bio 1.5.2 → 1.6.0.pre.20181210

data/{gemfiles/Gemfile.travis-ruby2.2 → Gemfile}

@@ -5,5 +5,4 @@ gem "rdoc"
 gem "test-unit"
 
 gem "xmlparser"
-gem "libxml-ruby"
 

data/KNOWN_ISSUES.rdoc

@@ -162,11 +162,6 @@ do not recommend to do so because the format is unstable.
 
 == 4. Compatibility issues with other libraries/extensions
 
-=== ActiveRecord
-
-BioSQL support uses ActiveRecord, but the author of the document does not
-know which version is suitable.
-
 === Ruby on Rails
 
 BioRuby Shell on Web uses Ruby on Rails, but the author of the document does

data/README.rdoc

@@ -30,7 +30,7 @@ Blast, Fasta, Hmmer and many other software packages for biological
 analysis can be executed within the BioRuby script, and the results
 can be fully parsed to extract the portion you need.  BioRuby supports
 major biological database formats and provides many ways for accessing
-them through flatfile indexing, SQL, web services etc.  Various web
+them through flatfile indexing, web services etc.  Various web
 services can be easily utilized by BioRuby.
 
 
@@ -140,23 +140,6 @@ Creating faster flatfile index using Berkley DB:
   (which took over {bdb}[https://github.com/ruby-bdb/bdb])
   (No RubyGems available)
   * {Oracle Berkeley DB}[http://www.oracle.com/technetwork/database/berkeleydb/index.html] and C compiler will be required.
-  
-Accessing BioSQL database created by other Open Bio* libraries:
-
-* {gem install activerecord}[http://rubygems.org/gems/activerecord]
-  and at least one driver (or adapter):
-  * {gem install mysql}[http://rubygems.org/gems/mysql] or
-    {gem install mysql2}[http://rubygems.org/gems/mysql2]
-  * {gem install pg}[http://rubygems.org/gems/pg]
-  * {gem install sqlite-ruby}[http://rubygems.org/gems/sqlite-ruby] or
-    {gem install sqlite3}[http://rubygems.org/gems/sqlite3]
-  * {gem install activerecord-oracle_enhanced-adapter}[http://rubygems.org/gems/activerecord-oracle_enhanced-adapter]
-
-For parsing PhyloXML format files:
-
-* {gem install libxml-ruby}[http://rubygems.org/gems/libxml-ruby]
-  * {GNOME Libxml2 XML toolkit}[http://xmlsoft.org/] and C compiler will be
-    required.
 
 
 == INSTALL
@@ -287,6 +270,16 @@ for list of plugins and related software utilizing BioRuby.
 
 * http://biogems.info/
 
+Plugins (Biogems) listed below had been included in BioRuby in former days,
+and were split to separate packages to reduce complexity and external
+dependencies.
+
+* {bioruby-phyloxml}[http://rubygems.org/gems/bioruby-phyloxml]
+  * NOTE: Please uninstall bio-phyloxml, that have been created as a
+    preliminary trial of splitting a module in 2012 and have not been
+    maintained after that.
+* {bio-biosql}[http://rubygems.org/gems/bio-biosql]
+
 To develop your own plugin, see "Plugins" pages of BioRuby Wiki.
 
 * http://bioruby.open-bio.org/wiki/Plugins

data/RELEASE_NOTES.rdoc

@@ -1,325 +1,68 @@
-= BioRuby 1.5.2 RELEASE NOTES
-
-Few bug fixes have been made to the BioRuby 1.5.2 after the version 1.5.1 is
-released.
-
-* Fixed a parser bug in Bio::Fasta::Report, FASTA output (-m 10) parser,
-  contributed by William Van Etten and Mark Wilkinson via GitHub.
-* HTTPS is used to access GenomeNet BLAST web service, contributed
-  by @ramadis via GitHub.
-* Bio::Shell#seqstat: Fix format specifier syntax error.
-* Bio::AAindex documentation fix, suggested by @kojix2 via GitHub.
-* Suppress warning messages in Ruby 2.4 and later.
-
-
 = BioRuby 1.5.1 RELEASE NOTES
 
