>%s</code> and
 1 for <code>%c</code>.
 Leading whitespaces are skipped with <code>%s</code> 
 <span class="new">
-except when the space flag is given</span>
+except when the space flag is used</span>
 but not with <code>%c</code>.
 The empty string matches.
 </p>
@ -283,23 +290,35 @@ The strings are separated by <code>|</code>.
 Example: <code>%{OFF|STANDBY|ON}</code> mapps the string <code>OFF</code>
 to the value 0, <code>STANDBY</code> to 1 and <code>ON</code> to 2.
 </p>
-<p>
+<p class="new">
 When using the <code>#</code> flag it is allowed to assign integer values
 to the strings using <code>=</code>.
+Unassigned strings increment their values by 1 as usual.
+</p>
+<p class="new">
+If one string is the initial substing of another, the substing must come
+later to ensure correct matching.
+In particular if one string is the emptry string, it must be the last one.
+Use <code>#</code> and <code>=</code> to renumber if necessary.
+</p>
+<p class="new">
+Use the assignment <code>=?</code> for the last string to make it the
+default value for output formats.
 </p>
 <p>
-Example: <code>%#{neg=-1|stop=0|pos=1|fast=10}</code>.
+Example: <code>%#{neg=-1|stop|pos|fast=10|rewind=-10}</code>.
 </p>
 <p>
 If one of the strings contains <code>|</code> or <code>}</code>
-(or <code>=</code>  if the <code>#</code> flag is used)
+<span class="new">(or <code>=</code>  if the <code>#</code> flag is used)</span>
 a <code>\</code> must be used to escape the character.
 </p>
 <p>
-In output, depending on the value, one of the strings is printed.
+<b>Output:</b> Depending on the value, one of the strings is printed,
+<span class="new">or the default if no value matches</span>.
 </p>
 <p>
-In input, if any of the strings matches the value is set accordingly.
+<b>Input:</b> If any of the strings matches, the value is set accordingly.
 </p>

 <a name="bin"></a>
--- a/doc/longin.html
+++ b/doc/longin.html
@ -10,7 +10,7 @@
 </head>
 <body>
 <iframe src="nav.html" id="navleft"></iframe>
-<h1>StreamDevice: longin Records</h1>
+<h1>longin Records</h1>

 <h2>Normal Operation</h2>
 <p>
--- a/doc/longout.html
+++ b/doc/longout.html
@ -10,7 +10,7 @@
 </head>
 <body>
 <iframe src="nav.html" id="navleft"></iframe>
-<h1>StreamDevice: longout Records</h1>
+<h1>longout Records</h1>

 <h2>Normal Operation</h2>
 <p>
--- a/doc/makepdf
+++ b/doc/makepdf
@ -0,0 +1,35 @@
+PAGES="
+index.html
+setup.html
+epics3_13.html
+protocol.html
+formats.html
+processing.html
+recordtypes.html
+aai.html
+aao.html
+ai.html
+ao.html
+bi.html
+bo.html
+mbbi.html
+mbbo.html
+mbbiDirect.html
+mbboDirect.html
+stringin.html
+stringout.html
+longin.html
+longout.html
+waveform.html
+calcout.html
+scalcout.html
+tipsandtricks.html
+recordinterface.html
+businterface.html
+formatconverter.html
+osinterface.html
+"
+
+rm -f stream.pdf
+wkhtmltopdf --print-media-type --dpi 1200 --zoom 0.85 --page-size Letter \
+$PAGES stream.pdf
--- a/doc/mbbi.html
+++ b/doc/mbbi.html
@ -10,7 +10,7 @@
 </head>
 <body>
 <iframe src="nav.html" id="navleft"></iframe>
-<h1>StreamDevice: mbbi Records</h1>
+<h1>mbbi Records</h1>

 <h2>Normal Operation</h2>
 <p>
--- a/doc/mbbiDirect.html
+++ b/doc/mbbiDirect.html
@ -10,7 +10,7 @@
 </head>
 <body>
 <iframe src="nav.html" id="navleft"></iframe>
-<h1>StreamDevice: mbbiDirect Records</h1>
+<h1>mbbiDirect Records</h1>

 <h2>Normal Operation</h2>
 <p>
--- a/doc/mbbo.html
+++ b/doc/mbbo.html
@ -10,7 +10,7 @@
 </head>
 <body>
 <iframe src="nav.html" id="navleft"></iframe>
-<h1>StreamDevice: mbbo Records</h1>
+<h1>mbbo Records</h1>

 <h2>Normal Operation</h2>
 <p>
--- a/doc/mbboDirect.html
+++ b/doc/mbboDirect.html
@ -10,7 +10,7 @@
 </head>
 <body>
 <iframe src="nav.html" id="navleft"></iframe>
-<h1>StreamDevice: mbboDirect Records</h1>
+<h1>mbboDirect Records</h1>

 <h2>Normal Operation</h2>
 <p>
--- a/doc/osinterface.html
+++ b/doc/osinterface.html
@ -10,7 +10,7 @@
 </head>
 <body>
 <iframe src="nav.html" id="navleft"></iframe>
-<h1>StreamDevice: Operating System API</h1>
+<h1>Operating System API</h1>

 <h2>Sorry, this documentation is still missing.</h2>

--- a/doc/processing.html
+++ b/doc/processing.html
@ -10,7 +10,7 @@
 </head>
 <body>
 <iframe src="nav.html" id="navleft"></iframe>
-<h1>StreamDevice: Record Processing</h1>
+<h1>Record Processing</h1>

 <a name="proc"></a>
 <h2>1. Normal Processing</h2>
--- a/doc/protocol.html
+++ b/doc/protocol.html
@ -10,7 +10,7 @@
 </head>
 <body>
 <iframe src="nav.html" id="navleft"></iframe>
-<h1>StreamDevice: Protocol Files</h1>
+<h1>Protocol Files</h1>
 <a name="gen"></a>
 <h2>1. General Information</h2>
 <p>
--- a/doc/recordinterface.html
+++ b/doc/recordinterface.html
@ -10,7 +10,7 @@
 </head>
 <body>
 <iframe src="nav.html" id="navleft"></iframe>
-<h1>StreamDevice: Record API</h1>
+<h1>Record API</h1>

 <h2>Sorry, this documentation is still missing.</h2>

--- a/doc/recordtypes.html
+++ b/doc/recordtypes.html
@ -10,7 +10,7 @@
 </head>
 <body>
 <iframe src="nav.html" id="navleft"></iframe>
-<h1>StreamDevice: Record Types</h1>
+<h1>Record Types</h1>

 <h2>Supported Record Types</h2>
 <p>
--- a/doc/scalcout.html
+++ b/doc/scalcout.html
@ -10,7 +10,7 @@
 </head>
 <body>
 <iframe src="nav.html" id="navleft"></iframe>
-<h1>StreamDevice: scalcout Records</h1>
+<h1>scalcout Records</h1>

 <p>
 <b>Note:</b> The scalcout record is part of the <i>calc</i> module of
--- a/doc/setup.html
+++ b/doc/setup.html
@ -10,7 +10,7 @@
 </head>
 <body>
 <iframe src="nav.html" id="navleft"></iframe>
-<h1>StreamDevice: Setup</h1>
+<h1>Setup</h1>

 <a name="pre"></a>
 <h2>1. Prerequisites</h2>
--- a/doc/stream.css
+++ b/doc/stream.css
@ -5,8 +5,9 @@ a:hover   {color: #FF0000;}
 body {
    margin-right:1em;
    margin-left:15em;
-    margin-top:11ex;
-    font-family: "HelveticaNeue-Light", "Helvetica Neue Light", "Helvetica Neue", Helvetica, Arial, "Lucida Grande", sans-serif;
+    margin-top:70px;
+    padding-top:1px;
+    font-family: Helvetica, Arial, sans-serif;
    font-size:14px;
    background-color:#ffffff;
 }
@ -30,26 +31,26 @@ dt {
 }

 h1 {
-    font-size:225%;
+    font-size:250%;
    margin-top:0;
    font-style:italic;
-    font-family:serif;
+    font-weight:bold;
+    font-family:"Times New Roman", serif;
    text-align:center;
    position:fixed;
    top:0;
    left:0;
    width:100%;
-    min-height:60px;
-    height: 4ex;
    line-height:190%;
    background-color:white;
-    border-top:6px solid #1b4486;
-    border-bottom:4px solid #1b4486;
+    border-width:0;
+    border-bottom:3px solid #1b4486;
    white-space:nowrap;
-    background-image:url(PSI.gif);
+    background-image:url(PSI.png);
    background-repeat:no-repeat;
-    background-position:10px 2px;
-    text-shadow: .1em .1em .1em lightgray;           
+    background-position:10px 5px;
+    text-shadow:.1em .1em .1em darkgray;
+    box-shadow:0 .3em .1em -.2em darkgray;
 }

 h2 {
@ -91,7 +92,7 @@ code {
    position:fixed;
    left:0;
    top:0;
-    padding-top:9ex;
+    padding-top:70px;
    width:14em;
    height:100%;
    border-style:solid;
@ -115,13 +116,9 @@ a[target=ex]:hover:after {
 }

@media print {
-  a:link    {color: black; text-decoration:none;}
-  a:visited {color: black; text-decoration:none;}
-  a:hover   {color: black; text-decoration:none;}
-  a[target=ex] {text-decoration:underline;}
+  a:link    {text-decoration:none;}
  a[target=ex]:after {content:" [" attr(href) "]";}
-  code      {color: black; }
-  body      {margin-left:10px;}
-  h1        {position:absolute;}
+  body      {margin:0 4em;}
+  h1        {position:relative; background-position:0 0;}
  #navleft  {display:none;}
 }
--- a/doc/stringin.html
+++ b/doc/stringin.html
@ -10,7 +10,7 @@
 </head>
 <body>
 <iframe src="nav.html" id="navleft"></iframe>
-<h1>StreamDevice: stringin Records</h1>
+<h1>stringin Records</h1>

