neurolibre 1.5.6 → 1.5.7

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 3a4fdc63012c7af7f469f312d4130205b9596a9a8f5a2367ef3a24ce74be2b51
-  data.tar.gz: 55d4caa8afc3e178c3b028d17ef9a2fb7b0282957026297e6706f3584975bd18
+  metadata.gz: d5b0fdff71e32a79ca2bd3d2ecfe6b8f670dcc0e1c509cc05d4f2c2c99123dc6
+  data.tar.gz: a65f2aedad14e0e4ca0d3152662a3376f53948bcc4fa1a23e6d8135955acdea8
 SHA512:
-  metadata.gz: 36012134ea8415dfd76632e0cb66c43fa646a1e7ed7a57d448d5cb68ebfe50f803c5657fabf25da324842702c83e7657d0e9a9b8be5e94ade4c39dc41cccf9af
-  data.tar.gz: 6e3233ed33201029b9d6efc7b45572c427b55d30c479ce546fc494da07fe9c92563832ff4f872b7ec2b684b5e484b5b906057309f5e5fde431eb147f95520933
+  metadata.gz: 7697da4e5b9007bc625499662ebd58b54f1e6bf1c36b1d25dc66f0753b4b02b1fd652c66021d2ebaa8bb8e434e668cbefb47df9c5b244247f006643ae57b9a41
+  data.tar.gz: 1c006318aaa671cedff181134ca7a100b4c82dd12925172f7182eb322f76b32eef1e00261183c04d26c27b02e589903662e6e4ceb21c956b3a7b149cef983988

data/CHANGELOG.md

@@ -1,5 +1,16 @@
 # Changelog
 
+## 1.5.7 (2026-08-19)
+
+- Paper metadata falls back to myst.yml when a key is absent from the paper's
+  own front matter. A MyST submission declares its title, authors and
+  affiliations in myst.yml for the living preprint, so paper.md need not repeat
+  them; depositing such a paper raised
+  `undefined method 'each' for nil` in `parse_affiliations`.
+- `parse_affiliations` tolerates a paper that names authors and no affiliations
+  instead of raising, and `parse_authors` reports a paper with no authors as a
+  failure rather than a NoMethodError.
+
 ## 1.5.2 (2022-10-29)
 
 - Updated journal data for JuliaCon

data/lib/neurolibre.rb

@@ -2,6 +2,7 @@ require_relative "theoj/version"
 require_relative "theoj/git"
 require_relative "theoj/github"
 require_relative "theoj/orcid"
+require_relative "theoj/myst_frontmatter"
 require_relative "theoj/published_paper"
 require_relative "theoj/submission"
 require_relative "theoj/journal"