-Few changes have been made to the BioRuby 1.5.1 after the version 1.5.0 is
-released.
-
-== NEWS
-
-=== HTTPS is used to access NCBI web services
-
-As you may know, NCBI announced that all HTTP resources will be switched
-to HTTPS on September 30, 2016. To follow the transition, all URLs for
-accessing NCBI E-utilities in BioRuby are changed to use HTTPS.
-
-In BioRuby, the following classes/modules are affected.
-
-* Bio::NCBI::REST and descending classes
-* Bio::PubMed
-
-In some rare cases (especially when building Ruby and/or OpenSSL by yourself
-from source code), Ruby does not include SSL/TLS support, or Ruby fails to
-detect SSL root certificates. In such cases, you may need to reinstall or
-upgrade Ruby, OpenSSL (or alternatives), and/or SSL root certificates with
-appropriate configuration options. Alternatively, installing binary packages
-is generally a good idea.
-
-
-= BioRuby 1.5.0 RELEASE NOTES
-
-A lot of changes have been made to the BioRuby 1.5.0 after the version 1.4.3
+A lot of changes have been made to the BioRuby 1.5.1 after the version 1.5.0
 is released. This document describes important and/or incompatible changes
-since the BioRuby 1.4.3 release.
+since the BioRuby 1.5.0 release.
 
 For known problems, see KNOWN_ISSUES.rdoc.
 
-== NEWS
-
-=== Full support of Ruby 2.0.0, 2.1, and 2.2
-
-Ruby 2.0.0, 2.1, and 2.2 are now recommended Ruby versions for running
-BioRuby codes.
-
-=== Support of Ruby 1.8 will be stopped
-
-This release is the final BioRuby version that can be run on Ruby 1.8.
-
-=== License is updated to the new Ruby's License
-
-BioRuby is distributed under the same license as Ruby's. In October 2011,
-Ruby's License was changed from a dual license with GPLv2 to a dual license
-with 2-clause BSDL. Since BioRuby 1.5.0, we have updated to the
-new version of Ruby's License. For details about the license, see COPYING
-or COPYING.ja and BSDL. In addition, please do not forget to see LEGAL for
-exception files that are subjected to different licenses.
-
-=== Semantic Versioning will be introduced
-
-We will adopt the Semantic Versioning since the next release version, which
-will be BioRuby 1.5.1. This means that BioRuby 1.5.0 is NOT subject to the
-Semantic Versioning.
+== New features
 
+=== (tbd)
 
-== New features and improvements
+(paragraph)
 
-=== New method Bio::FastaFormat#first_name
-
-Bio::FastaFormat#first_name method is added to get the first word in the
-definition line. This method was proposed by Ben J. Woodcroft.
-
-=== Accuracy of Bio::SiRNA
-
-Accuracy of siRNA designing algorithms in Bio::SiRNA is improved,
-contributed by meso_cacase.
-
-=== Speed up of Bio::ClustalW::Report
-
-Running speed of Bio::ClustalW::Report is optimized, contributed by
-Andrew Grimm.
-
-=== Many warning messages are squashed
-
-Most warning messages when running ruby with "-w" option, e.g. "assigned
-but unused variable", "instance variable @xxx not initialized", are
-suppressed. Fixes are contributed by Kenichi Kamiya, Andrew Grimm, and
-BioRuby core members.
-
-=== Refactoring of codes
-
-Many existing codes are reviewed and refactored. Patches are contributed
-by Iain Barnett, Kenichi Kamiya, and BioRuby core members.
+=== Other new features
 
+* (tbd)
+* (tbd)
 
