Sync ober_oid_cmp with ax_oid_cmp from libagentx.
This flips the returned signedness and adds the weight of 2 for
parent-child relationship in both direction.

This makes ober_oid_cmp consistent with the rest of the *_cmp based
functions.

OK tb@

Remove BER_TYPE_BOOLEAN, it's not part of the spec and I haven't seen it
used anywhere.

OK jan@

Implement displaying SMI enums as text.
For now limit to TruthValue implemented on ifPromiscuousMode and
ifConnectorPresent.

OK jan@

Clean up smi_mibtree.

It checks on o_oid[0] for 0, which prevents us from adding the ccitt oid
and its zeroDotZero subsidiary. Change this to check for o_name, which is
always set, because mibtree is filled by MIBDECL().

Somewhere during working on UTF-8 support I thought it was a good idea to
use sysContact as a testcase, because it's readily available in snmpd, but
forgetting it's actual textual convention as time went by, which is
DisplayString (which is ASCII).

Add support for the DisplayString textual convention, which is similar to
having no textual convention, except that invalid bytes (value 128 and up)
are printed as UTF-8 replacement characters, or question mark, depending on
LC_CTYPE.

Feedback and OK schwarze@

Add support for UTF-8 DISPLAY-HINTs with octet length. For now only
sysContact is supported with SnmpAdminString, but more shall follow
soon(tm).

Note that this will change output even in the C locale: Previously every
unprintable byte would be tranformed to a '.', with this diff a valid UTF-8
multibyte or unprintable character will be squished into a single dot to
give a better understanding of the intention of the original message.
Invalid bytes will now be printed as question marks.

Lot's of help and back and forth with schwarze@ who also kindly offered to
walk away when I mentioned looking through MIB files for more objects to
convert to this new code, which is understandable.

OK schwarze@

OIDs are unsigned. Reflect this when printing them.

Shave of some leftovers from snmpd's MIB initialization that are not needed
for snmp. Minus 503 LoC and about 200KB on binary size.

OK denis@

Add better support for exception contexts.
Now we can see proper nosuchinstance and endofmibtree messages.

OK jan@

The ber_* namespace is used by liblber since time immemorial,
so move our BER API to the unused ober_* prefix to avoid some
breakage in ports.

Problem diagnosed by jmatthew with ber_free() in samba, but
there are many others as pointed out by sthen.

tests & ok rob
ok sthen (who had an almost identical diff for libutil)
"go head hit it" deraadt

branches: 1.5.2;
don't use sizeof(pointer) to represent buffer size
ok martijn@ deraadt@

Use snprintf instead of sprintf.

Requested by and OK deraadt@

whitespace issues like a knife in my eyes

ugly whitespaces

Import snmp(1), a new snmp client which aims to be netsnmp compatible for
supported features. It only supports get, getnext, walk, bulkget, bulkwalk,
trap, mibtree, and is SNMPv1 and SNMPv2c for now.

This will shortly replace snmpctl entirely. People using snmpctl are encouraged
to test and migrate to this code as soon as possible.

Much help with the manpage from schwarze@ and jmc@
No objections from reyk@
"Roll it in" deraadt@

Remove BER_TYPE_BOOLEAN, it's not part of the spec and I haven't seen it
used anywhere.

OK jan@

Implement displaying SMI enums as text.
For now limit to TruthValue implemented on ifPromiscuousMode and
ifConnectorPresent.

OK jan@

Clean up smi_mibtree.

It checks on o_oid[0] for 0, which prevents us from adding the ccitt oid
and its zeroDotZero subsidiary. Change this to check for o_name, which is
always set, because mibtree is filled by MIBDECL().

Somewhere during working on UTF-8 support I thought it was a good idea to
use sysContact as a testcase, because it's readily available in snmpd, but
forgetting it's actual textual convention as time went by, which is
DisplayString (which is ASCII).

Add support for the DisplayString textual convention, which is similar to
having no textual convention, except that invalid bytes (value 128 and up)
are printed as UTF-8 replacement characters, or question mark, depending on
LC_CTYPE.

Feedback and OK schwarze@

Add support for UTF-8 DISPLAY-HINTs with octet length. For now only
sysContact is supported with SnmpAdminString, but more shall follow
soon(tm).

Note that this will change output even in the C locale: Previously every
unprintable byte would be tranformed to a '.', with this diff a valid UTF-8
multibyte or unprintable character will be squished into a single dot to
give a better understanding of the intention of the original message.
Invalid bytes will now be printed as question marks.

