swhid 0.2.1 → 0.3.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 78a2795583924a21b4aae1a173e528af68805dd89ad44e6c3db32b534ca55653
-  data.tar.gz: 1a65c99bade2ae1c4f00c69d2b49a31708da9fc8f15ee5c63d7d9806c876b7e5
+  metadata.gz: ab5130c9128fd1942d8a4a01eecb6d5f403f38fc2ce53d72323b08350e30bdf4
+  data.tar.gz: a7c5c116fcd54c08f7d51b4125d4871f25b98aacc4f7ab818ebae670fc0972c9
 SHA512:
-  metadata.gz: e9d04af133af2dca2b749ce7259a792dbed4622b7bfc587cc08be99d87e109d50be5926e0db2b653b3da7d0e74454fac3e4a9709b12fcd9306f4be20583eab97
-  data.tar.gz: 24affb1c2fa578ab1211b54540a3d2a51497e583b70804cf837c8999a2c8beb3c3546af6f0de35fe2743e26918fc775baf9f6ec0e75fad0d2d116b10730a412b
+  metadata.gz: ad64d5ffad60fde8ec7399abcd63af7c34ae203c54f469df9866283864f1850c0f9b1c112ab89300c7ea52f56d5809189bbe6ec0bd2c9fc2ed85510daa3b2d37
+  data.tar.gz: a44ef79ecd7a5d64f4d29c56f470b749ab61eac0d3f3ddd29c8b4798f334fa3f43222b8b463819f221e5687ca1355f7c23b7754be43a739009aafa4adba3e3cd

data/CHANGELOG.md

@@ -1,5 +1,20 @@
 ## [Unreleased]
 
+## [0.3.1] - 2025-11-23
+
+### Fixed
+- Snapshot implementation now includes HEAD symbolic reference
+- Extra headers extraction for signed commits (gpgsig, mergetag, etc.)
+- Tag-of-tag support (tags pointing to other tag objects)
+- Extra headers extraction for signed tags
+
+## [0.3.0] - 2025-11-23
+
+- `directory` CLI command - Generate SWHID for directory from filesystem
+- `revision` CLI command - Generate SWHID for git commit/revision
+- `release` CLI command - Generate SWHID for git tag/release
+- `snapshot` CLI command - Generate SWHID for git repository snapshot
+
 ## [0.2.1] - 2025-11-09
 
 - Package manager tests for PyPI, RubyGems, Maven, Cargo, and NPM artifacts (content and extracted directories)

data/README.md

@@ -151,12 +151,13 @@ swhid = Swhid::Identifier.new(
     visit: "swh:1:snp:d7f1b9eb7ccb596c2622c4780febaa02549830f9",
     anchor: "swh:1:rev:2db189928c94d62a3b4757b3eec68f0a4d4113f0",
     path: "/src/main.rb",
-    lines: "10-20"
+    lines: "10-20",
+    bytes: "0-100"
   }
 )
 
 puts swhid.to_s
-# => "swh:1:cnt:94a9ed024d3859793618152ea559a168bbcbb5e2;origin=https://github.com/example/repo;visit=swh:1:snp:...;anchor=swh:1:rev:...;path=/src/main.rb;lines=10-20"
+# => "swh:1:cnt:94a9ed024d3859793618152ea559a168bbcbb5e2;origin=https://github.com/example/repo;visit=swh:1:snp:...;anchor=swh:1:rev:...;path=/src/main.rb;lines=10-20;bytes=0-100"
 ```
 
 ### CLI Usage
@@ -183,6 +184,40 @@ $ echo "Hello, World!" | swhid content
 swh:1:cnt:96898574d1b88e619be24fd90bb4cd399acbc5ca
 ```
 
+**Generate SWHID from directory**
+
+```bash
+$ swhid directory /path/to/directory
+swh:1:dir:4b825dc642cb6eb9a060e54bf8d69288fbee4904
+```
+
+**Generate SWHID from git commit**
+
+```bash
+$ swhid revision /path/to/repo
+swh:1:rev:bc0195aad0daa2ad5b0d76cce22b167bc3435590
+
+$ swhid revision /path/to/repo main
+swh:1:rev:bc0195aad0daa2ad5b0d76cce22b167bc3435590
+
+$ swhid revision /path/to/repo abc123
+swh:1:rev:bc0195aad0daa2ad5b0d76cce22b167bc3435590
+```
+
+**Generate SWHID from git tag**
+
+```bash
+$ swhid release /path/to/repo v1.0.0
+swh:1:rel:2b10839e32c4c476e9d94492756bb1a3e1ec4aa8
+```
+
+**Generate SWHID from git snapshot**
+
+```bash
+$ swhid snapshot /path/to/repo