 == Bug fixes
 
-=== Bugs due to remote server changes
-
-==== Bio::PubMed
-
-Bio::PubMed#search, query, and pmfetch are re-implemented by using NCBI
-E-Utilities. They were broken because unofficial API was used.
-Paul Leader reported the bug and gave discussion.
-
-==== Bio::Hinv
-
-Bio::Hinv did not work because of the API server URL is changed.
-
-==== Bio::TogoWS::REST
-
-* Bio::TogoWS::REST#search with offset and limit did not work due to
-  TogoWS server spec change about URI escape.
-* Bio::TogoWS::REST#convert did not work because of the spec change of
-  TogoWS REST API.
-
-=== Bio::Fetch
-
-Bio::Fetch with default parameters did not work because BioRuby's default
-BioFetch server had been down. We have decided not to restore the service.
-For smooth migration of codes using BioRuby's BioFetch server, we provide
-"bio-old-biofetch-emulator" gem. See below "Imcompatible changes" for details.
-
-The bug was reported and discussed by Jose Irizarry, Robert A. Heiler, and
-others.
-
-=== BioSQL
-
-* Only do gsub on bio_ref.reference.authors if it exists.
-* Missing require when generating genbank output for BioSQL sequence.
-
-Contributed by Brynjar Smari Bjarnason.
+=== (tbd)
 
-=== Bugs found in data format parsers
+(paragraph)
 
-* Bio::PDB#seqres SEQRES serNum digits were extended in PDB v3.2 (2008).
-  Thanks to a researcher who sent the patch.
-* Bio::Blast::Default::Report parse error when subject sequence contains
-  spaces. Edward Rice reported the bug.
-* Bio::UniProtKB#gene_name raised NoMethodError when gene_names method
-  returns nil. It should return nil. Candidate fix sent by Jose Irizarry.
-* Bio::PhyloXML::Parser.open_uri did not return block return value when
-  giving a block.
-
-=== Other bugs
-
-* lib/bio/shell/plugin/seq.rb: String#step and #skip (extended by bioruby
-  shell) did not yield the last part of the string due to a change from
-  Ruby 1.8 to 1.9.
-* Documentation and typo fixes. Contributed by many persons, including
-  Iain Barnett and ctSkennerton.
-
-
-== Renamed features
-
-=== Bio::SPTR, Bio::UniProt, Bio::SwissProt, and Bio::TrEMBL => Bio::UniProtKB
-
-The classes for parsing UniProtKB (former SwissProt and TrEMBL) data,
-Bio::SPTR, Bio::UniProt, Bio::SwissProt, and Bio::TrEMBL, are unified into
-the new class name Bio::UniProtKB, and old names are deprecated.
-
-For keeping backward compatibility, old class names Bio::SPTR, Bio::UniProt,
-Bio::SwissProt, and Bio::TrEMBL are still available, but warning message
-will be shown when using the old class names. These old class names will be
-deleted in the future.
-
-The file contatining Bio::UniProtKB class definition is also changed to
-lib/bio/db/embl/uniprotkb.rb. For keeping backward compatibility, old
-files sptr.rb, uniprot.rb, tremble.rb are still kept, but they will be
-removed in the future.
-
-
-== Deprecated features
-
-=== Bio::RefSeq, Bio::DDBJ
-
-Bio::RefSeq and Bio::DDBJ are deprecated because they are only an alias of
-Bio::GenBank. Warning message will be shown when loading the classes and
-initializing new instances. Please use Bio::GenBank instead.
-
-lib/bio/db/genbank/ddbj.rb and lib/bio/db/genbank/refseq.rb which contain
-Bio::DDBJ and Bio::RefSeq, respectively, are also deprecated. For keeping
-backward compatibility, old file are still kept, but they will be removed
-in the future.
-
-
-== Removed features
-
-=== Bio::SOAPWSDL
-
-Bio::SOAPWSDL (lib/bio/io/soapwsdl.rb) is removed because SOAP4R
-(SOAP/WSDL library in Ruby) is no longer bundled with Ruby since Ruby 1.9.
-For Ruby 1.9 or later, some gems of SOAP4R are available, but we think they
-are not well-maintained. Moreover, many SOAP servers have been retired
-(see below). So, we give up maintaining Bio::SOAPWSDL and all SOAP client
-classes in BioRuby.
-
-=== Bio::EBI::SOAP
-
-Bio::EBI::SOAP (lib/bio/io/ebisoap.rb) is removed because Bio::SOAPWSDL
-is removed.
-
-=== Bio::KEGG::API
-
-Bio::KEGG::API is removed because KEGG API SOAP service was discontinued in
-December 31, 2012. See http://www.kegg.jp/kegg/rest/ for the announcement
-of service discontinuation.
-
-=== Bio::DBGET
-
-Bio::DBGET is removed because it only supports old original DBGET protocols
-that was discontinued in 2004. Note that the DBGET is still available via
-the web. See http://www.genome.jp/en/gn_dbget.html for details.
-
-=== Bio::Ensembl
-
-Bio::Ensembl is removed because it does not work after the renewal of Ensembl
-web site in 2008. Instead, bio-ensembl gem which supports recent ensembl API
-is available.
-
-=== Bio::DDBJ::XML, Bio::DDBJ::REST
-
-Bio::DDBJ::XML and Bio::DDBJ::REST are removed because DDBJ Web API (WABI)
-web services were suspended in 2012 and then they were completely renewed
-with incompatible APIs in 2013.
-
-=== Bio::HGC::HiGet
-
-Bio::HGC::HiGet (lib/bio/io/higet.rb) is removed because the HiGet web server
-http://higet.hgc.jp/ have been down since 2011, and we think that the server
-will not be restored again.
-
-=== Bio::NCBI::SOAP
-
-Bio::NCBI::SOAP is removed because it always raises error during the parsing
-of WSDL files provided by NCBI. In addition, NCBI announced that the SOAP
-web service for the E-utilities will be terminated on July 1, 2015. Instead,
-Bio::NCBI::REST, REST client for the NCBI E-utility web service, is available.
-
-=== Bio::KEGG::Taxonomy
-
-Bio::KEGG::Taxonomy is removed because it does not work correctly. It raises
-error, it falls into infinite loop, or it returns apparently broken data.
-Moreover, KEGG closed public FTP site and the file "taxonomy" could only be
-obtained by paid subscribers.
-
-=== Bio.method_missing
-
-Bio.method_missing, that aims to provide shortcuts of Bio::Shell methods
-with shorter name without typing "Shell", is removed because most of the
-methods raises error mainly due to bypassing of initialization procedure.
-In addition, we now think that the use of method_missing should generally
-be avoid unless it is really necessary.
-
-=== extconf.rb
-
-extconf.rb, an alternative way to install BioRuby to the system, is removed
-because of avoiding potential confusions. Nowadays, extconf.rb is usually
-only used for building native extensions, but no native extensions are
-included in this release. Use gem or setup.rb to install BioRuby.
+=== Other bug fixes
 