 <h2>Normal Operation</h2>
 <p>
--- a/doc/stringout.html
+++ b/doc/stringout.html
@ -10,7 +10,7 @@
 </head>
 <body>
 <iframe src="nav.html" id="navleft"></iframe>
-<h1>StreamDevice: stringout Records</h1>
+<h1>stringout Records</h1>

 <h2>Normal Operation</h2>
 <p>
--- a/doc/tipsandtricks.html
+++ b/doc/tipsandtricks.html
@ -10,7 +10,7 @@
 </head>
 <body>
 <iframe src="nav.html" id="navleft"></iframe>
-<h1>StreamDevice: Tips and Tricks</h1>
+<h1>Tips and Tricks</h1>

 <a name="argvar"></a>
 <h2>I have many almost identical protocols</h2>
@ -281,7 +281,7 @@ With some more records, you can clean the message record if SEVR is not INVALID.
 <code>
 record (calcout, "$(DEVICE):clean_1") {<br>
 &nbsp;&nbsp;field (INPA, "$(DEVICE):readcurrent.SEVR CP")<br>
-&nbsp;&nbsp;field (CALC, "A!=2")<br>
+&nbsp;&nbsp;field (CALC, "A#3")<br>
 &nbsp;&nbsp;field (OOPT, "When Non-zero")<br>
 &nbsp;&nbsp;field (OUT,  "$(DEVICE):clean_2.PROC")<br>
 }<br>
--- a/doc/waveform.html
+++ b/doc/waveform.html
@ -10,7 +10,7 @@
 </head>
 <body>
 <iframe src="nav.html" id="navleft"></iframe>
-<h1>StreamDevice: waveform Records</h1>
+<h1>waveform Records</h1>

 <h2>Normal Operation</h2>
 <p>
--- a/14
+++ b/14
@ -1,12 +1,10 @@
 include /ioc/tools/driver.makefile
 EXCLUDE_VERSIONS = 3.13.2
-PROJECT=stream2
+PROJECT=stream
 BUILDCLASSES += Linux

 #DOCUDIR = doc

-DBDS = stream.dbd
-
 BUSSES  += AsynDriver
 BUSSES  += Dummy

@ -42,9 +40,9 @@ HEADERS += StreamBuffer.h
 HEADERS += StreamError.h

 ifeq (${EPICS_BASETYPE},3.13)
-USR_INCLUDES += -include $(INSTALL_INCLUDE)/compat3_13.h
-endif
-ifeq (${EPICS_BASETYPE},3.14)
+# old gcc needs full path for -include
+CXXFLAGS += -include $(foreach d,${INCLUDES:-I%=%},$(wildcard $d/compat3_13.h))
+else
 RECORDTYPES += calcout
 endif

@ -54,5 +52,7 @@ streamReferences:
 	perl ../src/makeref.pl Interface $(BUSSES) > $@
 	perl ../src/makeref.pl Converter $(FORMATS) >> $@

-stream.dbd:
+# have to hack a bit to work with both versions of driver.makefile
+DBDFILES = O.$${EPICSVERSION}_$${T_A}/streamSup.dbd
+../O.${EPICSVERSION}_${T_A}/streamSup.dbd:
 	perl ../src/makedbd.pl $(RECORDTYPES) > $@
--- a/src/AsynDriverInterface.cc
+++ b/src/AsynDriverInterface.cc
@ -680,7 +680,7 @@ writeHandler()
            clientName());
        status = pasynOctet->read(pvtOctet, pasynUser,
            buffer, sizeof(buffer), &received, &eomReason);
-        if (received == 0) break;
+        if (status == asynError || received == 0) break;
 #ifndef NO_TEMPORARY
        if (received) debug("AsynDriverInterface::writeHandler(%s): flushing %ld bytes: \"%s\"\n",
            clientName(), (long)received, StreamBuffer(buffer, received).expand()());
--- a/src/BCDConverter.cc
+++ b/src/BCDConverter.cc
@ -31,9 +31,9 @@ class BCDConverter : public StreamFormatConverter
 };

 int BCDConverter::
-parse(const StreamFormat&, StreamBuffer&, const char*&, bool)
+parse(const StreamFormat& fmt, StreamBuffer&, const char*&, bool)
 {
-    return long_format;
+    return (fmt.flags & sign_flag) ? signed_format : unsigned_format;
 }

 bool BCDConverter::
--- a/src/BinaryConverter.cc
+++ b/src/BinaryConverter.cc
@ -19,6 +19,7 @@
 ***************************************************************/

 #include <ctype.h>
+#include <limits.h>
 #include "StreamFormatConverter.h"
 #include "StreamError.h"

@ -32,51 +33,59 @@ class BinaryConverter : public StreamFormatConverter
 };

 int BinaryConverter::
-parse(const StreamFormat& format, StreamBuffer& info,
+parse(const StreamFormat& fmt, StreamBuffer& info,
    const char*& source, bool)
 {
-    if (format.conv == 'B')
+    if (fmt.conv == 'b')
    {
-        // user defined characters for %B (next 2 in source)
+        // default characters 0 and 1 for %b
+        info.append("01");
+        return unsigned_format;
+    }
+    
+    // user defined characters for %B (next 2 in source)
+    if (*source)
+    {
+        if (*source == esc) source++;
+        info.append(*source++);
        if (*source)
        {
            if (*source == esc) source++;
            info.append(*source++);
-            if (*source)
-            {
-                if (*source == esc) source++;
-                info.append(*source++);
-                return long_format;
-            }
+            return unsigned_format;
        }
-        error("Missing characters after %%B format conversion\n");
-        return false;
    }
-    // default characters for %b
-    info.append("01");
-    return long_format;
+    error("Missing characters after %%B format conversion\n");
+    return false;
 }

 bool BinaryConverter::
-printLong(const StreamFormat& format, StreamBuffer& output, long value)
+printLong(const StreamFormat& fmt, StreamBuffer& output, long value)
 {
-    int prec = format.prec;
+    int prec = fmt.prec;
    if (prec == -1)
    {
-        // find number of significant bits
-        prec = sizeof (value) * 8;
-        while (prec && (value & (1L << (prec - 1))) == 0) prec--;
+        // Find number of significant bits is nothing is specified.
+        unsigned long x = (unsigned long) value;
+        prec = 32;
+#if (LONG_BIT > 32)
+        if (x > 0xFFFFFFFF)  { prec = 64;  x >>=32; }
+#endif 
+        if (x <= 0x0000FFFF) { prec -= 16; x <<=16; }
+        if (x <= 0x00FFFFFF) { prec -= 8; x <<=8; }
+        if (x <= 0x0FFFFFFF) { prec -= 4; x <<=4; }
+        if (x <= 0x3FFFFFFF) { prec -= 2; x <<=2; }
+        if (x <= 0x7FFFFFFF) { prec -= 1; }
    }
-    if (prec == 0) prec++; // print at least one bit
    int width = prec;
-    if (format.width > width) width = format.width;
-    char zero = format.info[0];
-    char one = format.info[1];
-    char fill = (format.flags & zero_flag) ? zero : ' ';
-    if (format.flags & alt_flag)
+    if (fmt.width > width) width = fmt.width;
+    char zero = fmt.info[0];
+    char one = fmt.info[1];
+    char fill = (fmt.flags & zero_flag) ? zero : ' ';
+    if (fmt.flags & alt_flag)
    {
        // little endian (least significant bit first)
-        if (!(format.flags & left_flag))
+        if (!(fmt.flags & left_flag))
        {
            // pad left
            while (width > prec)
@ -100,7 +109,7 @@ printLong(const StreamFormat& format, StreamBuffer& output, long value)
    else
    {
        // big endian (most significant bit first)
-        if (!(format.flags & left_flag))
+        if (!(fmt.flags & left_flag))
        {
            // pad left
            while (width > prec)
@ -124,18 +133,18 @@ printLong(const StreamFormat& format, StreamBuffer& output, long value)
 }

 int BinaryConverter::
-scanLong(const StreamFormat& format, const char* input, long& value)
+scanLong(const StreamFormat& fmt, const char* input, long& value)
 {
    long val = 0;
-    int width = format.width;
+    int width = fmt.width;
    if (width == 0) width = -1;
    int length = 0;
-    char zero = format.info[0];
-    char one = format.info[1];
+    char zero = fmt.info[0];
+    char one = fmt.info[1];
    if (!isspace(zero) && !isspace(one))
        while (isspace(input[length])) length++; // skip whitespaces
    if (input[length] != zero && input[length] != one) return -1;
-    if (format.flags & alt_flag)
+    if (fmt.flags & alt_flag)
    {
        // little endian (least significan bit first)
        long mask = 1;
--- a/src/EnumConverter.cc
+++ b/src/EnumConverter.cc
@ -36,7 +36,7 @@ class EnumConverter : public StreamFormatConverter

 int EnumConverter::
 parse(const StreamFormat& fmt, StreamBuffer& info,
-    const char*& source, bool)
+    const char*& source, bool scanFormat)
 {
    if (fmt.flags & (left_flag|sign_flag|space_flag|zero_flag))
    {
@ -56,7 +56,30 @@ parse(const StreamFormat& fmt, StreamBuffer& info,
        if (*source == '=' && (fmt.flags & alt_flag))
        {
            char* p;
-            index = strtol(++source, &p, 0);
+            
+            if (*++source == '?')
+            {
+                // default choice
+                if (scanFormat)
+                {
+                    error("Default value only allowed in output formats\n");
+                    return false;
+                }
+                if (*++source != '}')
+                {
+                    error("Default value must be last\n");
+                    return false;
+                }
+                source++;
+                numEnums = -(numEnums+1);
+                info.append('\0');
+                memcpy(info(n), &numEnums, sizeof(numEnums));
+                debug("EnumConverter::parse %ld choices with default: %s\n",
+                    -numEnums, info.expand()());
+                return enum_format;
+            }
+            
+            index = strtol(source, &p, 0);
            if (p == source || (*p != '|' && *p != '}'))
            {
                error("Integer expected after '=' "
@ -99,6 +122,9 @@ printLong(const StreamFormat& fmt, StreamBuffer& output, long value)
    const char* s = fmt.info;
    long numEnums = extract<long>(s);
    long index = extract<long>(s);
+    bool noDefault = numEnums >= 0;
+    
+    if (numEnums < 0) numEnums=-numEnums-1;    
    while (numEnums-- && (value != index))
    {
        while(*s)
@ -109,7 +135,7 @@ printLong(const StreamFormat& fmt, StreamBuffer& output, long value)
        s++;
        index = extract<long>(s);
    }
-    if (numEnums == -1)
+    if (numEnums == -1 && noDefault)
    {
        error("Value %li not found in enum set\n", value);
        return false;
--- a/src/RawConverter.cc
+++ b/src/RawConverter.cc
@ -31,23 +31,23 @@ class RawConverter : public StreamFormatConverter
 };

