Here is the simple debug parser and the debug2html converter. Still to do:

  * Debug message filtering.
  * I need to add all this to Makefile.in
    (If it looks at all strange I'll ask for help.)

If you want to compile debug2html, you'll need to do it by hand until I
make the changes to Makefile.in.  Sorry.

Chris -)-----
(This used to be commit fae161b9051c3ea8aa129bdb3df11d215007723d)

3 files changed, 635 insertions, 0 deletions

diff --git a/source3/include/debugparse.h b/source3/include/debugparse.h
new file mode 100644
index 0000000000..224141f75b
--- /dev/null
+++ b/source3/include/debugparse.h
@@ -0,0 +1,74 @@
+#ifndef DEBUGPARSE_H
+#define DEBUGPARSE_H
+/* ========================================================================== **
+ *                                debugparse.c
+ *
+ * Copyright (C) 1998 by Christopher R. Hertel
+ *
+ * Email: crh@ubiqx.mn.org
+ *
+ * -------------------------------------------------------------------------- **
+ * This module is a very simple parser for Samba debug log files.
+ * -------------------------------------------------------------------------- **
+ *
+ *  This library is free software; you can redistribute it and/or
+ *  modify it under the terms of the GNU Library General Public
+ *  License as published by the Free Software Foundation; either
+ *  version 2 of the License, or (at your option) any later version.
+ *
+ *  This library is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ *  Library General Public License for more details.
+ *
+ *  You should have received a copy of the GNU Library General Public
+ *  License along with this library; if not, write to the Free
+ *  Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ *
+ * -------------------------------------------------------------------------- **
+ * The important function in this module is dbg_char2token().  The rest is
+ * basically fluff.  (Potentially useful fluff, but still fluff.)
+ *
+ * NOTE:  Use this header if you are compiling with Samba headers.  See
+ *        debugparse-nonsamba.h for an alternate version.
+ *
+ * -------------------------------------------------------------------------- **
+ *
+ * $Log: debugparse.h,v $
+ * Revision 1.1  1998/10/26 23:21:37  crh
+ * Here is the simple debug parser and the debug2html converter.  Still to do:
+ *
+ *   * Debug message filtering.
+ *   * I need to add all this to Makefile.in
+ *     (If it looks at all strange I'll ask for help.)
+ *
+ * If you want to compile debug2html, you'll need to do it by hand until I
+ * make the changes to Makefile.in.  Sorry.
+ *
+ * Chris -)-----
+ *
+ * ========================================================================== **
+ */
+
+#include "includes.h"
+
+/* -------------------------------------------------------------------------- **
+ * These are the tokens returned by dbg_char2token().
+ */
+
+typedef enum
+  {
+  dbg_null = 0,
+  dbg_ignore,
+  dbg_header,
+  dbg_timestamp,
+  dbg_level,
+  dbg_sourcefile,
+  dbg_function,
+  dbg_lineno,
+  dbg_message,
+  dbg_eof
+  } dbg_Token;
+
+/* -------------------------------------------------------------------------- */
+#endif /* DEBUGPARSE_H */
diff --git a/source3/lib/debugparse.c b/source3/lib/debugparse.c
new file mode 100644
index 0000000000..b0f4e487ab
--- /dev/null
+++ b/source3/lib/debugparse.c
@@ -0,0 +1,322 @@
+/* ========================================================================== **
+ *                                debugparse.c
+ *
+ * Copyright (C) 1998 by Christopher R. Hertel
+ *
+ * Email: crh@ubiqx.mn.org
+ *
+ * -------------------------------------------------------------------------- **
+ * This module is a very simple parser for Samba debug log files.
+ * -------------------------------------------------------------------------- **
+ *
+ *  This library is free software; you can redistribute it and/or
+ *  modify it under the terms of the GNU Library General Public
+ *  License as published by the Free Software Foundation; either
+ *  version 2 of the License, or (at your option) any later version.
+ *
+ *  This library is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ *  Library General Public License for more details.