+* Fixed: (...)
+* Fixed: (...)
 
 == Incompatible changes
 
-Also see the above "Renamed features", "Deprecated features", and
-"Removed features" sections.
-
-=== Bio::Fetch
+=== Bio::Taxonomy is removed and merged to Bio::PhyloXML::Taxonomy
 
-The BioRuby default BioFetch server http://bioruby.org/cgi-bin/biofetch.rb,
-that was the default server for Bio::Fetch before BioRuby 1.4, is deprecated.
-Due to the service stop, default server URL in Bio::Fetch is removed, and
-we decide not to give any server URL by default for Bio::Fetch.
+Bio::Taxonomy in lib/bio/db/phyloxml/phyloxml_elements.rb was written for
+PhyloXML in 2009. It was intended to become general taxonomy data class in
+BioRuby in these days. However, no efforts have been made to improve the
+Bio::Taxonomy class, and it still remains to be a PhyloXML specific class.
+Because Bio::PhyloXML is split as a different Gem (Biogem) package, we now
+decide to remove Bio::Taxonomy and merge it to Bio::PhyloXML::Taxonomy.
 
-As an alternative, new class Bio::Fetch::EBI which uses the EBI Dbfetch
-server is added. When changing codes form Bio::Fetch to Bio::Fetch::EBI,
-be careful of the differences of database names, default and available
-data formats between the former BioRuby BioFetch server and the EBI Dbfetch
-server.
+In the codes using Bio::Taxonomy directly, changing Bio::Taxonomy to
+Bio::PhyloXML::Taxonomy or adding the following monkey patch may be needed.
 