 int RawConverter::
-parse(const StreamFormat&, StreamBuffer&,
+parse(const StreamFormat& fmt, StreamBuffer&,
    const char*&, bool)
 {
-    return long_format;
+    return (fmt.flags & (sign_flag|zero_flag)) ? signed_format : unsigned_format;
 }

 bool RawConverter::
-printLong(const StreamFormat& format, StreamBuffer& output, long value)
+printLong(const StreamFormat& fmt, StreamBuffer& output, long value)
 {
-    int prec = format.prec;      // number of bytes from value
+    int prec = fmt.prec;      // number of bytes from value
    if (prec == -1) prec = 1;    // default: 1 byte
    int width = prec;            // number of bytes in output
    if (prec > (int)sizeof(long)) prec=sizeof(long);
-    if (format.width > width) width = format.width;
+    if (fmt.width > width) width = fmt.width;
    
    char byte = 0;
-    if (format.flags & alt_flag) // little endian (lsb first)
+    if (fmt.flags & alt_flag) // little endian (lsb first)
    {
        while (prec--)
        {
@ -56,7 +56,7 @@ printLong(const StreamFormat& format, StreamBuffer& output, long value)
            value >>= 8;
            width--;
        }
-        if (format.flags & zero_flag)
+        if (fmt.flags & zero_flag)
        {
            // fill with zero
            byte = 0;
@ -73,7 +73,7 @@ printLong(const StreamFormat& format, StreamBuffer& output, long value)
    }
    else // big endian (msb first)
    {
-        if (format.flags & zero_flag)
+        if (fmt.flags & zero_flag)
        {
            // fill with zero
            byte = 0;
@ -97,17 +97,17 @@ printLong(const StreamFormat& format, StreamBuffer& output, long value)
 }

 int RawConverter::
-scanLong(const StreamFormat& format, const char* input, long& value)
+scanLong(const StreamFormat& fmt, const char* input, long& value)
 {
    long length = 0;
    long val = 0;
-    int width = format.width;
+    int width = fmt.width;
    if (width == 0) width = 1; // default: 1 byte
-    if (format.flags & skip_flag)
+    if (fmt.flags & skip_flag)
    {
        return width; // just skip input
    }
-    if (format.flags & alt_flag)
+    if (fmt.flags & alt_flag)
    {
        // little endian (lsb first)
        unsigned int shift = 0;
@ -118,7 +118,7 @@ scanLong(const StreamFormat& format, const char* input, long& value)
        }
        if (width == 0)
        {
-            if (format.flags & zero_flag)
+            if (fmt.flags & zero_flag)
            {
                // fill with zero
                val |= ((unsigned char) input[length++]) << shift;
@ -134,7 +134,7 @@ scanLong(const StreamFormat& format, const char* input, long& value)
    else
    {
        // big endian  (msb first)
-        if (format.flags & zero_flag)
+        if (fmt.flags & zero_flag)
        {
            // fill with zero
            val = (unsigned char) input[length++];
--- a/src/StreamCore.cc
+++ b/src/StreamCore.cc
@ -749,7 +749,7 @@ printSeparator()
 bool StreamCore::
 printValue(const StreamFormat& fmt, long value)
 {
-    if (fmt.type != long_format && fmt.type != enum_format)
+    if (fmt.type != unsigned_format && fmt.type != signed_format && fmt.type != enum_format)
    {
        error("%s: printValue(long) called with %%%c format\n",
            name(), fmt.conv);
@ -763,8 +763,8 @@ printValue(const StreamFormat& fmt, long value)
            name(), value);
        return false;
    }
-    debug("StreamCore::printValue(%s, long): \"%s\"\n",
-        name(), outputLine.expand()());
+    debug("StreamCore::printValue %s %%%c long %ld (%lx): \"%s\"\n",
+        name(), fmt.conv, value, value, outputLine.expand()());
    return true;
 }

@ -785,8 +785,8 @@ printValue(const StreamFormat& fmt, double value)
            name(), value);
        return false;
    }
-    debug("StreamCore::printValue(%s, double): \"%s\"\n",
-        name(), outputLine.expand()());
+    debug("StreamCore::printValue %s %%%c double %#g: \"%s\"\n",
+        name(), fmt.conv, value, outputLine.expand()());
    return true;
 }

@ -808,8 +808,8 @@ printValue(const StreamFormat& fmt, char* value)
            name(), buffer.expand()());
        return false;
    }
-    debug("StreamCore::printValue(%s, char*): \"%s\"\n",
-        name(), outputLine.expand()());
+    debug("StreamCore::printValue %s %%%c char* \"%s\"): \"%s\"\n",
+        name(), fmt.conv, value, outputLine.expand()());
    return true;
 }

@ -1216,7 +1216,8 @@ normal_format:
                    double ddummy;
                    switch (fmt.type)
                    {
-                        case long_format:
+                        case unsigned_format:
+                        case signed_format:
                        case enum_format:
                            consumed = StreamFormatConverter::find(fmt.conv)->
                                scanLong(fmt, inputLine(consumedInput), ldummy);
@ -1457,7 +1458,7 @@ matchSeparator()
 long StreamCore::
 scanValue(const StreamFormat& fmt, long& value)
 {
-    if (fmt.type != long_format && fmt.type != enum_format)
+    if (fmt.type != unsigned_format && fmt.type != signed_format && fmt.type != enum_format)
    {
        error("%s: scanValue(long&) called with %%%c format\n",
            name(), fmt.conv);
--- a/src/StreamEpics.cc
+++ b/src/StreamEpics.cc
@ -270,7 +270,7 @@ epicsExportRegistrar(streamRegistrar);
 struct stream_drvsup {
    long number;
    long (*report)(int);
-    DRVSUPFUN init;
+    long (*init)();
 } stream = {
    2,
    Stream::report,
@ -416,6 +416,7 @@ long streamInit(int after)
 long streamInitRecord(dbCommon* record, const struct link *ioLink,
    streamIoFunction readData, streamIoFunction writeData)
 {
+    long status;
    char filename[80];
    char protocol[80];
    char busname[80];
@ -441,13 +442,14 @@ long streamInitRecord(dbCommon* record, const struct link *ioLink,
    // scan the i/o link
    debug("streamInitRecord(%s): parse link \"%s\"\n",
        record->name, ioLink->value.instio.string);
-    pstream->parseLink(ioLink, filename, protocol,
+    status = pstream->parseLink(ioLink, filename, protocol,
        busname, &addr, busparam);
    // (re)initialize bus and protocol
    debug("streamInitRecord(%s): calling initRecord\n",
        record->name);
-    long status = pstream->initRecord(filename, protocol,
-        busname, addr, busparam);
+    if (status == 0)
+        status = pstream->initRecord(filename, protocol,
+            busname, addr, busparam);
    if (status != OK && status != DO_NOT_CONVERT)
    {
        error("%s: Record initialization failed\n", record->name);
@ -749,21 +751,16 @@ process()
 bool Stream::
 print(format_t *format, va_list ap)
 {
-    long lval;
-    double dval;
-    char* sval;
    switch (format->type)
    {
-        case DBF_ENUM:
+        case DBF_ULONG:
        case DBF_LONG:
-            lval = va_arg(ap, long);
-            return printValue(*format->priv, lval);
+        case DBF_ENUM:
+            return printValue(*format->priv, va_arg(ap, long));
        case DBF_DOUBLE:
-            dval = va_arg(ap, double);
-            return printValue(*format->priv, dval);
+            return printValue(*format->priv, va_arg(ap, double));
        case DBF_STRING:
-            sval = va_arg(ap, char*);
-            return printValue(*format->priv, sval);
+            return printValue(*format->priv, va_arg(ap, char*));
    }
    error("INTERNAL ERROR (%s): Illegal format type\n", name());
    return false;
@ -773,27 +770,22 @@ bool Stream::
 scan(format_t *format, void* value, size_t maxStringSize)
 {
    // called by streamScanfN
-    long* lptr;
-    double* dptr;
-    char* sptr;