+swh:1:snp:6e65b86363953b780d92b0a928f3e8fcdd10db36
+```
+
 **Add qualifiers**
 
 ```bash

data/exe/swhid

@@ -24,6 +24,14 @@ class SwhidCLI
       parse_swhid
     when "content"
       compute_content_swhid
+    when "directory"
+      compute_directory_swhid
+    when "revision"
+      compute_revision_swhid
+    when "release"
+      compute_release_swhid
+    when "snapshot"
+      compute_snapshot_swhid
     when "help"
       show_help
     else
@@ -41,9 +49,13 @@ class SwhidCLI
       opts.banner = "Usage: swhid [command] [options]"
       opts.separator ""
       opts.separator "Commands:"
-      opts.separator "  parse <swhid>     Parse and validate a SWHID"
-      opts.separator "  content           Generate SWHID for content from stdin"
-      opts.separator "  help              Show this help message"
+      opts.separator "  parse <swhid>          Parse and validate a SWHID"
+      opts.separator "  content                Generate SWHID for content from stdin"
+      opts.separator "  directory <path>       Generate SWHID for directory"
+      opts.separator "  revision <repo> [ref]  Generate SWHID for git revision/commit"
+      opts.separator "  release <repo> <tag>   Generate SWHID for git release/tag"
+      opts.separator "  snapshot <repo>        Generate SWHID for git snapshot"
+      opts.separator "  help                   Show this help message"
       opts.separator ""
       opts.separator "Options:"
 
@@ -104,6 +116,124 @@ class SwhidCLI
     end
   end
 
+  def compute_directory_swhid
+    path = @args.shift
+    unless path
+      puts "Error: Directory path required"
+      exit 1
+    end
+
+    unless File.exist?(path)
+      puts "Error: Path does not exist: #{path}"
+      exit 1
+    end
+
+    unless File.directory?(path)
+      puts "Error: Path is not a directory: #{path}"
+      exit 1
+    end
+
+    swhid = Swhid::FromFilesystem.from_directory_path(path)
+
+    unless @options[:qualifiers].empty?
+      swhid = Swhid::Identifier.new(
+        object_type: swhid.object_type,
+        object_hash: swhid.object_hash,
+        qualifiers: @options[:qualifiers]
+      )
+    end
+
+    case @options[:format]
+    when "json"
+      output_json(swhid)
+    else
+      puts swhid.to_s
+    end
+  end
+
+  def compute_revision_swhid
+    repo_path = @args.shift
+    unless repo_path
+      puts "Error: Repository path required"
+      exit 1
+    end
+
+    ref = @args.shift || "HEAD"
+
+    swhid = Swhid::FromGit.from_revision(repo_path, ref)
+
+    unless @options[:qualifiers].empty?
+      swhid = Swhid::Identifier.new(
+        object_type: swhid.object_type,
+        object_hash: swhid.object_hash,
+        qualifiers: @options[:qualifiers]
+      )
+    end
+
+    case @options[:format]
+    when "json"
+      output_json(swhid)
+    else
+      puts swhid.to_s
+    end
+  end
+
+  def compute_release_swhid
+    repo_path = @args.shift
+    unless repo_path
+      puts "Error: Repository path required"
+      exit 1
+    end
+
+    tag_name = @args.shift
+    unless tag_name
+      puts "Error: Tag name required"
+      exit 1
+    end
+
+    swhid = Swhid::FromGit.from_release(repo_path, tag_name)
+
+    unless @options[:qualifiers].empty?