-Methods directly affected are:
+   module Bio
+      unless defined? Taxonomy
+         Taxonomy = Bio::PhyloXML::Taxonomy
+      end
+   end
 
-* Bio::Fetch.new (Bio::Fetch#initialize) does not have default server URL,
-  and URL of a server must always be explicitly given as the first argument.
-* Bio::Fetch.query is removed.
+In the future, Bio::Taxonomy might be added as general taxonomy data class.
+The new Bio::Taxonomy might be incompatible with the current Bio::Taxonmy.
 
-For the purpose running old codes, it is recommended to install
-bio-old-biofetch-emulator gem. The bio-old-biofetch-emulator gem emulates
-old BioRuby's default BioFetch server by using other existing web services.
-See https://rubygems.org/gems/bio-old-biofetch-emulator for details.
-We think many codes can run with no changes by simply installing the gem and
-adding "-r bio-old-biofetch-emulator" into the command-line when executing
-ruby.
+=== (tbd)
 
+(paragraph)
 
 == Known issues
 
 The following issues are added or updated. See KNOWN_ISSUES.rdoc for other
 already known issues.
 
-=== Bio::PDB
-
-Bio::PDB should be updated to follow PDB format version 3.3.
-
-=== Bio::Blast::Report
-
-NCBI announces that that they are makeing a new version of BLAST XML data
-format. BioRuby should support it.
+* (tbd)
 
-=== Bio::Blast::Default::Report
+== Other important news
 
-Bio::Blast::Default::Report currently supports legacy BLAST only.
-It may be better to support BLAST+ text output format, although NCBI
-do not recommend to do so because the format is unstable.
+(tbd)
 

data/Rakefile

@@ -68,19 +68,23 @@ GEM_SPEC_FILE = "bioruby.gemspec"
 GEM_SPEC_TEMPLATE_FILE = "bioruby.gemspec.erb"
 
 # gets gem spec string
-gem_spec_string = File.open(GEM_SPEC_TEMPLATE_FILE, "rb") do |f|
-                    ERB.new(f.read).result
-                  end
+current_gem_spec_string = File.read(GEM_SPEC_FILE) rescue nil
+
+next_gem_spec_string = File.open(GEM_SPEC_TEMPLATE_FILE, "rb") do |f|
+  ERB.new(f.read).result
+end
 
 # gets gem spec object
-spec = eval(gem_spec_string)
+current_spec = eval(current_gem_spec_string || '')
+next_spec = eval(next_gem_spec_string)
+spec = (current_spec || next_spec)
 
 # adds notice of automatically generated file
-gem_spec_string = "# This file is automatically generated from #{GEM_SPEC_TEMPLATE_FILE} and\n# should NOT be edited by hand.\n# \n" + gem_spec_string
+next_gem_spec_string = "# This file is automatically generated from #{GEM_SPEC_TEMPLATE_FILE} and\n# should NOT be edited by hand.\n# \n" + next_gem_spec_string
 
 # compares current gemspec file and newly generated gemspec string
-current_string = File.read(GEM_SPEC_FILE) rescue nil
-if current_string and current_string != gem_spec_string then
+if current_gem_spec_string &&
+   current_gem_spec_string != next_gem_spec_string then
   #Rake::Task[GEM_SPEC_FILE].invoke
   flag_update_gemspec = true
 else
@@ -101,7 +105,7 @@ file GEM_SPEC_FILE => [ GEM_SPEC_TEMPLATE_FILE, 'Rakefile',
                         'lib/bio/version.rb' ] do |t|
   puts "creates #{GEM_SPEC_FILE}"
   File.open(t.name, 'wb') do |w|
-    w.print gem_spec_string
+    w.print next_gem_spec_string
   end
 end
 
@@ -188,7 +192,7 @@ task :rechangelog do
   # The tag name in the command line should be changed
   # after releasing new version, updating ChangeLog,
   # and doing "git mv ChangeLog doc/ChangeLog-X.X.X".
-  sh "git log --stat --summary 1.4.3..HEAD > ChangeLog"
+  sh "git log --stat --summary 1.5.0..HEAD > ChangeLog"
 end
 
 # define mktmpdir