+ *
+ *  You should have received a copy of the GNU Library General Public
+ *  License along with this library; if not, write to the Free
+ *  Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ *
+ * -------------------------------------------------------------------------- **
+ * The important function in this module is dbg_char2token().  The rest is
+ * basically fluff.  (Potentially useful fluff, but still fluff.)
+ * -------------------------------------------------------------------------- **
+ *
+ * $Log: debugparse.c,v $
+ * Revision 1.1  1998/10/26 23:21:37  crh
+ * Here is the simple debug parser and the debug2html converter.  Still to do:
+ *
+ *   * Debug message filtering.
+ *   * I need to add all this to Makefile.in
+ *     (If it looks at all strange I'll ask for help.)
+ *
+ * If you want to compile debug2html, you'll need to do it by hand until I
+ * make the changes to Makefile.in.  Sorry.
+ *
+ * Chris -)-----
+ *
+ * ========================================================================== **
+ */
+
+#include "debugparse.h"
+
+/* -------------------------------------------------------------------------- **
+ * Constants...
+ *
+ *  BSIZE - This internal constant is used only by dbg_test().  It is the
+ *          size of the read buffer.  I've tested the function using a
+ *          BSIZE value of 2.
+ */
+
+#define BSIZE 128
+
+/* -------------------------------------------------------------------------- **
+ * Functions...
+ */
+
+char *dbg_token2string( dbg_Token tok )
+  /* ------------------------------------------------------------------------ **
+   * Given a token, return a string describing the token.
+   *
+   *  Input:  tok - One of the set of dbg_Tokens defined in debugparse.h.
+   *
+   *  Output: A string identifying the token.  This is useful for debugging,
+   *          etc.
+   *
+   *  Note:   If the token is not known, this function will return the
+   *          string "<unknown>".
+   *
+   * ------------------------------------------------------------------------ **
+   */
+  {
+  switch( tok )
+    {
+    case dbg_null:
+      return( "null" );
+    case dbg_ignore:
+      return( "ignore" );
+    case dbg_header:
+      return( "header" );
+    case dbg_timestamp:
+      return( "time stamp" );
+    case dbg_level:
+      return( "level" );
+    case dbg_sourcefile:
+      return( "source file" );
+    case dbg_function:
+      return( "function" );
+    case dbg_lineno:
+      return( "line number" );
+    case dbg_message:
+      return( "message" );
+    case dbg_eof:
+      return( "[EOF]" );
+    }
+  return( "<unknown>" );
+  } /* dbg_token2string */
+
+dbg_Token dbg_char2token( dbg_Token *state, int c )
+  /* ------------------------------------------------------------------------ **
+   * Parse input one character at a time.
+   *
+   *  Input:  state - A pointer to a token variable.  This is used to
+   *                  maintain the parser state between calls.  For
+   *                  each input stream, you should set up a separate
+   *                  state variable and initialize it to dbg_null.
+   *                  Pass a pointer to it into this function with each
+   *                  character in the input stream.  See dbg_test()
+   *                  for an example.
+   *          c     - The "current" character in the input stream.
+   *
+   *  Output: A token.
+   *          The token value will change when delimiters are found,
+   *          which indicate a transition between syntactical objects.
+   *          Possible return values are:
+   *
+   *          dbg_null        - The input character was an end-of-line.
+   *                            This resets the parser to its initial state
+   *                            in preparation for parsing the next line.
+   *          dbg_eof         - Same as dbg_null, except that the character
+   *                            was an end-of-file.
+   *          dbg_ignore      - Returned for whitespace and delimiters.
+   *                            These lexical tokens are only of interest
+   *                            to the parser.
+   *          dbg_header      - Indicates the start of a header line.  The
+   *                            input character was '[' and was the first on
+   *                            the line.
+   *          dbg_timestamp   - Indicates that the input character was part
+   *                            of a header timestamp.
+   *          dbg_level       - Indicates that the input character was part
+   *                            of the debug-level value in the header.
+   *          dbg_sourcefile  - Indicates that the input character was part
+   *                            of the sourcefile name in the header.
