Package: libsmi2ldbl
Version: 0.4.8+dfsg-1
Severity: normal
File: libsmi2
Tags: patch
smistrip has problems correctly stripping certain RFCs:
- Some older RFCs have several blank lines between footer and header.
smistrip fails to detect the header line in this case. RFC2580 shows this
problem.
- RFC4780 (SIP-TC-MIB) has a line starting with "Internet-Draft". This line
incorrectly gets stripped from the MIB.
Net-SNMP has modified the script to get rid of these problems and
additionally stripping blank line more aggressively. These modifications are
however based on an older version of libsmi. I rebased these modifications
and made the aggressive blank line stripping optional.
--- smistrip 2009-11-25 13:01:14.000000000 +0100
+++ smistrip 2009-11-26 16:41:57.000000000 +0100
@@ -5,6 +5,13 @@
# Extract MIB and PIB modules from text files, like RFCs or I-Ds.
#
# Copyright (c) 1999 Frank Strauss, Technical University of Braunschweig.
+# Copyright (c) Niels Baggesen, Jochen Friedrich
+#
+# Modified by Niels Baggesen to be somewhat more aggressive in suppressing
+# blank lines, and support the -x option.
+#
+# Modified by Jochen Friedrich to merge the changes of libsmi back in and
+# make the aggressive suppressing of blank lines optional.
#
# See the file "COPYING" for information on usage and redistribution
# of this file, and for a DISCLAIMER OF ALL WARRANTIES.
@@ -27,13 +34,17 @@
do_usage () {
- echo "Usage: smistrip [-Vhn] [-i dir] [-d dir] [-m module] file1 [file2
[...]]"
+ echo "Usage: smistrip [-Vvhna] [-i dir] [-d dir] [-s suffix] [-m modules]
file1 [file2 [...]]"
echo "-V show version and license information"
+ echo "-v verbose"
echo "-h show usage information"
echo "-n do not write module files"
+ echo "-a strip blank lines more aggressively"
echo "-i dir try to read files from directory dir"
echo "-d dir write module to directory dir"
- echo "-m module strip only the specified module"
+ echo "-x suffix append suffix to the module file name"
+ echo "-m modules strip only the specified modules. For a list of modules"
+ echo " use : as a separator"
echo "file1 ... input files to parse (RFCs, I-Ds, ...)"
}
@@ -57,65 +68,90 @@
$CMD "$FILE" | \
tr -d '\015' | \
- grep -i -v '^[ ]*Internet[ \-]Draft' | \
- $AWK -vtest="$test" -vdir="$dir" -vsingle="$single" '
+ $AWK -vtest="$test" -vdir="$dir" -vsingle="$single" -vsuffix="$suffix"
-vverbose="$verbose" -vaggressive="$aggressive" '
+
+ BEGIN {
+ if (length(single) != 0) {
+ single = ":"single":"
+ }
+ else {
+ single = ""
+ }
+ }
+
+ END {
+ if (single != "" && single != ":") {
+ gsub(":", " ", single)
+ print "WARNING: Module(s) not found:" single
+ }
+ }
# start of module
/^[ \t]*[A-Za-z0-9-]* *(PIB-)?DEFINITIONS *(::=)? *(BEGIN)? *$/ {
module = $1
- skip = 9
- skipped = -1
+ collect = 1
macro = 0
+ skip = 0
n = 0
}
- # process each line
- {
- # at the end of a page we go back one line (which is expected to
- # be a separator line), and start the counter skipped to skip the
- # next few lines.
- if ($0 ~ /\[[pP]age [iv0-9]*\] */) {
- # some drafts do not use that separator line. so keep it if
- # there are non-blank characters.
- if (!(line[n] ~ /^[ \t]*$/)) { print "WARNING: the line
::"line[n]":: should be a separator before a page break. It was kept. " ; n-- }
- skipped = 0
- }
-
- # if we are skipping...
- if (skipped >= 0) {
- skipped++
-
- # if we have skipped enough lines to the top of the next page...
- if (skipped >= skip) {
- skipped = -1
- } else {
-
- # finish skipping, if we find a non-empty line, but not before
- # we have skipped four lines. remember the miminum of lines
- # we have ever skipped to keep empty lines in a modules that
- # appear near the top of a page.
- if ((skipped >= 4) && ($0 ~ /[^ \t]/)) {
- if (skipped < skip) { skip = skipped }
- skipped = -1
- }
+ # at the end of a page we go back one line (which is expected to
+ # be a separator line), and start the counter skipped to skip the
+ # next few lines.
+ /\[[pP]age [iv0-9]*\] */ {
+ # some drafts do not use that separator line. so keep it if
+ # there are non-blank characters.