+      swhid = Swhid::Identifier.new(
+        object_type: swhid.object_type,
+        object_hash: swhid.object_hash,
+        qualifiers: @options[:qualifiers]
+      )
+    end
+
+    case @options[:format]
+    when "json"
+      output_json(swhid)
+    else
+      puts swhid.to_s
+    end
+  end
+
+  def compute_snapshot_swhid
+    repo_path = @args.shift
+    unless repo_path
+      puts "Error: Repository path required"
+      exit 1
+    end
+
+    swhid = Swhid::FromGit.from_snapshot(repo_path)
+
+    unless @options[:qualifiers].empty?
+      swhid = Swhid::Identifier.new(
+        object_type: swhid.object_type,
+        object_hash: swhid.object_hash,
+        qualifiers: @options[:qualifiers]
+      )
+    end
+
+    case @options[:format]
+    when "json"
+      output_json(swhid)
+    else
+      puts swhid.to_s
+    end
+  end
+
   def output_text(swhid)
     puts "SWHID: #{swhid.to_s}"
     puts "Core:  #{swhid.core_swhid}"
@@ -134,8 +264,12 @@ class SwhidCLI
       swhid - Generate and parse SoftWare Hash IDentifiers
 
       Usage:
-        swhid parse <swhid>              Parse and validate a SWHID
-        swhid content [options]          Generate SWHID for content from stdin
+        swhid parse <swhid>                   Parse and validate a SWHID
+        swhid content [options]               Generate SWHID for content from stdin
+        swhid directory <path> [options]      Generate SWHID for directory
+        swhid revision <repo> [ref] [options] Generate SWHID for git revision/commit
+        swhid release <repo> <tag> [options]  Generate SWHID for git release/tag
+        swhid snapshot <repo> [options]       Generate SWHID for git snapshot
 
       Options:
         -f, --format FORMAT              Output format (text, json)
@@ -149,6 +283,20 @@ class SwhidCLI
         # Generate SWHID from file content
         cat file.txt | swhid content
 
+        # Generate SWHID from directory
+        swhid directory /path/to/dir
+
+        # Generate SWHID from git commit
+        swhid revision /path/to/repo
+        swhid revision /path/to/repo main
+        swhid revision /path/to/repo abc123
+
+        # Generate SWHID from git tag
+        swhid release /path/to/repo v1.0.0
+
+        # Generate SWHID from git snapshot
+        swhid snapshot /path/to/repo
+
         # Generate SWHID with qualifiers
         cat file.txt | swhid content -q origin=https://github.com/example/repo
 

data/lib/swhid/from_git.rb

@@ -0,0 +1,219 @@
+# frozen_string_literal: true
+
+require "rugged"
+
+module Swhid
+  module FromGit
+    def self.from_revision(repo_path, ref = "HEAD")
+      repo = Rugged::Repository.new(repo_path)
+      commit = repo.rev_parse(ref)
+
+      raise ArgumentError, "Reference #{ref} is not a commit" unless commit.is_a?(Rugged::Commit)
+
+      metadata = {
+        directory: commit.tree.oid,
+        parents: commit.parents.map(&:oid),
+        author: format_person(commit.author),
+        author_timestamp: commit.author[:time].to_i,
+        author_timezone: format_timezone(commit.author[:time]),
+        committer: format_person(commit.committer),
+        committer_timestamp: commit.committer[:time].to_i,
+        committer_timezone: format_timezone(commit.committer[:time]),
+        message: commit.message
+      }
+
+      # Extract extra headers if present (like gpgsig, svn headers, etc)
+      extra_headers = extract_extra_headers(repo, commit)
+      metadata[:extra_headers] = extra_headers unless extra_headers.empty?