    // first remove old value from inputLine (if we are scanning arrays)
    consumedInput += currentValueLength;
    currentValueLength = 0;
    switch (format->type)
    {
+        case DBF_ULONG:
        case DBF_LONG:
        case DBF_ENUM:
-            lptr = (long*)value;
-            currentValueLength = scanValue(*format->priv, *lptr);
+            currentValueLength = scanValue(*format->priv, *(long*)value);
            break;
        case DBF_DOUBLE:
-            dptr = (double*)value;
-            currentValueLength = scanValue(*format->priv, *dptr);
+            currentValueLength = scanValue(*format->priv, *(double*)value);
            break;
        case DBF_STRING:
-            sptr = (char*)value;
-            currentValueLength  = scanValue(*format->priv, sptr,
+            currentValueLength  = scanValue(*format->priv, (char*)value,
                maxStringSize);
            break;
        default:
@ -963,13 +955,13 @@ getFieldAddress(const char* fieldname, StreamBuffer& address)
    else
    {
        // FIELD in this record or VAL in other record
-        char fullname[PVNAME_SZ + 1];
-        sprintf(fullname, "%s.%s", name(), fieldname);
-        if (dbNameToAddr(fullname, &dbaddr) != OK)
+        StreamBuffer fullname;
+        fullname.print("%s.%s", name(), fieldname);
+        if (dbNameToAddr(fullname(), &dbaddr) != OK)
        {
            // VAL in other record
-            sprintf(fullname, "%s.VAL", fieldname);
-            if (dbNameToAddr(fullname, &dbaddr) != OK) return false;
+            fullname.clear().print("%s.VAL", fieldname);
+            if (dbNameToAddr(fullname(), &dbaddr) != OK) return false;
        }
    }
    address.append(&dbaddr, sizeof(dbaddr));
@ -977,9 +969,9 @@ getFieldAddress(const char* fieldname, StreamBuffer& address)
 }

 static const unsigned char dbfMapping[] =
-    {0, DBF_LONG, DBF_ENUM, DBF_DOUBLE, DBF_STRING};
+    {0, DBF_ULONG, DBF_LONG, DBF_ENUM, DBF_DOUBLE, DBF_STRING};
 static const short typeSize[] =
-    {0, sizeof(epicsInt32), sizeof(epicsUInt16),
+    {0, sizeof(epicsUInt32), sizeof(epicsInt32), sizeof(epicsUInt16),
        sizeof(epicsFloat64), MAX_STRING_SIZE};

 bool Stream::
@ -991,16 +983,16 @@ formatValue(const StreamFormat& format, const void* fieldaddress)
 // --  TO DO: If SCAN is "I/O Intr" and record has not been processed,  --
 // --  do it now to get the latest value (only for output records?)     --

+    format_s fmt;
+    fmt.type = dbfMapping[format.type];
+    fmt.priv = &format;
    if (fieldaddress)
    {
        // Format like "%([record.]field)..." has requested to get value
        // from field of this or other record.
        DBADDR* pdbaddr = (DBADDR*)fieldaddress;
-        long i;
-        long nelem = pdbaddr->no_elements;
-        size_t size = nelem * typeSize[format.type];
-        char* buffer = fieldBuffer.clear().reserve(size);
        
+        /* Handle time stamps special. %T converter takes double. */
        if (strcmp(((dbFldDes*)pdbaddr->pfldDes)->name, "TIME") == 0)
        {
            double time;
@ -1016,6 +1008,8 @@ formatValue(const StreamFormat& format, const void* fieldaddress)
                /* if getting time from own record, update timestamp first */
                recGblGetTimeStamp(record);
            }
+            /* convert EPICS epoch (1990) to unix epoch (1970) */
+            /* we are losing about 3 digits precision here */
            time = pdbaddr->precord->time.secPastEpoch +
                631152000u + pdbaddr->precord->time.nsec * 1e-9;
            debug("Stream::formatValue(%s): read %f from TIME field\n",
@ -1023,7 +1017,28 @@ formatValue(const StreamFormat& format, const void* fieldaddress)
            return printValue(format, time);
        }

-        if (dbGet(pdbaddr, dbfMapping[format.type], buffer,
+        /* convert type to LONG, ENUM, DOUBLE, or STRING */
+        long nelem = pdbaddr->no_elements;
+        size_t size = nelem * typeSize[format.type];
+
+        /* print (U)CHAR arrays as string */
+        if (format.type == string_format &&
+            (pdbaddr->field_type == DBF_CHAR || pdbaddr->field_type == DBF_UCHAR))
+        {
+            debug("Stream::formatValue(%s): format %s.%s array[%ld] size %d of %s as string\n",
+                name(),
+                pdbaddr->precord->name,
+                ((dbFldDes*)pdbaddr->pfldDes)->name,
+                nelem,
+                pdbaddr->field_size,
+                pamapdbfType[pdbaddr->field_type].strvalue);
+            fmt.type = DBF_CHAR;
+            size = nelem;
+        }
+        
+        char* buffer = fieldBuffer.clear().reserve(size);
+
+        if (dbGet(pdbaddr, fmt.type, buffer,
            NULL, &nelem, NULL) != 0)
        {
            error("%s: dbGet(%s.%s, %s) failed\n",
@ -1033,6 +1048,18 @@ formatValue(const StreamFormat& format, const void* fieldaddress)
                pamapdbfType[dbfMapping[format.type]].strvalue);
            return false;
        }
+        debug("Stream::formatValue(%s): got %ld elements\n",
+                name(),nelem);
+
+        /* terminate CHAR array as string */
+        if (fmt.type == DBF_CHAR)
+        {
+            if (nelem >= pdbaddr->no_elements) nelem = pdbaddr->no_elements-1;
+            buffer[nelem] = 0;
+            nelem = 1; /* array is only 1 string */
+        }
+
+        long i;
        for (i = 0; i < nelem; i++)
        {
            switch (format.type)
@ -1042,7 +1069,12 @@ formatValue(const StreamFormat& format, const void* fieldaddress)
                        (long)((epicsUInt16*)buffer)[i]))
                        return false;
                    break;
-                case long_format:
+                case unsigned_format:
+                    if (!printValue(format,
+                        (long)((epicsUInt32*)buffer)[i]))
+                        return false;
+                    break;
+                case signed_format:
                    if (!printValue(format,
                        (long)((epicsInt32*)buffer)[i]))
                        return false;
@ -1060,6 +1092,7 @@ formatValue(const StreamFormat& format, const void* fieldaddress)
                    error("%s: %%(FIELD) syntax not allowed "
                        "with pseudo formats\n",
                        name());
+                    return false;
                default:
                    error("INTERNAL ERROR %s: Illegal format.type=%d\n",
                        name(), format.type);
@ -1068,9 +1101,6 @@ formatValue(const StreamFormat& format, const void* fieldaddress)
        }
        return true;
    }
-    format_s fmt;
-    fmt.type = dbfMapping[format.type];
-    fmt.priv = &format;
    debug("Stream::formatValue(%s) format=%%%c type=%s\n",
        name(), format.conv, pamapdbfType[fmt.type].strvalue);
    if (!writeData)
@ -1096,7 +1126,10 @@ matchValue(const StreamFormat& format, const void* fieldaddress)
    char* buffer;
    int status;
    const char* putfunc;
+    format_s fmt;