data/lib/theoj/myst_frontmatter.rb

@@ -0,0 +1,250 @@
+require "date"
+require "yaml"
+
+module Theoj
+  # Fill gaps in a paper's front matter from the project's myst.yml.
+  #
+  # A NeuroLibre submission declares its title, authors, and affiliations in
+  # myst.yml for the living preprint, so paper.md need not repeat them. The
+  # deposit path still wants them in the shape paper.md front matter uses:
+  # affiliations numbered by index, and each author's affiliations as a
+  # comma-joined string of those indices.
+  #
+  # This is a port of inara's data/filters/myst-frontmatter.lua, which is the
+  # canonical statement of the mapping; full-stack-server's
+  # api/myst_frontmatter.py mirrors the same rules on the Python side. Keep the
+  # three in step -- an author whose affiliations resolve differently depending
+  # on which service is looking is worse than one that fails outright.
+  #
+  # The fallback is best-effort by design: a missing, unreadable, or malformed
+  # myst.yml leaves the metadata exactly as it was. It must never itself be the
+  # reason a deposit fails.
+  module MystFrontmatter
+
+    MYST_FILE = "myst.yml".freeze
+
+    # Parts of a myst.yml affiliation, joined into one name string. Department
+    # precedes institution to match the convention in existing NeuroLibre front
+    # matter.
+    NAME_PARTS = %w[
+      department institution address city region postal_code country
+    ].freeze
+
+    # MyST accepts these aliases for two of the parts.
+    ALIASES = { "institution" => "name", "region" => "state" }.freeze
+
+    # Keys that fill individually, unlike authors and affiliations.
+    SCALAR_KEYS = %w[title date tags bibliography].freeze
+
+    class << self
+
+      # Returns the paper's metadata with any gap filled from myst.yml.
+      #
+      # front_matter - the already-parsed paper.md front matter, or nil for a
+      #                paper that has none.
+      # myst_text    - the raw contents of myst.yml, or nil. Parsed here rather
+      #                than in the caller so a malformed file is tolerated in
+      #                one place.
+      def merge(front_matter, myst_text)
+        metadata = front_matter.is_a?(Hash) ? front_matter.dup : {}
+        return metadata if myst_text.to_s.strip.empty?
+
+        fallback = project_metadata(parse_project(myst_text))
+
+        # Authors and affiliations are filled as a pair. An affiliation index
+        # only means something relative to the list that defines it, so mixing
+        # front matter authors with myst.yml affiliations would silently attach
+        # authors to the wrong institutions.
+        if blank?(metadata["authors"]) || blank?(metadata["affiliations"])
+          unless blank?(fallback["authors"])
+            unless blank?(metadata["authors"])
+              warn "[neurolibre] #{MYST_FILE}: the paper names authors but no " \
+                   "affiliations, so its author list is replaced by the one in " \
+                   "#{MYST_FILE} rather than merged -- an affiliation index only " \
+                   "means something relative to the list that defines it."
+            end
+            metadata["authors"] = fallback["authors"]
+            metadata["affiliations"] = fallback["affiliations"] || []
+          end
+        end
+
+        SCALAR_KEYS.each do |key|
+          metadata[key] = fallback[key] if blank?(metadata[key]) && fallback.key?(key)
+        end
+
+        metadata
+      end
+
+      # Returns paper metadata derived from a myst.yml `project` mapping,
+      # holding only the keys the project actually defines so the caller can
+      # treat it as a set of defaults. Junk input yields an empty hash.
+      def project_metadata(project)
+        return {} unless project.is_a?(Hash)
+
+        metadata = {}
+        metadata["title"] = project["title"] unless blank?(project["title"])
+        unless blank?(project["date"])
+          # `date: 2024-01-15` -- unquoted ISO, the MyST-canonical form --
+          # parses to a Date. This value ends up in a deposit payload, and
+          # nothing downstream reads it structurally, so the string form is the
+          # right shape.
+          date = project["date"]
+          metadata["date"] = date.is_a?(String) ? date : date.to_s
+        end
+        metadata["tags"] = project["keywords"] unless blank?(project["keywords"])
+        metadata["bibliography"] = project["bibliography"] unless blank?(project["bibliography"])
+
+        affiliations, index_of = build_affiliations(project)
+        authors = build_authors(project, affiliations, index_of)
+
+        metadata["authors"] = authors unless authors.empty?
+        metadata["affiliations"] = affiliations unless affiliations.empty?
+        metadata
+      end
+
+      # The contents of the nearest myst.yml at or above the paper, or nil.
+      #
+      # paper_path  - path to the paper (paper.md, paper.tex ...).
+      # search_root - the directory the walk may climb to, normally the root of
+      #               a cloned repository. When nil the walk stops at the first
+      #               directory holding a .git, which is the repository root for
+      #               a plain checkout; failing that, at the paper's own
+      #               directory. myst.yml sits at the project root while the