+
+      Swhid.from_revision(metadata)
+    end
+
+    def self.from_release(repo_path, tag_name)
+      repo = Rugged::Repository.new(repo_path)
+      tag_ref = repo.references["refs/tags/#{tag_name}"]
+
+      raise ArgumentError, "Tag #{tag_name} not found" unless tag_ref
+
+      # Get the tag object
+      tag_obj = repo.lookup(tag_ref.target_id)
+
+      # Check if it's an annotated tag
+      if tag_obj.is_a?(Rugged::Tag::Annotation)
+        target_type = case tag_obj.target
+                      when Rugged::Commit then "rev"
+                      when Rugged::Tag::Annotation then "rel"
+                      when Rugged::Tree then "dir"
+                      when Rugged::Blob then "cnt"
+                      else "rev"
+                      end
+
+        metadata = {
+          name: tag_obj.name,
+          target: { hash: tag_obj.target.oid, type: target_type },
+          message: tag_obj.message
+        }
+
+        if tag_obj.tagger
+          metadata[:author] = format_person(tag_obj.tagger)
+          metadata[:author_timestamp] = tag_obj.tagger[:time].to_i
+          metadata[:author_timezone] = format_timezone(tag_obj.tagger[:time])
+        end
+
+        # Extract extra headers if present (like gpgsig for signed tags)
+        extra_headers = extract_tag_extra_headers(repo, tag_obj)
+        metadata[:extra_headers] = extra_headers unless extra_headers.empty?
+
+        Swhid.from_release(metadata)
+      else
+        # Lightweight tag - points directly to commit
+        raise ArgumentError, "Lightweight tags are not supported for release SWHIDs"
+      end
+    end
+
+    def self.from_snapshot(repo_path)
+      repo = Rugged::Repository.new(repo_path)
+      branches = []
+
+      # Check for HEAD first
+      head_path = File.join(repo.path, "HEAD")
+      if File.exist?(head_path)
+        head_content = File.read(head_path).strip
+        if head_content.start_with?("ref:")
+          # HEAD is a symbolic ref
+          target_ref = head_content.sub("ref: ", "")
+          branches << {
+            name: "HEAD",
+            target_type: "alias",
+            target: target_ref
+          }
+        end
+      end
+
+      # Get all references (branches and tags)
+      repo.references.each do |ref|
+        ref_name = ref.name
+
+        if ref.type == :symbolic
+          # This is an alias (symbolic ref)
+          target_ref_name = ref.target
+          branches << {
+            name: ref_name,
+            target_type: "alias",
+            target: target_ref_name
+          }
+        else
+          # Direct reference
+          target_obj = ref.target
+
+          # Determine target type and OID
+          target_type, target_oid = case target_obj
+                                     when Rugged::Commit
+                                       ["revision", target_obj.oid]
+                                     when Rugged::Tag::Annotation
+                                       ["release", target_obj.oid]
+                                     when Rugged::Tree
+                                       ["directory", target_obj.oid]
+                                     when Rugged::Blob
+                                       ["content", target_obj.oid]
+                                     else
+                                       ["revision", target_obj.oid]
+                                     end
+
+          branches << {
+            name: ref_name,
+            target_type: target_type,
+            target: target_oid
+          }
+        end
+      end
+
+      Swhid.from_snapshot(branches)
+    end
+
+    private
+
+    def self.format_person(person)
+      "#{person[:name]} <#{person[:email]}>"
+    end
+
+    def self.format_timezone(time)
+      offset = time.utc_offset
+      sign = offset >= 0 ? "+" : "-"
+      hours = offset.abs / 3600
+      minutes = (offset.abs % 3600) / 60
+      format("%s%02d%02d", sign, hours, minutes)
+    end
+
+    def self.extract_extra_headers(repo, commit)
+      # Rugged doesn't expose extra headers directly
+      # We need to parse the raw commit object
+      raw_data = repo.read(commit.oid).data
+      lines = raw_data.split("\n")
+
+      extra_headers = []
+      in_headers = true
+
+      lines.each do |line|
+        # Stop when we hit the blank line before the message
+        if line.empty?
+          in_headers = false
+          next
+        end
+
+        next unless in_headers
+
+        # Skip standard headers
+        next if line.start_with?("tree ", "parent ", "author ", "committer ")
+
+        # Extract extra headers (like gpgsig, mergetag, svn-repo-uuid, etc)
+        if line.start_with?(" ")
+          # Continuation of previous header
+          if extra_headers.any?
