nsml-parser 0.1.2

NSML++: Neuro-Symbolic Memory Lattice for Intelligent Log Parsing

NSML Parser

Neuro Symbolic Memory Lattice Parser (NSML) is a lightweight, cognitively inspired, fuzzy-intelligent log parser that combines heuristic reasoning, semantic backtracking, and fuzzy key recognition to extract structured information from semi-structured log lines.

---

🚀 Features

• 🔎 Fuzzy key matching: Autocorrects mistyped or aliased keys.
• 🧠 Heuristic inference: Guesses missing values based on patterns.
• 🧬 Semantic classifier-ready: Plug in vector similarity models.
• 📚 Confidence scoring: Score for every parsed key.
• 🔁 REPL-friendly: Can be wrapped into command-line tools.

---

🧪 Example

from nsml_parser import NSMLParser

log = '''Running HTTPS_PROXY='http://proxy.net:8080' /opt/radar/bin/apprise \
  -consumers "FUNNYTEAM_AUDIT" -cluster "POD109" -product "blah-base" -sp "KP" \
  -dc "MM" -caseNumber -release "rrrrr" -event "ERROR_RUNNING_PLSQL_TARGET" \
  -eventTime "2023-06-09_04:31:04" -stage "RUN_PLSQL_TARGET" -status "FAILED" \
  -type "ERROR" -source "ops0-release2-2-RR" "123456" --message "some issue here"'''

parser = NSMLParser()
parsed = parser.parse(log)
print(parsed)