Build a JSON index over compiled MIBs

Compile a few ASN.1 MIBs into JSON documents, then build an index over them.

The index answers the question the documents cannot: given an OID, which module defines it? It is what mibdump –build-index writes, and what the published index-v2.csv is.

The index is incremental. Each run merges its modules into whatever index is already in the destination directory, so a collection can be built up over many runs rather than in one pass. Run this script twice, with a different module in inputMibs the second time, and the index will name both.

import json
import os

from pysmi.codegen import JsonCodeGen
from pysmi.compiler import MibCompiler
from pysmi.parser import SmiStarParser
from pysmi.reader import FileReader, HttpReader
from pysmi.searcher import StubSearcher
from pysmi.writer import FileWriter

# from pysmi import debug

# debug.enableDebugLogging('compiler')

inputMibs = ["IF-MIB", "IP-MIB"]
srcDirectories = ["/usr/share/snmp/mibs"]
httpSources = ["https://pysnmp.github.io/mibs/asn1/@mib@"]
dstDirectory = os.path.join(os.path.expanduser("~"), ".pysnmp", "mibs", "json")

# Initialize compiler infrastructure

mibCompiler = MibCompiler(
    SmiStarParser(), JsonCodeGen(), FileWriter(dstDirectory).set_options(suffix=".json")
)

# search for source MIBs here
mibCompiler.add_sources(*[FileReader(x) for x in srcDirectories])

# search for source MIBs at Web sites
mibCompiler.add_sources(*[HttpReader(x) for x in httpSources])

# never recompile MIBs with MACROs
mibCompiler.add_searchers(StubSearcher(*JsonCodeGen.baseMibs))

# run recursive MIB compilation
results = mibCompiler.compile(*inputMibs)

print("# Results: " + ", ".join([f"{x}:{results[x]}" for x in results]))

# Build the index over everything just compiled.
#
# Any index already in dstDirectory is read back and merged into, which is
# why this may be called once per run rather than once per collection. Pass
# ignoreErrors=True to keep a failure here from losing the compilation
# above; pass dryRun=True to build the index without storing it.
mibCompiler.build_index(results, ignoreErrors=True)

# Read back what was written. mibCompiler.indexFile is the name used, and
# the writer's suffix option gives it its extension.
indexPath = os.path.join(dstDirectory, mibCompiler.indexFile + ".json")

with open(indexPath) as fileObj:
    index = json.load(fileObj)

print(f"\n# Index at {indexPath}, schema version {index['meta']['schema']}")

# The sections answer different questions, each mapping an OID to the
# modules that define it:
#
#   identity      MODULE-IDENTITY -- what a module calls itself
#   enterprise    the module's enterprise branch, where it has one
#   compliance    MODULE-COMPLIANCE -- what an implementation is held to
#   notification  NOTIFICATION-TYPE and TRAP-TYPE -- what a module may emit
#   oids          the top-level branches a module defines, shortest first
for section in ("identity", "enterprise", "compliance", "notification", "oids"):
    print(f"\n## {section}")

    for oid, modules in sorted(index[section].items())[:5]:
        print(f"   {oid} -> {', '.join(modules)}")

Download script.