+ if (!aggressive && n && collect) {
+ if (!(line[n-1] == ""))
+ print "WARNING: the line ::"line[n-1]":: should be a separator
before a page break. It was kept. ";
+ else n--;
+ skip = 3
+ }
+ collect = 0
+ next
+ }
+
+ /^[
\t]*(::=|DESCRIPTION|SYNTAX|MAX-ACCESS|MIN-ACCESS|ACCESS|STATUS|REFERENCE|INDEX|AUGMENTS|DEFVAL|UNITS|DISPLAY|")/
{
+ skip = 0
+ if (collect && aggressive)
+ if (line[n-1] == "") n--
+ }
+
+ # a blank line - suppress multiple
+ /^[ \t\r]*$/ {
+ if (collect && (skip == 0)) {
+ if (aggressive && n) {
+ if (line[n-1] != "" && line[n-1] !~ /,[ \t\r]*$/) line[n++] =
""
}
+ else line[n++] = ""
}
+ if (skip > 0) skip--;
+ next
+ }
- # so, if we are not skipping and inside a module, remember the line.
- if ((skipped == -1) && (length(module) > 0)) {
- line[n++] = $0
+ # collect non-blank line when inside mib module
+ /[^ \f\t]/ {
+ if (length(module) > 0) {
+ if (!collect)
+ collect = 1 # page header, stop skipping
+ else if (skip == 0)
+ line[n++] = $0
}
+ if (skip > 0) skip--;
}
# remember when we enter a macro definition
/^[ \t]*[A-Za-z0-9-]* *MACRO *::=/ {
macro = 1
+ skip = 0
}
# end of module
/^[ \t]*END[ \t]*$/ {
+ skip = 0
if (macro == 0) {
- if ((length(single) == 0) || (single == module)) {
+ if (single == "" || match(single, ":"module":")) {
+ sub(":"module, "", single)
strip = 99
for (i=0 ; i < n ; i++) {
# find the minimum column that contains non-blank characters
@@ -129,17 +165,21 @@
if (test != "1") {
if (dir) {
- f = dir"/"module
+ f = dir"/"module suffix
} else {
- f = module
+ f = module suffix
}
for (i=0 ; i < n ; i++) {
print substr(line[i], strip) >f
}
}
+ if (verbose) {
+ print module ": " n " lines."
+ }
- print module ": " n " lines."
}
+ else
+ print "NOTE: " module ": ignored."
module = ""
} else {
macro = 0
@@ -150,16 +190,22 @@
-while $GETOPTS Vhnm:i:d: c ; do
+while $GETOPTS Vvhnam:i:d:x: c ; do
case $c in
+ v) verbose=1
+ ;;
n) test=1
;;
+ a) aggressive=1
+ ;;
m) single=$OPTARG
;;
i) indir=$OPTARG
;;
d) dir=$OPTARG
;;
+ x) suffix=$OPTARG
+ ;;
h) do_usage
exit 0
;;
-- System Information:
Debian Release: squeeze/sid
APT prefers unstable
APT policy: (500, 'unstable'), (500, 'testing'), (500, 'stable')
Architecture: sparc
Kernel: Linux 2.6.31-1-sparc64
Locale: LANG=de_DE.UTF-8, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash
Versions of packages libsmi2ldbl depends on:
ii libc6 2.10.1-7 GNU C Library: Shared libraries
Versions of packages libsmi2ldbl recommends:
pn libsmi2-common <none> (no description available)
libsmi2ldbl suggests no packages.
-- no debconf information
--- smistrip 2009-11-25 13:01:14.000000000 +0100
+++ smistrip 2009-11-26 16:41:57.000000000 +0100
@@ -5,6 +5,13 @@
# Extract MIB and PIB modules from text files, like RFCs or I-Ds.
#
# Copyright (c) 1999 Frank Strauss, Technical University of Braunschweig.
+# Copyright (c) Niels Baggesen, Jochen Friedrich
+#
+# Modified by Niels Baggesen to be somewhat more aggressive in suppressing
+# blank lines, and support the -x option.
+#
+# Modified by Jochen Friedrich to merge the changes of libsmi back in and
+# make the aggressive suppressing of blank lines optional.
#
# See the file "COPYING" for information on usage and redistribution
# of this file, and for a DISCLAIMER OF ALL WARRANTIES.
@@ -27,13 +34,17 @@
do_usage () {
- echo "Usage: smistrip [-Vhn] [-i dir] [-d dir] [-m module] file1 [file2
[...]]"
+ echo "Usage: smistrip [-Vvhna] [-i dir] [-d dir] [-s suffix] [-m modules]
file1 [file2 [...]]"
echo "-V show version and license information"
+ echo "-v verbose"
echo "-h show usage information"
echo "-n do not write module files"
+ echo "-a strip blank lines more aggressively"
echo "-i dir try to read files from directory dir"
echo "-d dir write module to directory dir"
- echo "-m module strip only the specified module"
+ echo "-x suffix append suffix to the module file name"
+ echo "-m modules strip only the specified modules. For a list of modules"
+ echo " use : as a separator"
echo "file1 ... input files to parse (RFCs, I-Ds, ...)"