Lot's of help and back and forth with schwarze@ who also kindly offered to
walk away when I mentioned looking through MIB files for more objects to
convert to this new code, which is understandable.

OK schwarze@

OIDs are unsigned. Reflect this when printing them.

Shave of some leftovers from snmpd's MIB initialization that are not needed
for snmp. Minus 503 LoC and about 200KB on binary size.

OK denis@

Add better support for exception contexts.
Now we can see proper nosuchinstance and endofmibtree messages.

OK jan@

The ber_* namespace is used by liblber since time immemorial,
so move our BER API to the unused ober_* prefix to avoid some
breakage in ports.

Problem diagnosed by jmatthew with ber_free() in samba, but
there are many others as pointed out by sthen.

tests & ok rob
ok sthen (who had an almost identical diff for libutil)
"go head hit it" deraadt

branches: 1.5.2;
don't use sizeof(pointer) to represent buffer size
ok martijn@ deraadt@

Use snprintf instead of sprintf.

Requested by and OK deraadt@

whitespace issues like a knife in my eyes

ugly whitespaces

Import snmp(1), a new snmp client which aims to be netsnmp compatible for
supported features. It only supports get, getnext, walk, bulkget, bulkwalk,
trap, mibtree, and is SNMPv1 and SNMPv2c for now.

This will shortly replace snmpctl entirely. People using snmpctl are encouraged
to test and migrate to this code as soon as possible.

Much help with the manpage from schwarze@ and jmc@
No objections from reyk@
"Roll it in" deraadt@

Implement displaying SMI enums as text.
For now limit to TruthValue implemented on ifPromiscuousMode and
ifConnectorPresent.

OK jan@

Clean up smi_mibtree.

It checks on o_oid[0] for 0, which prevents us from adding the ccitt oid
and its zeroDotZero subsidiary. Change this to check for o_name, which is
always set, because mibtree is filled by MIBDECL().

Somewhere during working on UTF-8 support I thought it was a good idea to
use sysContact as a testcase, because it's readily available in snmpd, but
forgetting it's actual textual convention as time went by, which is
DisplayString (which is ASCII).

Add support for the DisplayString textual convention, which is similar to
having no textual convention, except that invalid bytes (value 128 and up)
are printed as UTF-8 replacement characters, or question mark, depending on
LC_CTYPE.

Feedback and OK schwarze@

Add support for UTF-8 DISPLAY-HINTs with octet length. For now only
sysContact is supported with SnmpAdminString, but more shall follow
soon(tm).

Note that this will change output even in the C locale: Previously every
unprintable byte would be tranformed to a '.', with this diff a valid UTF-8
multibyte or unprintable character will be squished into a single dot to
give a better understanding of the intention of the original message.
Invalid bytes will now be printed as question marks.

Lot's of help and back and forth with schwarze@ who also kindly offered to
walk away when I mentioned looking through MIB files for more objects to
convert to this new code, which is understandable.

OK schwarze@

OIDs are unsigned. Reflect this when printing them.

Shave of some leftovers from snmpd's MIB initialization that are not needed
for snmp. Minus 503 LoC and about 200KB on binary size.

OK denis@

Add better support for exception contexts.
Now we can see proper nosuchinstance and endofmibtree messages.

OK jan@

The ber_* namespace is used by liblber since time immemorial,
so move our BER API to the unused ober_* prefix to avoid some
breakage in ports.

Problem diagnosed by jmatthew with ber_free() in samba, but
there are many others as pointed out by sthen.

tests & ok rob
ok sthen (who had an almost identical diff for libutil)
"go head hit it" deraadt

branches: 1.5.2;
don't use sizeof(pointer) to represent buffer size
ok martijn@ deraadt@

Use snprintf instead of sprintf.

Requested by and OK deraadt@

whitespace issues like a knife in my eyes

ugly whitespaces

Import snmp(1), a new snmp client which aims to be netsnmp compatible for
supported features. It only supports get, getnext, walk, bulkget, bulkwalk,
trap, mibtree, and is SNMPv1 and SNMPv2c for now.

This will shortly replace snmpctl entirely. People using snmpctl are encouraged
to test and migrate to this code as soon as possible.

Much help with the manpage from schwarze@ and jmc@
No objections from reyk@
"Roll it in" deraadt@

Clean up smi_mibtree.

It checks on o_oid[0] for 0, which prevents us from adding the ccitt oid
and its zeroDotZero subsidiary. Change this to check for o_name, which is
always set, because mibtree is filled by MIBDECL().

