InstallAnywhere Specification Collection Process

InstallAnywhere Specification Collection Process

STATUS: DRAFT v0.1 -- This process file is iterative. Each research session should refine the process itself, update findings, and record what worked and what didn't. See Self-Improvement Log at the bottom.

Created: 2026-02-19 Last Updated: 2026-02-19 Goal: Collect all specification sheets, file format documentation, manifest structures, and configuration file references for Revenera InstallAnywhere -- sufficient to interpret a Windows InstallAnywhere manifest and install the target application on Linux.

1. Objective

Given a Windows-targeted InstallAnywhere installer (.exe), produce a complete understanding of its internal structure such that we can:

1. Extract the application payload
2. Reconstruct the installation sequence on Linux
3. Configure LaunchAnywhere (.lax) files for a Linux JVM
4. Handle silent/unattended installation via response files
5. Resolve platform-specific rules and conditions

2. Primary Documentation Sources

2.1 Official Revenera Documentation (Verified 2026-02-19)

2.2 Key Documentation Pages

All under https://docs.revenera.com/installanywhere/Content/helplibrary/:

File Formats & Manifests:

LaunchAnywhere & JVM:

Silent/Unattended Installation:

Command Line:

Platform & Cross-Platform:

XML & Actions:

3. File Format Quick Reference (Known So Far)

4. Architecture Understanding (Verified 2026-02-19)

4.1 Built Installer Structure

A built InstallAnywhere installer is a self-extracting archive containing:

<installer>.exe / <installer>.bin
├── Native launcher stub (platform-specific)
├── Bundled JRE (optional, platform-specific)
├── install.xml (runtime descriptor)
├── Compressed payload (application files)
└── LAX configuration templates

4.2 Runtime Execution Sequence

1. Native launcher self-extracts to temp dir (configurable via -tempdir)
2. Locates JVM: bundled > LAX_VM param > system-installed (order configurable)
3. Launches Java-based installer engine
4. Java engine reads install.xml
5. Processes actions/panels/rules in sequence
6. Creates LaunchAnywhere executables + .lax files
7. Optionally generates response file (installer.properties)

4.3 LaunchAnywhere Runtime (Post-Install)

1. User runs LaunchAnywhere executable
2. Reads companion .lax file for configuration
3. Locates JVM per .lax settings
4. Configures classpath, system properties, heap, env vars
5. Launches the Java application

5. Cross-Platform Strategy: Windows -> Linux

5.1 Known Approaches

5.2 Information Still Needed

• Exact binary structure of the self-extracting archive (ZIP header? custom format?)
• install.xml schema / DTD -- what actions are defined, what is the XML structure?
• Full .lax property reference with all valid keys and value formats
• Platform rule evaluation logic -- how does the installer decide what to skip on Linux?
• Registry action equivalents -- what maps to Linux (config files? symlinks?)
• Service creation actions -- Windows services vs Linux systemd/init.d
• File permission handling differences (NTFS ACLs vs POSIX permissions)
• Response file encoding requirements per platform (UTF-8 no BOM for Linux, verified)
• How $INSTALL_DIR$ and other path variables resolve on Linux vs Windows

6. Collection Process (Iterative)

Phase 1: Harvest Official Docs (Current)

Method: Fetch each page listed in Section 2.2 and extract structured content.

For each page in Section 2.2:
  1. Fetch page content from docs.revenera.com
  2. Extract spec-relevant information (schemas, property names, value formats)
  3. Record in corresponding reference file under research/installer-tools/references/
  4. Update Collection Status in this file to "Collected" with date
  5. Note any linked pages not yet in our list -> add to Section 2.2

Output: One reference file per topic area:

• references/file-formats.md -- All file format specifications
• references/lax-properties.md -- Complete LAX property reference
• references/response-files.md -- Response file format and encoding
• references/manifest-structure.md -- Manifest file structure
• references/cli-arguments.md -- All command-line arguments
• references/platform-rules.md -- Platform-specific rules and conditions
• references/jvm-handling.md -- JRE bundling, VM selection, LAX_VM
• references/install-xml-schema.md -- install.xml structure and actions

Phase 2: Community Knowledge

Method: Search for community experiences with cross-platform InstallAnywhere usage.

1. Search GitHub for "installanywhere extract" / "installanywhere linux" / ".lax file"
2. Search Stack Overflow [installanywhere] tag for cross-platform Q&A
3. Search Revenera Community forums for Linux installation topics
4. Document any open-source tools that parse IA formats
5. Record practical tips and gotchas

Output: references/community-knowledge.md

Phase 3: Hands-On Exploration

Method: Given an actual Windows InstallAnywhere installer, reverse-engineer its structure.

1. Attempt extraction with 7z, unzip, binwalk
2. Document the actual archive structure found
3. Locate and parse install.xml
4. Locate and document .lax templates
5. Map actions to Linux equivalents
6. Test silent install with response file under Wine
7. Test native Linux execution with extracted payload + Linux JVM

Output: references/extraction-guide.md and references/action-mapping.md

Phase 4: Synthesis

Method: Combine all collected specs into an actionable installation procedure.

1. Write step-by-step Linux installation procedure from Windows IA manifest
2. Document all assumptions and platform-specific translations
3. Create decision tree for handling each IA action type on Linux
4. Validate procedure against actual installer

Output: references/linux-installation-procedure.md

7. CLI Arguments Reference (Collected 2026-02-19)

Installer Arguments

Launcher Arguments

All other arguments pass through to the launched application.

8. Key LAX Properties (Known So Far)

TODO: Full property list requires fetching ia_ref_laxprop.htm.

9. Self-Improvement Log

Track what changed with each iteration so the process gets better over time.

Process Improvements Backlog

• Determine if Sitemap.xml can be parsed to auto-discover additional pages
• Find alternative search approach for community content (GitHub API? gh search?)
• Evaluate whether the PDF user guide can be parsed section-by-section for faster collection than fetching individual HTML pages
• Add validation step: after collecting each spec, verify it against the PDF to catch HTML-only or PDF-only content
• Consider creating a structured schema (JSON/YAML) for the file format specs to enable programmatic use
• Add a "confidence level" column to the File Format Quick Reference table
• Determine if older version docs contain information removed from current version

10. Session Resumption Checklist

When resuming work on this process:

1. Read this file first to understand current state
2. Check Collection Status columns in Section 2.2 for next pending items
3. Check the Self-Improvement Log for known issues
4. Check the Process Improvements Backlog for meta-improvements
5. After each collection session, update:
   • Collection Status for pages fetched
   • File Format Quick Reference for new formats discovered
   • Self-Improvement Log with what worked/didn't
   • Process Improvements Backlog with new ideas
   • Section 5.2 checklist for information gaps closed

References

• Revenera InstallAnywhere Documentation (accessed 2026-02-19)
• InstallAnywhere 2025 R2 User Guide PDF (accessed 2026-02-19)
• Revenera Community Forums (not yet accessed)

Freshness Tracking