+      #               paper is often nested (content/paper.md), so the walk has
+      #               to happen -- but it must never wander out of the tree the
+      #               caller meant.
+      def config_text(paper_path, search_root: nil)
+        path = config_path(paper_path, search_root)
+        path.nil? ? nil : File.read(path)
+      end
+
+      private
+
+      def config_path(paper_path, search_root)
+        return nil if paper_path.to_s.strip.empty?
+
+        directory = File.expand_path(File.dirname(paper_path))
+        root = search_root.nil? ? nil : File.expand_path(search_root)
+
+        loop do
+          candidate = File.join(directory, MYST_FILE)
+          return candidate if File.file?(candidate)
+
+          break if root.nil? && File.directory?(File.join(directory, ".git"))
+          break if !root.nil? && (directory == root || !directory.start_with?(root))
+
+          parent = File.dirname(directory)
+          break if parent == directory
+
+          directory = parent
+        end
+
+        nil
+      end
+
+      def parse_project(myst_text)
+        data = YAML.safe_load(myst_text, permitted_classes: [Date, Time], aliases: true)
+        data.is_a?(Hash) ? data["project"] : nil
+      rescue Psych::Exception, ArgumentError => error
+        warn "[neurolibre] could not parse #{MYST_FILE}: #{error.message}"
+        nil
+      end
+
+      # Builds the indexed affiliation list and an id => index map.
+      def build_affiliations(project)
+        affiliations = []
+        index_of = {}
+
+        as_list(project["affiliations"]).each do |source|
+          index = (affiliations.length + 1).to_s
+          if source.is_a?(Hash)
+            affiliations << { "index" => index, "name" => affiliation_name(source) }
+            index_of[source["id"].to_s] = index unless source["id"].nil?
+          else
+            # MyST's validator accepts a bare string where an affiliation
+            # mapping is expected. It becomes an affiliation named after that
+            # string, with no id, and it still consumes its index position --
+            # the Lua filter and the Python port apply the same rule, so all
+            # three agree on every author's index.
+            affiliations << { "index" => index, "name" => source.to_s.strip }
+          end
+        end
+
+        [affiliations, index_of]
+      end
+
+      # Builds the author list, resolving affiliation ids to indices. Appends to
+      # affiliations for any token matching no declared id: MyST permits ad-hoc
+      # affiliations, and inventing an entry beats dropping the author's.
+      def build_authors(project, affiliations, index_of)
+        as_list(project["authors"]).map do |source|
+          # Same MyST rule for authors: `authors: [Ada Lovelace]` is valid. A
+          # bare string becomes a named author with no affiliations.
+          next { "name" => source.to_s.strip } unless source.is_a?(Hash)
+
+          author = { "name" => source["name"] }
+          { "email" => "email",
+            "orcid" => "orcid",
+            "corresponding" => "corresponding",
+            "equal-contrib" => "equal_contributor" }.each do |target, key|
+            author[target] = source[key] unless source[key].nil?
+          end
+
+          indices = affiliation_tokens(source["affiliations"] || source["affiliation"]).map do |token|
+            index = index_of[token]
+            if index.nil?
+              index = (affiliations.length + 1).to_s
+              affiliations << { "index" => index, "name" => token }
+              index_of[token] = index
+            end
+            index
+          end
+          author["affiliation"] = indices.join(",") unless indices.empty?
+
+          author
+        end
+      end
+
+      # Joins an affiliation's parts into a single display string.
+      def affiliation_name(affiliation)
+        NAME_PARTS.map { |key|
+          value = affiliation[key]
+          value = affiliation[ALIASES[key]] if blank?(value) && ALIASES.key?(key)
+          blank?(value) ? nil : value.to_s.strip
+        }.compact.join(", ")
+      end
+
+      # Normalises an author's `affiliations` value to a list of tokens. MyST
+      # accepts a list, a single id, or several ids in one ';'-separated string.
+      def affiliation_tokens(value)
+        return [] if blank?(value)
+        return value.map { |entry| entry.to_s.strip }.reject(&:empty?) if value.is_a?(Array)
+
+        value.to_s.split(";").map(&:strip).reject(&:empty?)
+      end
+
+      # Normalises a myst.yml sequence to an Array. `affiliations: harvard` is
+      # legal MyST; without this, iterating the string would walk its
+      # characters.
+      def as_list(value)
+        return [] if value.nil?
+        return value if value.is_a?(Array)
+
+        [value]
+      end
+
+      # Is a value absent, or present but carrying nothing?
+      #
+      # A front matter of `title:` parses to nil, not to a missing key, and
+      # "", [] and {} say the same thing. All of them must count as absent or a
+      # key that was merely typed out defeats the fallback.
+      def blank?(value)
+        return true if value.nil?
+        return value.strip.empty? if value.is_a?(String)
+        return value.empty? if value.respond_to?(:empty?)
+
+        false
+      end
+    end
+  end
+end