Somewhere during working on UTF-8 support I thought it was a good idea to
use sysContact as a testcase, because it's readily available in snmpd, but
forgetting it's actual textual convention as time went by, which is
DisplayString (which is ASCII).

Add support for the DisplayString textual convention, which is similar to
having no textual convention, except that invalid bytes (value 128 and up)
are printed as UTF-8 replacement characters, or question mark, depending on
LC_CTYPE.

Feedback and OK schwarze@

Add support for UTF-8 DISPLAY-HINTs with octet length. For now only
sysContact is supported with SnmpAdminString, but more shall follow
soon(tm).

Note that this will change output even in the C locale: Previously every
unprintable byte would be tranformed to a '.', with this diff a valid UTF-8
multibyte or unprintable character will be squished into a single dot to
give a better understanding of the intention of the original message.
Invalid bytes will now be printed as question marks.

Lot's of help and back and forth with schwarze@ who also kindly offered to
walk away when I mentioned looking through MIB files for more objects to
convert to this new code, which is understandable.

OK schwarze@

OIDs are unsigned. Reflect this when printing them.

Shave of some leftovers from snmpd's MIB initialization that are not needed
for snmp. Minus 503 LoC and about 200KB on binary size.

OK denis@

Add better support for exception contexts.
Now we can see proper nosuchinstance and endofmibtree messages.

OK jan@

The ber_* namespace is used by liblber since time immemorial,
so move our BER API to the unused ober_* prefix to avoid some
breakage in ports.

Problem diagnosed by jmatthew with ber_free() in samba, but
there are many others as pointed out by sthen.

tests & ok rob
ok sthen (who had an almost identical diff for libutil)
"go head hit it" deraadt

branches: 1.5.2;
don't use sizeof(pointer) to represent buffer size
ok martijn@ deraadt@

Use snprintf instead of sprintf.

Requested by and OK deraadt@

whitespace issues like a knife in my eyes

ugly whitespaces

Import snmp(1), a new snmp client which aims to be netsnmp compatible for
supported features. It only supports get, getnext, walk, bulkget, bulkwalk,
trap, mibtree, and is SNMPv1 and SNMPv2c for now.

This will shortly replace snmpctl entirely. People using snmpctl are encouraged
to test and migrate to this code as soon as possible.

Much help with the manpage from schwarze@ and jmc@
No objections from reyk@
"Roll it in" deraadt@

Clean up smi_mibtree.

It checks on o_oid[0] for 0, which prevents us from adding the ccitt oid
and its zeroDotZero subsidiary. Change this to check for o_name, which is
always set, because mibtree is filled by MIBDECL().

Somewhere during working on UTF-8 support I thought it was a good idea to
use sysContact as a testcase, because it's readily available in snmpd, but
forgetting it's actual textual convention as time went by, which is
DisplayString (which is ASCII).

Add support for the DisplayString textual convention, which is similar to
having no textual convention, except that invalid bytes (value 128 and up)
are printed as UTF-8 replacement characters, or question mark, depending on
LC_CTYPE.

Feedback and OK schwarze@

Add support for UTF-8 DISPLAY-HINTs with octet length. For now only
sysContact is supported with SnmpAdminString, but more shall follow
soon(tm).

Note that this will change output even in the C locale: Previously every
unprintable byte would be tranformed to a '.', with this diff a valid UTF-8
multibyte or unprintable character will be squished into a single dot to
give a better understanding of the intention of the original message.
Invalid bytes will now be printed as question marks.

Lot's of help and back and forth with schwarze@ who also kindly offered to
walk away when I mentioned looking through MIB files for more objects to
convert to this new code, which is understandable.

OK schwarze@

OIDs are unsigned. Reflect this when printing them.

Shave of some leftovers from snmpd's MIB initialization that are not needed
for snmp. Minus 503 LoC and about 200KB on binary size.

OK denis@

Add better support for exception contexts.
Now we can see proper nosuchinstance and endofmibtree messages.

OK jan@

The ber_* namespace is used by liblber since time immemorial,
so move our BER API to the unused ober_* prefix to avoid some
breakage in ports.

Problem diagnosed by jmatthew with ber_free() in samba, but
there are many others as pointed out by sthen.

tests & ok rob
ok sthen (who had an almost identical diff for libutil)
"go head hit it" deraadt

branches: 1.5.2;
don't use sizeof(pointer) to represent buffer size
ok martijn@ deraadt@

Use snprintf instead of sprintf.