+   *          dbg_function    - Indicates that the input character was part
+   *                            of the function name in the header.
+   *          dbg_lineno      - Indicates that the input character was part
+   *                            of the DEBUG call line number in the header.
+   *          dbg_message     - Indicates that the input character was part
+   *                            of the DEBUG message text.
+   *
+   * ------------------------------------------------------------------------ **
+   */
+  {
+  /* The terminating characters that we see will greatly depend upon
+   * how they are read.  For example, if gets() is used instead of
+   * fgets(), then we will not see newline characters.  A lot also
+   * depends on the calling function, which may handle terminators
+   * itself.
+   *
+   * '\n', '\0', and EOF are all considered line terminators.  The
+   * dbg_eof token is sent back if an EOF is encountered.
+   *
+   * Warning:  only allow the '\0' character to be sent if you are
+   *           using gets() to read whole lines (thus replacing '\n'
+   *           with '\0').  Sending '\0' at the wrong time will mess
+   *           up the parsing.
+   */
+  switch( c )
+    {
+    case EOF:
+      *state = dbg_null;   /* Set state to null (initial state) so */
+      return( dbg_eof );   /* that we can restart with new input.  */
+    case '\n':
+    case '\0':
+      *state = dbg_null;   /* A newline or eoln resets to the null state. */
+      return( dbg_null );
+    }
+
+  /* When within the body of the message, only a line terminator
+   * can cause a change of state.  We've already checked for line
+   * terminators, so if the current state is dbg_msgtxt, simply
+   * return that as our current token.
+   */
+  if( dbg_message == *state )
+    return( dbg_message );
+
+  /* If we are at the start of a new line, and the input character 
+   * is an opening bracket, then the line is a header line, otherwise
+   * it's a message body line.
+   */
+  if( dbg_null == *state )
+    {
+    if( '[' == c )
+      {
+      *state = dbg_timestamp;
+      return( dbg_header );
+      }
+    *state = dbg_message;
+    return( dbg_message );
+    }
+
+  /* We've taken care of terminators, text blocks and new lines.
+   * The remaining possibilities are all within the header line
+   * itself.
+   */
+
+  /* Within the header line, whitespace can be ignored *except*
+   * within the timestamp.
+   */
+  if( isspace( c ) )
+    {
+    /* Fudge.  The timestamp may contain space characters. */
+    if( (' ' == c) && (dbg_timestamp == *state) )
+      return( dbg_timestamp );
+    /* Otherwise, ignore whitespace. */
+    return( dbg_ignore );
+    }
+
+  /* Okay, at this point we know we're somewhere in the header.
+   * Valid header *states* are: dbg_timestamp, dbg_level,
+   * dbg_sourcefile, dbg_function, and dbg_lineno.
+   */
+  switch( c )
+    {
+    case ',':
+      if( dbg_timestamp == *state )
+        {
+        *state = dbg_level;
+        return( dbg_ignore );
+        }
+      break;
+    case ']':
+      if( dbg_level == *state )
+        {
+        *state = dbg_sourcefile;
+        return( dbg_ignore );
+        }
+      break;
+    case ':':
+      if( dbg_sourcefile == *state )
+        {
+        *state = dbg_function;
+        return( dbg_ignore );
+        }
+      break;
+    case '(':
+      if( dbg_function == *state )
+        {
+        *state = dbg_lineno;
+        return( dbg_ignore );
+        }
+      break;
+    case ')':
+      if( dbg_lineno == *state )
+        {
+        *state = dbg_null;
+        return( dbg_ignore );
+        }
+      break;
+    }
+
+  /* If the previous block did not result in a state change, then
+   * return the current state as the current token.
+   */
+  return( *state );
+  } /* dbg_char2token */
+
+void dbg_test( void )
+  /* ------------------------------------------------------------------------ **
+   * Simple test function.
+   *
+   *  Input:  none.
+   *  Output: none.
+   *  Notes:  This function was used to test dbg_char2token().  It reads a
+   *          Samba log file from stdin and prints parsing info to stdout.
+   *          It also serves as a simple example.