data/lib/theoj/paper.rb

@@ -88,6 +88,10 @@ module Theoj
         if path.to_s.strip.empty?
           setup_local_repo
           @paper_path = Theoj::Paper.find_paper_path(local_path)
+          # myst.yml lives at the project root while the paper is often nested
+          # (content/paper.md), so the lookup may walk up -- but no further
+          # than the tree we cloned.
+          @metadata_search_root = local_path
         else
           @paper_path = path
         end
@@ -107,7 +111,19 @@ module Theoj
       def load_metadata
         @paper_metadata ||= if paper_path.nil?
           {}
-        elsif paper_path.include?('.tex')
+        else
+          # A NeuroLibre submission may declare its title, authors and
+          # affiliations in myst.yml instead of repeating them in the paper's
+          # own front matter, so anything absent here is filled from there.
+          Theoj::MystFrontmatter.merge(
+            front_matter_metadata,
+            Theoj::MystFrontmatter.config_text(paper_path, search_root: @metadata_search_root)
+          )
+        end
+      end
+
+      def front_matter_metadata
+        if paper_path.include?('.tex')
           YAML.load_file(paper_path.gsub('.tex', '.yml'))
         else
           YAML.load_file(paper_path)
@@ -119,6 +135,8 @@ module Theoj
         authors_metadata = @paper_metadata['authors']
         affiliations_metadata = parse_affiliations(@paper_metadata['affiliations'])
 
+        failure "Cannot find the authors of this paper" if authors_metadata.nil?
+
         # Loop through the authors block and build up the affiliation
         authors_metadata.each do |author|
           author['name'] = author.dup if author['name'].nil?
@@ -138,7 +156,12 @@ module Theoj
       def parse_affiliations(affiliations_yaml)
         affiliations_metadata = {}
 
-        affiliations_yaml.each do |affiliation|
+        # A paper may legitimately name authors and no affiliations: myst.yml
+        # permits it and so does a hand-written paper.md. Iterating nil here
+        # raised NoMethodError and took the whole deposit down with it.
+        Array(affiliations_yaml).each do |affiliation|
+          next unless affiliation.is_a?(Hash)
+
           affiliations_metadata[affiliation['index']] = affiliation['name']
         end
 

data/lib/theoj/version.rb

@@ -1,3 +1,3 @@
 module Theoj
-  VERSION = "1.5.6"
+  VERSION = "1.5.7"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: neurolibre
 version: !ruby/object:Gem::Version
-  version: 1.5.6
+  version: 1.5.7
 platform: ruby
 authors:
 - Juanjo Bazán
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2023-06-19 00:00:00.000000000 Z
+date: 2026-08-20 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: octokit
@@ -152,6 +152,7 @@ files:
 - lib/theoj/github.rb
 - lib/theoj/journal.rb
 - lib/theoj/journals_data.rb
+- lib/theoj/myst_frontmatter.rb
 - lib/theoj/orcid.rb
 - lib/theoj/paper.rb
 - lib/theoj/published_paper.rb
@@ -185,7 +186,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.4.6
+rubygems_version: 3.5.11
 signing_key:
 specification_version: 4
 summary: Editorial objects used by NeuroLibre