Requested by and OK deraadt@

whitespace issues like a knife in my eyes

ugly whitespaces

Import snmp(1), a new snmp client which aims to be netsnmp compatible for
supported features. It only supports get, getnext, walk, bulkget, bulkwalk,
trap, mibtree, and is SNMPv1 and SNMPv2c for now.

This will shortly replace snmpctl entirely. People using snmpctl are encouraged
to test and migrate to this code as soon as possible.

Much help with the manpage from schwarze@ and jmc@
No objections from reyk@
"Roll it in" deraadt@

Add support for UTF-8 DISPLAY-HINTs with octet length. For now only
sysContact is supported with SnmpAdminString, but more shall follow
soon(tm).

Note that this will change output even in the C locale: Previously every
unprintable byte would be tranformed to a '.', with this diff a valid UTF-8
multibyte or unprintable character will be squished into a single dot to
give a better understanding of the intention of the original message.
Invalid bytes will now be printed as question marks.

Lot's of help and back and forth with schwarze@ who also kindly offered to
walk away when I mentioned looking through MIB files for more objects to
convert to this new code, which is understandable.

OK schwarze@

OIDs are unsigned. Reflect this when printing them.

Shave of some leftovers from snmpd's MIB initialization that are not needed
for snmp. Minus 503 LoC and about 200KB on binary size.

OK denis@

Add better support for exception contexts.
Now we can see proper nosuchinstance and endofmibtree messages.

OK jan@

The ber_* namespace is used by liblber since time immemorial,
so move our BER API to the unused ober_* prefix to avoid some
breakage in ports.

Problem diagnosed by jmatthew with ber_free() in samba, but
there are many others as pointed out by sthen.

tests & ok rob
ok sthen (who had an almost identical diff for libutil)
"go head hit it" deraadt

branches: 1.5.2;
don't use sizeof(pointer) to represent buffer size
ok martijn@ deraadt@

Use snprintf instead of sprintf.

Requested by and OK deraadt@

whitespace issues like a knife in my eyes

ugly whitespaces

Import snmp(1), a new snmp client which aims to be netsnmp compatible for
supported features. It only supports get, getnext, walk, bulkget, bulkwalk,
trap, mibtree, and is SNMPv1 and SNMPv2c for now.

This will shortly replace snmpctl entirely. People using snmpctl are encouraged
to test and migrate to this code as soon as possible.

Much help with the manpage from schwarze@ and jmc@
No objections from reyk@
"Roll it in" deraadt@

OIDs are unsigned. Reflect this when printing them.

Shave of some leftovers from snmpd's MIB initialization that are not needed
for snmp. Minus 503 LoC and about 200KB on binary size.

OK denis@

Add better support for exception contexts.
Now we can see proper nosuchinstance and endofmibtree messages.

OK jan@

The ber_* namespace is used by liblber since time immemorial,
so move our BER API to the unused ober_* prefix to avoid some
breakage in ports.

Problem diagnosed by jmatthew with ber_free() in samba, but
there are many others as pointed out by sthen.

tests & ok rob
ok sthen (who had an almost identical diff for libutil)
"go head hit it" deraadt

branches: 1.5.2;
don't use sizeof(pointer) to represent buffer size
ok martijn@ deraadt@

Use snprintf instead of sprintf.

Requested by and OK deraadt@

whitespace issues like a knife in my eyes

ugly whitespaces

Import snmp(1), a new snmp client which aims to be netsnmp compatible for
supported features. It only supports get, getnext, walk, bulkget, bulkwalk,
trap, mibtree, and is SNMPv1 and SNMPv2c for now.

This will shortly replace snmpctl entirely. People using snmpctl are encouraged
to test and migrate to this code as soon as possible.

Much help with the manpage from schwarze@ and jmc@
No objections from reyk@
"Roll it in" deraadt@

Shave of some leftovers from snmpd's MIB initialization that are not needed
for snmp. Minus 503 LoC and about 200KB on binary size.

OK denis@

Add better support for exception contexts.
Now we can see proper nosuchinstance and endofmibtree messages.

OK jan@

The ber_* namespace is used by liblber since time immemorial,
so move our BER API to the unused ober_* prefix to avoid some
breakage in ports.

Problem diagnosed by jmatthew with ber_free() in samba, but
there are many others as pointed out by sthen.

tests & ok rob
ok sthen (who had an almost identical diff for libutil)
"go head hit it" deraadt

branches: 1.5.2;
don't use sizeof(pointer) to represent buffer size
ok martijn@ deraadt@

Use snprintf instead of sprintf.