+   *
+   * ------------------------------------------------------------------------ **
+   */
+  {
+  char bufr[BSIZE];
+  int  i;
+  int  linecount  = 1;
+  dbg_Token old   = dbg_null,
+            new   = dbg_null,
+            state = dbg_null;
+
+  while( fgets( bufr, BSIZE, stdin ) )
+    {
+    for( i = 0; bufr[i]; i++ )
+      {
+      old = new;
+      new = dbg_char2token( &state, bufr[i] );
+      switch( new )
+        {
+        case dbg_header:
+          if( linecount > 1 )
+            (void)putchar( '\n' );
+          break;
+        case dbg_null:
+          linecount++;
+          break;
+        case dbg_ignore:
+          break;
+        default:
+          if( old != new )
+            (void)printf( "\n[%05d]%12s: ", linecount, dbg_token2string(new) );
+          (void)putchar( bufr[i] );
+        }
+      }
+    }
+  (void)putchar( '\n' );
+  } /* dbg_test */
+
+
+/* -------------------------------------------------------------------------- **
+ * This simple main line can be uncommented and used to test the parser.
+ */
+/*
+int main( void )
+  {
+  dbg_test();
+  return( 0 );
+  }
+*/
+
+/* ========================================================================== */
diff --git a/source3/utils/debug2html.c b/source3/utils/debug2html.c
new file mode 100644
index 0000000000..ff891f0385
--- /dev/null
+++ b/source3/utils/debug2html.c
@@ -0,0 +1,239 @@
+/* ========================================================================== **
+ *                                debug2html.c
+ *
+ * Copyright (C) 1998 by Christopher R. Hertel
+ *
+ * Email: crh@ubiqx.mn.org
+ *
+ * -------------------------------------------------------------------------- **
+ * Parse Samba debug logs (2.0 & greater) and output the results as HTML.
+ * -------------------------------------------------------------------------- **
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ *
+ * -------------------------------------------------------------------------- **
+ * This program provides an example of the use of debugparse.c, and also
+ * does a decent job of converting Samba logs into HTML.
+ * -------------------------------------------------------------------------- **
+ *
+ * $Log: debug2html.c,v $
+ * Revision 1.1  1998/10/26 23:21:37  crh
+ * Here is the simple debug parser and the debug2html converter.  Still to do:
+ *
+ *   * Debug message filtering.
+ *   * I need to add all this to Makefile.in
+ *     (If it looks at all strange I'll ask for help.)
+ *
+ * If you want to compile debug2html, you'll need to do it by hand until I
+ * make the changes to Makefile.in.  Sorry.
+ *
+ * Chris -)-----
+ *
+ * ========================================================================== **
+ */
+
+#include "debugparse.h"
+
+/* -------------------------------------------------------------------------- **
+ * The size of the read buffer.
+ */
+
+#define BSIZE 1024
+
+/* -------------------------------------------------------------------------- **
+ * Functions...
+ */
+
+static dbg_Token modechange( dbg_Token new, dbg_Token mode )
+  /* ------------------------------------------------------------------------ **
+   * Handle a switch between header and message printing.
+   *
+   *  Input:  new   - The token value of the current token.  This indicates
+   *                  the lexical item currently being recognized.
+   *          mode  - The current mode.  This is either dbg_null or
+   *                  dbg_message.  It could really be any toggle
+   *                  (true/false, etc.)
+   *
+   *  Output: The new mode.  This will be the same as the input mode unless
+   *          there was a transition in or out of message processing.
+   *
+   *  Notes:  The purpose of the mode value is to mark the beginning and end
+   *          of the message text block.  In order to show the text in its
+   *          correct format, it must be included within a <PRE></PRE> block.
