Generate a Parser Package

Create a Tenzir package that parses raw log data into structured events.

Prerequisites: Read these pages before starting:

• /tutorials/write-a-package - Package structure, operators, testing patterns
• /explanations/packages - Package concepts
• /reference/test-framework - Test framework reference

Execute the phases below in order. Do not skip phases.

Phase 1: Input Schema Analysis

Objective: Learn about the input data and understand its structure.

Steps:

1. Ask the user to provide sample log data (file path or pasted content)
2. Identify the data source format (CSV, JSON, YAML, syslog, etc.)
3. Identify vendor and product that may have generated this data
4. Document the complete input schema in terms of fields and types

Completion: State “Phase 1 complete” before proceeding.

Phase 2: Package Scaffolding

Objective: Create the package structure for iterative development.

Steps:

1. Confirm the package ID with the user (typically vendor name, e.g., fortinet, cisco, microsoft). In the instructions below, replace <pkg> with the chosen package ID.

2. Create the package directory structure as described in the Write a Package tutorial. Include operators/parse.tql, tests/parse.tql, and tests/inputs/sample.txt.

3. Create package.yaml with the package metadata.

4. Create the initial parse operator in operators/parse.tql with just read_lines as a starting point.

5. Create a test file tests/parse.tql that reads from the input:

   from_file f"{env("TENZIR_INPUTS")}/sample.txt" {
     <pkg>::parse
   }

6. Save the sample log data to tests/inputs/sample.txt.

7. Create the initial baseline: uvx tenzir-test --root <pkg> -u --summary

Completion: State “Phase 2 complete” before proceeding.

Phase 3: Iterate and Test

Objective: Refine the parser until all fields are parsed and properly typed.

Prerequisites: Read these guides for transformation patterns:

• /guides/data-shaping/transform-basic-values - Type conversion, null handling, sentinel values
• /guides/data-shaping/extract-structured-data-from-text - Nested structures, delimited data
• /guides/data-shaping/manipulate-strings - String cleanup, splitting, extraction

Loop until all fields are properly parsed:

1. Make ONE modification to the parse operator. Work through these categories:
   • Type conversion: Parse timestamps, IPs, subnets (see transform-basic-values)
   • Structure extraction: Parse nested JSON, CSV, key-value pairs (see extract-structured-data-from-text)
   • Data cleaning: Normalize sentinel values to null, trim whitespace, extract substrings (see transform-basic-values and manipulate-strings)
2. Observe the impact of your change by re-running:
   • uvx tenzir-test --root <pkg> --summary
3. If the diff looks good, update the baseline: uvx tenzir-test --root <pkg> -u --summary
4. Go back to Step 1 and continue with the next modification

Completion: State “Phase 3 complete” before proceeding.

Phase 4: Summarize

Provide a final summary of the parser’s functionality:

• Input: Description of the input format and source
• TQL: The complete parsing logic
• Output: Description of the parsed schema with types
• Package structure: Tree view of the package
• Noteworthy findings: Any interesting discoveries or caveats