+    fmt.type = dbfMapping[format.type];
+    fmt.priv = &format;
    if (fieldaddress)
    {
        // Format like "%([record.]field)..." has requested to put value
@ -1112,12 +1145,24 @@ matchValue(const StreamFormat& format, const void* fieldaddress)
                name(), fieldBuffer.expand()());
            switch (format.type)
            {
-                case long_format:
+                case unsigned_format:
+                {
+                    consumed = scanValue(format, lval);
+                    if (consumed >= 0) ((epicsUInt32*)buffer)[nord] = lval;
+                    debug("Stream::matchValue(%s): %s.%s[%li] = %lu\n",
+                            name(), pdbaddr->precord->name,
+                            ((dbFldDes*)pdbaddr->pfldDes)->name,
+                            nord, lval);
+                    break;
+                }
+                case signed_format:
                {
                    consumed = scanValue(format, lval);
                    if (consumed >= 0) ((epicsInt32*)buffer)[nord] = lval;
-                    debug("Stream::matchValue(%s): %s[%li] = %li\n",
-                            name(), pdbaddr->precord->name, nord, lval);
+                    debug("Stream::matchValue(%s): %s.%s[%li] = %li\n",
+                            name(), pdbaddr->precord->name,
+                            ((dbFldDes*)pdbaddr->pfldDes)->name,
+                            nord, lval);
                    break;
                }
                case enum_format:
@ -1125,8 +1170,10 @@ matchValue(const StreamFormat& format, const void* fieldaddress)
                    consumed = scanValue(format, lval);
                    if (consumed >= 0)
                        ((epicsUInt16*)buffer)[nord] = (epicsUInt16)lval;
-                    debug("Stream::matchValue(%s): %s[%li] = %li\n",
-                            name(), pdbaddr->precord->name, nord, lval);
+                    debug("Stream::matchValue(%s): %s.%s[%li] = %li\n",
+                            name(), pdbaddr->precord->name,
+                            ((dbFldDes*)pdbaddr->pfldDes)->name,
+                            nord, lval);
                    break;
                }
                case double_format:
@ -1139,18 +1186,34 @@ matchValue(const StreamFormat& format, const void* fieldaddress)
                    if (consumed >= 0)
                        memcpy(((epicsFloat64*)buffer)+nord,
                            &f64, sizeof(f64));
-                    debug("Stream::matchValue(%s): %s[%li] = %#g %#g\n",
-                            name(), pdbaddr->precord->name, nord, dval,
+                    debug("Stream::matchValue(%s): %s.%s[%li] = %#g %#g\n",
+                            name(), pdbaddr->precord->name,
+                            ((dbFldDes*)pdbaddr->pfldDes)->name,
+                            nord, dval,
                            ((epicsFloat64*)buffer)[nord]);
                    break;
                }
                case string_format:
                {
-                    consumed = scanValue(format,
-                        buffer+MAX_STRING_SIZE*nord, MAX_STRING_SIZE);
-                    debug("Stream::matchValue(%s): %s[%li] = \"%.*s\"\n",
-                            name(), pdbaddr->precord->name, nord,
-                            MAX_STRING_SIZE, buffer+MAX_STRING_SIZE*nord);
+                    if (pdbaddr->field_type == DBF_CHAR)
+                    {
+                        // string to char array
+                        consumed = scanValue(format, buffer, nelem);
+                        debug("Stream::matchValue(%s): %s.%s = \"%.*s\"\n",
+                                name(), pdbaddr->precord->name,
+                                ((dbFldDes*)pdbaddr->pfldDes)->name,
+                                (int)consumed, buffer);
+                        nord = nelem;
+                    }
+                    else
+                    {
+                        consumed = scanValue(format,
+                            buffer+MAX_STRING_SIZE*nord, MAX_STRING_SIZE);
+                        debug("Stream::matchValue(%s): %s.%s[%li] = \"%.*s\"\n",
+                                name(), pdbaddr->precord->name,
+                                ((dbFldDes*)pdbaddr->pfldDes)->name,
+                                nord, MAX_STRING_SIZE, buffer+MAX_STRING_SIZE*nord);
+                    }
                    break;
                }
                default:
@ -1187,15 +1250,17 @@ matchValue(const StreamFormat& format, const void* fieldaddress)
                    name());
                return false;
            }
+            /* convert from Unix epoch (1 Jan 1970) to EPICS epoch (1 Jan 1990) */
            dval = dval-631152000u;
            pdbaddr->precord->time.secPastEpoch = (long)dval;
            // rouding: we don't have 9 digits precision
            // in a double of today's number of seconds
            pdbaddr->precord->time.nsec = (long)((dval-(long)dval)*1e6)*1000;
-            debug("Stream::matchValue(%s): writing %i.%i to TIME field\n",
+            debug("Stream::matchValue(%s): writing %i.%i to %s.TIME field\n",
                name(),
                pdbaddr->precord->time.secPastEpoch,
-                pdbaddr->precord->time.nsec);
+                pdbaddr->precord->time.nsec,
+                pdbaddr->precord->name);
            pdbaddr->precord->tse = epicsTimeEventDeviceTime;
            return true;
 #else
@ -1204,18 +1269,19 @@ matchValue(const StreamFormat& format, const void* fieldaddress)
            return false;
 #endif
        }
-        
+        if (format.type == string_format &&
+            (pdbaddr->field_type == DBF_CHAR || pdbaddr->field_type == DBF_UCHAR))
+        {
+            /* write strings to [U]CHAR arrays */
+            nord = consumed;
+            fmt.type = DBF_CHAR;
+        }
        if (pdbaddr->precord == record || INIT_RUN)
        {
            // write into own record, thus don't process it
            // in @init we must not process other record
-            debug("Stream::matchValue(%s): dbPut(%s.%s,%s)\n",
-                name(),
-                pdbaddr->precord->name,
-                ((dbFldDes*)pdbaddr->pfldDes)->name,
-                fieldBuffer.expand()());
            putfunc = "dbPut";
-            status = dbPut(pdbaddr, dbfMapping[format.type], buffer, nord);
+            status = dbPut(pdbaddr, fmt.type, buffer, nord);
            if (INIT_RUN && pdbaddr->precord != record)
            {
                // clean error status of other record in @init
@ -1227,41 +1293,44 @@ matchValue(const StreamFormat& format, const void* fieldaddress)
        else
        {
            // write into other record, thus process it
-            debug("Stream::matchValue(%s): dbPutField(%s.%s,%s)\n",
-                name(),
-                pdbaddr->precord->name,
-                ((dbFldDes*)pdbaddr->pfldDes)->name,
-                fieldBuffer.expand()());
            putfunc = "dbPutField";
-            status = dbPutField(pdbaddr, dbfMapping[format.type],
+            status = dbPutField(pdbaddr, fmt.type,
                buffer, nord);
        }
+        debug("Stream::matchValue(%s): %s(%s.%s, %s, %s) status=0x%x\n",
+            name(), putfunc,
+            pdbaddr->precord->name,
+            ((dbFldDes*)pdbaddr->pfldDes)->name,
+            pamapdbfType[fmt.type].strvalue,
+            fieldBuffer.expand()(),
+            status);
        if (status != 0)
        {
            flags &= ~ScanTried;
-            switch (format.type)
+            switch (fmt.type)
            {
-                case long_format:
-                case enum_format:
+                case DBF_ULONG:
+                case DBF_LONG:
+                case DBF_ENUM:
                    error("%s: %s(%s.%s, %s, %li) failed\n",
-                        putfunc, name(), pdbaddr->precord->name,
+                        name(), putfunc, pdbaddr->precord->name,
                        ((dbFldDes*)pdbaddr->pfldDes)->name,
-                        pamapdbfType[dbfMapping[format.type]].strvalue,
+                        pamapdbfType[fmt.type].strvalue,
                        lval);
                    return false;
-                case double_format:
+                case DBF_DOUBLE:
                    error("%s: %s(%s.%s, %s, %#g) failed\n",
-                        putfunc, name(), pdbaddr->precord->name,
+                        name(), putfunc, pdbaddr->precord->name,
                        ((dbFldDes*)pdbaddr->pfldDes)->name,
-                        pamapdbfType[dbfMapping[format.type]].strvalue,
+                        pamapdbfType[fmt.type].strvalue,
                        dval);
                    return false;
-                case string_format:
-                    error("%s: %s(%s.%s, %s, \"%s\") failed\n",
-                        putfunc, name(), pdbaddr->precord->name,
+                case DBF_STRING:
+                    error("%s: %s(%s.%s, %s, \"%.*s\") failed\n",
+                        name(), putfunc, pdbaddr->precord->name,
                        ((dbFldDes*)pdbaddr->pfldDes)->name,
-                        pamapdbfType[dbfMapping[format.type]].strvalue,
-                        buffer);
+                        pamapdbfType[fmt.type].strvalue,
+                        (int)consumed, buffer);
                    return false;
                default:
                    return false;
@ -1270,9 +1339,6 @@ matchValue(const StreamFormat& format, const void* fieldaddress)
        return true;
    }
    // no fieldaddress (the "normal" case)
-    format_s fmt;
-    fmt.type = dbfMapping[format.type];
-    fmt.priv = &format;
    if (!readData)
    {
        error("%s: No readData() function provided\n", name());
--- a/src/StreamFormat.h
+++ b/src/StreamFormat.h
@ -34,11 +34,12 @@ typedef enum {
 } StreamFormatFlag;

 typedef enum {
-    long_format   = 1,
-    enum_format   = 2,
-    double_format = 3,
-    string_format = 4,
-    pseudo_format = 5
+    unsigned_format = 1,
+    signed_format,
+    enum_format,
+    double_format,
+    string_format,
+    pseudo_format
 } StreamFormatType;

 extern const char* StreamFormatTypeStr[];
--- a/src/StreamFormatConverter.cc
+++ b/src/StreamFormatConverter.cc
@ -21,6 +21,7 @@

 #include <stdlib.h>
 #include <ctype.h>
+#include <limits.h>
 #include "StreamFormatConverter.h"
 #include "StreamError.h"

@ -310,12 +311,16 @@ parse(const StreamFormat& fmt, StreamBuffer& info,
        info.append('l');
        info.append(fmt.conv);
    }
-    return long_format;
+    if (fmt.conv == 'd' || fmt.conv == 'i') return signed_format;
+    return unsigned_format;
 }

 bool StdLongConverter::
 printLong(const StreamFormat& fmt, StreamBuffer& output, long value)
 {
+    // limits %x/%X formats to number of half bytes in width.
+    if (fmt.width && (fmt.conv == 'x' || fmt.conv == 'X') && fmt.width < 2*sizeof(long))
+        value &= ~(-1L << (fmt.width*4));
    output.print(fmt.info, value);
    return true;
 }
@ -327,6 +332,7 @@ scanLong(const StreamFormat& fmt, const char* input, long& value)
    int length;
    bool neg;
    int base;
+    long v;

    length = prepareval(fmt, input, neg);
    if (length < 0) return -1;
@ -336,14 +342,11 @@ scanLong(const StreamFormat& fmt, const char* input, long& value)
            base = 10;
            break;
        case 'o':
-            base = 8;
-            goto signcheck;
        case 'x':
        case 'X':
-            base = 16;
-signcheck:
-        // allow negative hex and oct numbers with - flag
+            // allow negative hex and oct numbers with - flag
            if (neg && !(fmt.flags & left_flag)) return -1;
+            base = (fmt.conv == 'o') ? 8 : 16;
            break;
        case 'u':
            if (neg) return -1;
@ -352,10 +355,10 @@ signcheck:
        default:
            base = 0;
    }
-    value = strtoul(input, &end, base);
-    if (neg) value = -value;
+    v = strtoul(input, &end, base);
    if (end == input) return -1;
    length += end-input;
+    value = neg ? -v : v;
    return length;
 }

@ -543,7 +546,7 @@ parse(const StreamFormat& fmt, StreamBuffer& info,
        info.append("%n");
        return string_format;
    }
-    return long_format;
+    return unsigned_format;
 }

 bool StdCharsConverter::
--- a/src/StreamProtocol.cc
+++ b/src/StreamProtocol.cc
@ -26,7 +26,8 @@
 #include "StreamError.h"

 const char* StreamFormatTypeStr[] = {
-    "none", "long", "enum", "double", "string", "pseudo"
+    // must match the order in StreamFormat.h
+    "none", "unsigned", "signed", "enum", "double", "string", "pseudo"
 };

 class StreamProtocolParser::Protocol::Variable
@ -1438,7 +1439,7 @@ compileFormat(StreamBuffer& buffer, const char*& formatstr,
        // parsing failed
        return false;
    }
-    if (type < long_format && type > pseudo_format)
+    if (type < 1 && type > pseudo_format)
    {
        error(line, filename(),
            "Illegal format type %d returned from '%%%c' converter\n",
--- a/src/TimestampConverter.cc
+++ b/src/TimestampConverter.cc
@ -64,6 +64,13 @@ int timezone = 0;
 } while (0)
 #endif

+#if defined(__MINGW32__)
+/* MinGW has no re-entrant localtime. How about other environments? */
+/* Just let's hope for the best */
+#undef localtime_r
+#define localtime_r(timet,tm) (*(tm)=*localtime(timet))
+#endif
+
 class TimestampConverter : public StreamFormatConverter
 {
    int parse(const StreamFormat&, StreamBuffer&, const char*&, bool);
@ -384,7 +391,7 @@ startover:
                        debug ("TimestampConverter::scantime: %s hour = %d\n", pm?"PM":"AM", tm->tm_hour);
                        break;
                    case 'M': /* minute */
-                        i = nummatch(input, 1, 59);
+                        i = nummatch(input, 0, 59);
                        if (i < 0)
                        {
                            error ("error parsing minute: '%.20s'\n", input);
@ -394,7 +401,7 @@ startover:
                        debug ("TimestampConverter::scantime: min = %d\n", tm->tm_min);
                        break;
                    case 'S': /* second */
-                        i = nummatch(input, 1, 60);
+                        i = nummatch(input, 0, 60);
                        if (i < 0)
                        {
                            error ("error parsing week second: '%.20s'\n", input);
--- a/src/devaaiStream.c
+++ b/src/devaaiStream.c
@ -20,9 +20,9 @@