+   *
+   * ------------------------------------------------------------------------ **
+   */
+  {
+  switch( new )
+    {
+    case dbg_null:
+    case dbg_ignore:
+      return( mode );
+    case dbg_message:
+      if( dbg_message != mode )
+        {
+        /* Switching to message mode. */
+        (void)printf( "<PRE>\n" );
+        return( dbg_message );
+        }
+      break;
+    default:
+      if( dbg_message == mode )
+        {
+        /* Switching out of message mode. */
+        (void)printf( "</PRE>\n\n" );
+        return( dbg_null );
+        }
+    }
+
+  return( mode );
+  } /* modechange */
+
+static void newblock( dbg_Token old, dbg_Token new )
+  /* ------------------------------------------------------------------------ **
+   * Handle the transition between tokens.
+   *
+   *  Input:  old - The previous token.
+   *          new - The current token.
+   *
+   *  Output: none.
+   *
+   *  Notes:  This is called whenever there is a transition from one token
+   *          type to another.  It first prints the markup tags that close
+   *          the previous token, and then the markup tags for the new
+   *          token.
+   *
+   * ------------------------------------------------------------------------ **
+   */
+  {
+  switch( old )
+    {
+    case dbg_timestamp:
+      (void)printf( ", " );
+      break;
+    case dbg_level:
+      (void)printf( "</FONT>]</B>\n   " );
+      break;
+    case dbg_sourcefile:
+      (void)printf( ":" );
+      break;
+    case dbg_lineno:
+      (void)printf( ")" );
+      break;
+    }
+
+  switch( new )
+    {
+    case dbg_timestamp:
+      (void)printf( "<B>[" );
+      break;
+    case dbg_level:
+      (void)printf( "<FONT COLOR=MAROON>" );
+      break;
+    case dbg_lineno:
+      (void)printf( "(" );
+      break;
+    }
+  } /* newblock */
+
+static void charprint( dbg_Token tok, int c )
+  /* ------------------------------------------------------------------------ **
+   * Filter the input characters to determine what goes to output.
+   *
+   *  Input:  tok - The token value of the current character.
+   *          c   - The current character.
+   *
+   *  Output: none.
+   *
+   * ------------------------------------------------------------------------ **
+   */
+  {
+  switch( tok )
+    {
+    case dbg_ignore:
+    case dbg_header:
+      break;
+    case dbg_null:
+    case dbg_eof:
+      (void)putchar( '\n' );
+      break;
+    default:
+      switch( c )
+        {
+        case '<':
+          (void)printf( "&lt;" );
+          break;
+        case '>':
+          (void)printf( "&gt;" );
+          break;
+        case '&':
+          (void)printf( "&amp;" );
+          break;
+        case '\"':
+          (void)printf( "&#34;" );
+          break;
+        default:
+          (void)putchar( c );
+          break;
+        }
+    }
+  } /* charprint */
+
+int main( int argc, char *argv[] )
+  /* ------------------------------------------------------------------------ **
+   * This simple program scans and parses Samba debug logs, and produces HTML
+   * output.
+   *
+   *  Input:  argc  - Currently ignored.
+   *          argv  - Currently ignored.
+   *
+   *  Output: Always zero.
+   *
+   *  Notes:  The HTML output is sent to stdout.
+   *
+   * ------------------------------------------------------------------------ **
+   */
+  {
+  int       i;
+  int       len;
+  char      bufr[BSIZE];
+  dbg_Token old   = dbg_null,
+            new   = dbg_null,
+            state = dbg_null,
+            mode  = dbg_null;
+
+  (void)printf( "<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 3.2//EN\">\n" );
+  (void)printf( "<HTML>\n<HEAD>\n" );
+  (void)printf( "  <TITLE>Samba Debug Output</TITLE>\n</HEAD>\n\n<BODY>\n" );
+
+  while( (len = fread( bufr, 1, BSIZE, stdin )) > 0 )
+    {
+    for( i = 0; i < len; i++ )
+      {
+      old = new;
+      new = dbg_char2token( &state, bufr[i] );
+      if( new != old )
+        {
+        mode = modechange( new, mode );
+        newblock( old, new );
+        }
+      charprint( new, bufr[i] );
+      }
+    }
+  (void)modechange( dbg_eof, mode );
+
+  (void)printf( "</BODY>\n</HTML>\n" );
+  return( 0 );
+  } /* main */