}
@@ -57,65 +68,90 @@
$CMD "$FILE" | \
tr -d '\015' | \
- grep -i -v '^[ ]*Internet[ \-]Draft' | \
- $AWK -vtest="$test" -vdir="$dir" -vsingle="$single" '
+ $AWK -vtest="$test" -vdir="$dir" -vsingle="$single" -vsuffix="$suffix"
-vverbose="$verbose" -vaggressive="$aggressive" '
+
+ BEGIN {
+ if (length(single) != 0) {
+ single = ":"single":"
+ }
+ else {
+ single = ""
+ }
+ }
+
+ END {
+ if (single != "" && single != ":") {
+ gsub(":", " ", single)
+ print "WARNING: Module(s) not found:" single
+ }
+ }
# start of module
/^[ \t]*[A-Za-z0-9-]* *(PIB-)?DEFINITIONS *(::=)? *(BEGIN)? *$/ {
module = $1
- skip = 9
- skipped = -1
+ collect = 1
macro = 0
+ skip = 0
n = 0
}
- # process each line
- {
- # at the end of a page we go back one line (which is expected to
- # be a separator line), and start the counter skipped to skip the
- # next few lines.
- if ($0 ~ /\[[pP]age [iv0-9]*\] */) {
- # some drafts do not use that separator line. so keep it if
- # there are non-blank characters.
- if (!(line[n] ~ /^[ \t]*$/)) { print "WARNING: the line
::"line[n]":: should be a separator before a page break. It was kept. " ; n-- }
- skipped = 0
- }
-
- # if we are skipping...
- if (skipped >= 0) {
- skipped++
-
- # if we have skipped enough lines to the top of the next page...
- if (skipped >= skip) {
- skipped = -1
- } else {
-
- # finish skipping, if we find a non-empty line, but not before
- # we have skipped four lines. remember the miminum of lines
- # we have ever skipped to keep empty lines in a modules that
- # appear near the top of a page.
- if ((skipped >= 4) && ($0 ~ /[^ \t]/)) {
- if (skipped < skip) { skip = skipped }
- skipped = -1
- }
+ # at the end of a page we go back one line (which is expected to
+ # be a separator line), and start the counter skipped to skip the
+ # next few lines.
+ /\[[pP]age [iv0-9]*\] */ {
+ # some drafts do not use that separator line. so keep it if
+ # there are non-blank characters.
+ if (!aggressive && n && collect) {
+ if (!(line[n-1] == ""))
+ print "WARNING: the line ::"line[n-1]":: should be a separator
before a page break. It was kept. ";
+ else n--;
+ skip = 3
+ }
+ collect = 0
+ next
+ }
+
+ /^[
\t]*(::=|DESCRIPTION|SYNTAX|MAX-ACCESS|MIN-ACCESS|ACCESS|STATUS|REFERENCE|INDEX|AUGMENTS|DEFVAL|UNITS|DISPLAY|")/
{
+ skip = 0
+ if (collect && aggressive)
+ if (line[n-1] == "") n--
+ }
+
+ # a blank line - suppress multiple
+ /^[ \t\r]*$/ {
+ if (collect && (skip == 0)) {
+ if (aggressive && n) {
+ if (line[n-1] != "" && line[n-1] !~ /,[ \t\r]*$/) line[n++] =
""
}
+ else line[n++] = ""
}
+ if (skip > 0) skip--;
+ next
+ }
- # so, if we are not skipping and inside a module, remember the line.
- if ((skipped == -1) && (length(module) > 0)) {
- line[n++] = $0
+ # collect non-blank line when inside mib module
+ /[^ \f\t]/ {
+ if (length(module) > 0) {
+ if (!collect)
+ collect = 1 # page header, stop skipping
+ else if (skip == 0)
+ line[n++] = $0
}
+ if (skip > 0) skip--;
}
# remember when we enter a macro definition
/^[ \t]*[A-Za-z0-9-]* *MACRO *::=/ {
macro = 1
+ skip = 0
}
# end of module
/^[ \t]*END[ \t]*$/ {
+ skip = 0
if (macro == 0) {
- if ((length(single) == 0) || (single == module)) {
+ if (single == "" || match(single, ":"module":")) {
+ sub(":"module, "", single)
strip = 99
for (i=0 ; i < n ; i++) {
# find the minimum column that contains non-blank characters
@@ -129,17 +165,21 @@
if (test != "1") {
if (dir) {
- f = dir"/"module
+ f = dir"/"module suffix
} else {
- f = module
+ f = module suffix
}
for (i=0 ; i < n ; i++) {
print substr(line[i], strip) >f
}
}
+ if (verbose) {
+ print module ": " n " lines."
+ }
- print module ": " n " lines."
}
+ else
+ print "NOTE: " module ": ignored."
module = ""
} else {
macro = 0
@@ -150,16 +190,22 @@
-while $GETOPTS Vhnm:i:d: c ; do
+while $GETOPTS Vvhnam:i:d:x: c ; do
case $c in
+ v) verbose=1
+ ;;
n) test=1
;;
+ a) aggressive=1
+ ;;
m) single=$OPTARG
;;
i) indir=$OPTARG
;;
d) dir=$OPTARG
;;
+ x) suffix=$OPTARG
+ ;;
h) do_usage
exit 0
;;