 #include <string.h>
 #include <stdlib.h>
-#include "devStream.h"
-#include <aaiRecord.h>
 #include <errlog.h>
+#include <aaiRecord.h>
+#include "devStream.h"
 #include <epicsExport.h>

 static long readData (dbCommon *record, format_t *format)
@ -57,6 +57,7 @@ static long readData (dbCommon *record, format_t *format)
                }
                break;
            }
+            case DBF_ULONG:
            case DBF_LONG:
            case DBF_ENUM:
            {
@ -191,6 +192,7 @@ static long writeData (dbCommon *record, format_t *format)
                    return ERROR;
                break;
            }
+            case DBF_ULONG:
            case DBF_LONG:
            case DBF_ENUM:
            {
--- a/src/devaaoStream.c
+++ b/src/devaaoStream.c
@ -20,9 +20,9 @@

 #include <string.h>
 #include <stdlib.h>
-#include "devStream.h"
-#include <aaoRecord.h>
 #include <errlog.h>
+#include <aaoRecord.h>
+#include "devStream.h"
 #include <epicsExport.h>

 static long readData (dbCommon *record, format_t *format)
@ -57,6 +57,7 @@ static long readData (dbCommon *record, format_t *format)
                }
                break;
            }
+            case DBF_ULONG:
            case DBF_LONG:
            case DBF_ENUM:
            {
@ -191,6 +192,7 @@ static long writeData (dbCommon *record, format_t *format)
                    return ERROR;
                break;
            }
+            case DBF_ULONG:
            case DBF_LONG:
            case DBF_ENUM:
            {
--- a/src/devaiStream.c
+++ b/src/devaiStream.c
@ -18,10 +18,18 @@
 *                                                              *
 ***************************************************************/

-#include "devStream.h"
+#include <math.h>
+#include <menuConvert.h>
 #include <aiRecord.h>
+#include "devStream.h"
 #include <epicsExport.h>

+#ifdef vxWorks
+#include <private/mathP.h>
+#define isinf(x) isInf(x)
+#define isnan(x) isNan(x)
+#endif
+
 static long readData (dbCommon *record, format_t *format)
 {
    aiRecord *ai = (aiRecord *) record;
@ -32,20 +40,28 @@ static long readData (dbCommon *record, format_t *format)
        {
            double val;
            if (streamScanf (record, format, &val)) return ERROR;
-            if (ai->aslo != 0.0) val *= ai->aslo;
+            if (ai->aslo != 0.0 && ai->aslo != 1.0) val *= ai->aslo;
            val += ai->aoff;
-            if (!INIT_RUN && ai->smoo != 0.0)
-            {
+            if (!(ai->smoo == 0.0 || ai->init || ai->udf || isinf(ai->val) || isnan(ai->val)))
                val = ai->val * ai->smoo + val * (1.0 - ai->smoo);
-            }
            ai->val = val;
            return DO_NOT_CONVERT;
        }
+        case DBF_ULONG:
        case DBF_LONG:
        {
            long rval;
            if (streamScanf (record, format, &rval)) return ERROR;
            ai->rval = rval;
+            if (ai->linr == menuConvertNO_CONVERSION)
+            {
+                /* allow more bits than 32 */
+                if (format->type == DBF_ULONG)
+                    ai->val = (unsigned long)rval;
+                else    
+                    ai->val = rval;
+                return DO_NOT_CONVERT;
+            }
            return OK;
        }
    }
@ -55,19 +71,32 @@ static long readData (dbCommon *record, format_t *format)
 static long writeData (dbCommon *record, format_t *format)
 {
    aiRecord *ai = (aiRecord *) record;
-    double val;

    switch (format->type)
    {
        case DBF_DOUBLE:
        {
-            val = ai->val - ai->aoff;
-            if (ai->aslo != 0) val /= ai->aslo;
+            double val = ai->val - ai->aoff;
+            if (ai->aslo != 0.0 && ai->aslo != 1.0) val /= ai->aslo;
            return streamPrintf (record, format, val);
        }
+        case DBF_ULONG:
+        {
+            if (ai->linr == menuConvertNO_CONVERSION)
+            {
+                /* allow more bits than 32 */
+                return streamPrintf (record, format, (unsigned long)ai->val);
+            }
+            return streamPrintf (record, format, (unsigned long)ai->rval);
+        }
        case DBF_LONG:
        {
-            return streamPrintf (record, format, (long) ai->rval);
+            if (ai->linr == menuConvertNO_CONVERSION)
+            {
+                /* allow more bits than 32 */
+                return streamPrintf (record, format, (long)ai->val);
+            }
+            return streamPrintf (record, format, (long)ai->rval);
        }
    }
    return ERROR;
--- a/src/devaoStream.c
+++ b/src/devaoStream.c
@ -18,8 +18,9 @@
 *                                                              *
 ***************************************************************/

-#include "devStream.h"
+#include <menuConvert.h>
 #include <aoRecord.h>
+#include "devStream.h"
 #include <epicsExport.h>

 static long readData (dbCommon *record, format_t *format)
@ -32,16 +33,26 @@ static long readData (dbCommon *record, format_t *format)
        {
            double val;
            if (streamScanf (record, format, &val)) return ERROR;
-            if (ao->aslo != 0) val *= ao->aslo;
+            if (ao->aslo != 0.0 && ao->aslo != 1.0) val *= ao->aslo;
            ao->val = val + ao->aoff;
            return DO_NOT_CONVERT;
        }
+        case DBF_ULONG:
        case DBF_LONG:
        {
            long rval;
            if (streamScanf (record, format, &rval)) return ERROR;
            ao->rbv = rval;
-            if (INIT_RUN) ao->rval = rval;
+            ao->rval = rval;
+            if (ao->linr == menuConvertNO_CONVERSION)
+            {
+                /* allow more bits than 32 */
+                if (format->type == DBF_ULONG)
+                    ao->val = (unsigned long)rval;
+                else    
+                    ao->val = rval;
+                return DO_NOT_CONVERT;
+            }
            return OK;
        }
    }
@ -56,16 +67,27 @@ static long writeData (dbCommon *record, format_t *format)
    {
        case DBF_DOUBLE:
        {
-            double val;
-            if (INIT_RUN) val = ao->val;
-            else val = ao->oval;
-            val -= ao->aoff;
-            if (ao->aslo != 0) val /= ao->aslo;
+            double val = (INIT_RUN ? ao->val : ao->oval) - ao->aoff;
+            if (ao->aslo != 0.0 && ao->aslo != 1.0) val /= ao->aslo;
            return streamPrintf (record, format, val);
        }
+        case DBF_ULONG:
+        {
+            if (ao->linr == menuConvertNO_CONVERSION)
+            {
+                /* allow more bits than 32 */
+                return streamPrintf (record, format, (unsigned long)(INIT_RUN ? ao->val : ao->oval));
+            }
+            return streamPrintf (record, format, (unsigned long)ao->rval);
+        }
        case DBF_LONG:
        {
-            return streamPrintf (record, format, (long) ao->rval);
+            if (ao->linr == menuConvertNO_CONVERSION)
+            {
+                /* allow more bits than 32 */
+                return streamPrintf (record, format, (long)(INIT_RUN ? ao->val : ao->oval));
+            }
+            return streamPrintf (record, format, (long)ao->rval);
        }
    }
    return ERROR;
--- a/src/devbiStream.c
+++ b/src/devbiStream.c
@ -18,9 +18,9 @@
 *                                                              *
 ***************************************************************/

-#include "devStream.h"
-#include <biRecord.h>
 #include <string.h>
+#include <biRecord.h>
+#include "devStream.h"
 #include <epicsExport.h>

 static long readData (dbCommon *record, format_t *format)
@ -30,6 +30,7 @@ static long readData (dbCommon *record, format_t *format)

    switch (format->type)
    {
+        case DBF_ULONG:
        case DBF_LONG:
        {
            if (streamScanf (record, format, &val)) return ERROR;
@ -69,13 +70,14 @@ static long writeData (dbCommon *record, format_t *format)

    switch (format->type)
    {
+        case DBF_ULONG:
        case DBF_LONG:
        {
            return streamPrintf (record, format, bi->rval);
        }
        case DBF_ENUM:
        {
-            return streamPrintf (record, format, (long) bi->val);
+            return streamPrintf (record, format, (long)bi->val);
        }
        case DBF_STRING:
        {
--- a/src/devboStream.c
+++ b/src/devboStream.c
@ -18,9 +18,9 @@
 *                                                              *
 ***************************************************************/

-#include "devStream.h"
-#include <boRecord.h>
 #include <string.h>
+#include <boRecord.h>
+#include "devStream.h"
 #include <epicsExport.h>

 static long readData (dbCommon *record, format_t *format)
@ -30,6 +30,7 @@ static long readData (dbCommon *record, format_t *format)

    switch (format->type)
    {
+        case DBF_ULONG:
        case DBF_LONG:
        {
            if (streamScanf (record, format, &val)) return ERROR;
@ -70,13 +71,14 @@ static long writeData (dbCommon *record, format_t *format)

    switch (format->type)
    {
+        case DBF_ULONG:
        case DBF_LONG:
        {
            return streamPrintf (record, format, bo->rval);
        }
        case DBF_ENUM:
        {
-            return streamPrintf (record, format, (long) bo->val);
+            return streamPrintf (record, format, (long)bo->val);
        }
        case DBF_STRING:
        {
--- a/src/devcalcoutStream.c
+++ b/src/devcalcoutStream.c
@ -17,9 +17,9 @@
 *                                                              *
 ***************************************************************/