Requested by and OK deraadt@

whitespace issues like a knife in my eyes

ugly whitespaces

Import snmp(1), a new snmp client which aims to be netsnmp compatible for
supported features. It only supports get, getnext, walk, bulkget, bulkwalk,
trap, mibtree, and is SNMPv1 and SNMPv2c for now.

This will shortly replace snmpctl entirely. People using snmpctl are encouraged
to test and migrate to this code as soon as possible.

Much help with the manpage from schwarze@ and jmc@
No objections from reyk@
"Roll it in" deraadt@

Add better support for exception contexts.
Now we can see proper nosuchinstance and endofmibtree messages.

OK jan@

The ber_* namespace is used by liblber since time immemorial,
so move our BER API to the unused ober_* prefix to avoid some
breakage in ports.

Problem diagnosed by jmatthew with ber_free() in samba, but
there are many others as pointed out by sthen.

tests & ok rob
ok sthen (who had an almost identical diff for libutil)
"go head hit it" deraadt

branches: 1.5.2;
don't use sizeof(pointer) to represent buffer size
ok martijn@ deraadt@

Use snprintf instead of sprintf.

Requested by and OK deraadt@

whitespace issues like a knife in my eyes

ugly whitespaces

Import snmp(1), a new snmp client which aims to be netsnmp compatible for
supported features. It only supports get, getnext, walk, bulkget, bulkwalk,
trap, mibtree, and is SNMPv1 and SNMPv2c for now.

This will shortly replace snmpctl entirely. People using snmpctl are encouraged
to test and migrate to this code as soon as possible.

Much help with the manpage from schwarze@ and jmc@
No objections from reyk@
"Roll it in" deraadt@

The ber_* namespace is used by liblber since time immemorial,
so move our BER API to the unused ober_* prefix to avoid some
breakage in ports.

Problem diagnosed by jmatthew with ber_free() in samba, but
there are many others as pointed out by sthen.

tests & ok rob
ok sthen (who had an almost identical diff for libutil)
"go head hit it" deraadt

don't use sizeof(pointer) to represent buffer size
ok martijn@ deraadt@

Use snprintf instead of sprintf.

Requested by and OK deraadt@

whitespace issues like a knife in my eyes

ugly whitespaces

Import snmp(1), a new snmp client which aims to be netsnmp compatible for
supported features. It only supports get, getnext, walk, bulkget, bulkwalk,
trap, mibtree, and is SNMPv1 and SNMPv2c for now.

This will shortly replace snmpctl entirely. People using snmpctl are encouraged
to test and migrate to this code as soon as possible.

Much help with the manpage from schwarze@ and jmc@
No objections from reyk@
"Roll it in" deraadt@

don't use sizeof(pointer) to represent buffer size
ok martijn@ deraadt@

Use snprintf instead of sprintf.

Requested by and OK deraadt@

whitespace issues like a knife in my eyes

ugly whitespaces

Import snmp(1), a new snmp client which aims to be netsnmp compatible for
supported features. It only supports get, getnext, walk, bulkget, bulkwalk,
trap, mibtree, and is SNMPv1 and SNMPv2c for now.

This will shortly replace snmpctl entirely. People using snmpctl are encouraged
to test and migrate to this code as soon as possible.

Much help with the manpage from schwarze@ and jmc@
No objections from reyk@
"Roll it in" deraadt@

Use snprintf instead of sprintf.

Requested by and OK deraadt@

whitespace issues like a knife in my eyes

ugly whitespaces

Import snmp(1), a new snmp client which aims to be netsnmp compatible for
supported features. It only supports get, getnext, walk, bulkget, bulkwalk,
trap, mibtree, and is SNMPv1 and SNMPv2c for now.

This will shortly replace snmpctl entirely. People using snmpctl are encouraged
to test and migrate to this code as soon as possible.

Much help with the manpage from schwarze@ and jmc@
No objections from reyk@
"Roll it in" deraadt@

whitespace issues like a knife in my eyes

ugly whitespaces

Import snmp(1), a new snmp client which aims to be netsnmp compatible for
supported features. It only supports get, getnext, walk, bulkget, bulkwalk,
trap, mibtree, and is SNMPv1 and SNMPv2c for now.

This will shortly replace snmpctl entirely. People using snmpctl are encouraged
to test and migrate to this code as soon as possible.

Much help with the manpage from schwarze@ and jmc@
No objections from reyk@
"Roll it in" deraadt@