+            extra_headers.last[1] += "\n#{line[1..]}"
+          end
+        elsif line.include?(" ")
+          key, value = line.split(" ", 2)
+          extra_headers << [key, value]
+        end
+      end
+
+      extra_headers
+    end
+
+    def self.extract_tag_extra_headers(repo, tag)
+      # Parse raw tag object for extra headers
+      raw_data = repo.read(tag.oid).data
+      lines = raw_data.split("\n")
+
+      extra_headers = []
+      in_headers = true
+
+      lines.each do |line|
+        # Stop when we hit the blank line before the message
+        if line.empty?
+          in_headers = false
+          next
+        end
+
+        next unless in_headers
+
+        # Skip standard tag headers
+        next if line.start_with?("object ", "type ", "tag ", "tagger ")
+
+        # Extract extra headers (like gpgsig for signed tags)
+        if line.start_with?(" ")
+          # Continuation of previous header
+          if extra_headers.any?
+            extra_headers.last[1] += "\n#{line[1..]}"
+          end
+        elsif line.include?(" ")
+          key, value = line.split(" ", 2)
+          extra_headers << [key, value]
+        end
+      end
+
+      extra_headers
+    end
+  end
+end

data/lib/swhid/objects/directory.rb

@@ -69,7 +69,10 @@ module Swhid
         sorted_entries = entries.sort_by(&:sort_key)
 
         sorted_entries.map do |entry|
-          "#{entry.perms} #{entry.name}\0#{entry.target_hash}"
+          # Convert name to binary UTF-8 to match target_hash encoding
+          name_binary = entry.name.encode(Encoding::UTF_8).force_encoding(Encoding::BINARY)
+          perms_binary = entry.perms.encode(Encoding::UTF_8).force_encoding(Encoding::BINARY)
+          "#{perms_binary} #{name_binary}\0#{entry.target_hash}"
         end.join
       end
 

data/lib/swhid/objects/release.rb

@@ -39,6 +39,11 @@ module Swhid
           lines << author_line
         end
 
+        extra_headers = metadata[:extra_headers] || []
+        extra_headers.each do |key, value|
+          lines << format_header_line(key, value)
+        end
+
         result = lines.join("\n") + "\n"
 
         if metadata[:message]
@@ -54,6 +59,11 @@ module Swhid
         "#{prefix} #{person_escaped} #{timestamp} #{tz}"
       end
 
+      def self.format_header_line(key, value)
+        value_escaped = value.gsub("\n", "\n ")
+        "#{key} #{value_escaped}"
+      end
+
       def self.extract_target(target)
         case target
         when Hash

data/lib/swhid/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Swhid
-  VERSION = "0.2.1"
+  VERSION = "0.3.1"
 end

data/lib/swhid.rb

@@ -8,6 +8,7 @@ require_relative "swhid/objects/revision"
 require_relative "swhid/objects/release"
 require_relative "swhid/objects/snapshot"
 require_relative "swhid/from_filesystem"
+require_relative "swhid/from_git"
 
 module Swhid
   class Error < StandardError; end

metadata

@@ -1,14 +1,28 @@
 --- !ruby/object:Gem::Specification
 name: swhid
 version: !ruby/object:Gem::Version
-  version: 0.2.1
+  version: 0.3.1
 platform: ruby
 authors:
 - Andrew Nesbitt
 bindir: exe
 cert_chain: []
 date: 1980-01-02 00:00:00.000000000 Z
-dependencies: []
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: rugged
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.9'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.9'
 description: A Ruby library and CLI for generating and parsing SoftWare Hash IDentifiers
   (SWHIDs). Supports all object types (content, directory, revision, release, snapshot)
   and qualifiers. Compatible with Git object hashing.
@@ -27,6 +41,7 @@ files:
 - exe/swhid
 - lib/swhid.rb
 - lib/swhid/from_filesystem.rb
+- lib/swhid/from_git.rb
 - lib/swhid/identifier.rb
 - lib/swhid/objects/content.rb
 - lib/swhid/objects/directory.rb