-#include "devStream.h"
 #include <postfix.h>
 #include <calcoutRecord.h>
+#include "devStream.h"
 #include <epicsExport.h>

 static long readData (dbCommon *record, format_t *format)
@ -32,13 +32,17 @@ static long readData (dbCommon *record, format_t *format)
        {
            return streamScanf (record, format, &co->val);
        }
+        case DBF_ULONG:
        case DBF_LONG:
        case DBF_ENUM:
        {
            long lval;

            if (streamScanf (record, format, &lval)) return ERROR;
-            co->val = lval;
+            if (format->type == DBF_LONG)
+                co->val = lval;
+            else
+                co->val = (unsigned long)lval;
            return OK;
        }
    }
@ -55,8 +59,12 @@ static long writeData (dbCommon *record, format_t *format)
        {
            return streamPrintf (record, format, co->oval);
        }
-        case DBF_LONG:
+        case DBF_ULONG:
        case DBF_ENUM:
+        {
+            return streamPrintf (record, format, (unsigned long)co->oval);
+        }
+        case DBF_LONG:
        {
            return streamPrintf (record, format, (long)co->oval);
        }
--- a/src/devlonginStream.c
+++ b/src/devlonginStream.c
@ -18,20 +18,25 @@
 *                                                              *
 ***************************************************************/

-#include "devStream.h"
 #include <longinRecord.h>
+#include "devStream.h"
 #include <epicsExport.h>

 static long readData (dbCommon *record, format_t *format)
 {
    longinRecord *li = (longinRecord *) record;

-    if (format->type == DBF_LONG || format->type == DBF_ENUM)
+    switch (format->type)
    {
-        long val;
-        if (streamScanf (record, format, &val)) return ERROR;
-        li->val = val;
-        return OK;
+        case DBF_ULONG:
+        case DBF_LONG:
+        case DBF_ENUM:
+        {       
+            long val;
+            if (streamScanf (record, format, &val)) return ERROR;
+            li->val = val;
+            return OK;
+        }
    }
    return ERROR;
 }
@ -40,9 +45,13 @@ static long writeData (dbCommon *record, format_t *format)
 {
    longinRecord *li = (longinRecord *) record;

-    if (format->type == DBF_LONG || format->type == DBF_ENUM)
+    switch (format->type)
    {
-        return streamPrintf (record, format, (long) li->val);
+        case DBF_ULONG:
+        case DBF_ENUM:
+            return streamPrintf (record, format, (unsigned long)li->val);
+        case DBF_LONG:
+            return streamPrintf (record, format, (long)li->val);
    }
    return ERROR;
 }
--- a/src/devlongoutStream.c
+++ b/src/devlongoutStream.c
@ -19,20 +19,25 @@
 *                                                              *
 ***************************************************************/

-#include "devStream.h"
 #include <longoutRecord.h>
+#include "devStream.h"
 #include <epicsExport.h>

 static long readData (dbCommon *record, format_t *format)
 {
    longoutRecord *lo = (longoutRecord *) record;

-    if (format->type == DBF_LONG || format->type == DBF_ENUM)
+    switch (format->type)
    {
-        long val;
-        if (streamScanf (record, format, &val)) return ERROR;
-        lo->val = val;
-        return OK;
+        case DBF_ULONG:
+        case DBF_LONG:
+        case DBF_ENUM:
+        {       
+            long val;
+            if (streamScanf (record, format, &val)) return ERROR;
+            lo->val = val;
+            return OK;
+        }
    }
    return ERROR;
 }
@ -41,9 +46,13 @@ static long writeData (dbCommon *record, format_t *format)
 {
    longoutRecord *lo = (longoutRecord *) record;

-    if (format->type == DBF_LONG || format->type == DBF_ENUM)
+    switch (format->type)
    {
-        return streamPrintf (record, format, (long) lo->val);
+        case DBF_ULONG:
+        case DBF_ENUM:
+            return streamPrintf (record, format, (unsigned long)lo->val);
+        case DBF_LONG:
+            return streamPrintf (record, format, (long)lo->val);
    }
    return ERROR;
 }
--- a/src/devmbbiDirectStream.c
+++ b/src/devmbbiDirectStream.c
@ -19,16 +19,16 @@
 *                                                              *
 ***************************************************************/

-#include "devStream.h"
 #include <mbbiDirectRecord.h>
+#include "devStream.h"
 #include <epicsExport.h>

 static long readData (dbCommon *record, format_t *format)
 {
    mbbiDirectRecord *mbbiD = (mbbiDirectRecord *) record;
-    long val;
+    unsigned long val;

-    if (format->type == DBF_LONG)
+    if (format->type == DBF_ULONG || format->type == DBF_LONG)
    {
        if (streamScanf (record, format, &val)) return ERROR;
        if (mbbiD->mask)
@ -40,7 +40,7 @@ static long readData (dbCommon *record, format_t *format)
        else
        {
            /* No MASK, (NOBT = 0): use VAL field */
-            mbbiD->val = (short)val;
+            mbbiD->val = (unsigned short)val;
            return DO_NOT_CONVERT;
        }
    }
@ -50,9 +50,9 @@ static long readData (dbCommon *record, format_t *format)
 static long writeData (dbCommon *record, format_t *format)
 {
    mbbiDirectRecord *mbbiD = (mbbiDirectRecord *) record;
-    long val;
+    unsigned long val;

-    if (format->type == DBF_LONG)
+    if (format->type == DBF_ULONG || format->type == DBF_LONG)
    {
        if (mbbiD->mask) val = mbbiD->rval & mbbiD->mask;
        else val = mbbiD->val;
--- a/src/devmbbiStream.c
+++ b/src/devmbbiStream.c
@ -19,19 +19,20 @@
 *                                                              *
 ***************************************************************/

-#include "devStream.h"
-#include <mbbiRecord.h>
 #include <string.h>
+#include <mbbiRecord.h>
+#include "devStream.h"
 #include <epicsExport.h>

 static long readData (dbCommon *record, format_t *format)
 {
    mbbiRecord *mbbi = (mbbiRecord *) record;
-    long val;
+    unsigned long val;
    int i;

    switch (format->type)
    {
+        case DBF_ULONG:
        case DBF_LONG:
        {
            if (streamScanf (record, format, &val)) return ERROR;
--- a/src/devmbboDirectStream.c
+++ b/src/devmbboDirectStream.c
@ -19,16 +19,16 @@
 *                                                              *
 ***************************************************************/

-#include "devStream.h"
 #include <mbboDirectRecord.h>
+#include "devStream.h"
 #include <epicsExport.h>

 static long readData (dbCommon *record, format_t *format)
 {
    mbboDirectRecord *mbboD = (mbboDirectRecord *) record;
-    long val;
+    unsigned long val;

-    if (format->type == DBF_LONG)
+    if (format->type == DBF_ULONG || format->type == DBF_LONG)
    {
        if (streamScanf (record, format, &val)) return ERROR;
        if (mbboD->mask)
@ -53,7 +53,7 @@ static long writeData (dbCommon *record, format_t *format)
    mbboDirectRecord *mbboD = (mbboDirectRecord *) record;
    long val;

-    if (format->type == DBF_LONG)
+    if (format->type == DBF_ULONG || format->type == DBF_LONG)
    {
        if (mbboD->mask) val = mbboD->rval & mbboD->mask;
        else val = mbboD->val;
--- a/src/devmbboStream.c
+++ b/src/devmbboStream.c
@ -19,19 +19,20 @@
 *                                                              *
 ***************************************************************/

-#include "devStream.h"
-#include <mbboRecord.h>
 #include <string.h>
+#include <mbboRecord.h>
+#include "devStream.h"
 #include <epicsExport.h>

 static long readData (dbCommon *record, format_t *format)
 {
    mbboRecord *mbbo = (mbboRecord *) record;
-    long val;
+    unsigned long val;
    int i;

    switch (format->type)
    {
+        case DBF_ULONG:
        case DBF_LONG:
        {
            if (streamScanf (record, format, &val)) return ERROR;
@ -76,11 +77,12 @@ static long readData (dbCommon *record, format_t *format)
 static long writeData (dbCommon *record, format_t *format)
 {
    mbboRecord *mbbo = (mbboRecord *) record;
-    long val;
+    unsigned long val;
    int i;

    switch (format->type)
    {
+        case DBF_ULONG:
        case DBF_LONG:
        {
            /* print VAL or RVAL ? */
--- a/src/devstringinStream.c
+++ b/src/devstringinStream.c
@ -18,8 +18,8 @@
 *                                                              *
 ***************************************************************/

-#include "devStream.h"
 #include <stringinRecord.h>
+#include "devStream.h"
 #include <epicsExport.h>

 static long readData (dbCommon *record, format_t *format)
--- a/src/devstringoutStream.c
+++ b/src/devstringoutStream.c
@ -18,8 +18,8 @@
 *                                                              *
 ***************************************************************/

-#include "devStream.h"
 #include <stringoutRecord.h>
+#include "devStream.h"
 #include <epicsExport.h>

 static long readData (dbCommon *record, format_t *format)
--- a/src/devwaveformStream.c
+++ b/src/devwaveformStream.c
@ -18,10 +18,10 @@
 *                                                              *
 ***************************************************************/

-#include "devStream.h"
-#include <waveformRecord.h>
 #include <string.h>
 #include <errlog.h>
+#include <waveformRecord.h>
+#include "devStream.h"
 #include <epicsExport.h>

 static long readData (dbCommon *record, format_t *format)
@ -57,6 +57,7 @@ static long readData (dbCommon *record, format_t *format)
                }
                break;
            }
+            case DBF_ULONG:
            case DBF_LONG:
            case DBF_ENUM:
            {
--- a/streamApp/tests/streamtestlib.tcl
+++ b/streamApp/tests/streamtestlib.tcl
@ -55,7 +55,7 @@ proc startioc {} {
        if [info exists asynversion] {
            puts $fd "require asyn,$asynversion"
        }
-        puts $fd "require stream2,$streamversion"
+        puts $fd "require stream,$streamversion"
    } else {
        puts $fd "#!../O.$env(EPICS_HOST_ARCH)/streamApp"
        puts $fd "dbLoadDatabase ../O.Common/streamApp.dbd"
@ -77,7 +77,9 @@ proc startioc {} {
    }
    fconfigure $ioc -blocking yes -buffering none
    debugmsg "waiting to connect"
+    set timer [after 1000 {puts stderr "\033\[31;7mCannot start IOC.\033\[0m"; exit 1}]
    vwait sock
+    after cancel $timer
 }

 set lastcommand ""
--- a/streamApp/tests/testDouble
+++ b/streamApp/tests/testDouble
@ -41,12 +41,10 @@ set protocol {
 set startup {
 }

-set debug 0
-
 startioc

 set inf [format %f inf]
-set nan [format %f nan]
+if [catch {set nan [format %f nan]}] {set nan nan}

 ioccmd {dbpf DZ:test1.PROC 1}
 send "3.14159265359\n"
--- a/streamApp/tests/testEnum
+++ b/streamApp/tests/testEnum
@ -50,7 +50,7 @@ set protocol {
    in  {in  "%{zero|one|two}bla"; out "%d";}
    out1 {out "%#{zero|one|two}bla";}
    in1  {in  "%#{zero|one|two}bla"; out "%d";}
-    out2 {out "%#{zero=-1|one|two=5}bla";}
+    out2 {out "%#{zero=-1|one|two=5|default=?}bla";}
    in2  {in  "%#{zero=-1|one|two=5}bla"; out "%d";}
    in3  {in  "%{\x00|\r|}bla"; out "%d";}
 }
@ -81,6 +81,8 @@ ioccmd {dbpf DZ:testout2 0}
 assure "onebla\n"
 ioccmd {dbpf DZ:testout2 5}
 assure "twobla\n"
+ioccmd {dbpf DZ:testout2 17}
+assure "defaultbla\n"

 ioccmd {dbpf DZ:testin.PROC 1}
 send "zerobla\n"
--- a/streamApp/tests/testFormats
+++ b/streamApp/tests/testFormats
@ -68,15 +68,15 @@ ioccmd {dbpf DZ:lo 12345}
 assure "12345 12345 012345 3039 003039 11000000111001 11000000111001 111001 !!......!!!..! !!......!!!..!\n"
 ioccmd {dbpf DZ:lo -1}
 if {$tcl_platform(machine) == "x86_64"} {
-assure "-1 -1 -00001 ffffffffffffffff FFFFFFFFFFFFFFFF 1111111111111111111111111111111111111111111111111111111111111111 1111111111111111111111111111111111111111111111111111111111111111 111111 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!\n"
+assure "-1 -1 -00001 ffffffffffffffff FFFFFF 11111111111111111111111111111111 11111111111111111111111111111111 111111 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!\n"
 } else {
-assure "-1 -1 -00001 ffffffff FFFFFFFF 11111111111111111111111111111111 11111111111111111111111111111111 111111 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!\n"
+assure "-1 -1 -00001 ffffffff FFFFFF 11111111111111111111111111111111 11111111111111111111111111111111 111111 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!\n"
 }
 ioccmd {dbpf DZ:lo -1234}
 if {$tcl_platform(machine) == "x86_64"} {
-assure "-1234 -1234 -01234 fffffffffffffb2e FFFFFFFFFFFFFB2E 1111111111111111111111111111111111111111111111111111101100101110 1111111111111111111111111111111111111111111111111111101100101110 101110 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!.!!..!.!!!. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!.!!..!.!!!.\n"
+assure "-1234 -1234 -01234 fffffffffffffb2e FFFB2E 11111111111111111111101100101110 11111111111111111111101100101110 101110 !!!!!!!!!!!!!!!!!!!!!.!!..!.!!!. !!!!!!!!!!!!!!!!!!!!!.!!..!.!!!.\n"
 } else {
-assure "-1234 -1234 -01234 fffffb2e FFFFFB2E 11111111111111111111101100101110 11111111111111111111101100101110 101110 !!!!!!!!!!!!!!!!!!!!!.!!..!.!!!. !!!!!!!!!!!!!!!!!!!!!.!!..!.!!!.\n"
+assure "-1234 -1234 -01234 fffffb2e FFFB2E 11111111111111111111101100101110 11111111111111111111101100101110 101110 !!!!!!!!!!!!!!!!!!!!!.!!..!.!!!. !!!!!!!!!!!!!!!!!!!!!.!!..!.!!!.\n"
 }
 ioccmd {dbpf DZ:lo 255}
 assure "255 255 000255 ff 0000FF 11111111 11111111 111111 !!!!!!!! !!!!!!!!\n"
--- a/streamApp/tests/testInteger
+++ b/streamApp/tests/testInteger
@ -10,16 +10,19 @@ set records {
    record (ai, "DZ:test1")
    {
        field (DTYP, "stream")
+        field (LINR, "LINEAR")
        field (INP,  "@test.proto test1 device")
    }
    record (ai, "DZ:test2")
    {
        field (DTYP, "stream")
+        field (LINR, "LINEAR")
        field (INP,  "@test.proto test2 device")
    }
    record (ai, "DZ:test3")
    {
        field (DTYP, "stream")
+        field (LINR, "LINEAR")
        field (INP,  "@test.proto test3 device")
    }
    record (ai, "DZ:test4")
@ -30,26 +33,31 @@ set records {
    record (ai, "DZ:test5")
    {
        field (DTYP, "stream")
+        field (LINR, "LINEAR")
        field (INP,  "@test.proto test5 device")
    }
    record (ai, "DZ:test6")
    {
        field (DTYP, "stream")
+        field (LINR, "LINEAR")
        field (INP,  "@test.proto test6 device")
    }
    record (ai, "DZ:test7")
    {
        field (DTYP, "stream")
+        field (LINR, "LINEAR")
        field (INP,  "@test.proto test7 device")
    }
    record (ai, "DZ:test8")
    {
        field (DTYP, "stream")
+        field (LINR, "LINEAR")
        field (INP,  "@test.proto test8 device")
    }
    record (ai, "DZ:test9")
    {
        field (DTYP, "stream")
+        field (LINR, "LINEAR")
        field (INP,  "@test.proto test9 device")
    }
    record (ai, "DZ:test10")
--- a/streamApp/tests/testStreamBuffer
+++ b/streamApp/tests/testStreamBuffer
@ -53,13 +53,22 @@ int main () {
 }
 EOF

-g++ -I ../../src ../../src/O.$EPICS_HOST_ARCH/StreamBuffer.o test.cc -o test.exe
-
-test.exe
-if [ $? != 0 ]
+if [ "$1" = "-sls" ]
 then
-    echo -e "\033[31;7mTest failed.\033[0m"
-    exit 1
+    O=../../O.*_$EPICS_HOST_ARCH/StreamBuffer.o
+else
+    O=../../src/O.$EPICS_HOST_ARCH/StreamBuffer.o
 fi
+
+for o in $O
+do
+    g++ -I ../../src $o test.cc -o test.exe
+    test.exe
+    if [ $? != 0 ]
+    then
+        echo -e "\033[31;7mTest failed.\033[0m"
+        exit 1
+    fi
+done
 rm test.*
 echo -e "\033[32mTest passed.\033[0m"
Author	SHA1	Message	Date
zimoch	e2af7e3149	typo fixed	2015-09-15 07:21:41 +00:00
zimoch	950d75ba43	Split long_format into signed_format and unsigned_format to hangle sign extension from EPICS LONG or ULONG to long or unsigned long correctly on systems with 64 bit long.	2015-08-31 15:25:23 +00:00
zimoch	e651c99697	fix tests	2015-08-18 14:34:11 +00:00
zimoch	9694628d3f	limit number of significant half bytes in %x formats to avoid problems on 64 bit machines	2015-08-18 14:33:48 +00:00
zimoch	046582d1a0	add default choice in enum output	2015-08-18 14:32:24 +00:00
zimoch	f6f0ea9ae5	fix redirection of strings to char arrays	2015-08-18 14:30:54 +00:00
zimoch	018547cdf3	allow and document direct conversion from VAL/OVAL to integer to get more bits on 64 bit systems	2015-08-18 14:28:26 +00:00
zimoch	d763a64f03	fix for 64 bit machines	2015-08-17 13:34:00 +00:00
zimoch	cefa460a0d	find correct object code	2015-08-17 13:33:35 +00:00
zimoch	cc5948ad0b	If ai uses no conversion, transport long i/o to and from val directly. This preverves more bits on 64 bit systems than going through the 32 bit rval field.	2015-07-17 13:24:37 +00:00
zimoch	be1943a66d	fix for version check	2015-06-25 16:06:39 +00:00
zimoch	d9acb47afe	make it work with old and new version of driver.makefile	2015-06-25 16:03:49 +00:00
zimoch	b721d26fb8	avoid buffer overflow	2015-04-08 09:35:00 +00:00
zimoch	7530c7c67d	check for error condifiton	2015-04-08 09:34:46 +00:00
zimoch	824cd4d283	debug message improved	2015-04-02 13:16:22 +00:00
zimoch	2e30e7aee6	fix message when INP/OUT link is empty fix bug with formatting arrays of char as strings in redirect	2015-04-02 13:15:46 +00:00
zimoch	3a5a3e22b9	Stupid error: Minutes and seconds start from 0, not 1.	2015-02-10 16:48:39 +00:00
zimoch	51bd766ba6	Fix or rather workaround for MinGW and maybe other systems without localtime_r function	2014-06-24 13:20:19 +00:00
zimoch	0ab0d6db07	function pointer type problem fixed	2014-06-24 13:19:36 +00:00
zimoch	a12de614d6	script to make a pdf from the hrml pages	2014-05-02 15:10:18 +00:00
zimoch	240abd2788	new logo and design cleanup, printing improved	2014-05-02 15:09:49 +00:00
zimoch	e131fcb32f	new logo, png instead of gif, better quality	2014-05-02 15:08:36 +00:00
zimoch	f88f594e87	shorter header, better for printing	2014-05-02 15:07:06 +00:00